4.3.7.
. 1..."...

"mum,
‘ 3.

.
. .. . nﬁxxnulhnnﬂ»)
.. S 33......

.- n: 116»
.9531“;

. 23......
2» :35:
i
.3‘ .-
l}
h

. ... ..........i...vW. wry

 

 

 

 

i URRARY
I Michigan State
I U ‘wwez‘m

it!V\L:\4ILJI

This is to certify that the
thesis entitled

THE ECONOMICS OF ANAEROBIC DIGESTION UNDER
DIFFERENT ELECTRICITY PURCHASE AGREEMENTS

presented by
David F. Binkley

has been accepted towards fulﬁllment
of the requirements for the

MASTER OF degree in Agricultural Economics

 

 

r. rat/M

Major Professor’s STgnature

7 Macro

Date

 

MSU is an Afﬁrmative Action/Equal Opportunity Employer

-.-'--

.n-.---»-n-n----—--o---------c--n-_.-n--.-L-n-t-o-n-n-o-------------.-._.--—.-. V

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProi/Acc&Pres/ClRC/DateDue,indd

THE ECONOMICS OF ANAEROBIC DIGESTION UNDER DIFFERENT
ELECTRICITY PURCHASE AGREEMENTS
By

David F. Binkley

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE
Agricultural Economics

2010

ABSTRACT

THE ECONOMICS OF ANAEROBIC DIGESTION UNDER DIFFERENT UTILITY
PURCHASE AGREEMENTS

By

David Binkley

Anaerobic Digestion is receiving a great deal of attention as a viable alternative in
supporting residuals management for livestock operations. In contrast to conventional
liquid and slurry management systems, anaerobic digesters provide multiple
environmental beneﬁts such as odor control, improved air and water quality, improved
nutrient management ﬂexibility, and the opportunity to capture biogas for heat and
electricity production. The digester system is a process which includes: collection and
handling, anaerobic digestion, by-product recovery and efﬂuent use, and biogas recovery
and use. Although energy production alone has not been cited as the primary motivation
for the installation of anaerobic digesters, state policies on distributed power pricing can
greatly affect the economic viability of digesters. The model developed in this study
incorporates a variety of system parameters to examine the economics of a digester
system under three different electricity purchase agreements. The results suggest that
making Speciﬁc changes to Michigan’s energy policy will improve digester return on

investment over a range of herd sizes.

ACKNOWLEDGEMENTS
I would like to thank my committee, namely Dr. Stephen Harsh, Dr. Steve Safferman and
Dr. Chris Wolf for their guidance and support throughout this research. Special thanks
go to Dr. Harsh, my major professor. His encouragement and advice during our weekly

meetings have been invaluable over the course of this project.

I would also like to thank all of the graduate students and faculty with whom I
have worked with at Michigan State, and who have been so helpful and supportive.
Speciﬁcally, Dana Kirk has been a tremendous help in putting together this research. He
has always made himself available to answer any questions and provided valuable
feedback. He also went out of his way to involve me with the Green Meadow Farms
digester project which was key to improving my understanding of digester systems. Eric
Wittenberg deserves special thanks as well for helping me obtain digester price
information for Chapter 3. I also thank Nicole Olynk for her willingness to help and for

being a good friend.

iii

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... vii
LIST OF FIGURES ......................................................................................................... ix
Chapter 1: Introduction ................................................................................................... 1
Chapter 2: The Basics of Anaerobic Digestion ............................................................. 10
2.1 The Microbiological Process ................................................................................ 10
2.2 The Digester System ............................................................................................. 12
2.2.1 Collection and Handling .................................................................................. 13
2.2.2 Anaerobic Digestion ........................................................................................ 13

2. 2. 2A Complete Mix ............................................................................................. 14

2.2.3 By-Product Recovery and Efﬂuent Use ........................................................... 16
2.2.4 Biogas Recovery .............................................................................................. 17
2.2.7 Biogas Use ....................................................................................................... 17

2. 2. 7A Electrical Generation ................................................................................ 18

2. 2. 73 Key Considerations ................................................................................... 19

2. 2. 7C Other Energy Uses .................................................................................... 21

2.8 Annual Operation and Maintenance Costs ........................................................ 22
2.9 Electricity Contracts ............................................................................................. 23
2.9.1 Surplus Sale ..................................................................................................... 23
2.9.2 Buy-All Sell-All ............................................................................................... 26
2.9.3 Net Metering .................................................................................................... 26
2.10 Renewable Energy Credits ................................................................................. 29
2.11 Carbon Credits .................................................................................................... 29
2.12 Co-digestion ......................................................................................................... 31
2.13 Non-Energy Benefits ........................................................................................... 34
Chapter 3: The Model .................................................................................................... 38
3.1 Default Value Investment ..................................................................................... 43
3.1.1 Capital Investment ............................................................................................. 43
3.1.2 Design Study and Engineering ..................................................................... 45

3.1.3 Excavation .................................................................................................... 45

3.1.4 Tanks ............................................................................................................ 45

3.1.4A Desired Tank Volume (Digester) ........................................................... 45

3.1.43 Tank Quantity (Digester) ....................................................................... 46

3.1.4C Tank Unit Cost ...................................................................................... 46

3. 1. 4D T anks-(Post Storage) ............................................................................. 47

3.1.4E Tanks-(Equalization) ............................................................................. 48

3.1.4F Roofs-(Digester) .................................................................................... 48

3.1.4G Insulation .............................................................................................. 48

3.1.5 Boiler ............................................................................................................ 49

iv

3.1.6 Heating ......................................................................................................... 49
3.1.7 Plumbing, Valves, Mixing and other Miscellaneous Components .............. 50
3.1.8 Water-to-Manure Heat Exchangers ............................................................. 50

3. l .9 Instrumentation ............................................................................................ 50
3.1.10 Contingency ............................................................................................... 50
3.1.11 Engine-Generator ....................................................................................... 51
3.1.12 Building ...................................................................................................... 51
3.1.13 Switchgear and Additional Engine Components ....................................... 51
3.1.14 Interconnection .......................................................................................... 52
3.1.15 Salvage Value ............................................................................................ 52

3.2 Operation and Maintenance Costs .................................................................... 53
3.2.1 Digester ............................................................................................................ 53
3.2.2 Engine-Generator ............................................................................................. 55
3.3 Depreciation ........................................................................................................... 57
3.4 Property Taxes ...................................................................................................... 57
3.5 Biogas Production ................................................................................................. 59
3.5.1 Inﬂuent Flow .................................................................................................... 59
3.5. IA Manure ...................................................................................................... 59

3.6 Utilization Analysis ............................................................................................... 63
3.6.1 Digester Heating .............................................................................................. 64
3.6.2 Co-digestion ..................................................................................................... 67

3. 6. 2A Feedstock Cost ........................................................................................... 67

3. 6. 23 Feedstock Revenue .................................................................................... 68

3. 6. 2C Amount of Feedstock Entering the Digester per Day ............................... 69

3. 6. 2D Feedstock Characteristics and Biogas Yield ............................................ 69
3.6.3 Energy Uses ..................................................................................................... 70

3. 6. 3A Propane Offsets ......................................................................................... 71

3. 6. 38 Electricity Generation ............................................................................... 72

3.7 Electricity Purchase Agreements ......................................................................... 73
3.7.1 Surplus Sale ..................................................................................................... 74
3.7.2 Buy-All Sell-All ............................................................................................... 76
3.7.3 Net Metering .................................................................................................... 77
3.8 Utility Meter Bills .................................................................................................. 78
3.9 Standby Charges ................................................................................................... 79
3.9.1 Power Supply Standby Charges ....................................................................... 80
3.9.2 Delivery Standby Charges ............................................................................... 80
3.10 Carbon Credits .................................................................................................... 82
3.10.1 Methane Combustion Method ........................................................................ 83
3.10.2 Ex-Ante Method ............................................................................................. 84
3.11 Financing Options ............................................................................................... 86
3.12 Capital Budgeting Analysis ................................................................................ 87
3.13 Veriﬁcation of Model .......................................................................................... 90
3.13.1 Test Farm 1 .................................................................................................... 90
3.13.2 Test Farm2 .................................................................................................... 92
Chapter 4: Results and Analysis .................................................................................... 94

4.1 Section One - 1,000 cow Example ........................................................................ 95

4.1.1 Part I Price Sensitivity Analyses ...................................................................... 95
4.1.1A Retail Electricity ........................................................................................ 98

4.1.18 Value of Electricity Production ............................................................... 100

4.1.1C Carbon Credits ........................................................................................ 103

4.1.1D Renewable Energy Credits (RECs) ......................................................... 105

4.1.2 Part II Co-digestion of Ethanol Syrup ........................................................... 106
4.1.2A Scenario ................................................................................................... 107

4.3 Section Two - Engineering ................................................................................. 111
4.3.1 Total Solids Concentration ............................................................................ 111
4.3.2 Volatile Solids Loss ....................................................................................... 116
4.3.3 Online Time ................................................................................................... 119

4.4 Section Three - Policy ......................................................................................... 121
4.4.1 Part I - Current Policy .................................................................................... 121

4. 4. 1A Value of Odor Reduction ......................................................................... 125

4.4.2 Part II - Recommendations ............................................................................ 127

4. 4. 2A Standby Charges ...................................................................................... 127

4. 4. 23 “Net Metering Components ” .................................................................. 129
Chapter 5: Conclusions ................................................................................................ 135
5.1 Areas for Future Research ................................................................................. 139
REFERENCES .............................................................................................................. 140

vi

LIST OF TABLES

Table 1. Tank Quantity ..................................................................................................... 46
Table 2. Total Tank Volume ............................................................................................. 47
Table 3. Boiler Size Ranges .............................................................................................. 49
Table 4. Engine-Generator Costs ...................................................................................... 51
Table 5. Direct-Entry Operation and Maintenance Costs ................................................. 54
Table 6. 0&M Costs with Late Project Period Repair and Replacement ......................... 55
Table 7. Cost Weight Factor for 0&M Costs Based Upon Percentage of Total Capital
Costs .......................................................................................................................... 56
Table 8. Operation and Maintenance Costs as a Percentage of Total Capital Costs ........ 57

Table 9. Manure Characteristics from the American Society for Biological Engineers .. 61

Table 10. Carbon Credit Summary ................................................................................... 85
Table 11. Assumptions for 1,000 Cow Example .............................................................. 88
Table 12. Sheland Farms Inc. Case Study Inputs ............................................................. 91
Table 13. Sheland Farms Inc. Case Study Outputs ........................................................... 91
Table 14. Model Assumptions for Test Farm 1 ................................................................ 91
Table 15. Penn England Farm (Mixed Loop) Case Study Inputs ..................................... 92
Table 16. Penn England Farm (Mixed Loop) Case Study Outputs .................................. 93
Table 17. Model Assumptions for Test Farm 2 ................................................................ 93
Table 18. Section One Assumptions - 1,000 Cow Dairy Sensitivity Analysis ................. 96
Table 19. Section I Electricity Purchase Agreement Assumptions .................................. 97
Table 20. Co-digestion Scenario Assumptions Summary .............................................. 108
Table 21. Co-digestion of Ethanol Syrup and Electricity Production (7 ton loads) ....... 109
Table 22. Section Two Assumptions - Engineering ....................................................... 111

vii

Table 23. Section Three Assumptions- Policy ................................................................ 121
Table 24. Current Policy Summary ................................................................................ 122

Table 25. Scenario 1 Policy Summary ............................................................................ 130

viii

LIST OF FIGURES

Figure 1. Simpliﬁed Process of Biogas Production .......................................................... 12
Figure 2. Digester System ................................................................................................. 15
Figure 3. Model Layout .................................................................................................... 41
Figure 4. Model Cost Estimation Compared with AgSTAR Cost Curve ......................... 44
Figure 5. Sensitivity of NPV to Retail Electricity Prices ................................................. 98
Figure 6. The Sensitivity of NPV to the Value of Electricity Produced ......................... 101
Figure 7. Sensitivity of NPV to Carbon Credit Prices .................................................... 104
Figure 8. Sensitivity of NPV to Renewable Energy Credit Prices ................................. 106
Figure 9. The Effect of Each Tuck Load (7 tons) of Ethanol Syrup on NPV ................. 110

Figure 10. Total Solids Concentration vs. Average Yearly Electricity Production ........ 113

Figure 11. The Effect of Total Solids Concentrations on NPV ...................................... 115
Figure 12. Volatile Solids Loss vs. Average Yearly Electricity Production .................. 117
Figure 13. NPV Compared to Herd Sizes (500 to 4,000 cows) ...................................... 125
Figure 14. The Cost per Cow/Day of a Digester Across a Range of Herd Sizes ............ 126
Figure 15. Estimated Engine-Generator Nameplate Capacities and Herd Size .............. 129
Figure 16. NPV vs Herd Size Using Net Metering Components ................................... 131
Figure 17. NPV vs Herd Size with Standby Charge Threshold of 400 kW .................... 132

Figure I8. NPV vs. Herd Size with Increased Standby Charge Threshold of 800 kW .. 133

Figure 19. “Modiﬁed” vs “True” Net Metering Across a Range of Herd Sizes ............. 134

ix

MW
Agricultural production in the United States annually discharges large amounts of
nitrogen and phosphorus, some of which eventually end up in ground and surface waters.
According to the Environmental Protection Agency (EPA), these nutrients from crop and
animal production are found in 50% of impaired lakes and 20% of impaired rivers in the
US. (Kaplan et al., 2004). AS livestock operations continue to increase in Size, one of the
most signiﬁcant challenges that producers face iS managing manure and process water in
a way that controls odors and protects environmental quality (U .3. EPA, 2002). In
Michigan, state law limits the land application of manure based upon phosphorous levels
and farms must ﬁnd ways of handling the excess manure in order to avoid violations. In
addition to the restrictions on the land application of manure, livestock producers are
facing an increased risk of odor complaints as people move into rural areas (Safferman
and Faivor, 2008). New residents are generally less tolerant of odors and the number of
reported complaints has increased in recent years. In some cases, these actions may seek

millions of dollars in damages and injunctions to close the operation (Miner, 1997).

AS a result of these pressures, anaerobic digestion is receiving a great deal of attention as
a viable alternative in supporting residuals management for livestock operations (MDA,
2009). The process itself involves the controlled breakdown of organic wastes by bacteria
in the absence of oxygen (Lazarus and Rudstrom, 2007). In contrast to conventional
liquid and Slurry management systems, anaerobic digesters provide multiple
environmental beneﬁts such as odor control, improved air and water quality, improved

nutrient management ﬂexibility, and the opportunity to capture biogas for heat and

electricity production (US. EPA, 2002). When properly running, a digester has the
potential to turn a waste liability into a proﬁt center that generates annual revenues and
diversiﬁes farm income (Lusk, 1998). “Farmers have found that the returns provided
from electricity and co-product sales from the digester, however limited, are preferred to
the sunk-cost of conventional disposal that provides zero return on investment. In
addition, without the environmental beneﬁts provided by digester technology, some
might be forced out of livestock production and a digester is sometimes the only

technology that allows growth in the livestock production business” (Lusk,l998, p. 1-2).

Anaerobic digestion, however, is not a new technology. During and immediately after
the energy crisis caused by the oil embargo in 1973, many anaerobic systems were built
to produce energy. At least 71 were installed on commercial livestock or poultry
operations, but with lower energy prices many of these systems were abandoned. The
limited long-term success in the United States can be attributed to poor system design,
improper system installation, and unsatisfactory system management (Lusk, 1998). While
interest in digesters was initially driven by energy concerns during the 1970’s oil crisis,
they are fairly capital-intensive when viewed primarily as an energy source (Lazarus,
2008). The cost for digesters varies depending on the design, an estimated $500/cow to
$750/cow on average (N RCSa, Undated). AS a result of this high capital investment,
when considering the economics of anaerobic digestion, it is necessary to perform a
thorough examination of potential revenue streams, system design and the extent to

which it satisﬁes the objectives of the individual farm operator.

Although energy production alone has not been cited as the primary motivation for the
installation of anaerobic digesters, state policies on distributed power pricing and
interconnection can greatly affect the economic viability of digesters (Lazarus, 2008).
Contractual agreements with utility companies tend to be of three types: “buy all-sell all,”
“surplus sale” and “net metering.” The type of utility contract utilized may have a
Signiﬁcant impact on digester economics and the speciﬁcs of each type of agreement vary
depending on the utility company and state energy policy. According to a survey of 64
producers across the US. and 10 in California, negotiating these contracts were cited as
the biggest challenge faced by the producer (Lazarus, 2008). In addition, some utilities
impose “demand” or “standby” charges to pay for the availability of electricity to the
farm when the digester system is not rtmning. In many cases, difﬁculties related to
negotiating with utility companies discouraged farmers from installing digesters that had

been planned (Lazarus, 2008).

At the same time, the area of energy policy represents a signiﬁcant opportunity to
improve digester proﬁtability. For example, Michigan has recently enacted a new Net
Metering Law as part of Public Act 295 of 2008 which also speciﬁed a Renewable
Portfolio Standard (RPS) requiring the state to produce 10% of its electricity from
renewable sources by 2015. Although this is a step in the right direction for encouraging
digester installations, even more favorable energy policies for anaerobic digesters could
improve their economic feasibility and increase their adoption among farmers.
Nationally, AgSTAR estimates that around 7,000 large dairy and swine operations could

operate proﬁtable biogas systems with a generating potential of 722 megawatts—0.1

percent of total US. electrical generating capacity or enough to supply almost 1 million
homes (Lazarus 2008; US. EPA, 2006). AS energy policy continues to develop, it is
expected that the sale of Carbon Credits and Renewable Energy Credits will also

increasingly contribute to digester revenues.

The focus of this research was to develop a model to examine the economics of anaerobic
digestion under different policy and system design scenarios using capital budgeting
methods. Among the policy options evaluated were different electricity purchase
agreements as well as the impact of renewable energy credits (RECS), carbon credits and
the value of electricity produced by the digester. Additional elements analyzed include
the impacts of propane offsets and the co-digestion of added feedstocks. Although non-
energy beneﬁts are Signiﬁcant in the decision to install a digester, they have not been

quantiﬁed in previous research and therefore are not included in this study.

The concept of a model to examine digester economics, however, has been explored in
several other studies. Dr. Brent Gloy from Cornell University developed a ﬁnancial
analysis model of anaerobic digestion systems on diary farms along with a corresponding
paper (Gloy and Enahoro, 2008). This model was illustrated with two sources of data; a
“base” case using parameters developed from a wide range of sources and direct
comparison with values calculated from the F armWare 3.1 simulator]. The analysis
includes many key digester parameters and examines the effects of ﬁnancial incentives

- offered under the New York State Energy Research and Development Authority’s

Customer-Sited Tier Anaerobic Digester Gas-to-Electricity Program. Based upon

 

' Developed by the US. Environmental Protection Agency (EPA) AgSTAR Program (US. EPA, Undated)

estimates of the costs of a 1,000 cow dairy operation, the study found such a system to be
only marginally proﬁtable. A sensitivity analysis was then run under a variety of
scenarios in which the proﬁtability of the system was improved. In particular, increased
biogas production and higher retail electricity prices resulted in a higher net present value

(N PV) for the digester investment.

Gloy’s model, however, does not account for potential seasonality in biogas production
or other production related parameters such as: heat loss, parasitic energy requirements
and variations in manure characteristics. It was also only programmed to consider proﬁts
under a “surplus sale” utility contract and ignores the cost of standby charges and other

fees typically imposed by utility companies.

In a study out of Pennsylvania State University (Leuer et al., 2008) another model was
developed which also compared the proﬁtability of several digester scenarios. Similar to
the study from Cornell, the scenarios examine the application of beneﬁts realized by
digester systems. In contrast to the study by Gloy, however, this model focused on the
effect of the Pennsylvania net metering law and examined the added beneﬁt of solids
separation and the resulting revenues from either compost sales or livestock bedding
offsets. Here, a stochastic capital budgeting model was used which incorporated Monte
Carlo Simulations to derive the probability of the NPV being either greater than or equal
to zero. The study also tested digester proﬁtability under three herd Sizes: 500, 1,000 and

2,000 cows. System income was valued in the form of electricity sales and offsets,

bedding savings, separated solids sales, carbon credits and renewable energy credits.

Non-energy beneﬁts of the system, however, were not included.

The work showed that larger dairy farms, in the range of 1,000 or 2,000 cows, have the
most potential to be proﬁtable with a digester system. In addition, a solids separator to
produce digestate for bedding as well as new policies and regulations can further increase
proﬁtability. Items such as the sale of carbon credits and Pennsylvania net metering
regulations also affected the project’s proﬁtability but neither, by itself, turned an
unproﬁtable scenario into a proﬁtable one (Leuer et al., 2008). This study is valuable in
its ability to illustrate the effects of state speciﬁc energy policy as well as a variety of
system beneﬁts and revenues. For example, it compared the revenues associated with the
sale of digestate solids for compost and their possible use as a source of livestock
bedding. On the other hand, it did not deal with the many variables affecting biogas

production and examines only one type of electricity purchase agreement.

Yet another model was developed by Lazarus (Lazarus, UM), which helps users make
rough initial calculations of the annual costs and returns associated with owning an on-
farm anaerobic digester. The main issues it intended to address were: herd Size, digester
installation cost, amount and value of electricity produced, value of co-products and
ﬁnancing (e. g., grants). It does not, however, address engineering, design issues or

expected biogas output.

A separate paper by Lazarus (Lazarus, 2003) is a case study of the Haubenschild
demonstration digester in Minnesota which has exhibited relatively high biogas
production levels compared to similar digesters (Lazarus and Rudstrom, 2003). Here,
Lazarus constructed a model in Microsoft Excel with two main objectives. The ﬁrst was
to document the economics of the case farm’s digester system utilizing actual data from
the Haubenschild digester in Minnesota. The other objective was to compare the
demonstration farm against future scenarios involving the installation of a digester.
Scenarios for future digester installations included reduced biogas production, lower
payments for electricity produced and decreased public funding when compared to the
Haubenschild digester. Due to the detailed data recorded by the farm operator, an
estimate of the various non-energy beneﬁts was also examined including avoided pit pre-
agitation, reduced herbicide costs and fertilizer beneﬁts. The results highlighted the
importance of non-energy beneﬁts on digester proﬁtability, particularly under scenarios
with decreased production levels and reduced grant funding. The model in this study,
however, was not developed as a decision support tool and did not predict monthly
biogas production, capital costs or analyze different electricity purchase agreements.

Instead, data recorded by the Haubenschild Dairy operator was input directly.

A study by Mehta (Metha, 2002) did not develop a model, but did attempt to examine
optimal energy use and sale to utility companies based upon various electricity pricing
scenarios. In particular, it explored the economics and feasibility of electricity generation
using digesters on small and mid-size dairy farms. Essentially, the paper concluded that

if the electricity sale price was greater than or equal to the purchase price, then larger

farms would be able to make a larger proﬁt margin on each kWh of electricity sold. They
would gain more of a competitive edge through the introduction of a digester than smaller
farms. Conversely, if the sale price is less than the purchase price, then farms would
utilize generated electricity to offset their own electric bills with smaller farms
experiencing a relative competitive edge since they can utilize more energy on-farm. Due
to a lack of data, however, only rough conclusions or recommendations could be drawn

from the analysis.

Although, several digester models have already been developed by various economists,
the model developed in this research is unique in the following aspects.

0 Provides ﬂexibility for either direct input of data or values can be calculated by

the model. AS a result, analyses can be performed with minimal system details

which increases its value as an outreach tool.

0 Models biogas production on a monthly basis while accounting for heat loss,
parasitic energy requirements, variations in manure characteristics and the
possibility of co-digestion (to be further discussed in Section 2.12). Through this
method, digester electricity production and performance is more accurately
represented. Estimations of digester output also help to decide which electricity

purchase agreement is most favorable for a given farm.

0 Allows for evaluating engineering designs and changes in system parameters.
This capability helps engineers to examine the effects of changes to on-farm

operations.

0 Estimates the capital costs of a digester (complete mix) based upon herd Size.
The ability to evaluate a range of herd Sizes increases its value as a research tool
since the effects of policy changes can be examined for all farm Sizes that are

potentially impacted.

o Compares three different electricity purchase agreement policies based upon
actual utility rates and rules. Proposed changes in Michigan Speciﬁc energy

policy can then be evaluated.

The hypothesis of this study is that under current Michigan energy policy, an anaerobic
digester is unlikely to Show a positive return on investment based solely upon its energy
beneﬁts. On the other hand, it is expected that feasible scenarios exist in which the
system could actually become a viable source of proﬁt. The scenarios examined include
changes in the prices of electricity, renewable energy credits and carbon credits as well as
changes in digester performance parameters (e. g., operational online time, volatile solids
loss, and total solids concentration). In addition, speciﬁc aspects of Michigan’s energy
policy are analyzed in hypothetical situations over a range of herd sizes.
Recommendations are then made based upon the results in order to make current energy

policies more favorable for anaerobic digesters.

Chapter 2: The Basics of Anaerobic Digestion

The purpose of this chapter is to provide an overview of anaerobic digestion from the
microbiological level to more speciﬁc engineering concepts and policy issues. Chapter 3,
which describes the speciﬁcs of the model, builds upon the material covered in the

following sections.

2.1 The Microbiological Process

Anaerobic digestion is the breakdown of animal manure by bacteria in the absence of
oxygen resulting in the production of biogas (Bracmort et al., 2008). “It tends to occur
naturally wherever high concentrations of wet organic matter accumulate in the absence
of dissolved oxygen. Most often, this is in the bottom sediments of lakes and ponds,
swamps, peat bogs, intestines of animals, and in the anaerobic interiors of landﬁll sites”
(Lusk, 1998, p. 2-1). As a technology, it has been around for centuries with anecdotal
evidence indicating that biogas was used for heating bath water in Assyria during the 10th
century BC. and in Persia during the 16th century BC (Lusk, 1998). The nation’s ﬁrst
farm-based digester, however, wasn’t initiated until 1972 and was constructed as a
response to urban encroachment (Lusk, 1998). During the 1970’s, a number of digesters
were constructed, but many failed due to poor system design, improper system
installation and unsatisfactory system management. Since around 1984, however,
digester designs have improved (Lusk, 1998). The need for odor control and residuals
management combined with these improved designs has led to a recent resurgence of
interest in anaerobic digester technology. According to the EPA AgSTAR program, there

are currently 135 operational digesters in the US.

10

The biogas produced in anaerobic digestion is the result of microbial degradation of
carbon-containing compounds. These compounds are present in all organic matter (all
cells, living and dead) and are often measured as volatile solids (VS). VS is a standard
measurable parameter that allows different types of organic substances to be compared
for degradability. Higher levels of volatile solids indicate that the material has more
organic carbon and may be more degradable. Therefore, the amount of biogas that can be
produced is directly proportional to the amount of volatile solids in the feedstocks being
digested (Crook and Gould, 2009). Another frequently used term is total solids (TS). In
contrast to the VS measurement, it accounts for all solids including both the volatile and

non-volatile compounds and elements.

The anaerobic digestion process can occur at a wide temperature range, but generally
occurs at either the mesophilic (95°-105°F) or thermophilic temperature ranges (125°-
135°F) (Lusk, 1998). First, the volatile solids in manure are broken down to produce a
series of fatty acids. This step is called the acid-forming stage and is carried out by a

group of bacteria called acid formers.

In the second stage, a highly Specialized group of bacteria, called methane formers,
convert the acids to biogas (Fulhage et al.,l993). “These bacteria are slower growing than
acid-forrning bacteria and are extremely ph-sensitive (pH 6.8-7.4 optimum). The acid
formers will grow rapidly if an excess of organic material is fed to a digester, producing V
an excess of volatile acids. If this happens, the accumulated acids will lower the pH,

inhibiting the methane bacteria and stopping gas production. To help buffer the system

11

against increases in acids, high alkalinity must be maintained. Lime can be added to

digesters during start-up periods of slug loading to maintain pH control” (Fulhage et al.,
1993, p. 1). In addition, a variety of materials such as salts, heavy metals, ammonia and
antibiotics can become toxic to anaerobic bacteria and must also be carefully monitored

(Fulhage et al., 1993).

The resulting biogas is a combination of methane (60-70%), carbon dioxide (30-40%),
water vapor and trace amounts of other gases such as hydrogen sulﬁde (H28) (US EPA,
2002). H28 is very corrosive and can cause damage to engines, boilers and other digester

components. Only the methane component of biogas has energy value.

Figure 1. Simplified Process of Biogas Production

 

 

 

 

 

acid-forming methane-forming
bacteria bacteria
Protein Volatile Biogas
Carbohydrates Organic _ (60-70% CH4, 30-40%
Complex Organic ACldS * 7 C02, water vapor and
Residuals other gases)

 

 

 

 

 

 

 

 

 

2.2 The Digester System

The digester system is a process which includes collection and handling, anaerobic
digestion, by-product recovery and efﬂuent use, and biogas recovery and use. There is
Signiﬁcant variability in digesters from one farm to another and it is difﬁcult to make

generalizations and comparisons.

12

2.2.1 Collection and Handling

The starting point for a digester system is manure collection and handling, with the key
considerations in the system being the amount of water and inorganic solids mixing with
the manure. With dairy farms, the manure is generally scrape collected from freestall
barns two or three times a day. Following collection, the manure may undergo
pretreatment prior to introduction into the digester. The pretreatment involved varies
depending on the farm and the type of digester technology used and may consist of
screening, sand and/or grit removal, mixing and/or ﬂow equalization (Krich et al., 2005).
Some forms of pretreatment (e. g., sand removal), however, may not be beneﬁcial to

energy production as a portion of the volatile solids which produce biogas are removed

(Burke, 2001).

At this point in the process, water from the milking parlor and other sources may be
added in order to dilute the manure, but this varies from farm to farm and the solids
requirements of the digester. “Dilution also reduces concentrations of nitrogen and sulfur
which convert into ammonia and hydrogen sulﬁde during anaerobic digestion. Ammonia
is inhibitory to the process and hydrogen sulﬁde is an undesirable component in biogas

due to its highly corrosive characteristics” (Burke, 2001 ).

2.2.2 Anaerobic Digestion
After pretreatment, the manure along with any added water is pumped into the digester.
Manure characteristics and collection technique determine the type of anaerobic digester

technology used. U.S. livestock operations currently use four types of anaerobic digester

13

technologies: plug-ﬂow, complete-mix, ﬁxed ﬁlm and covered lagoons (Lusk, 1998).
“However, the parameters of any waste management system are site-Speciﬁc and may
vary signiﬁcantly from one livestock operation to the next. Effective implementation of
anaerobic digestion technology, therefore, demands that the digester be integrated with
the existing or planned manure management system. This requires an understanding of
the technology and of the impact that other site-Speciﬁc management practices can have
on both the energy potential of the feedstock and the efﬁcient operation of the digester
unit” (Wilkie, 2005, p.311-312). For these reasons, few generalizations or comparisons
can be made between digesters and each system must be evaluated on a case by case
basis. In this study, the model is based on the use of a complete mix digester since they
are the best suited for Michigan’s climate. Figure 2 is a representation of a typical

complete digester system.

2. 2. 2A Complete Mix

Complete-mix digesters can handle manures with TS concentrations of 2.5%-10%2, and
generally can handle substantial manure volumes. “The reactor is a large, vertical,
poured concrete or steel circular container and the manure is collected in a mixing pit by
either a gravity-ﬂow or pump system. If needed, the TS concentration can be diluted, and

the manure preheated before it is introduced to the digester tank. Within the digester, the

 

2 (NRCSb, 2005)

14

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bow :8: 883 iiiiii
3cm mewomm . l . . l

Soc 98qu

22m .:

wEEBp couﬁocowbﬁwcm .0
zooms Begum .m

x53 owecowmdmom .m

x53 coamowa d

83 “5832on 0552 .O
8:8 32:: .m

9:3 :88on .<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-----------------d—-----------

l
v

589$ .323»:— .N 0.5»:—

 

 

 

 

 

 

 

 

I LV

 

 

 

 

15

manure is mixed creating a homogeneous mixture that prevents the formation of a surface
crust and keeps solids in suspension. Mixing is important to ensure contact between the
bacteria and the waste and also to help release gas out of the liquid. Complete-mix
digesters can operate at either the mesophilic or thermophilic temperatures range with a
hydraulic retention time (HRT) of 10-20 days” (Lusk, 1998, p. 2-7). HRT refers to the
average length of time that a particle of manure or other feedstock remains in the

digester.

“A ﬁxed cover is placed over the complete-mix digester to maintain anaerobic conditions
and to contain the biogas that is produced. The biogas produced is then removed from
the digester, processed, and transported to the Site of end-use application. The most
common application for methane produced by the digestion process is electricity

generation using a modiﬁed internal combustion engine” (Lusk, 1998, p. 2-7).

2.2.3 By-Product Recovery and Efﬂuent Use

Once digestion is complete, it is possible to recover digested ﬁber from the efﬂuent of
some dairy manure digesters with the use of a solid/liquid separator. This material can
then be used for livestock bedding, sold as a soil amendment or marketed for other uses.
The efﬂuent (either separated or non-separated) can also be used as a high value
fertilizer. A further discussion on the beneﬁts of digester efﬂuent and separated solids is

included in Section 2.13 “Non-Energy Beneﬁts.”

16

2.2.4 Biogas Recovery

“The composition and digestibility of the manure is the primary determinant of maximum
methane yield (Wilkie, 2005, p.306 ).” “Biogas formed in the anaerobic digester bubbles
to the surface, is collected (typically with plastic piping), and then directed to gas
handling subsystems (Krich et al., 2005). “It is then pumped or compressed to the
operating pressure required by speciﬁc applications and then metered to the gas use

equipment” (Krich eta1.,2005, p. 30 ).

2.2.7 Biogas Use

Recovered biogas can be used directly as fuel for heating, combusted in an engine to
generate electricity, upgraded to natural gas or ﬂared (Krich et al., 2005). Since biogas
is only roughly 60-70% methane it has a lower energy content that either natural gas or
propane. “With equipment modiﬁcations to account for its reduced energy potential and
other constituent components, however, biogas can be used in all energy-consuming
applications designed for natural gas or propane (Lusk, 1998, p. 2-11).” Sending biogas
to a ﬂare to be burned is considered the least attractive option for biogas since it does not
generate energy revenues for the farm. Carbon credits may be possible, however,
depending on the amount that is ﬂared (Section 2.11). In general, ﬂaring is limited to
disposing excess biogas which cannot be used by the engine-generator as a result of

downtime or overproduction.

17

2. 2. 7A Electrical Generation

Most farm-based digesters use the biogas output to generate electricity (Lazarus, 2008).
The most common electrical generator system used at farm biogas facilities today is a
stationary internal combustion engine that has been modiﬁed to 1.) run on biogas, 2.)
drive a generator and 3.) produce Single or three phase electrical power (Ciolkosz et al.,
2009). An induction generator is generally used since it can run off the signal from the

utility and will allow parallel hook up with the grid (Lazarus, 2008).

In the process of electric production, the exhaust from internal combustion engines
(waste heat) can be used to pre-heat and maintain the temperature of the manure. Since
electricity production is only roughly 35% efﬁcient, the remaining 65% could be
characterized as “waste heat.” (Gebremedhin, 2006). “Usually, the equipment installed
for capturing this waste heat consists of a heat exchanger and recirculating pump
connected to a system of pipe lines immersed in hot water which deliver raw slurry or
feedstocks to the digester. Another system of pipe lines then pick up slurry from the
digester and return it to the heat exchanger to be recirculated” (Crook and Gould, 2009,

p.45 ).

Due to the high hydrogen sulﬁde (H28) content of biogas, engine-generators may require
more frequent maintenance. Biogas from dairy manure typically contains 0.2-0.4%
hydrogen sulﬁde H28 (Jones et al., 1980). “The sulfuric acid from the H28 can
accumulate in the engine oil, resulting in accelerated corrosion and early failure of engine

components. As a result, engine oil from biogas generators must be changed more often

18

which results in higher operation and maintenance costs. Some digester systems may
even purchase an H28 scrubber to extend the life of the generator. An oil analysis can
also be valuable tool for assessing the condition of the oil and can alert the operator to
engine problems before they cause serious damage (Ciolkosz et al., 2009).

Electrical generation equipment can be very expensive, however, and some operators
who have installed digester systems are unable to recoup the installation and operation
costs through the sale of electricity (Beddoes et al., 2007). An analysis of 38 existing
U.S. manure anaerobic digestion systems indicates that 36% of the total cost of the
system isattributed to the electrical equipment (Bracmort et al., 2008). Due to this high
capital cost, digesters which produce electricity are generally only feasible on larger farm
operations. Recent studies have shown that a herd size of around 800 cows to be the
lower limit (Jewell et al., 1997). The EPA AgSTAR program estimates this number to be

approximately 500 (US. EPA, 2002).

2. 2. 78 Key Considerations

From an engineering perspective, there are also several key considerations that must be
taken into account to determine the feasibility of installing an anaerobic digester. In terms
of electricity production, the ﬁrst element is calculating the energy demand of the
digester. In order to accomplish this, there are two major energy requirements that must
be analyzed: 1.) the amount of energy to bring the inﬂuent manure up to the digester
operating temperature, and 2.) the amount of energy required to maintain the digester at
the operating temperature. (Gebremedhin, 2006). The optimal temperature is maintained

by using either waste-heat captured off the engine-generator or biogas which is used to

19

run a boiler. If heat loss exceeds the waste heat produced from the engine-generator, then

biogas must be diverted from electricity production to the boiler.

Two main performance parameters which affect the amount of biogas used in the boiler
are the total solids concentration and the loss of volatile solids. A low total solids
concentration means that higher amounts of water are present in the inﬂuent which
increases the heat requirements to maintain the optimal digester temperature. In the case
of volatile solids loss, the energy potential of the inﬂuent is reduced which decreases
biogas production. Decreased biogas production results in leSS waste heat capture from
the engine-generator and therefore increases the need to use a boiler to maintain the
optimal digester temperature. The overall result in both Situations is that electricity

production form the digester system decreases.

Another key consideration is the type of animal bedding used. For example, with sand
bedding, the sand-laden manure presents a problem for conventional digester designs.
This is because the sand will settle out in the digester, reduce digester volume and create
excessive wear on components. Over time, the HRT along with the biogas yield will be
reduced (Wilkie, 2005). “Mechanical sand-manure separators, however, can extend the
potential application of anaerobic digestion to scrape operations using sand bedding.
During the sand-separation process, dilution water is added which produces a sand-free
manure stream of low solids content more similar to ﬂushed manure”(Wilkie, 2005,

p.310). In addition to sand, other bedding materials have been identiﬁed as potentially

20

‘ﬁ. - "I. '~ ’.I

troublesome to digester function such as large amounts of straw and wood Shavings

(Steffen et al., 1998).

2. 2. 7C Other Energy Uses

“Given the increase in natural gas prices over the past ﬁve years, the direct use of biogas
as a replacement for natural gas or propane for on-Site heating purposes (e. g., heating
water, heating animal housing, etc.) would provide economic beneﬁts to animal
producers with a consistent year-round requirement for the biogas. The direct use on the
farm for biogas produced via a manure anaerobic digestion system appears to be
economically feasible when the on-farm heating requirement are high enough to utilize
the biogas produced by the system” (Bracmort et al., 2008, p.1). The energy utilized for
the natural gas/propane offsets can come either from waste heat (if using electrical engine
generator) or be taken directly from the biogas production once the digester heating
requirements are met. In both cases, the energy is used to heat water which would
typically be accomplished with either propane or natural gas in the absence of the

digester.

“AS with electrical engine generators, boilers are also adversely affected by the corrosive
characteristics of biogas. One way around this problem is to operate the boiler
continuously at a temperature above dew point” (Beddoes et al., 2007). This is because it
prevents sulfuric acid (H2804) from forming which causes corrosion. If the boiler is only
used on an “as needed” basis, however, this may not be an effective strategy. “It should

also be noted that most farm heat requirements are seasonal and the problem of how to

21

best use the gas in the “off” season must be dealt with. Storage of the gas in large
amounts is largely impractical because of the relatively low heat value of methane
(compared to propane and other liquid fuels) and its difﬁculty to liquefy under reasonable
pressures. Most storage applications would likely involve only short-term accumulations
of methane (Fulhage et al., 1993, pp. 4, 7).” This is a main reason why biogas is
generally consumed on-site continuously either for electricity production or for heating

needs.

Biogas can also be upgraded to biomethane for retail sale by processing it to remove
moisture, H28 and CO2, none of which possess any energy value. In order to be
economical, the cost of upgrading must be less than the incremental difference between
the biogas and natural gas cost (Bracmort et al., 2008). Currently, this is not a routine

practice on dairy farms and therefore biogas upgrading was not examined in this study.

2.8 Annual Operation and Maintenance Costs

For digesters with electrical equipment, “Operation and Maintenance (0&M) costs
include daily operator labor to pump the manure and perform routine maintenance;
expenses for engine oil changes and minor repairs; and periodic major repairs and
maintenance such as engine overhauls, Sludge removal, and ﬂexible cover repair or
replacement” (Bracmort, 2008, p.8 ). In addition, all digesters require some management
and labor to control the process. Successful operation for a typical on-farm digester will

require a minimum of 1-2 hours per day for monitoring, loading, unloading and

22

performing general maintenance (Jones et al., 1980). This estimate will vary depending

on the digester design and may ﬂuctuate in times of repair and overhaul.

2.9 Electricitv Com

“Producing electricity is only part of the challenge with making an anaerobic digester
cost effective. In particular, selecting a favorable electricity purchase agreement is vital
and has a signiﬁcant impact on the proﬁtability of the digester system. In some cases,
variables such as the number of different electricity meters may limit the farm’s ability to
utilize electricity on-farm where it has the highest value. Speciﬁc requirements for
insurance, demand charges for the use of electricity when the on—site generator is down”,
and other rules may also make it difﬁcult to deal with the utility company (Wright, 2001,
p.10). Electricity contracts for anaerobic digesters are of three types: surplus sale, buy-all
sell-all and net metering. While the speciﬁcs of each will vary depending of the power
provider, the basic concept behind each agreement is the same. The EPA AgSTAR
handbook also makes reference to these three agreements and provides sample contract
language for reference. In this study, the purchase agreement speciﬁcs were taken from a

Michigan utility which has been involved with several digester projects in the state.

2.9.1 Surplus Sale

The concept of a surplus sale agreement is that only excess electricity production is sold
back to the utility company. Under this agreement, the farm will ﬁrst utilize their
electricity on-Site where they have the ability to offset their usage at the retail rate. The

amount of energy available for resale will then depend on the rate at which biogas can be

23

produced on a continuous basis as well as the amount and timing of electricity use for the

farm’s dairy operation (the load curve) (Mehta, 2002).

According to the Department of Energy (DOE) Energy Information Administration, the
rate (commercial) at which electricity is offset is approximately $0.093 nationwide. Any
electricity that is sold back to the utility, however, is valued at the hourly real-time
locational marginal price (LMP) of the particular utility’s load node as determined by the
Midwest Independent Transmission System Operator (M180). M180 manages one of the
world’s largest energy markets using complex computer programs. The prices obtained
from MISO represent the wholesale price of electricity and tend to be roughly half of the
retail rate on average. Upon examination of historical prices, the LMP ﬂuctuates with
demand and can even be a negative value during certain periods. This is due to the fact
that the utility cannot Shut down its operations in times of lower demand. Dming these

periods, the farm would actually be paying the utility to put their electricity on the grid.

In addition, depending on the particular rate plan for customer generation, some utilities
will impose an administrative charge per kWh for the purchase of the electricity from the
power provider as well as a system access charge. The administrative charge
compensates the utility company for time and labor time associated with administering
the agreement. Minimum and maximum monthly charge amounts are also stipulated in
the purchase agreement and are adjusted for inﬂation based upon consumer price indexes

from the US. Bureau of Labor and Statistics. The system access charge recovers the

 

3 This price does not include inﬂation.

24

costs of metering equipment, meter reading, billings and other customer-related operating

COSIS.

An interval data meter is usually required for the utility to monitor the customer’s
generator and electricity ﬂow. The company typically reads the meter electronically via
telecommunication links or electronic data methods to obtain the information needed for

billing.

Generally, standby charges will also apply if the farm wishes to purchase electricity from
the utility when their engine-generator is down. Standby charges have been cited as a
signiﬁcant obstacle when a farm attempts to offset their energy use with electricity
produced from their engine-generator. The concept is that these charges are used to
compensate the utility company for providing electricity when the digester engine-

generator is not running.

A typical digester engine-generator with “good” performance will generally be
operational 90% of the time (US. EPA, Undated). This means that the farm will need
energy from the utility the remaining 10% of the time or when peak demand exceeds
digester output. The speciﬁcs of standby service will vary depending on the utility
company, but generally are composed of two parts, power supply and delivery standby
charges. Power supply refers to the cost of fuel and power producing investments.
Delivery charges compensate the utility for transporting the electricity from the plant to

the customer.

25

2.9.2 Buv-All Sell-All

Under this agreement, all electricity produced by the digester is sold to the utility. The
farm must then purchase all their electricity needs from the utility at the applicable retail
rate. AS with the surplus sale, the rate of compensation by the utility is the hourly real-
time LMP. At ﬁrst, this may appear to be an inferior agreement when compared to the
surplus sale. The tradeoff, however, is that standby charges do not apply since no on-farm
energy usage is being offset by the digester system. On the other hand, any energy
demand created by the digester system itself will need to be purchased from the utility
and would represent an incremental project cost. As a result, this cost must be considered
in the capital budgeting analysis and can have a potential impact on the economics of this
agreement depending on the amount of energy used by the digester. In addition, the same
guidelines for system access and administrative charges will apply as under the surplus

sale agreement.

2.9.3 Net Metering

With the passing of Michigan Public Act 295 in October of 2008, a new net metering law
was approved by the state legislature. The concept of net metering is that the utility
allows a customer to offset only their electrical “need” and receive credits for any energy
produced which exceeds that need. When a customer is under a demand based rate
system, the electrical need is usually established based upon the peak demand over a
twelve month period. If that information is not available, an appropriate level is
negotiated with the utility. This limits the engine-generator Size that can be used with a

net metering agreement, Since a typical complete mix digester in Michigan is capable of

26

producing more electricity than the need of the farm. Without the use of net metering, a
farm would normally use an engine-generator sized to match the full biogas production

potential of the digester system.

Another distinguishing aspect of net metering agreement is that fact that no actual
payment is made for the electricity produced by the customer. The revenue is
represented by the electricity purchases offset by self-generation and the credits which
can be carried over from month to month. Most utilities, however, will put a cap on the
period of time that credits can accumulate (e. g., one year). Net metering is not intended

to be a proﬁt making mechanism for the digester owner.

Under Michigan law, net metering consists of four main categories based upon the
nameplate capacity of the generator. For digester projects, only categories 2 through 4
will realistically apply.

0 Category 1- projects 520 kW

0 Category 2- projects >20 kW and 3150 kW
0 Category 3- projects >150 kW and 5550 kW
0 Category 4- projects >550 kW and S 2 MW
0 Category 5- projects > 2 MW

While net metering is not a new concept, Public Act 295 made the agreement more
favorable for category 1 projects such as small wind and solar. For projects in category

1, the utility will credit the customer for their excess generation at the retail rate, which is

27

referred to a “true” net metering. On the other hand, categories 2-5 receive “modiﬁed”
net metering in which customer credits are valued at either the monthly average LMP
(similar to the surplus sale agreement) or the power supply component4 of their electric
bill. While the act signed into law gave utilities the choice between the two pricing
schemes, recently published utility guidelines from the largest Michigan utilities indicate
that they have chosen to compensate at the power supply component price. The power
supply component of the customer’s bill is approximately $0.06 whereas the monthly
average LMP is roughly $0.04. While not as favorable as “true net metering,” the higher
value for electricity produced by the digester system represents an improvement on

previous policies.

There are also other beneﬁts of the new net metering for digester projects. One beneﬁt is
a provision that category 2 projects do not pay standby charges. Additionally, category 3
projects do not pay standby charges unless the engine-generator used has a nameplate
capacity greater than 150 kW. At the current time, however, the speciﬁcs of standby
charges under the net metering law have yet to be established. In the absence of speciﬁc
rates, standby charges are assumed to be the same as under the surplus sale agreement.
Another beneﬁt is that category 2 projects do not incur the cost of any additional
metering equipment. Also, there is no mention of system access or administrative
charges for projects in categories 2 or 3. In contrast, surplus sale and buy-all sell-all

agreements are subject to these charges.

 

4 The power supply component refers the cost of fuel and power producing investments.

28

2.10 Renewable Energy Credits

If a farm with a digester generates alternative energy, it can receive a Renewable Energy
Credit (REC) for every megawatt hour (1,000 kWh) of energy it produces (Leuer et al.,
2008). Farms in Michigan may sell these to utility companies, if allowable under
contract, or sell them as carbon credits (as explained in section 2.11). A farm may still
sell RECS to one utility and have a purchase agreement established with another. RECS
are completely separate form electricity purchase agreements and represent different

revenue sources.

AS a result of the recent Renewable Portfolio Standard, PA 295, a number of Michigan
utility companies are currently seeking RECS in order to comply with the new law. .

Recent contracts from large Michigan utility companies indicate that RECS in the future

will be purchased for $30 to $50 per credit (1,000 kWh).

2.11 Carbon Credits

Carbon credits have the potential to be an essential revenue stream for an anaerobic
digester system. This model is based upon the rules established by the Chicago Climate
Exchange (CCX), which is North America’s only active voluntary, legally binding
integrated trading system to reduce emissions of all six greenhouse gases (GHG’S)
(CCXa, 2009). Credits are issued based upon an emission baseline calculation which
calculates the amount of methane that would be emitted to the atmosphere during the

crediting period in the absence of the anaerobic digester project (CCXa, 2009).

29

There are two methods to calculate the baseline methane emissions. The ﬁrst is by the
actual monitored amount of methane captured and destroyed by the project activity using
existing CCX monitoring protocols and a global warming potential (GWP) for methane
of 21. The GWP for a particular greenhouse gas is deﬁned as the ratio of heat trapped by
one unit mass of the greenhouse gas to that of one unit mass of carbon dioxide (CO2)
over a speciﬁed time period. The second is calculated “Ex Ante” which refers to the
amount of the animal manure that would decay anaerobically in the absence of the project
activity. CCX calculates the amount of methane destroyed or avoided using the method
which results in the lowest level of methane. Each credit traded represents a reduction of
one metric ton of carbon dioxide (CO2) and the value of each credit ﬂuctuates with the
current market price. For example, in the summer of 2007, the price of a carbon credit

was roughly $7.50. In contrast, the price dropped to only $0.25 in the summer of 2009.

In order to receive carbon credits, certain guidelines also apply which place certain
requirements on participants. The guidelines (CCXb, 2009) applicable to anaerobic
digesters are highlighted below.

0 Only renewable energy systems activated on or after January 1, 2003 qualify.

0 Project proponents need to demonstrate clear ownership rights of the emission

reductions from the destruction of methane.

o All projects must be independently veriﬁed by a CCX-Approved Veriﬁer.
Speciﬁc guidelines on the equipment and record keeping required are Speciﬁed in

the agricultural methane offset protocol.

30

A detailed description of the calculation of carbon credits is included in Chapter 3.

2.12 Co-digestion

While livestock manure is the main feedstock for farm-based digesters, other feedstocks
(e.g., crop residues, leaves, food processing waste, ethanol syrup) can be added to
potentially increase biogas production (co-digestion). The goal of co-digestion is to
maximize the amount of carbon in the mixture while staying within the correct C: N ratio.
The overall nutrient ratio in waste materials is of major importance for the microbial

biodegradation process (Steffen et al., 1998).

Large amounts of agricultural raw materials are processed in the food industries. During
processing, wastes and wastewater are produced which can often be co-digested in
agricultural digesters (Steffen et al., 1998). Not every feedstock, however, is suitable for
anaerobic digestion. They vary considerably in composition, homogeneity, ﬂuid
dynamics and biodegradability. When selecting wastes for digestion, the total solids
content, the percentage of volatile solids, the C:N-ratio and the biodegradability have to
be carefully considered (Steffen et al., 1998). For farm-based anaerobic digesters, mainly
feedstocks that have characteristics (moisture content, total solids, etc.) similar to animal

manure should be considered for anaerobic digestion (Scott and Ma, 2004).

Due to the variability among digester systems, no established guideline exists for the

best percentage of feedstock to mix. Some anaerobic digesters have been built to process

100% food waste and the perception is that it is possible to go as high as 75% (Scott and

31

Ma, 2004). In contrast, however, the Ministry of Agriculture, Food and Rural Affairs in
Ontario, Canada recommends that a farm-based digester can only blend up to 10-25%
non-farm source material and work effectively (Ontario, 2009). In general, gradual
loading of the digester can give the system time to adjust to the new feedstock and prOper
maintenance can help to prevent most problems associated with co-digestion (Scott and

Ma, 2004).

“Non-farm industries that have organic wastes to dispose of will sometimes pay tipping
fees to a farm digester to accept the waste. For the digester enterprise, the tipping fees
can be an important side beneﬁt of accepting this feedstock, making the difference
between proﬁt and loss. A concern, however, is that on livestock farms with small land
bases, the livestock manure alone may already have too much nitrogen and phosphorus
for the cropland available. Imported non-farm organic wastes would contain additional
nutrients, which could exacerbate the cropland nutrient imbalance. The tipping fees and
added gas output need to be weighed against potentially greater manure disposal costs to
take the efﬂuent to more distant cropland” (Lazarus, 2008, p.15 ). The model in this
study allows the user to account for this potential tradeoff between increased gas

production and increased disposal costs.

It may also be the case that a farm decides to purchase non-farm organic wastes (e. g.,
ethanol syrup), in which case no tipping fee is received. In this situation, the farm incurs
the cost per unit for the feedstock purchase and possibly the hauling cost from the point

of pickup to the farm. The ﬁnal disposal cost will be a function of the amount of solids

32

remaining after digestion and the distance to the disposal site (either to a landﬁll or for
land application). The beneﬁts of increased biogas production, however, may justify the

increased costs depending on the situation.

With co-digestion, it is also important to be aware of any applicable state laws which may
prohibit or place restrictions on the mixing of non-farm organic waste with livestock
manure. In Michigan, the Michigan Department of Environmental Quality (MDEQ) has
issued an “Organic Residuals Exemption” for on-farm anaerobic digestion. It only
allows food processing residuals (as deﬁned in Section 324.11503(9) of Part 115 in the
Natural Resources and Environmental Protection Act (NREPA)), syrup from ethanol
production and ﬁsh wastes to be added. Any other materials used must receive written
approval by the MDEQ.
Per NREPA, "Food processing residuals" means any of the following:
(a) Residuals of fruits, vegetables, aquatic plants, or ﬁeld crops.
(b) Otherwise unusable parts of links, vegetables, aquatic plants, or ﬁeld crops
from the processing thereof.
(0) Otherwise unusable food products which do not meet size, quality, or other
product speciﬁcations and which were intended for human or animal
consumption.
The approval, however, is limited to a maximum 20 percent substitution rate, by volume,
of the material going into the digester unless an alternative substitution rate is approved

in writing. Digester operators must also be approved by the MDA as a “Certiﬁed

33

Operator for Agricultural Anaerobic Digesters.” Other provisions and permit

requirements under NREPA pertaining to air, water and hazardous waste are not exempt.

Speciﬁcally in regard to land application, the digester efﬂuent (digestate) may be land
applied provided several conditions are met (MDEQ, 2009).
1.) “The owner/operator must ensure that the digestate is managed according to the
Nutrient Utilization Generally Accepted Agricultural Management Practices
(GAAMPS) or the Manure Management GAAMP developed under the Right to

Farm Act.”

2.) “The operator of the farm must ensure that the concentration of contaminants in
the soil, after land application, shall not cause the creation of a “facility” as

deﬁned by Part 201 , Environmental Remediation, of the NREPA.”

3.) “If the digestate is not used on the farm where it was generated, it must then be

licensed with the MDA under Part 85, Fertilizers, of the NREPA.”

This organic residuals exemption represents a recent change in the state’s previous

policies regarding the co-digestion of waste for on-farm anaerobic digesters.

2.13 Non-Energy Beneﬁts

Beneﬁts unrelated to the production of heat and electricity are present yet not always
easily quantiﬁable. The process of digestion itself converts volatile organic compounds in
manure to more stable forms that can be land-applied with fewer objectionable odors

(Lazarus, 2008). For example, if the manure is spread on the operator’s own cropland,

34

the reduced odor potential may have economic value to the livestock operation by
minimizing the chance of neighbors’ complaints or nuisance lawsuits. The reduced odor
of the digestate itself may make it more marketable to crop farms (Lazarus, 2008).
Quantifying this value has been notoriously difﬁcult, however, since factors unique to an
individual farm will determine the exact value of the odor reduction. In Kramer’s
Anaerobic Digester (AD) casebook nearly all the system owners mentioned odor
reduction as an important beneﬁt of their AD system (Kramer, 2004). For new farms,
some means of odor control is often either implicitly or explicitly required for the facility
to be sited and built. Some owners of ongoing operations reported that the encroachment
of residential developments near their farms have put increasing pressure on them over
time (sometimes in the form of lawsuits) to reduce odor emissions (Kramer, 2004).
Because digested manure (digestate) has much lower odor than raw manure, owners also
have more ﬂexibility in when and where they ﬁeld-apply it (e. g., they do not have to wait

until the wind is blowing the right way or avoid applying it on weekends) (Kramer,

2004)

In addition to odor control, increased ﬂexibility of nutrient management is also cited
among the non-energy beneﬁts of anaerobic digestion. “Nearly all animal manures are
land applied, which means that some of their nutrient value is returned to the soil for
plant growth. However, much of the nutrient value contained in manure can be lost
before it is recycled or before the nutrients can be presented in a plant usable form” under
traditional manure handling practices” (Lusk, 1998, p. 2-16). Since digestate is in a more

stable form, crops are able to absorb more nutrients from the manure which increases its

35

value as a substitute for commercial fertilizer. The increase in uptake reduces the
possibility of nutrient run-off into surface waters given proper land application (e.g.,
direct soil injection) (MDA FAQ, 2009). Pathogens in digested manure are also reduced

by as much as 99.99% (Lusk, 1998).

Further beneﬁts are also possible through the use of mechanical solids separation. For
example, the digestate solids (bioﬁbers) can be utilized as a bedding material in dairy
farm free-stall barns, sold as a soil amendment or used in other applications such as: the
production of construction products (wall board), decking, and greenhouse pots
(Safferman and Faivor, 2008). The solids separation is also made easier after the

digestion process.

Since separated solids have less volume, they can also be hauled to fertilize distant ﬁelds
at less cost than hauling the original manure (Lazarus, 2008). The phosphorus portion of
the manure is also primarily sequestered in the solids which aids in residuals
management. In contrast, the liquid fraction (ﬁltrate) is high in nitrogen and low in
phosphorus thus enabling irrigation of ﬁelds that may be phosphorus limiting. The
ﬁltrate is also low in volatile fatty acids and therefore does not stick to the leaves and can
be spread on growing plants (e.g., on corn as tall as 20 inches) with only minor risk of
burning (Lusk, 1998). Studies which quantify the amount of these beneﬁts are not

presently available.

36

One concern about using digestate solids as bedding, however, is that pathogens might
remain to cause increased mastitis problems (Lazarus and Rudstrom, 2007). Although,
the Agricultural Biogas Casebook cites various examples of farms utilizing digestate
solids without adverse effects, “the literature review suggests that more research is
needed to clarify the impact of bedding type on mastitis, in the context of the many
management factors on a typical dairy farm” (Lazarus, 2008, p. 16). Although solids
separation is easier (e. g., reduced use of chemicals) with digestion, the beneﬁts of solids
separation can also be achieved without incorporating a digester in the system. For that
reason, the AgSTAR digester protocol recommends setting boundary conditions for
digester evaluations that leave out the separator part of the system (Martin, 2006). For
these reason, the model developed for this study does not include the use of a solids

separator or the use of digestate solids as livestock bedding.

37

Chapter 3: The Model

The model developed in this study has two distinct functions. The ﬁrst is its use as an
outreach tool with a variety of modules allowing for the analysis of existing systems as
well as those in the planning stage. In the absence of speciﬁc information, its default
value modules allow the user to assume the initial cost of investment, operation and
maintenance (O&M) costs as well as monthly biogas production. In order to construct the
default value investment module, the budget from a large Michigan dairy farm was used
as a reference. Component line items were estimated as either a percentage of total costs
or sized based on related elements of the digester system. Other modules such as O&M

costs and biogas production were based upon values from the published literature.

If data is available from an operational system, these same values may be input directly.
In terms of biogas production, a separate module is included for the addition of
feedstocks which can increase output and improve digester proﬁtability. Biogas
production is then converted into revenue streams under surplus sale, buy-all sell-all and
net metering electricity purchase agreements. RECS, carbon credits and potential
propane offsets are also included as typical sources of income which may be available for

on-farm digesters.

Due to the high capital costs of an anaerobic digester, a separate module is used to
account for the various ﬁnancing options which generally come as a combination of
grants, loans and equity. All revenues and expenses are then analyzed in a capital

budgeting model using net present value, internal rate of return and simple payback

38

period methods. The user can then alter the various system parameters to examine

performance under current and hypothetical circumstances.

A second purpose for this model, and the primary focus of this study, is its use as a
research tool to analyze energy policies as they relate to anaerobic digesters. Three
different Michigan utility purchase agreements (surplus sale, buy-all sell-all and net
metering) are examined in detail with rules and rates taken directly from one of the
state’s largest utility companies. Of Special interest is the new net metering law and its
effect on the economics of a typical digester in Michigan. While customers installing
small wind and solar systems clearly beneﬁt under this law, it is unclear how it compares
to the other electricity purchase agreement options currently offered by utility companies.
Also related to energy policy is the issue of standby charges. According to the literature,
these charges are a major obstacle when using a digester to offset on—farm electricity use.
Since the rules for standby charges are often rather complex, however, very little
information is available which analyzes their impact on digester proﬁts. The high cost of
interconnection is also recognized as a barrier for digester owners, but an analysis of the
details of interconnection is beyond the scope of this study. The primary research focus
of this paper was a comparison of the three different electricity purchase agreements
including standby charges under a variety of scenarios for a representative Michigan

dairy farm.

This chapter explains the details of the model using an example dairy farm with a

lactating herd of 1,000 Holstein cows. The model also assumes the digester to be a

39

complete mix design with a total solids content of 8% and a hydraulic retention time
(HRT) of 20 days. Data related to on-farm energy usage and investment costs were taken
from the Michigan case farm and scaled to match the needs of the 1,0005 cow example
dairy farm for analysis. Examples are provided with each equation to illustrate the

included formulas. Figure 3 illustrates the layout of the modeling process.

 

5 dry cows and young stock not included

40

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m G O <
r r r _
. :
_ _
32.: 03%: 526m 32: “swam:
him 825 - - - - . 530: 2:33am
3582 32 =<-=om :23: 2% Beam scan. €285
K
3:9: AI 3:9: 3:9: 4 383:0 9:985": F:
3:582 82 =<-=om =<r§m 2mm "239:6 $335 I a
z : IIIIIIIII 329m
max—31¢. 5:355 couemooaom l
K
2:: as: new: .
553:8: 9 5:36:85 380 33:03:32 I
3*on bam— 825 I . mewomm 2:.on on 5:825
A _ METER _
8:30:35 mama: .8532: mmwomm a. moan: 3658:: A
Eam— 825 8.: 3:9: 2:33am .5: 3:9: / 3:28.332: 3:96 I
:memowMUIOU /l
3:08:82: 3:: 825 ..... E08626: o:_m> ::3oQ

 

 

 

 

 

 

3:955 33: .m Paar:

41

 

 

inseam 33%:

 

 

 

 

 

mix—SE 35033: 3:93

 

 

 

 

Cow—A B 30 01:an Hog—mm H rrrrrrrrrrr

 

 

 

 

 

 

 

 

 

 

 

wﬂOEQ—Sm 0800::

 

 

 

 

 

 

 

 

 

 

 

 

£86 :85

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<

 

 

8.55.58 :53 352 .m 2.9...

42

3.1 Default Value Investment
The model has the option to utilize either a default value investment or directly entered
investment information based upon the user’s own information. The ﬂexibility allows

the user to provide decision support in the absence of speciﬁc cost data.

3.1.1 Capital Investment

Since digesters are often engineered to ﬁt the individual needs of a particular farm, it is
difﬁcult to standardize the exact components included in each system. The task is further
complicated by the fact that itemized budgets with each component (e.g., pumps, valves,
mixers) separated out by cost are generally not publicly available. For example, the listed
cost of a digester tank will generally aggregate the cost of the tank, roof, insulation,
heating and related components together. The most valuable piece of data in cost
determination, however, was the Michigan case farm budget which separated each
component out by quantity and price. In order to design a default value investment
module, component line items were estimated as either a percentage of total costs or

sized based upon related elements of the digester system.

The total project costs formulated in the model were then benchmarked against complete
mix digester cost curves created by the EPA AgSTAR program (US. EPA, 2009).
AgSTAR analyzed AD system cost data for 10 complete mix digesters on dairy farms for
which itemized cost estimates were available. Using SAS 9.1, regression analyses were
performed for the complete mix digester costs versus the number of dairy cows. Included

in the cost of each system was the cost of the digester, the engine-generator set,

43

engineering design and installation. The analyzed systems did not include those designed

for co-digestion and were based on quotes for systems in 2005-2008.

In order to validate the default value investment module, the model was then run for herd
sizes ranging from 500 to 4,000 cows and compared to the AgSTAR cost curves (Figure

4).

Figure 4. Model Cost Estimation Compared with AgSTAR Cost Curve

 

Model Cost Estimation vs AgSTAR Cost Curve

 

 

 

 

 

 

 

 

’s‘

2 Model
.35. *

E ' ----- AgSTAR
8

O

l-

 

 

 

Herd Size

 

 

 

When predicting the cost of the digester, the total solids content, hydraulic retention time
(HRT) and amount of solids lost in collection are important in calculating the size of the
tank and biogas production. For the model veriﬁcation (Figure 4), the follow assumptions
were made: scrape collected free stall barns, manure removed 3 times a day, Holstein
cows (1,400 lbs.), parlor water added, a total solids (TS) content of 8%, a HRT of 20 days

and a modest amount straw bedding.

44

3.1.2 Design Study and Engineering
This is not a capital cost, but was included in a majority of case budgets examined.
Engineering and design fees were assumed to be to be 8% of the total capital costs (State

of Louisiana, 2007).

3.1.3 Excavation
Excavation was estimated to be 4% of total capital costs and was taken from the case

farm budget.

3.1.4 Tanks

The cost of tanks represents a signiﬁcant portion of the investment and must be Sized
according to system parameters, as covered in sections 3.2.3A through 3.2.3 C. The post
digestion storage, equalization tanks, roofs and insulation are sized and priced based upon
the values determined for the digester tanks and are included in sections 3.2.3D through

3.2.3G.

3.1.4A Desired Tank Volume (Digester)

The desired tank volume (DTV) in gallons was estimated based upon the average daily
ﬂow rate (gallons per day) and the hydraulic retention time (HRT). An additional
volume of 10% was then added for freeboard to arrive at the desired tank volume.
Freeboard Space is necessary to provide room for extra inﬂuent and gas storage in the
case of a manure pump malﬁmction or engine-generator downtime. Equation 1 is the

calculation for DTV.

45

DTV = (VAD x HRT) x (l + F8) (Equation 1)

DTV : Desired tank volume of the digester tank (gallons)

V A D : Average daily ﬂow rate (gallons/day)

HRT : Hydraulic retention time (days)
F B : Freeboard space

For E136: 658, 900 gallons = [29,950 gpdx 20 days] x (1.10)

31.48 Tank Ouantity (Digester)

In order to determine the quantity of tanks needed, the assumption was made that 870,000
gallons would be the cut off point at which two digester tanks would be needed. This
volume was selected based on an examination of all operational complete mix digesters
listed on the EPA AgSTAR digester database (U .S. EPAb, 2009). For dairy farms
utilizing more than one tank, the average tank volume was found to be approximately
870,000 gallons. Therefore, if the desired tank volume is less than 870,000 gallons, the

assumption is made that one tank is used (Table 1).

 

 

 

 

Table 1. Tank Quantity

Desired Tank Volume Tank Quantity
(gallons)

0 - 870,000 1

> 870,000 DTV / 870,000 gallon?

 

 

 

3.1.4C Tank Unit Cost
To determine the unit cost of each digester tank, a similar methodology was used.

Corrections are made, however, if the desired tank volume is greater than 870,000

¥

6 EF refers to a 1,000 cow example farm used to demonstrate the use of the model.
Quantity rounded to the nearest whole number

46

gallons, but less than 1,000,000 gallons. For this speciﬁc interval, two 500,000 gallon
tanks were utilized to avoid unnecessary capacity. Similarly, if the desired tank volume
is greater than 1,000,000 gallons, but less than 1,400,000 gallons, then two 700,000
gallon tanks were used. The volume of the tank multiplied by a cost per gallon of $.20
becomes the unit cost (Equation 2). The tank quantity multiplied by the unit cost is the
total digester tank cost (Equation 3). The value of $.20 was obtained from the case farm
budget. Table 2 Shows the determination of the Total Tank Volume.

Table 2. Total Tank Volume

 

 

 

 

 

 

 

 

 

 

 

 

DTV(gallons) Selected Tank Total Tank Volume
Tank Size Quantity (gallons)
(gallons)
0 - 870,000 DTV 1 DTV
870,001 — 1,000,000 500,000 2 1,000,000
1,000,001 - 1,400,000 700,000 2 1,400,000
1,400,001 - 2,100,000 700,000 3 2,100,000
> 2,100,000 870,000 DTV/ 870,000 x Tank
870,000 Quantity
gallons
Unit Cost = Total Tank Volume (gallons) x Cost per Gallon (Equation 2)
For EF: $131, 780 = 658, 900 gallons x $0.20 per gallon
Total Digester Tank Cost = Tank Quantity x Unit C 0st (Equation 3)

For EF.‘ $131,780 = I Tankx $131,780

3. 1. 4D Tanks- {Post Storage)

Post digestion storage was obtained by ﬁrst calculating the post storage tank volume from
the case farm and comparing it to the volume of the digester tanks. The post digestion
storage volume was found to be 20% of the digester tank volume for the case farm. This

Size relationship was then used to estimate the post digestion storage volume for all

47

digester Sizes in the model. The storage volume was then multiplied by a cost per gallon

of 8.30 which was also taken from the case farm budget (Equation 4).

C Storage = (DTV x .20) x C P G (Equatron 4)
C Storage : Cost of post-storage tanks
C P G : Cost per gallon for post storage tanks ($/ gallon)

For EF.‘ $39,534 = (658,900 gallons x .20) x $0.30

3. 1. 4E Tanks- (Equalization)
Two equalization tanks are used to stabilize the inﬂow and outﬂow of manure and are
assumed to be a ﬁxed cost of $8,000 per tank. This value was obtained from the case

farm budget.

3.1.4FRoofs-(Digester)
The number of roofs needed is a function of the number of digester tanks and is assumed
to be 60% of the cost of the digester tank. This value was obtained from the case farm

budget.

3. 1. 4G Insulation
The number of units of insulation needed is a ftmction of the number of digester tanks
and is assumed to be 17% of the cost of the digester tank. This value was obtained from

the case farm budget.

48

3.1.5 Boiler

To estimate the cost of the boiler, the size needed was determined by the total heat input
required for the coldest average ambient temperature of the year (Perssen etal., 1979).
The heat input required is a function of the heat loss through the digester and varies
depending on the dimensions of the digester and the insulation material used. Two
different size boilers were priced and the model was programmed to Select the correct
size based on the rated heat capacity (Btu/hr) of each boiler. The price of $90,000 was
obtained from the case farm budget and the $45,000 was an estimate based on a boiler

capacity need half the size of the case farm (Table 3).

Table 3. Boiler Size Ranges

 

Boiler Size Ranges Cost

 

Boiler Size (300 MMBtu/hr to 2,000 MMBtu/hr) $45,000

 

 

 

 

Boiler Size (2,001 MMBtu/hr to 5,000 MMBtu/hr) $90,000

 

3.1.6 Heating

Heating refers to the tubes within the ﬂoors and walls of the digester tanks which
circulate water. This water is warmed by waste heat recovered from the engine-generator
and helps raise the inﬂuent temperature to the (95°-105°F) necessary for optimal
mesophilic digestion. The number of units of heating was assumed to be a function of the
number of digester tanks and a value of 5.5% of the cost of the digester tank was taken

from the case farm budget.

49

3.1.7 Plumbing. Valves, Mixing_and other Miscellaneous Components
The cost of plumbing, valves, mixing units and miscellaneous items were combined and
were assumed to be 14% of total capital costs. This value was obtained from the case

farm budget.

3.1.8 Wager-to-Manure Heat Exchangers

Water-to-manure heat exchangers capture heat from the exhaust of the engine-generator
and were assumed to be 4.5% of the total capital costs. This percentage was taken from

the case farm budget. For the purposes of this study, a water-to-manure heat exchanger
was not considered a piece of equipment to be included with all complete mix digesters.
The module was programmed to include this piece of equipment, however, if additional

heating ability is needed in addition to a boiler.

3.1.9 Instrumentation

Instrumentation refers to the cost of the computer equipment and ﬂow meters to monitor
the system and record the necessary information to receive carbon credits. This cost is
assumed to be a ﬁxed $28,000 and will not vary with herd Size. This value was taken

from the case farm budget.
3.1.10 Contingency

Contingency refers to funds set aside for unexpected budget overspends and increases in

construction costs and were assumed to be 5% of total capital costs.

50

3.1 .l 1 Engine-Generator

The cost of the generator was a function of the average yearly electricity output in kWh.
A range of generator Sizes was established based on case farm data and other published
case studies (Table 4). When the average yearly electricity output falls between a
particular engine-generator Size range, the model selects the average cost associated with

the corresponding generator.

Table 4. Engine-Generator Costs

 

 

 

 

 

 

 

 

Engine-Generator Size Range Average Cost
701 kW to 900 kW $400,000
501 kW to 700 kW $350,000
301 kW to 500 kW $300,000
201 kW to 300 kW $250,000
121 kW to 200 kW $200,000
50 kW to 120 kW $150,000

 

 

 

3.1.12 Buildigg
The cost of the building was assumed to be a function of the generator Size and was
estimated at 10% of the cost of the generator. The 10% was obtained from an actual

itemized budget from a project developer.

3.1.13 Switchgear and Addition_al Engine Components

The switchgear and additional engine components were assumed to be a function of the
generator size and were estimated at 70% of the generator cost. The additional engine
components also include items such as heat exchangers which capture heat from the
exhaust of the engine-generator. The 70% was based upon an actual itemized component

list from a project developer (Equation 5). .

51

C 0.70 (Equation 5)

SC = CE — Generator x

C S C : Cost of switchgear and components

: Cost of the en ine- enerator
E — Generator g g

For EF.‘ $140,000 = $200,000 x 0. 70

3.1.14 Interconnection

The process of interconnection involves connecting an electricity producing generator to
the grid. The cost can vary greatly depending on the location of the farm and the size of
the generator (U.S. EPAa, 2009). The EPA AgSTAR program has estimated this cost to
be 7.9% of the total project capital costs on average. This percentage is also used in the

model.

3.1.15 Salvage Val_u_e

To determine the salvage value of the digester, the user must enter the value on each
component that will remain at the end of the project period as a percentage of the
purchase price. With both the direct-entry and the default value investment modules, the
salvage value of the initial investment system components were not determined by the
resulting book value under the Modiﬁed Accelerated Cost Recovery System (MACRS).
In this study, the engine-generator was assumed to be worth 10% of the purchase price

and the digester tanks (heat, insulation, rooﬁng, concrete) were valued at 2%.

If the direct-entry investment module is being used, however, the book value (MACRS)

of any repair and replacement parts purchased after the initial investment (year zero) are

52

included in the salvage value. In this module, the total salvage value then consists of the
percentage of value on each component from the initial investment combined with the
book value of the repair and replacement parts remaining at the end of the project period.
The direct-entry of repair and replacement costs is explained in detail in Section 3.3

which describes the operation and maintenance costs module.

3.2 Operation and Maintenance Costs

Operation and maintenance (O&M) costs were determined in the model by either
directly entering Speciﬁc values or using the default value investment module. The
default value module calculates costs as a percentage of total capital costs obtained from
the published literature. The values used in case farm examples are from the default value

investment module since actual 0&M data was not available.

3.2.1 Digester
When the user is analyzing an already existing system, speciﬁc data may be entered in
the model. The costs are separated apart between the digester and the engine-generator

unit.

For both the digester and generator maintenance, the labor costs were calculated as
follows in Equation 6.

LYC = L. Hr x L R x 52 weeks (Equation 6)

LYC : Yearly labor costs ($)
L m: Total hours to service and operate per week (hrs)
L R : Rate of pay per hour ($/hr)

53

For EF: $3,276 = 3.5 hours x $18/hour x 52 weeks

For repairs and replacements, the costs were estimated. The sum of the annual costs for
the ﬁrst 5 years were determined and weighted for the ﬂow of expenditures (Table 5).
Expenditures in years four and ﬁve were weighted progressively higher. After year ﬁve,
the costs increased at a rate of 3%. This approach was taken from the American Society
of Agricultural and Biological Engineers (ASABE) in which downtime and reliability are

calculated as a logarithmic cost function increasing with accumulated use and age.

Table 5. Direct-Entry Operation and Maintenance Costs

 

 

 

 

 

 

 

 

 

 

 

 

Cost for First 5 Years: $150,000
Rate Increase After Year 5: 3%
Year 1 2 3 4 5
Cost Weight Factor: 10% 15% 20% 25% 30%
Repair and Replacement Cost: $15,000 $22,500 $30,000 $37,500 $45,000

 

 

In addition to estimated costs, the model has the ﬂexibility for the user to enter Speciﬁc
repair data for large purchases. Assets that are likely to need replacement or overhaul
such as pumps and valves, mixing units, roofs and instrumentation are preprogrammed
into the model. The model selects the higher value between the repair and replacement
costs baseline estimation and the speciﬁc repair/replacement costs directly entered by the
user. In the module, this is referred to the “adjusted digester repair and replacement cost.”
Since the baseline costs are estimated based on the amount spent for the ﬁrst ﬁve years,
they do not completely capture the cost ﬂuctuations associated with large

repairs/replacements occurring later on in the project period (e. g., year 10). In some

cases, they may exceed the originally estimated amount. By selecting the higher value of

54

the two, the model is not only accounting for the steady growth of O&M costs over time,
but also incorporating the signiﬁcant cost increases associated with large overhauls and

major repairs in a given year.

Table 6. O&M Costs with Late Project Period Repair and Replacement

 

Year 6 7 8 9 10

 

Repair and Replacement Costs: $46,350 $47,741 $49,173 $50,648 $52,167

 

 

Adj. Repair and Replacement $46,350 $47,741 $79,000 $50,648 $100,000
Costs:

 

 

 

 

 

 

 

The adjusted cost of $79,000 in year 8 is for roof repairs and the $100,000 in year 10 are
for repairing pumps and valves. In both cases, these values exceed the amount originally

estimated from the ﬁrst ﬁve years of costs (Table 6).

These repair and replacement costs may be so costly that they exceed the available net
working capital and need to be externally ﬁnanced. In this case, a loan amortization
schedule is also calculated in the ﬁnancing module. In addition, if the purchase includes
. a depreciable asset, then the Modiﬁed Accelerated Cost Recovery System (MACRS) is

applied according to the proper project life outlined in the IRS Publication 946.

3.2.2 Egine-Generator
Labor costs were calculated the same as with the digester. The cost of oil, however, is
Speciﬁc to the engine-generator and was calculated using the following equation.

OYC = OGallons x O N X 0C / G (Equation 7)

OYC: Yearly engine generator oil costs ($)
OGallons: Oil required (gallons)

55

O C / G : Cost per gallon ($/gallon)

O N : Number of times oil is changed (times/yr)

For EF.‘ $462 = 22 gallons x 7 times/yr x $3 per gallon
The frequency of oil changes can be directly entered by the user to account for varying

concentrations of hydrogen sulﬁde in the biogas.

With repairs and replacements, the costs were estimated the same as with the digester.
The sum of the annual costs for the ﬁrst 5 years is determined and weighted for the ﬂow
of expenditures. After year 5, the costs were increased at 3% each year until the end of
the project period. Incidental repair, replacement and overhaul costs can also be directly
entered and are depreciated in the same manner as with the digester. If the user is
analyzing a system in the absence of Speciﬁc cost information, the 0&M costs for both
the digester and engine-generator can also be calculated using a percentage of total
capital costs. Based upon estimates from the literature, this percentage was estimated to
be between 3% and 7% (Beddoes et al., 2007; Bracmort, 2008). Due to this wide
variability, an average value of 5% was spread out over the ﬁrst ﬁve years and weighted

progressively higher in years four and ﬁve (Table 7).

Table 7. Cost Weight Factor for O&M Costs Based Upon Percentage of Total
Capital Costs

 

Year 1 2 3 4 5

 

 

 

 

 

 

 

Cost Weight F actor: 45% 55% 65% 80% 100%

 

 

The method of weighting the percentage of costs for the ﬁrst ﬁve years was to account
for the fact that digester and engine generator repair costs are likely to be minimal in the

ﬁrst several years of operation. After year ﬁve, however, the costs are increased at a rate

56

of 3% until the end of the project period. By year 15, the 0&M costs reach nearly seven
percent of total capital costs which captures a balance between the range of values found
in the literature. A steady growth of the O&M cost percentage will also help account for
the large repair and replacement costs of the digester and engine-generator as they age.
The total 0&M costs for years 1 through 5 are calculated for the 1,000 cow example

digester in Table 8.

Table 8. Operation and Maintenance Costs as a Percentage of Total Capital Costs

Year 1 2 3 4 5
% of Total Capital Costs: 2.25% 2.75% 3.25% 4.00% 5.00%
Total 0&M Cost: $18,700 $22,855 $27,011 $33,244 $41,555

 

 

 

 

 

 

 

 

 

 

 

3.3 Depreciation
The depreciation schedule is calculated using the (MACRS) 150% Declining Balance

Method (Half-Year Convention). The recovery periods used for digester components
were assumed to be either 15 years (single purpose livestock structure), 20 years (farm
building) or 7 years (Farm Machinery and Equipment). The corresponding depreciation
schedules were taken from the IRS Publication 946. The IRS Section 179 direct

expensing option, however, was not considered in this model.

3.4 Proper_ty Taxes

In Michigan, on-farm anaerobic digester facilities (including the engine-generator) can be
exempt from real and personal property taxes. In order to be eligible for the exemption,
methane digester equipment must be certiﬁed by the Michigan Department of Agriculture
(MDA) and the farm must be veriﬁed as compliant under the Michigan Agriculture

Environmental Assurance Program (MAEAP). In addition, the facility owner must allow

57

"access for not more than 2 universities to collect information regarding the effectiveness
of the methane digester and the methane digester electric generating system in generating

electricity and processing animal waste and production area waste” (DSIRE, 2009).

Currently, the Michigan Department of Treasury (MDT) has not dealt with the issue of
how to properly value an anaerobic digester system for property tax purposes.8 The
assumption was made that the six operating digesters in the state must be taking
advantage of the tax exemption. Since little information was available in this area,
certain assumptions regarding the taxable project cost and fair market value of a digester
were used. For example, all ﬁxed structures were considered to be real property (taxable)
and the value of 25% (fair market value as percent of total taxable project cost) was a
“best guess” estimation. Since the MDT has indicated that no farms are currently paying

property taxes on their digester systems, none of the analyses in this study include this

expense.

FMV .-. (PFMV x TC) For EF: $207,774 = (0.25 x $831,097) (Equation 8)
A V = (FMV x P A V) For EF: $103,887 = ($207, 774 x 0.50) (Equation 9)
PT = (A V x TR) For EF: 8311 7 = ($103,887 x 0. 03) (Equation 10)

TC : Total taxable project cost
PFMV : Fair market value as percentage of total taxable project cost

P A V : Assessed value as a percentage of fair market value

A V : Assessed value

F M V : Fair market value
PT : Property tax

TR : Tax rate

 

8 Conversations with Michigan Department of Treasury

58

3.5 Biogas Production

Biogas production is estimated on a month to month basis over a period of one year. This
allows aspects of seasonality (e. g., heat loss, manure freezing) to be accounted for which
gives a more accurate view of digester performance. The biogas production modules of
the model consist of three main elements: inﬂuent ﬂow, digester heating and biogas and
electricity production. The model is also set up to allow for the use of baseline or directly
entered values when calculating monthly biogas production. For situations where a
digester is not installed or daily inﬂuent values are unknown, biogas production can be
calculated utilizing baseline values. Alternatively, if a pre—existing system is being
evaluated, the user can enter more Site Speciﬁc data. In addition, hypothetical or existing
situations with the co-digestion of additional feedstocks can also be modeled on a

monthly basis.

3.5.1 Inﬂuent Flow
Inﬂuent ﬂow is deﬁned as the amount of material (e. g., water, manure, food processing
water) which enters the digester per period. It is often measured by the daily ﬂow rate

and has implications on digester Sizing, heating needs and biogas production.

3.5.1A Manure
In order to calculate the amount of manure inﬂuent entering the digester on a monthly

basis, the following steps (Equations 11-14) are used based upon the daily ﬂow rate. This

59

rate includes the manure and any added dilution water from the milking parlor or other
sources entering the digester.
Step (1)

QD =QC/DXH (Equationll)

Q D : Daily ﬂow rate (gallons per day)

. 9
QC / D . Flow rate per cow/day (gallons)

H : Herd Size (lactating and dry)

For EF: 29, 950 gpd = 29. 95 gallons x 1, 000 lactating cows

Step (2)

V = Q D x LG (Equation 12)

VD: Daily volume (ft3/day)

LG : Liquid gallons per ft3
For EF: 4, 004 ﬁ3/day = 29,950 gpd x 0. 1 33 680556 gallons per ft3

Step (3)

MD = VD x MD (Equatron 13)

MD: Daily mass (lb/d)

MD: Manure density (lb/ft3) [0

For EF.‘ 249, 937 Ib/d = 4, 004 ft3/day x 62.4

Step (4)

DMT = [HLC x DMLC/D]+[HDC x DMDC/D] (Equatron l4)

 

9 The most common values are 10 to 30 gallons of fresh water per milk cow (Burke, 2001).
‘° Manure Density= 62.4, (ASAE, 2005)

60

DM T : Total dry matter (lb/d)

H L C : Number of lactatrng cows

DM L C / D : Dry matter per lactatrng cow/day

H D C : Number of dry cows

DM D C / D : Dry matter per dry cow/day

For EF.‘ 20, 000 lb/day = {1, 000 Lactating Cows x 20 lb/day/cow} + {0' 1 Dry Cows x 11

lb/day/cow}

The amount of manure per animal varies by animal type and production grouping. For the
1,000 cow example, the assumption is that all dry cows are kept in a separate barn and do
not contribute manure to the digester. Table 9 contains a listing of manure characteristics
from the American Society for Biological Engineers (ASAE, 2005). In addition to the
animal manure produced, dry matter also includes organic animal bedding incorporated

in the inﬂuent stream.

Table 9. Manure Characteristics from the American Society for Biological
Engineers

 

 

 

 

 

 

 

 

Animal Type and Production Grouping Total Solids (Dry Matter)
(lbs/day/cow)

Beef-Cow (Conﬁnement) 15
Beef-Growing Calf (Conﬁnement) 6
Dairy-Lactating Cow 20

Dairy-Dry Cow ll
Swine-Gestating Sow (440 lb) 1.1
Swine-Lactating Sow (423 lb) 2.5

 

 

 

Once the daily dry matter per period has been determined, the percent of total solids (TS)

is calculated by dividing the dry matter by the daily mass (lb/d) (Equation 15). The

 

u The assumption is made that dry cows are kept in a separate barn and are not contributing manure to the
digester.

61

percentage of total solids concentration is important to monitor since it affects the heating
needs of the digester and the type of digester design which is selected. It is not a constant
due to water spill, humidity and the type of manure handling (Gebremedhin, 2006). In

contrast to the dry matter which consists of only solid material, daily mass includes solids

as well as added dilution water from the milking parlor or other sources entering the

digester.

TS C = DM T / MD (Equatron 15)
TS C : Total solids concentration (%)
DM T : Total dry matter (lb/d)
MD: Daily mass (lb/d)

For CF: 8% = 20, 000 lb/d/249, 937 lb/d

From the amount of collectable total solids (dry matter), the actual volatile solids content
is calculated using baseline values from both the dry and lactating dairy cows. The
volatile solids (V 8) content determines the amount of degradable solids which can
produce biogas from manure or any other feedstock suitable for anaerobic digestion
(Equation 16). Lost solids as a result of biodegradation during the pretreatment process
are also accounted for in the equation, Since it can have a signiﬁcant impact on biogas
production (Equation 17).

M VS = DM T x VS C (Equation 16)

M VS : Volatile solids mass (lb/d)

VS C : Volatile solids concentration (%)/2

 

‘2 vs (%) = 35%, (Steffen et al., 1998)

62

For CF: 17,0001b/d = 20, 000 lb/d x 0.85

M M (Equation 17)

VS(i) = VS(i) ’ (1 ' LVS)

M VS (1.) : Volatile solids mass (lb/perrod I)

LVS: Volatile solids loss (%)

1': Period of time (c. g. hour, day, month)

For CF: 1 7, 000 lb/d = [1 7, 000 lb/d — (1 - 0%”)]

3.6 Utilization Analysis

To calculate the biogas yield from either manure or additional feedstocks, Equation 18 is
used. The result is multiplied by the biogas methane concentration to obtain the energy
potential (Equation 19). The biogas production levels were assumed to be constant for the
15 year project period.

YB“) = M VS (i) x BVS. (Equatron l8)

. . . 3 . .
Y B (1.) . Brogas yreld (ft / period 1)
M VS (i) : Volatile solids entering the digester (lb/period i)
BVS: Biogas produced per (ft3/lb of VS destroyed)

For EF: 73,100ft3/d = 17, 000 lb/dx 4. 3 ft3/Ib”

EADQ’) : YB(i) X MC (Equation 19)

E A D : Energy produced by the system (Btu/period i)

 

‘3 The amount of solids lost depends on the system and a value of 0% was chosen in this example.
'4 This value is an average taken from various literature sources (Steffen et al., 1998; Bracmort et al.,
2008)

63

 

F 01

I'-”

116

C0

Iir

C

F 01‘

M C : Methane concentration (%)

For EF: 43,860 = 73, 100 ftde 60%l5

3.6.1 Digester Heating

In order to account for the amount of heat leaving the digester, the model was
programmed to either calculate heat loss values based upon the Speciﬁc dimensions of the
digester tank and construction material or assume heat loss as a percentage of energy
potential. Regardless of the method, the ﬁrst step iS to calculate the amount of heat
needed to warm the inﬂuent (manure and feedstock) to the target temperature. For
complete mix digesters operating in the mesophillic range, this is generally between 95°F
and 105°F (Lusk, 1998). The same formula is applied to all inﬂuent entering the digester

(including additional feedstocks) (Equation 20).

Qi = me(To _ Ti) (Equation 20)

Q1. : Energy needed to heat the digester to optimal temperature (Btu for period i)

m :Mass flow rate (lbs/period i)
To :Efﬂuent temperature which is equal to the digester temperature (°F)

Ti :Inﬂuent temperature for period i (°F )16

Cp: Speciﬁc heat of feedstock (assumed to be equal to that of water, 1 Btu/lb)

i : Period of analysis (e.g. January)

For EF: 759,183 Btu/hr = 10, 414 lb/hr x 1 Btu/lb x (95 °F — 22.1 °F’7

 

'5 55-65% (us. EPA, 2002)
'6 Assumed to be ambient temperature
'7 Average ambient temperature for the month of January in central Michigan (US. EPA, Undated)

64

When using Speciﬁc digester dimensions to calculate heat loss through the ﬂoor, walls
and roof, Equation 21 was used (Persson et al., 1979). The thermal conductivity
coefﬁcient varies depending on the construction material. Increased insulation will
decrease this value. When calculating surface area, a distinction is made between the
portion of the wall buried in the soil and that which is exposed to ambient temperatures.
The soil can provide some insulation in the winter months with temperatures assumed to
be 55°F year-round at a depth of Six feet (NREL, 2009). The depth of digester (if buried
at all) will be input by the user. Ambient temperatures from the FarmWare 3.1 simulator

were used which obtains its data from the National Climate Data Center.

17
QH= 2

U .A .(t. —t(,) (Equation 21)
j—l J J 1

Q H : Digester Heat Loss (Btu/hr)

U : Thermal conductivity coefﬁcient
A : Surface Area (ﬁz)
ti : Inside temperature (°F)

to : Outside temperature (°F)

i : Period of analysis

j : Type of surface being insulated
If digester heat loss is not being calculated based upon speciﬁc construction materials and
surface area dimensions, it can also be determined based upon a percentage of the
digester energy potential (Btu/hr) (Equation 22). A value of 5% was taken from the

literature (Liu et al., 2008). Since speciﬁc digester tank data from a 1,000 cow dairy was

not available, the analysis in this study uses the 5% estimate to determine heat loss.

65

 

F0.

The
less
exha
Waist
recor
digest

Upon n

QH = EAD x 5% (Equation 22)

Q H : Digester Heat Loss (Btu/hr)
E A D : Energy produced by the system (Btu/hr)

For EF.‘ 91,375 Btu/hr = 1,827,500 Btu/hrx 5%

The minimal energy needed for static warmth (E ,Btu/hr) then becomes the heat

AD, Min

needed to warm the inﬂuent to the target temperature plus digester heat loss (Btu/hr)

(Equation 23).

EAD, Min = mCP(To ‘ Ti) + Q x EAD (Equation 23)
E A D, Min : Mrnrmum heat needed for static warmth (Btu/hr)
E A D : Energy produced by the system (Btu/hr)

Q: Heat loss (%)

For EF: 850,558 Btu/hr = [10,414 lbs/hr x 1 Btu/lb x (95 °F — 22.1°F)] + (5% x
1,82 7,500 Btu/hr)

The energy required from the boiler is the minimum energy required for static warmth
less the energy captured from waste heat (Equation 24). Waste heat comes from both the
exhaust of the engine-generator and the water-to-manure heat exchanger (if one is used).
Waste heat captured from engine-generator is determined by multiplying the heat
recovery efﬁciency]8 of the heat exchangers by the energy potential (Btu/period) of the
digester. The amount of heat recovered from the water-to-manure heat exchanger is based

upon manufacturing speciﬁcations and must be input directly.

W

EB = (EAD, Min — EG _ WManure ) / BEF (Equatron 24)

 

‘3 The industry standard of 40% was assumed in the model

66

E B : Energy required from boiler (Btu/hr)19

W E G : Waste heat captured from exhaust off the engine-generator (Btu/hr)

: Waste heat captured from the water-to-manure heat exchanger

(Btu/hr)

B E F : The efﬁciency of the boiler (%)

Manure

For EF: 299,302 Btu/hr = [850,558 Btu/hr — 611,116 Btu/hr - 02” Btu/hr]/80%

The net energy potential of the system is the energy produced by the digester reduced by
the energy required to run the boiler (Equation 25).

ENet = EAD _ EB (Equation 25)

E Ne t : Net energy potential of the system (Btu/hr)

For EF: 1,827,500 Btu/hr = 1,827,500 Btu/hr - 299,302 Btu/hr

3.6.2 Co-digestion
The amount of additional feedstock inﬂuent is accounted for in a separate module. The
co-digestion module accounts for the speciﬁc characteristics of each feedstock and then

combines them with the biogas production from the manure.

3. 6. 2A Feedstock Cost

The ﬁnal cost of feedstock includes the cost of the feedstock itself, fuel, labor and
disposal. Unless the amount of feedstock added is signiﬁcant (e. g., 50% of total volume),
the cost of disposal is not expected to be different than what is normally Spent on the

disposal of manure without added feedstock. The user may enter a disposal cost ﬁgure if

 

'9 Based upon the month of January
2° This example does not include a water-to-manure heat exchanger.

67

it is considered appropriate. The transportation fuel cost is inﬂuenced by the distance
traveled, Speed, vehicle fuel efﬁciency and the cost of ﬁrel. The transportation labor cost
is inﬂuenced by the duration of the trip and the hourly pay rate of the worker as described
in Equation 25.

CT = TD + T1. + TF + (FQ + CU) (Equation 25)

F : Feedstock quantity (tons)

IO

C : Cost per unit ($)

Q

T : Transportation fuel costs ($)

"1':

T : Transportation labor costs ($)

1‘

T : Transportation cost for disposal ($)

CT: Total feedstock cost per truck load ($/load)

D

For EF.‘ $129=$0+$61 +$33+ (7 tons x $5.00)

3. 6. 28 Feedstock Revenue

The cost of transportation is not considered in this equation, since the farm generally will
not incur the delivery cost when tipping fees are involved. Tipping fees are a payment to
the farm from an outside entity (e.g., restaurant, food processor) for the ability to dispose
of their organic waste in the digester. The feedstock revenue is the amount of feedstock
delivered to the farm multiplied by the revenue per unit (Equation 26). The disposal cost
is considered at the discretion of the user and will depend on the amount of feedstock

added.

R = F R —
T ( x U) TD (Equation 26)

RU: Revenue per unit ($)

RT: Total feedstock revenue ($)

68

For EF.’ $400 = (10 tons x $40 per ton) - $0

3. 6.2C Amount of Feedstock Entering the Digestergper Day

It is important to determine the amount of feedstock entering the digester in order to
accurately determine the energy potential and, digester heating requirements. In addition,
it is convenient in planning a digester to consider the amount of additional feedstock
added when determining the digester tank size. The average percentage of feedstock
added each day will depend on the management practices of the digester operator and it
varies between systems. The model assumed that a constant percentage of each truck load

was fed to the digester each day. In practice, however, this percentage will vary widely.

3. 6. 2D Feedstock Characteristics and Biogas Yield

Due to the variability among different feedstocks, the energy potential of each must be
calculated separately. Some agro-industrial wastes may contain less than 1% total solids
(TS), while others contain high TS contents of more than 20% (Steffen et al., 1998).
There is also wide variability in the content of V8 and resulting conversion to methane.

Equations 27-30 Show an example of adding 7 tons of ethanol syrup three times per

month throughout the year”.
Mom) = FAD(,°) x ”or For EF: 3,1501b/d = 42, 000 lb/dx 15% (Eq. 27)
VSF(,~) = Mot-(t) X VSCF For EF: 2, 698 lb/d = 3,150 lb/dx 85.66% (Eq. 28)

YBF(1') = VSFU) X YBF For EF: 40,474 ftj/d = 2, 698 [Mix 15 ft3/lb of VS (Eq. 29)

 

2' Values pertaining the characteristics of ethanol syrup taken from (Rosentrater et al., 2006).

69

Era) = YBF(i) x M CF For EF: 29,951 ft3/d = 40,474 ftde 74% (Eq. 30)

M D F (1.): Feedstock mass (lb/period I)

F A D (i) : Feedstock added to the d1gester(lb/perrod 1)

TS : Total solids concentration in feedstock (%)

CF
VS F (i) : Volatile solids (lb/period 1)
VS C F : Volatile solids concentration in feedstock (%)
YBFU)‘
Y B F : Biogas Yield (ft3/lb of VS F destroyed)
E F( .) : Energy produced from added feedstock (ft3/period)
r

M C F : Methane concentration of feedstock (%)

Biogas Yield per period (ft3/period)

3.6.3 Energy Uses

The actual energy potential sent to the engine-generator must be not of any biogas that is
used to offset the use of propane. Offsets achieved using waste heat from the engine
generator (net of digester heating needs) will not have an effect on the actual energy
potential. The user has the Option whether to include these offsets and which energy

source will be used (biogas or waste heat).

In addition, the actual energy potential (E A ) is adjusted for when the methane produced

exceeds the capacity of the engine-generator (Equation 31). The excess methane is then

sent to a ﬂare where it is burned. In this circumstance, if E A exceeds the rated capacity of

the engine-generator, the model will use the rated capacity as the adjusted actual energy

potential.

70

A Net P (Equation 31)

E A : Actual energy potential net of all other uses (Btu/period)

E Ne t : Energy potential net of digester heating requirements (Btu/period)

E P : Energy potential from biogas used as a propane offset (Btu/period)

For EF: 1,82 7,500 Btu/hr = 1,82 7,500 Btu/hr — 0 Btu/hr”

3. 6. 3A Propane Ofﬁsets

The user has the option whether to include propane offsets as a potential energy use. The
energy required to replace propane can come either from waste heat or from biogas net of
any boiler use to warm the digester. Since farm heating needs are seasonal, the model

also gives the option to select which months are included.

Assuming a value of 92,000 Btu’s per gallon, the energy available from the digester is

converted to a gallon equivalent in Equations 32-33. The examples provided are for the

month of January.
PP = E Ne t /Btu’s per gallon (Equation 32)

1.23 gallons/hr = 113,351 Btu/hr / 92, 000 Btu/gallon

01'

PP = Ewaste /Btu’s per gallon (Equation 33)

0 gallons/hr = 0 Btu/hr / 92, 000 Btu/gallon

P P : Propane offset potential (gallons/period)
E Ne t : Energy potential net of digester heating requirements (Btu/period)

 

22 -
Thrs example assumes no propane offsets.

71

waste: Energy potential from waste heat (Btu/period)

To estimate the on-farrn propane use, 11 gallons per cow/year was assumed (Lazarus,
2003). This is based upon the heating needs of the milking parlor and cow holding area.
The actual on-farm propane offset is either the on-farrn need for the selected time period.
(based upon 11 gallons per cow/year) or the propane offset potential produced by the
digester. For example, if the pr0pane offset potential from the digester is less than the
farm’s need, the actual propane offset becomes the energy potential produced from the
digester and vice versa. The example provided is for the month of January and assumes
waste heat as the energy source (Equation 34).

P = min(E

AP P ’ PFarm) (Equation 34)

P A P : Actual propane offset (gallons/period)

P Farm :On-farm propane need (gallons/period)

E P : Energy potential available (per period) for propane offsets (either E Ne t or
waste )

For EF.‘ 0 gallons/month: min (0 gallons/month, 91 7 gallons/month)

It is important to note that excess waste heat may not always be available during the
winter months when it is needed most. This is due to the fact that the majority of the

waste heat will be used for digester heating requirements.

3. 6. 3B Electriciﬂ Generation

Electricity generation (E G ) is calculated by multiplying the actual energy potential by

the recovery and engine efﬁciency of the generator (Equation 35). The generator is

72

assumed to be an internal combustion engine. The E value is also adjusted for the

G

parasitic energy load of the tank mixers.

EG = [(EA x GRE x GEE) x (1 — P)]/3,412 Btu/kWh (Equation 35)
E G : Electricity generation (kWh/hr)
E A : Actual energy potential net of all other uses (Btu/period)
G R E : Online time of the engine generator (%)
G E E : Engine efﬁciency of the engine generator (%)

P: Parasitic energy requirement (% of EC)

For EF.°138 kWh/hr = [(1,528,198 Btu/hrx 90% x 0.35) x (1-0.02)]/3,412 Btu/kWh

The parasitic energy requirement (P) refers to the amount of energy needed to power the
tank mixers. From case farm data, this value was calculated to be 2% and is used as a
constant in the model.

To arrive at this value, Equation 36 was used:

P=[(MERxMN)xMHr]/EG (Equation 36)

M E R : Energy requirement per mixer (kW)

MN : Number of mixers

M Hr : Number of hours run per day (hrs)

3.7 Electricig Purchase Agreements

Each sub-module has its own corresponding set of inputs (prices, proﬁt retention, meters
included and sales charges) which can be altered for scenario comparison purposes.
Additionally, with all three agreements, the model is programmed for the user to select

which electricity meters are to be included in the analysis.

73

3.7.1 Surplus Sale

Since a surplus sale agreement involves the offsetting of on-farm energy needs, a power
usage index23 is used which speciﬁes the amount of electricity used per hour for each
month. This is necessary because a farm can only offset the amount of energy they are
using at any given moment. Once power usage ﬂows are established, the amount of I
electricity that is available for sale to the utility can be determined. The example provided

in Equation 37 is for 7 am in the month of January.

K =U><U

UP P A (Equation 37)
U A : Annual electricity usage (kWh)
U P : Percentage of energy used per hour per month (%lhr/month)
K UP : kWh used per period (kWh)

For EF: 5,125 kWh = 0.4545% x 833,288 kWh

If the electricity generated exceeds the farm’s need for a particular hour, then electricity
is sold back to the utility company. Likewise, if a farm is at a deﬁcit for a given hour, the

needed electricity must be purchased.

Under a surplus sale agreement, on-farm electricity offset by the system is valued at the
commercial retail rate.24 The commercial rate used in the model was obtained from
utility bills at the Michigan case farm. The electricity pricing for the 15 year project

period was then determined by creating an index from the DOE Energy Information

 

23 The index was taken from a representative dairy farm different from the case farm used in the calibration
pf the default value investment module.
The commercial rate is used as a constant for all three agreements for a given year.

74

 

 

It?

de

01‘

ele

St

For

Administration. In Equation 38, the kilowatt hours offset by the system are multiplied by
the commercial rate and totaled for the month.

n

Voﬂset = [E10010 ﬂs etc) x Emmi!) (Equation 38)

Value of on-farm electricity offset ($/kWh)
Hourly kWh offset by the system (kWh)

Vojfet

KHoﬁCsetU)

Ere t il' The commercial retail rate for electricity ($/kWh)

i . Period of time (e. g., hour, month)

For EF.‘ $374 = 3, 787 kth $0988 kWh

In contrast, any electricity that is sold back to the utility is valued at the 2009 average
real-time Locational Marginal Price (LMP) of the particular utility’s load node as
determined by the Midwest Independent Transmission System Operator (MISO). These
prices are forecasts used in planning activities involving Michigan’s PA 295 and energy
optimization. The hourly values are then totaled up to arrive at monthly and yearly

electricity sales and savings.

V=sold [E] (KH excess(i) x ELMP) (Equation 39)

Vso l d : Value of on-farm electricity sold ($/kWh)

ex cess (r) Hourly kWh produced in excess of on-farm electricity need (kWh)

E L MP . The average monthly real-time locational marginal price (LMP) ($/kWh)

For EF: $22 = 499 kWh x $0. 04392 kWh

75

 

SUI

be

isx

eng

311‘

DJ
\J

Ind

calci

$36
: 1C0]
" E5111

The electricity produced is also eligible to be sold for Renewable Energy Credits (RECS)
and is separate from the surplus sale electricity purchase agreement. Equation 40 Shows

the total value of RECS earned for the month of January. One REC is worth 1,000 kWh.

VREC = i §£(KHG(1') /1,000kWh) x PREC] (Equation 40)
$2, 725 = (102,864 kWh / 1, 000) x $265025

V R E C .' Value of a renewable energy credit ($)

KHGQ') .' Hourly kWh generated by the system (kWh)

PR E C .' Price of a renewable energy credit ($/MW)

In addition to the revenues associated with electricity production, several selling expenses
such as a monthly administrative expense and the cost of an additional phone line must
be included. The administrative expense is valued at $0.0010 per kWh26of electricity sold
and the phone line is assumed to be a monthly ﬁxed rate of $30.27 The phone line charge
is variable, however, and can be much higher depending of the nameplate capacity of the
engine-generator and rules of the utility company. A system access charge of $100 also

applies with an engine-generator nameplate capacity over 100 kW.

3.7.2 Buy-All Sell-All

Under a buy-all sell-all agreement, no on-farm electricity use is offset by the system. The

calculation of electricity sales is equal to the number of kWh’s generated per month

¥

:2 $26.50 was obtained from the Michigan case farm
(Consumers Energy, 2009)
Estimate from Conversations with Consumers Energy

76

 

3‘;
5

usag

IIOIII

multiplied by the average real-time LMP rate per kWh (Equation 41). The calculation is
the same as under the surplus sale agreement.

11

Vgenerated = i :21 (KH

G(i) X ELMP )) (Equation 41)

Vgenerated .' Value of electricrty generated ($)

For EF: $4,518 =102,864 kWh x $0. 04392 kWh

The farm, however, must purchase all electricity needs from the utility company at the

full commercial retail value.

The formulas regarding the sale of RECS are the same as under the surplus sale
agreement. The selling expenses associated with the sale of energy (administrative
charge, phone line charges) are also calculated the same way as under the surplus sale
agreement. In addition, the farm must pay a system access charge of $100 per month to

sell back to the utility.

3.7.3 Net Metering

Under a net metering agreement, the farm uses electricity produced by the digester to
offset on-farm usage. Excess electricity is then credited to next month’s bill. In order to
determine the amount of energy credited and used to offset purchases, the same power
usage ﬂow index is used as under the surplus sale agreement (Equation 42). In contrast

however, excess energy on an hourly basis is credited instead of sold and is valued at the

77

 

 

n3

power supply component (see Section 12) of the customer’s electricity bill (Equation 43).

n

Voffset = i §1(KHoﬂset(I-) x Eretail) (Equation 42)

V0 ﬁ’S e t : Value of on-farm electric1ty offset ($/kWh)
KHoffsetU) :

. : The commercial retail rate for electricity ($/kWh)
retail

i: Period of time (e.g., hour, month)

Hourly kWh offset by the system (kWh)

For EF: $374 = 3,787 kWh x $. 0988 kWh

n

Vcredit = i §1(KHexceSS(i) x EPSC )

(Equation 43)

Vere dit .' Value of hourly electrrcrty credit ($)

. : Hourly kWh’S produced in excess of the on-farm electricity need
excess(r)

E P S C : The power supply component of the customer’s bill ($/kWh)

For EF: 8251 = (3,787 kWh x $0. 06615 kWh”)

All formulas regarding RECS, proﬁt retention and selling expenses are the same as under

the surplus sale agreement.

3.8 Utility Meter Bills

A key element of the model is the ability to examine the effect of various electricity
agreements based upon actual utility bills provided by the farm. The rates and tariffs will

vary depending on the utility company and will need to be entered at the beginning of the

 

28 Value taken from utility bills at case farm

78

 

Se

C0}

act

util

Ser

29 M

analysis. For this study, the rates and tariffs used under a demand rate schedule typical
for medium to large dairies are used in the model. The model estimates demand by using
actual data from the Michigan case farm and scaling the values to the needs of the farm

size analyzed.

Under a demand rate schedule, the farm is charged for their peak demand usage in
addition to monthly kWh charges. The charge per kWh (e.g., $0.056), however, is
considerably lower than the typical commercial rate (e.g., $0.098)”. If a farm can
maintain low peak demand, they will end up actually paying less than the typical
commercial rate depending on the meter. Due to the fact that farms often have multiple
meters with varying levels of peak demand, however, the rate charged per meter will
vary. In order to determine a single rate to use in the model, the rates charged for
electricity usage were averaged together. To calculate electricity charges for the entire

year, the user must enter the historical usage information included on the monthly bill.

3.9 Standby Charges

The Speciﬁcs of standby service will vary depending on the utility company, but
generally are composed of two parts, Power Supply and Delivery Standby Charges (see
Section 2.9.1). The rates used in the research are taken from a large representative utility
company in Michigan and should be considered as a close approximation of what the
actual charges would be. Information regarding the speciﬁc charges can be located in the
utility company’s rates and tariffs book and are available from the Michigan Public

Service Commission (MPSC) or the company’s website.

 

’9 Michigan case farm utility bill

79

 

|';)

Ft

\i‘ii

loll

this

SC

3.9.1 Power Supply Standby Charges

Power supply standby charges ﬂuctuate month to month reﬂecting the changing MISO
costs from variations in electricity demand. In this model, the average monthly LMP was
used in order to increase the ﬂexibility of the model.30 Equation 44 summarizes the
power and delivery standby charges for one meter in the month of January.

SC KH x E (Equation 44)

PS(m) = C(m) LMP(m)

SC PS (m) : Monthly power supply standby charges ($)

KH C (m) : Monthly on-farm electricity consumption (kWh)

E L MP : Monthly location marginal price for electrrcrty

For EF.‘ $398 = 9,063 kWh x $004392

3.9.2 Delivery Standby Charges
Similar to charges for power supply, delivery standby charges only apply to customers
with engine-generator nameplate capacities greater than 100 kW, as in Equation 45. The

following rates are taken from a Michigan utility company and are used in all analyses in

this study.
SCD(m) = (MSD(m) x Ccapacity) + (KHC (m) x Cdistribution) (Equation 45)
SC D (m) .' Delivery standby charges ($)
capacity .° Capacity charge per kw of max standby demand ($/kW)

MSD(m) : Maximum standby demand (kW)

C dis tri bu tion : Distribution charge per kWh of consumption ($)

m : Month of the year

 

3° (Consumers Energy, 2009)

80

For EF.‘ $65 = (27 ka $1.77) + (9,063 kth $0. 003009)

The total standby charges then become the sum of the power supply and distribution
charges.
SC

S (Equation 46)

SCT(m) = PS(m) + CD(m)

SC T (m) : Total standby charges (power supply and distribution)

For EF: $463 = $398 + $65

Where:

The Capacity Charge is a Michigan Public Service Commission(MPSC)-authorized
charge applicable to most nonresidential customers which recovers system costs for
transporting electricity from the transmission (high voltage) lines over the distribution
(lower voltage) lines to the customer's premises. A value of $1.77 was used in the

analysis (Consumers Energy, 2009).

Maximum Standby Demand is the highest 15-minute kW demand created during the
previous 11 months (“Historical Max” on the customer bill) minus the contracted demand
created during the current month (“Max Demand” on the customer bill) (Equation 47).
The max standby demand, however, cannot be less than 80% of the monthly max demand
(kW). This prevents the farm from avoiding standby charges for a given month in which
the historical maximum demand over the past 11 months is very close or equal to the

maximum demand.

81

 

'lh
(kl
chm
0pc

kVV

his

send
capat
Net)
lhanl

111fo I

 

MSD (Equation 47)

(m) = Wham) ‘ MXdemand(m)

MSD(m) : Maximum standby demand (kW)

MX his I (m) : Historical maxrmum demand (kW)

demand (m) : Maxrmum demand (kW)

The Distribution Charge is a MPSC-authorized charge based upon the electric energy
(kWh) used by the customer. This charge allows the utility to recover costs for delivering
electric energy from the transmission system to the customer's premises, including
operating and maintenance expenses of the distribution plant. A value of $0.003009 per

kWh was used in the analysis (Consumers Energy, 2009).

It is not always the case, however, that the customer must pay standby charges to receive
service. Under a typical surplus sale agreement, if the generator installed has a nameplate
capacity under 100 kW, then no standby charges are paid. Similarly, under the Michigan
Net Metering Law, no standby charges are paid if the generator nameplate capacity is less
than 150 kW. The model developed as part of this study is programmed to account for

these differences and can be altered depending on the rules of the individual utility.

3.10 Carbon Credits
Once it has been determined that the digester system is eligible to receive carbon credits,
there are two methods for calculating a methane emission baseline. The method yielding

the lowest value is used for the issuing of credits.

82

3.10.1 Methgne Combustion Method

Actual monitored amount of methane captured and destroyed by the project activity using
existing CCX monitoring protocols and a Global Warming Potential (GWP) for methane
of 21. The GWP for a particular greenhouse gas is deﬁned as the ratio of heat trapped by
one unit mass of the greenhouse gas to that of one unit mass of CO2 over a speciﬁed time
period.31 The ﬁnal amount of methane destroyed includes biogas that is either combusted
in an electric generator or ﬂare. The examples provided are for the month of January and
are shown in Equation 48.

CH4(R) = (BGenerator + BFlare ) x CH4(Average) (Equation 48)

. 3
CH 4 ( R) . Methane recovered (ft /year)

. - 3
B Genera tor . Biogas to the generator (ft /year)

. - 3
B Flare . Biogas to ﬂare (ft / year)

. O
CH 4 ( Average) . Average monthly methane (/o)

For EF: 15,426,108 (ft3/yr) = [23,139,162 (ﬁ3/yr) + 2,571,018(f13/ yr) ] x 60%

The monthly methane ﬂows are then summed on a yearly basis and converted to metric
tons per year (Mg/yr). Lastly, the methane combusted (Mg/yr) is converted to carbon
credits through a conversion factor established by CCX (Equations 49-50).

CH 4 ( C) = CH x [16 (molecular weight of CH4) ] x [1Mg/106 g] x [1mol/24. 04L

4(R)
@STP] x [28.32 L/Icf] (Equation 49)

CH 4 ( C) : Methane combusted (Mg/yr)

 

3'(us. EPAd, 2009)

83

 

For

(C

F 01'

J)

“'0.
par

lib

Cl

F01

III?

C 0.

4.23

3 (CC
" Carl

For EF: 291 (Mg/yr) = [15,426,108 (ft3/yr)] x [16 (molecular weight of CH4) ] x [1Mg/106
g] x [1mol/24. 04L @3717 x [28.32 L/Icf]32

CCC = CH 4 ( C) x 18. 25 (Mg/yr) 33 (Equation 50)

C C C : Carbon credits from combustion (credits/year)

For EF.‘ 5,311 Credits =29] (Mg/yr) x 18.25 (Mg/yr)

3.10.2 Ex-Ante Method

The methane emission calculated ex-ante based on the amount of the animal manure that

would decay anaerobically in the absence of the project activity. Values for the listed
parameters are quoted directly from the Chicago Climate Exchange (CCX) guideline
tables (CCXa, 2009). Equations 51-52 Show the calculation of the ex-ante method.
(Step 1)

CH x SSCF s

()XMS(T,S)XPd

4Manure = TZS N(T) x EF(T, S, St) ays

(Equation 51)

For EF: 301,673 Kg CHa/periodl= 1,500 Dairy Cows x .551 x1.00x1.00x 365 days
Where:

CH4Manure = CH4 emissions from manure management (kg CH4 per period!)

 

 

 

(Step 2)
CO = CH4Manure x GWPMethane
2Baseline 1,000 (Equation 52)
4 223 C / 201,115KgC1‘14 X 21
, M O r =
( g 2 y ) 1,000
3’ (CCXa, 2009)

Carbon credits per ton of methane combusted

84

 

Sptt
H I
mar
opti
SS(
(0111
grit
SIUC
rule
.115
mar
0 ll

esti

Nm= The number of animals in livestock species/category T. The possible livestock
species are listed in the CCX guidelines.

EF(T,S,St)= A methane emission factor for livestock where (T)=livestock category, (8)
manure management system, (St)=state. The baseline manure management system
options are anaerobic lagoon, pit storage below animal conﬁnement, liquid/Slurry.
SSCF(S) = A solids separation correction factor for manure management system S
(unitless fraction). For those systems that do not separate solids, or that utilize simple
gravity separation of sand or other non-manure solids, the SSCF is equal to 1. Since this
study does not include solids separation, a SSCF of 1 was used. CCX guidelines provide
rules for other manure management systems.

MSmg,= The fraction of livestock category T’s manure handled using manure
management system S (unitless fraction)

G WPMMW= The global warming potential of methane (kg CO2c to Mg'lCO2c). The CCX
estimates it to be 21.

Pam: Number of days in the reporting period (days)

1,000= Mass conversion factor (kg CO2e to Mg'lCO2c)

Table 10. Carbon Credit Summary

Tarbon Credits

Tombustion Method 5,31 1
ﬁreditS/year)

 

 

 

 

"ET-ante Method 4,223
3 credits/ year)

 

 

 

Lowest Value 4,223

 

 

 

85

Table 10 summarizes the number of carbon credits that would be earned using both
methods. The model is programmed to automatically select the lower of the two carbon
credit values. In the present example, the ex-ante method is the lowest value and
revenues would be based upon 4,223 carbon credits a year. Over the range of herd sizes
tested in this study, the ex-ante method consistently produced the lowest carbon credit

values.

3.11 Financing Options

The model is programmed to automatically calculate the loan amortization schedule for
all items ﬁnanced with debt. This includes any debt incurred in the initial investment as
well as incidental purchases associated with 0&M costs. The user must input the
speciﬁc annual percentage rate (APR), loan term and total loan principal. The model
calculates the total periodic loan payment (interest and principal) using the formula for
the present value of an annuity. The assmnption is made that the payments will be
yearly. Any loans made for O&M are programmed to begin in the year they are entered

from the O&M module. The user must enter the speciﬁcs of the major 0&M loans.

Grant payments are also included in this section and are handled by the model based
Upon the period of the grant and the number of payments per year. Any quarterly
payments are aggregated into a yearly payment using Equation 53. The example provided
is for the USDA Renewable Energy for American Program (REAP) which awards a

maximum of 25% of the total project cost.

GYP = (GA “GNP X GD» X GNP (Equation 53)

86

GYP : Yearly grant payments

G A : Total grant award

G N P : Number of grant payments

GD : Duration of grant

For EF.‘ $259,510 = ($259,510 / (3 payments x 1 year)) x 3 payments

In the example, the yearly grant payments are equal to the total award amount since the
grant is only for one year. Certain federal grants, however, may have a longer duration

which makes the formula necessary to calculate the yearly payments.

3.12 Capital Budgeting Analysis

The costs and revenues from the various modules are then connected to an income
statement assuming a 15 year project period. The required return to total capital was
assmned to be 10.0% and then adjusted for taxes (1- Tax Rate). This made the adjusted
after-tax return to total capital 6.66%. The weighted average cost of capital (WACC) was
not used in this model in order to account for the lower interest rates on debt available to
renewable energy projects receiving federal loan guarantees. Instead, interest was
included as an operating expense and principal payments were deducted from after-tax
cash ﬂows. All analyses in this study assume a USDA grant covering 25% of the total
investment cost and a loan guarantee covering 50%. The assumption is made that the loan
guarantee would allow a farm to obtain debt at a 6% annual rate. This was considered a

typical funding scenario for a Michigan digester.

87

In order to analyze the proﬁtability of the digester system the net present value (NPV),
internal rate of return (IR) and payback period were calculated. The tax rate was
calculated by aggregating all applicable tax rates in the state that would apply to a dairy

farm.

TR=(TFP +TS)+TS

E (Equation 54)

TR: Tax rate (%)

. O
T FP . Federal personal tax rate (At)

T S E : Self employment tax rate (%)

For EF: 33.45% = 15% + 4.35% + 14.1%

The model was then run on the 1,000 cow example dairy using NPV, IR and payback
period. Revenue streams include: electricity sales, RECS, carbon credits and propane
offsets (from waste heat). In addition, this scenario assumes only dairy manure as a

feedstock. All assumptions are listed in Table 11.

Table 11. Assumptions for 1,000 Cow Example

 

 

 

Influent
Herd Size 1,000
Daily Flow Rate (gpd) 29,950

 

 

Biogas Production
_Methane concentration (%) 6O
_Biogas Yield (ﬁ3/lb V8) 4.3

 

 

 

‘

ﬂectricity Generation
£nline Time (%) 90
Engine Efficiency (%) 35
Eggie Gengzytor Size (kW)
Surplus Sale and Buy-All Sell-All 160
Net Metering 105
Heat Recovery Efﬁciency (%) 4O
Parasitic Energy Requirement (%) 2

 

 

 

 

 

  
   
  

 

 

 

 

 

 

 

88

Table. 11 Assumptions for 1,000 Cow Example (Continued)

 

Digester Tank and Heating

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Heat Loss (%) 5

Total Solids (%) 8

Design Temp (°F) 95

Hydraulic Retention Time (days) 20

Boiler Efficiency (%) 8O

Pricing

Carbon credits ($/credit) 2

REC's ($/credit) 26.5

Propane gas34 ($/gallon) 2.31

Retail Electricity ($/kWh) 0.0988

Table Continued

Surplus Sale Buy-All Sell-All Net Metering

Simple Payback Period, years 15+ 15+ 15+

IRR on Equity N/A N/A -l7.83%
Net Present Value on Equity -$734,075 -$760,599 -$560,519
After Tax Required Return to Equity 6.66% 6.66% 6.66%
Tax Rate 33.45% 33.45% 33.45%
Total Initial Investment $1,038,040 $1,038,040 $909,893
Borrowed Capital for Initial Investment $519,020 $519,020 $454,946
Debt as Percentage of Initial Investment 50% 50% 50%
APR 6% 6% 6%
Grant Funds $259,510 $259,510 $227,473
Grants as Percentage of Investment 25% 25% 25%
Equity Invested Net of Grant Payments $259,510 $259,510 $227,473

 

 

 

 

In this example, the digester system produces negative returns on investment under all

three electricity purchase agreements. The internal rate of return is unable to be

calculated under the surplus sale and buy-all sell-all agreements due to negative after-tax
cash ﬂows later in the project period. The payback period exceeded the project period of
15 years for all three agreements. Also, note that a digester system under a net metering
agreement involves a lower initial investment cost. As explained in Chapter 2 Section

2.9.3, this is because a net metering agreement limits the size of the engine-generator to

g

34 All propane offsets in this study were achieved using waste heat.

89

 

match the electrical need of the farm. Even with a smaller engine-generator and lower
electricity production, net metering is the most favorable agreement at a herd size of

1,000 cows. This is because of policy beneﬁts involving less utility company charges
and a higher value for electricity produced by the digester. A comparison of the three

electricity purchase agreements will be covered in more detail in Chapter 4.

3.13 Verification of Model

The default value investment and baseline biogas production modules were veriﬁed
against several case studies which are publicly available through Cornell and
Pennsylvania State Universities. Speciﬁc values for the methane concentration, online
time, electrical conversion and heat recovery efﬁciency were not provided in the Study
and were taken from industry standards. The values were set as a constant in the model

for all three digester systems examined.

3.13.1 Test Farml

The Sheland Farms Digester in New York is a vertical complete mix design with manure
scraped from free stall barns. The farm installed a vertical complete mix digester as a
solution to both increasing electrical and purchased bedding costs. The digester

35 as well as pre-digested

processes 14,000 gallons per day of barn efﬂuent from 560 cows
solid-liquid separator liquid (SLS) efﬂuent. The system has a 125-kW Caterpillar engine
generator and was recorded as using an average 30 113/minute of biogas to generate 80 to

85 kW of power (Pronto and Gooch, 2009). Details are provided in Tables 12-14.

 

35 No breakdown given between lactating and dry cows

90

Table 12. Sheland Farms Inc. Case Study Inputs

 

 

 

 

Herd Size: 560
Design Temp, (°F): 100
Estimated Daily Flow Rate, (gpd): 14,000
Hydraulic Retention Time, (days): 17

 

 

Inﬂuent Composition

 

manure and solid-liquid separator liquid

 

 

 

 

 

 

 

 

efﬂuent

Table 13. Sheland Farms Inc. Case Study Outputs

Case Study Model
Tank Volume, (gallons) 238,000 262,000
Average Electricity Production, (kW) 80-85 92
Average Biogas Used, (ft3/minute) 30 28
T8 Content, (%) Not Given 9.6
Total Project Cost $1,199,717 $669,592

 

 

 

 

Table 14. Model Assumptions for Test Farm 1

 

 

 

 

 

Methane Concentration, (%) 60
Engine Efﬁciency, (%) 35
Online Time, (%) 90
Heat Recovery Efﬁciency, (%) 40

 

 

 

In terms of the model outputs, the electricity production predicted by the model is slightly

higher than the recorded values. This could be explained by either a lower engine

efﬁciency on the actual farm engine-generator or possible volatile solids loss from the

pre-digestion solid-liquid separator included in the system. The difference in predicted

tank size could be due to the amount of freeboard space added. The model assumes a

value of 10%.

One main discrepancy, however, is the difference in the total project costs. When

examining the budget provided in the case study, the engineering design is budgeted at

$200,000, whereas the model estimated it at $42,888. In addition, the biogas utilization

91

 

building in the case study cost an estimated $232,917, whereas the model estimates the
building to cost $15,000. Lastly, the case study includes $100,000 for miscellaneous costs
which include construction supplies and materials, employee travel and shipping charges
for equipment and materials. The default value module does not include a miscellaneous
category, but does allocate $26,805 for contingency costs to cover cost overruns and
delays. AS explained in Section 3.1, all investment costs calculated in this study are
benchmarked against the EPA AgSTAR cost curve. It is unclear whether this particular

digester was overpriced or if the model undervalued the investment.

3.13.2 Test Farm 2

The Penn England Digester in New York is a mixed loop, partially above ground tank
design. The manure collection system is connected to free stall barns which are scraped
continuously. The milk parlor and holding pen are ﬂushed 3 times per day and this ﬂush
water is also combined with the digester inﬂuent. The digester is fed with 18,200 gallons
of raw manure per day in batches and an additional 6,000 gallons of parlor water. After
10 months operation the production was recorded at between 110 and 140 kW, averaging

120 kW (Pennsylvania, 2007). Tables 15-17 provide details.

Table 15. Penn England Farm (Mixed Loop) Case Study Inputs

 

 

 

 

 

 

 

Herd Size ‘ 720 Lactating and 80 Dry
Design Temp,(°F) 100
Hydraulic Retention Time,(days) 20

mnﬂuent Composition manure and parlor water

 

92

Table 16. Penn England Farm (Mixed Loop) Case Study Outputs

 

 

 

 

 

 

 

Case Study Model
Average Electricity Production, (kW) 110-140 119
Daily Flow Rate, (gpd) 24,200 22,480
Tank Volume, (gallons) 525,000 494,560
TS Content, (%) 8 to 9 8.2
Total Project Cost $1,140,000 $810,444

 

 

 

 

Table 17. Model Assumptions for Test Farm 2

 

 

 

 

 

 

Model Assumptions

Methane Concentration, (%) 60
Engine Efﬁciency, (%) 35
Online Time, (%) 90
Heat Recovery Efﬁciency, (%) 40

 

 

 

The predicted electricity production is within the ranges listed in the case study. The tank
size estimated by the model is 5.8% less than the actual volume, but could also be

explained by a difference in freeboard Space added.

In terms of the total project costs, the values calculated by the model are relatively close
to this case study. The difference in cost is attributed to a power prime mover (CHP) for
$135,000 and biogas conditioning equipment for $50,000 which is not included in the

default value investment module.

93

Chapter 4: Results and Analysis

This chapter is separated into three main sections. Section one is a continuation of the
1,000 cow example described in Chapter 3 and is separated into two parts. The ﬁrst is a
series of sensitivity analyses which examine the price of retail electricity, the value of
excess electricity produced, carbon credits and RECS. The second is a scenario examining
the co-digestion of ethanol syrup to increase electricity production. Section two is

focused on demonstrating the model’s use as a tool for engineers seeking to optimize
and/or estimate digester performance. Key variables which have a Signiﬁcant impact on
performance were chosen for analysis including the total solids concentration, volatile
solids loss and operational online time. The ﬁnal section analyzes the digester return on
investment from each of the three electricity purchase agreements over a range of herd
sizes. In particular, speciﬁc aspects of Michigan’s net metering policy and standby
charges were analyzed in hypothetical situations. Recommendations were then made
based on the results in order to make current energy policies more favorable for anaerobic

digesters.

All sensitivity analyses and scenarios were tested using the default investment and
baseline biogas production modules of the model. In each scenario/analysis, the after-tax
net present value (N PV) of the digester investment was calculated and the results
compared under the three electricity purchase agreements (surplus sale, buy-all sell-all
and net metering). Although the model is also capable of measuring the return on
investment using the internal rate of return and payback period, the after-tax NPV was

best suited for the purposes of this study.

94

4.1 Section One - 1,000 cow Example

This section is separated into two parts and is based on the 1,000 cow example described
in Chapter 3. Part I is a series of sensitivity analyses which examine the price of retail
electricity, the value of excess electricity produced, carbon credits and RECS. Part II is a

scenario examining the co-digestion of ethanol syrup to increase electricity production.

4.1.1 Part I Price Sensitivity Analyses

With the 1,000 cow herd example, the total project cost was estimated to be
approximately $1,038,040 with an average yearly output of 159 kWh/hr for a total output
of 1,395,616 kWh/year. The average yearly output accounts for engine-generator
downtime, the parasitic energy load required to run the digester system and seasonal
variations in electricity production. For modeling purposes, the average monthly output
was used as an estimation of the engine-generator size needed for the surplus sale and
buy-all sell-all agreements. Therefore, an engine-generator size of 160 kW was assumed
in this section. With not metering, however, the engine-generator must be sized to match
the electricity consumption needs of the farm. With the 1,000 cow dairy, a nameplate
capacity of approximately 105 kW would be used based upon the maximum hourly
electricity demand (kW) over a 12-month period estimated by the model. The model
estimated demand by using actual data from the Michigan case farm and scaling the
values to the needs of a 1,000 cow dairy. With the 105 kW engine-generator, the average
yearly output would then be 93 kWh/hr assruning a 90% online time and a parasitic
energy load that consumes 2% of the electrical output. All assumptions used in section

one (Parts I and II) analyses are listed in Tables 18-19.

95

Table 18. Section One Assumptions - 1,000 Cow Dairy Sensitivity Analysis

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lnfluent

Herd Size 1,000
Daily Flow Rate (gpd) 29,950
Biogas Production

Methane concentration (%) 60
Biogas Yield (ﬁ3/Ib V8) 4.3
Electricity Generation

Online Time (%) 90
Engine Efﬁciency (%) 35
Engine Generator Size (kW)

Surplus Sale and Buy-All Sell-All 160
Net Metering 105
Co-digestion 500
Heat Recovery Efﬁciency (%) 40
Parasitic Energy Requirement 2

%)

Digester Tank and Heating
Heat Loss (%) 5
Total Solids (%) 8
Design Temp (°F) 95
Hydraulic Retention Time (days) 20
Boiler Efficiency (%) 80
Pricing
Carbon credits ($/credit) 2
REC's ($/credit) 26.5
Propane gas36 (S/gallon) 2.31
Retail Electricity ($1kWh) 0.0988
Financial Inputs

Return on Equity (%) 10
Tax Rate (%) 33.45

 

 

36 All propane offsets in this study are achieved using waste heat.

 

96

Table 18. Section One Assumptions - 1,000 Cow Dairy Sensitivity Analysis

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Continued)

160 kW Engine-Generator Total Project Cost

$ 1 ,03 8,040
USDA REAP Funding Total Principal Term APR Percent of lnvestrnent
Loan Guarantee $519,020 15 6% 50%
Table Continued

Amount Duration (yr) Percent of Investment
Grants $259,510 1 25%
105 kW Engine-Generator Total Project Cost

$909,893
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $454,946 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $227,473 1 25%
500 kW Engine-Generator Total Project Cost

$1,294,335
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $647,168 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $323,584 1 25%

 

 

 

 

 

Table 19. Section I Electricity Purchase Agreement Assumptions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Policy Summary Surplus Buy-All Net
Sale Sell-All MeterﬁiL

Standby Charge Threshold (kW) 100 100 150

Average Value of Excess Electricity ($/kWh) 80.0435 $0.0435 $0.061237

Administrative Charges ($/kWh purchased) $0.001 O $00010 N/A

8 stem Access Charge ($/month) $100 $100 N/A

 

 

 

 

 

 

_

37 This is referred to as the power supply component of the customer’s bill.

97

 

4. 1.1A Retail Electricitv

The term retail electricity is referring to the price that the farm pays to purchase
electricity from the utility company. This is also the electricity price that is offset with
digester production under the surplus sale and net metering agreements. In this analysis,
the value of the electricity produced was increased proportionately with the retail
electricity price (Figure 5). This is because a high retail electricity price implies a higher
cost of production which would also be reﬂected in the sell price (value of the electricity

produced).

Figure 5. Sensitivity of NPV to Retail Electricity Prices

 

NPV vs Retail Electricity Price

 

 

 

 

+ Surplus Sale

NPV

-I— Buy-All Sell-

All
- ,.__ Net Metering

 

 

 

 

 

 

 

 

 

Electricity Price (S/kWh)

 

 

 

 

The breakeven price for retail electricity price was found to be $02135 per kWh under a
surplus sale agreement and $02245 per kWh under net metering. Note that the
breakeven price under net metering is very close to the breakeven price with a surplus
sale agreement, even though the electricity production under net metering is 41% less
than that of the other two agreements which utilize a larger engine-generator. There are

three explanations for these results. First, a 1,000 cow dairy with a 105 kW engine-

98

generator nameplate capacity will not pay standby charges. Under category 3 (Chapter 2)
net metering, only farms with nameplate capacities greater than 150 kW will pay standby
charges. Since these charges can represent 16% to 54%38 of a farm’s electricity
revenues”, the policy beneﬁt actually offsets the decreased production levels from the
smaller generator. Second, net metering customers receive a higher price for credited40
electricity than under the other two agreements. Lastly, the net metering agreement is not
subject to administrative charges per kWh of electricity purchased by the utility company

or monthly system access charges. These differences are summarized in Table 19.

The breakeven price with net metering would also be the preferred agreement up to a
price of $01850 per kWh. A surplus sale agreement, however, shows the steepest
increase in NPV as the prices rise. This is because the larger engine-generator used with
the surplus sale agreement allows the customer to offset more of their on-farm electricity
consumption. The beneﬁt of offsetting larger amounts of electricity usage becomes
increasingly proﬁtable at higher retail electricity prices despite the burden of standby

charges.

Under a buy-all sell-all agreement, the after-tax NPV increases due to a rise in the value
of electricity sold, but does not reach a breakeven price. This is because agreements

which allow the farm to offset retail electricity purchases will experience the greatest

 

38 The exact percentage depends on the retail electricity price. The higher the price, the lower the
percentage that standby charges represent.

Electricity revenues are deﬁned as electricity offset by the digester, sold to the utility company or
credited to the farm’s utility bill in the case of net metering.

4° Electricity that is credited to next month’s bill is considered a revenue for the customer.

99

increase in return on investment from higher retail electricity prices. Note that a buy-all
sell-all agreement is preferable to a surplus sale purchase agreement with retail electricity
prices up to $00900 per kWh. This is due to the fact that no standby charges are required
with a buy-all sell-all agreement4lsince on-farm electricity is not being offset. With prices
higher than $00900, however, the beneﬁts of offsetting on-farm electricity make buy-all

sell-all a less desirable agreement for a farm with a milking herd of 1,000 cows.

The Department of Energy’s Energy Information Administration predicts electricity costs
to increase steadily for the next 15 years. By the end of the project period, the average
nominal price of commercial electricity in the US. is expected to be as high $0.134 per
kWh on average. In states like California, the commercial electricity rate reached an
average of $01638 per kWh during the summer of 2009 (U S EIA, 2009)). The trend for
higher electricity prices around the country suggests that a breakeven price of $02135
per kWh might be feasible in the future. In addition, this breakeven price is considered
with the price of carbon credits and RECS held constant. In reality, increases in other

revenues sources will likely decrease the breakeven price.

41.13 Value of Electricity Production

The value of electricity production refers to the price that the utility pays/credits the
customer for electricity produced by the digester. Currently, customers are paid either
the locational marginal price or the power supply component of their bill (net metering).
This analysis considers a range of prices from $0.020 per kWh to $0.250 per kWh to

examine the effect on the NPV of the digester investment (Figure 6).

 

4' Buy-all sell-all agreements do not pay standby charges under any engine generator size since no on-farm
electricity is offset.

100

Figure 6. The Sensitivity of NPV to the Value of Electricity Produced

 

NPV vs Value of Electricity Produced

 

 
  
    

 

 

 

   
   
 

 

 

 

 

 

 

 

 

 

$1,500,000
$1,000,000 a- r e .— _2_ , - «a v r: L-__ 4
$500,000 ——- — -~—- —-— ————- — —
+Surplus Sale
at $0 IIIiI—iliiiiriw—rrit lllllllllllllII‘IAI‘W +Buy-AIlSeIl-All
AAAAAAAAAAA . -+- Net Metering
‘ — W
4500,0001: .7 , w , “WM; ,_, ﬂ,
-$1,000,000 - ————_——---— 4 ~44» -_ at -9 .4 44
-$1,500,000
0 l0 O 1-0 LO 0 LO LO 0 l0 1.0 O In
seesessgiate§aaa
O o O O O O O o o O Q o O O o C
Price ($/kWh)

 

 

 

The results Show distinct effects under each agreement. The buy-all sell-all agreement is
the most sensitive to increases in the value of digester electricity with a breakeven price
of $01315 per kWh. This is because all electricity is sold and therefore the customer will
beneﬁt the most from increased compensation from the utility company. The surplus sale
agreement has a higher breakeven price of $02530 per kWh since only the excess
electricity is sold back to the grid. Even though a surplus sale agreement involves
offsetting on-farm electricity at the higher retail rate, standby charges signiﬁcantly
decrease proﬁtability. The model calculates that 42%42 of the farm’s electricity revenues
will be used to pay standby charges under a surplus sale agreement. Net metering is
clearly the least favorable agreement as the price increases and does not reach a
breakeven point within the range of prices tested. This is because the 105 kW engine-

generator allowed under net metering reduces the amount of excess electricity that can be

42 At a retail electricity price of $00988 per kWh

101

credited at the higher prices. Note that net metering shows higher returns to investment
than a buy-all sell-all agreement for prices up to $00650 per kWh. This is because
offsetting electricity at the retail rate and no administrative or system access charges
makes the returns under net metering higher in this range. Above $00650 per kWh,
however, it becomes more proﬁtable to sell the entire output from the digester than to
offset on-farm consumption and pay standby charges. In comparing net metering to a
surplus sale agreement at lower prices, net metering is also preferable, but up to the
higher price of $01000. This is because a surplus sale agreement is less sensitive to
increases in the value of electricity produced. Above a price of $01000 per kWh, the
surplus sale agreement allows the farm to earn more revenues from the increasing value
of the digester electricity produced. This is due to the larger engine-generator permitted

under this agreement (160 kW).

When considering these results in terms of energy policy, the breakeven prices can be
used as supporting evidence in the setting of feed-in-tariffs. Feed-in-tariffs are prices
paid per kilowatt-hour of electricity generated by a renewable energy system and are
currently utilized in several European countries and Canada. The price that is paid is
based on the cost of the electricity produced plus a reasonable proﬁt for the producer.
They are differentiated by technology and can also be further differentiated within each

technology by project size or productivity (Gipe, 2009).

Michigan also is pursuing similar legislation and in February, 2009, House Bill 4137 was

proposed which speciﬁes feed-in-tariff prices by technology. The bill identiﬁes a tariff of

102

$0.125 for methane digesters with an engine generator nameplate capacity between 150
kW and 500 kW. The breakeven price of $01315 calculated by the model supports the
price in the legislation only if considering a buy—all sell-all agreement. Under a surplus

sale agreement, however, the tariff would need to be at least $02530 for the investment
to breakeven. Therefore, a farm owner would want to consider future energy legislation

before deciding which electricity purchase agreement to select.

4.1.1C Carbon Credits

Methane is a greenhouse gas that is approximately twenty-three times more powerful per unit
at trapping heat than carbon dioxide (CCXb, 2009). Its destruction or avoidance has created
potential for anaerobic digesters to earn additional revenue in the form of carbon credits.
Under CCX rules methane is credited at a rate of 21 metric tons CO2 for each metric ton
of methane avoided. With the 1,000 cow example, the breakeven price for carbon credits
is $30.06 per credit with a surplus sale agreement, $31.07 per credit with a buy-all sell-all
agreement and $23.43 with net metering (Figure 7). With carbon credits, net metering is
Slightly less than the other two agreements due to the policy beneﬁts cited in the retail
electricity price sensitivity analysis. In this scenario, the number of carbon credits is
determined by the amount of methane that is prevented from entering the atmosphere
through the installation of the digester (Methane Combustion Method). Through this
method, the amount of revenue earned from carbon credits will be constant for a given
herd size, climate and manure management practice and will not be affected by the
amount of biogas ﬂared or sent to the engine-generator. The alternative method (Ex-Ante
Method) of earning credits involves the combustion of methane through the engine-

generator or the ﬂare. In this example, the combustion method calculates a higher level of

103

methane emission reductions. Since the CCX selects the method which calculates the

lowest level of reductions, the combustion method was not used in the analysis.

Figure 7. Sensitivity of NPV to Carbon Credit Prices

 

NPV vs Carbon Credit Price

 

$400, 000

$200,000 - —

 

 

$0

 

$200,000 ~ — ,, - . -—+— Surplus Sale

NPV

.$400'000 .. _ I +Buy-All Sell-All

 

-0— Net Metering

$600,000 -
$800,000 - 1

 

  

 

 

 

-$1,000,000

.3“ 5,59 (0° 09° g9° 69° 99° ,9“ 69° go° (9° ,9°
‘9 “’4‘ 65‘ at" so“ 60' 60' s1" 5'5 ‘39

Price (SICrcrit)

 

 

 

While the price of carbon credits over the past year has been as low as $0.15, current
climate change legislation has the potential to raise the price of carbon credits much
higher. Perhaps a predictor of future prices is the European Climate Exchange which had
an average price of $17.81 in 2009. If these prices are indicative of the future carbon
market in the US, the breakeven values calculated could become feasible. It is
important to note that a cap and trade system will not only have the effect of increasing
the price of carbon credits, but will also increase the price of electricity produced from
fossil fuels (e.g., coal). If an increased price of both carbon credits and electricity were
considered together in the same scenario, the breakeven price for carbon credits depicted

in Figure 7 would be lower.

104

4.1.1D Renewable Energy Credits (RECs)

A Renewable Energy Credit (REC) is a payment per Megawatt (1,000 kWh) of electricity
produced from a renewable energy system. The breakeven prices for RECS are
$111.40/MW for a surplus sale agreement, $114.50/MW for a buy-all sell-all agreement
and $138.10/MW for net metering (Figure 8). From Figure 8, it is evident that increasing
REC prices have less of an effect on the NPV of a digester under net metering. This is
consistent with the fact that net metering requires a farm to install a smaller engine-
generator that is Sized to meet the energy consumption needs of the farm. With a smaller
generator, less electricity is produced which limits the quantity of RECS that can be sold
by the farm. Under the surplus sale and buy-all sell-all agreements, a larger engine-
generator allows for a larger increase in revenues from higher REC prices. Note that net
metering actually has a higher after-tax NPV than under a surplus sale agreement for
REC prices up to $74.50/MW. It is also higher then under a buy-all sell-all agreement for
REC prices up to $81 .SO/MW. This is due to the policy beneﬁts cited in Section 4.1.1A

Retail Electricity.

105

Figure 8. Sensitivity of NPV to Renewable Energy Credit Prices

 

 

NPV vs REC Price

 

$500,000
$400,000
$200,000 — -

$0
-$200,000 4“
-$400,000 4
-$500,000 -
-$800,000 4

-$1,000,000 ~L~———-—~~

circa» '\° engages

 

 

+ Surplus Sale
”I“ Buy-All Sell-All
+ Net Metering

 

NPV

 

 

 

 

 

 

 

REC Price ($/Credit)

 

 

 

The market for RECS is largely dedicated by utility companies which decide the price
they are willing to offer for each REC. The market for RECS exists primarily as a means
for companies or other entities to comply with renewable portfolio standards. In this
study, a price of $26.50/MW was taken from a Michigan case farm, but recent contracts
from Consumers Energy list prices of roughly $30 to $50/MW per REC. Within that
price range, the model suggests that a net metering agreement would be preferable. When
considering the long run, however, higher REC prices would make net metering a less

favorable option.

4.1.2 Part II Co-digestion of Ethanol Syrup
This scenario examines the number of truck loads of ethanol syrup that are needed in
order for the digester system to reach a positive NPV on a 1,000 cow dairy. When a farm

decides to co-digest ethanol syrup, the amount added is determined by the capacity of the

106

truck and number of times the farm is willing to travel to the ethanol plant for pickup. In
this model, the assumption is made that trips will only be made to the plant when the
truck is loaded to full capacity. The price of fuel, labor costs and the distance traveled
are also likely to be factors which inﬂuence a farmer’s decision. Therefore, when
calculating the NPV, the cost of the syrup, fuel (including vehicle gas mileage), labor,
truck capacity, and trip duration (a function of distance, speed and time needed for

loading) are used to calculate the full cost of the ethanol syrup.

4.1.2A Scenario

Once the feedstock has been purchased and brought back to the farm, another key
decision is the amount added each day. Adding the full amount at once may shock the
system and result in increased acid levels and foaming. In order to account for this
factor, the assumption was made that 50% of the syrup would be added over a period of
two days”. It is important to consider that a farm may not want to purchase additional
syrup if excess supply exists. This scenario is set up under the assumption that all

existing supply will be utilized before new truck loads are purchased.

In order to take advantage of the increased biogas production, the assumption was made
that at larger engine-generator Size would be necessary than if manure was the sole
feedstock. For surplus sale and buy-all sell-all agreements, a 500 kW engine-generator
was assumed compared to 160 kW with no syrup added in Part I. The 500 kW size was
chosen as an example Since it would allow for enough electricity production to produce a

positive digester NPV. The initial investment cost for the electrical generation equipment

 

‘3 As an example, if 100% of the syrup were added at each feeding then there would be no excess supply.

107

was adjusted accordingly. With not metering, however, the utility company limits the size
to ﬁt the customer’s consumption needs and the model was run assuming a 105 kW

engine-generator.

The scenario assumes that 7 tons of ethanol syrup are loaded into the truck each trip
which is approximately 5.60% of the total inﬂuent to the digester. The value of 7 tons
was determined from scaling back actual values obtained from the Michigan case farm.
Other assumptions such as trip duration, truck speed, fuel prices and labor costs, are

highly variable and an attempt was made to choose the most reasonable values possible

for the analysis (Table 20).

Table 20. Co-digestion Scenario Assumptions Summary

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Assumptions
Quantity (tons) 7
Price ($/ton) 5
Feedstock Added (%/day) 50
Days of Supply 2
Percentage of Mixture (%) 5.60
Distance (miles) 60
Average Speed (mph) 65
Time for Loading (hrs) 1
Fuel Economy (mpg) 15
Fuel Price ($/gallon) 2.50
Total Trip Duration (hrs) 1.92

 

 

 

Table 21 compares the number of truck loads per month and the percentage of ethanol
syrup added on a daily basis to the corresponding increase in average yearly electricity
production (kWh/hr) from the digester. The column “Surplus Sale & Buy-All Sell-All”
represents the electricity production assuming no limitation on the engine-generator size.

As mentioned in Part 1, net metering limits the engine-generator size to the electricity

108

consumption needs of the farm. Therefore, increased biogas production from the co-
digestion of additional feedstocks has an extremely limited effect on electricity
production. The only noticeable effect is that the parasitic energy load from the digester

no longer reduces the average yearly electricity production after one truck load of ethanol
syrup.

Table 21. Co-digestion of Ethanol Syrup and Electricity Production (7 ton loads)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Average Yearly Electricity Production (kWh/month)
Truck % Added Surplus Sale & Net Metering
Loads Daily Buy-All Sell-All

0 0 159 93
1 2.72 203 95
2 5.30 241 95
3 7.75 278 95
4 10.07 316 95
5 12.28 353 95
6 14.39 391 95
7 16.39 428 95
8 18.30 450 95
9 20.13 450 95
10 21.88 450 95
11 23.55 450 95
12 25.1 450 95
13 26.69 450 95
14 28.17 450 95

 

 

 

 

 

In Figure 9 on average, one truck load of ethanol syrup increases the average yearly
electricity production by 36 kWh/hr. At 8 truck loads per month, the engine generator is
producing at its maximum level of 500 kW and sufﬁcient waste heat is produced to avoid
the use of the boiler. This results in an average yearly electrical production of 450

kWh/hr given an online time of 90%.

109

Figure 9. The Effect of Each Tuck Load (7 tons) of Ethanol Syrup on NPV

T

 

NPV vs Truck Loatk of Ethanol Syrup (7 tons)

 

 

 

 

 

 

 

 

 

 

 

$200,000 "I
$0
-$200,000 ~
$400 000 "Ar— Surplus Sale
> ' . ‘
n. , +3 -All Sell-
2 , _ , _ , “Y
-$600,000 , All
$800 000 4 +Net Metering
-$1,000,000 .'
-$1,200,000 4*”:-

 

01234567891011121314
TruckLoads

 

 

 

Given the assumptions made, the system will require 7 truck loads of ethanol syrup per
month with a surplus sale agreement and 8 loads per month with a buy-all sell-all
agreement in order to reach a positive NPV. The difference in the number of truck loads
is attributed to a slight advantage with the surplus agreement given the retail electricity
prices and the value of electricity produced assumed in the model. Since a net metering
agreement prevents a farm from realizing increased electricity production from co-
digestion, adding ethanol syrup actually decreases the return on investment. This is
because the farm would be incru'ring feedstock related costs without increasing revenues.
When comparing the results of this analysis to the Michigan Department of
Environmental Quality (MDEQ) organics residuals exemption discussed in Chapter 2, the
hypothetical farm operation could have a proﬁtable investment and still be in compliance
with state laws. Note that at 8 truck loads per month, the ethanol syrup entering the
digester each day comprises roughly 18% of the total mixture (manure, parlor/dilution

water and ethanol syrup). This ﬁts within the 20% limit stated in the MDEQ exemption.

110

4.3 Section Two - Engineering

The next section demonstrates the model’s use as a tool for engineers to analyze and
predict digester performance. While a variety of parameters can be tested with this
model, three key variables (total solids concentration, volatile solids loss, and online
time) were chosen which can Signiﬁcantly affect digester performance. This will help
provide insight into how system performance directly relates to proﬁtability. All
examples in Section Two involve a herd of 1,000 lactating cows and many of the same
assumptions as Section One (Table 18). In this section, however, a larger engine-
generator was selected for analysis in order to more effectively demonstrate the use of the

model when other assumptions are changed. A list of assumptions is presented in Table

22.

4.3.1 Total Solids Concentration

Table 22. Section Two Assumptions - Engineering

 

 

 

Inﬂuent
Herd Size: 1,000
Daily Flow Rate (gpd) 29,950

 

 

Biogas Production

 

Methane concentration (%) 6O

 

Biogas Yield (113/lb V8) 4.3

 

 

Electricity Generation

 

Online Time (%) 90

 

Engine Efﬁciency (%) 35

 

Engine Generator Size (kW)

 

Surplus Sale and Buy-A ll Sell-All 180

 

Net Metering 95

 

Heat Recovery Efﬁciency (%) 40

 

 

 

 

Parasitic Energy Requirement (%) 2

 

111

Table 22. Section Two Assumptions - Engineering (Continued)

 

Digester Tank and Heating

 

 

 

 

 

 

 

 

 

 

 

 

Heat Loss (%) 5
Volatile Solids Loss (%) 0
Total Solids (%) Base Case Variable
Design Temp (°F) 95
Hydraulic Retention Time (days) 20
Boiler Efﬁciency (%) 80
Pricing

Carbon credits ($) 2
REC's ($) 26.5
Propane gas ($/gallon) 2.31
Retail Electricity ($/kWh) 0.0988

 

 

Financial Inputs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Return on Equity (%) 10
Tax Rate (%) 33.45
180 kW Engine-Generator Total Project Cost

$1 ,038,040
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $519,020 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $259,510 1 25%
105 kW Engine-Generator Total Project Cost

$909,893
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $454,946 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $227,473 1 25%

 

 

 

 

 

Figure 10 shows the relationship between the total solids concentration of digester
inﬂuent and average yearly electricity production. AS mentioned in Chapter 2, the total
solids concentration decreases as water (parlor, rain) are mixed with the manure. In
Michigan’s climate, more liquid in the inﬂuent mixture increases the heating

requirements of the digester and can decrease electricity production. In addition,

112

 

investment costs increase due to the need for larger tanks and more related components.

The following scenario analyzes this relationship between the total solids concentration

of digester inﬂuent and electricity production.

Figure 10. Total Solids Concentration vs. Average Yearly Electricity Production

 

 

Total Solids Concentration vs Average Yearly Electricity Production

 

 

 

 

 

 

180
160 -
140 -
120 - + 180 kW Generator
'5 100 $— _ _ _ Average Yearly
2 80 3 Production
5 60 - (kWh/hr)
40 -. “t“- 105 kW Generator
Average Yearly
20 _ Production
0 I I I I I I I I T I I I I If (kWh/hr)

 

 

 

‘5 b4: ‘3' ‘0' ’\' ‘b' 0."
Total Solids Concentration (%)

 

 

 

In Figure 10, average yearly electricity production with the 180 kW engine-generator
represents the surplus sale and buy-all sell-all agreements. A 105 kW engine generator is
assumed with net metering. Note that the average yearly production from the 180 kW
engine generator increases with the total solids (TS) concentration. Speciﬁcally, between
T8 concentrations of 2.5% and 5.5%, every 1% increase in T8 concentration raises the
average yearly electricity production (kWh/hr) by an average of 30 kWh/hr. For TS
concentrations greater than 5.50%, an increase in the TS concentration has a decreasing
effect on the average yearly electricity production. This is because at higher TS
concentrations, less biogas is required to run the boiler allowing more biogas to be sent to

the engine-generator.

113

With the 105 kW engine-generator, at T8 concentrations between 2.5% and 4.0%, every
% increase in the TS concentration increases the average yearly electricity production by
20 kWh/hr. In this range, the digester system will experience signiﬁcant waste heat
deﬁcits and the boiler will need to be used a minimum of 8 months out of the year.
Beyond a concentration of 4.0%, electricity production is a constant 93 kWh/hr for T8
concentrations ranging from 5.00% to 10%. This is because at T8 concentrations above
5.00%, extra biogas can be burned in the boiler to meet digester heating needs without
diverting biogas from the engine-generator. The extra biogas exists because the engine-

generator is undersized for the biogas production of a 1,000 cow herd.

While Figure 10 only considers the effect of TS on average yearly electricity production,
Figure 11 considers the entire digester investment. For example, at lower TS
concentrations, higher quantities of water are present in the digester inﬂuent which
requires a larger tank size and more related components. In addition to the higher
investment costs, lower electricity production levels due to heat deﬁciencies also

contribute to lower returns on investment at lower TS concentrations.

114

Figure 11. The Effect of Total Solids Concentrations on NPV

 

 

NPV vs. Total Solids Content (°/o)

$0 I I I i 1 I r
-$500, 000 -
-$1 .000, 000 r

 

 

“f“ Surplus Sale
—I— Buy-All Sell—A ll
—0— Net Metering

-$1.500,000 -

NPV

-$2, 000, 000 .

 

 

 

-$2,500,000 4
—$3,000,000 --./
-$3,500,000

 

 

 

 

3 50%
4 00%
4 50%
5 00%
5 50%
6 00%
6 50%
7.00%
7 50%
8 00%
8 50%
9 00%
9 50%
10 00%

 

 

Across the levels tested, a buy-all sell-all agreement Shows higher returns at lower total
solids concentrations (2.50% to 6.0%) than the surplus sale agreement. This is because
as the digester electrical output decreases with lower TS concentrations, the system is
offsetting less on-farm electricity. On—farm electricity is valued at the commercial retail
rate of $00988 which is roughly double the average monthly LMP which ranges from
$0.037 to $0.052 in the model. AS a result, standby charges, which are determined by the
farm’s peak usage, represent a larger percentage of the electricity revenues. Therefore,
the cash ﬂows at lower TS levels are Signiﬁcantly reduced. For example, at a T8
concentration of 4%, the standby charge will use up 81% of electricity revenue44
generated by the digester. At this level, a buy-all sell-all agreement would be preferable

to offsetting on-farm electricity with a surplus sale agreement. Buy-all sell-all

agreements are not subject to these charges.

 

‘4 Revenue is deﬁned as offsets and sales

115

 

Despite the lower electricity production under the net metering, it produces higher returns
than the other two agreements. This is due to the policy beneﬁts mentioned in previous
sections. It is important to keep in mind, however, that the results of this analyze are
speciﬁc to a 1,000 cow milking herd and Should not be used to make generalizations
across a range of herd sizes. An analysis of each purchase agreement across a range of

herd Sizes is covered later in Section Three.

4.3.2 Volatile Solids Loss

The loss of volatile solids (VS) can often be attributed to manure pretreatment in which
certain processes (e. g., mechanical separation, sand removal) can cause the loss of these
solids. Since volatile solids are the energy producing portion of the manure or any other
added substrate, their loss has a direct effect on the performance of the digester. The
issues are similar to that experienced with low total solids concentrations, except that
volatile solids losses will not increase the capital costs of the system. All assumptions
from Table 22 are held constant expect that the total solids concentration is no longer the
tested variable and is assumed to be 8%. Figure 12 shows the relationship between

volatile solids loss and average yearly electricity production.

116

Figure 12. Volatile Solids Loss vs. Average Yearly Electricity Production

 

Volatile Solids Loss vs Average Yearly Electricity Production

 

 

 

 

 

 

  
 

 

 

 

 

180 ’ “I
160 - ._ "I
140 ~» _
120 ~ -- _ _._.ul
"' 100 --. -__ A —- — —— _ a: _- MEI +180kWGenerator
S aaaaaaaaaaaaaaaaaasash-autos; Asian-A9. AverageYearly
E 80 -,-._ , ______ __ i F T. “'1‘ Production
60. H _ _ __ - - h a _. W... _. (kWh/hr)
40-«'- e , , 7.. - .- -2, ,L,,,, pg? “105kWGenerator
I AverageYearly
20 l ” 'T T '" " ” " * ‘ ' ' ‘" ‘ r ’ *7 - 7 v -: Production
I
i (kWh/hr)
0 w I I I W I T r l I I I I I i
o

 

o\°
o_
.— a
Volatile Solids Loss (%)

 

 

 

Note that with the 105 kW engine-generator, average yearly electricity production is not
affected by volatile solids loss throughout the ranges tested. This is because since a 105
kW engine-generator is under sized for a 1,000 cow milking heard, the system can lose a

minimum of 35% of its V8 and still run the generator at its maximum capacity.

With a 180 kW engine-generator Size, electricity production is more sensitive to losses in
VS. Speciﬁcally, between the entire range of 0% to 35%, every 1% decrease in VS
results in a 2.07% decrease in electricity production on average. Slightly larger decreases
were observed at V8 losses above 21%. Figure 13 examines the effect of VS losses on
the NPV of the digester investment. On average, a 1% increase in VS loss decreases the
NPV of the digester by 1.23% with a surplus sale agreement and 1.00% with a buy-all
sell-all agreement. Net metering remains unaffected since extra biogas is available to
achieve digester heating requirements without decreasing electricity production. Unlike

the situation with TS concentrations, the capital investment does not change which

117

explains a more linear trend in NPV across the range of solids losses tested. Even at a
V8 loss level of 0%, the net metering agreement still shows higher returns. This is

explained by policy advantages outlined in section one of this chapter.

Figure 13. The Effect of Volatile Solids Loss on NPV

 

NPV vs Volatile Solids Loss

 

-5400_000__-_- “—9— W.” --__a---._.-:-_- M-.- -. a;

 

t Surplus Sale

 

 

 

 

 

 

 

 

 

E '$600'000 “T. _- —* #7” TE 7-7”” “FT—'TTT'" “T774” "'7” “T ”" ' ' I +Buy-All Sell-All
, I +Net Metering
-$800.000 -- ,
-$1,000,000 *
-$1.200,000 J~~——--— -— ___,_.‘ -——--—--———
Q o\o o\o o\o o\o o\o o\o o\o o\o o\0 o\° o\o

Volatile Solids Loss (%)

 

 

As with the analysis of TS concentrations, a buy-all sell-all agreement shows slightly
higher returns than a surplus sale agreement at higher levels of VS loss. The explanation
is the same in that as electricity production decreases under a surplus sale agreement,
standby charges represent a larger percentage of electricity revenues. For example, with
a V8 loss of 30%, the standby charge uses up 65% of the electricity revenues generated
by the digester. This suggests that a system with a potential for large losses in volatile

solids would be not be well suited for a surplus sale agreement.

118

 

4.3.3 Online Time

The third key variable in the engineering section of the analysis deals with the operational
online time of the engine generator. While 90% is considered as “good” performance by
EPA AgSTAR’s Farmware simulator, actual digester systems may experience a range of
online times depending on the condition of the engine-generator and quality of the
maintenance by farm operators. All assumptions from Table 23 are held constant expect
that the total solids concentration is no longer the tested variable and is assumed to be
8%. Figure 14 shows the relationship between online time and average yearly electricity
production.

Figure 14. Online Time vs Average Yearly Electricity Production

 

Online Time vs Ave rage Yearly Electricity Production

 

 

 

 

+ 180 kW Generator

h Average Yearly

g Production (kWh/hr)

3

a: --—A.~—- 105 kW Generator
Average Yearly
Production (kWh/hr)

 

 

 

 

 

 

 

l 111 T lfl ll 1
gasasaeeesseesaaa

Online Time (%)

 

 

 

 

In Figure 14, there is a linear decrease in the average yearly electricity production for
both the 180 kW and 105 kW engine-generators. Every 10% decrease in online time,
decreases the average yearly electricity production by 18 kWh/hr and 10 kWh/hr,

respectively.

119

In Figure 15, note that the buy-all sell-all agreement shows higher returns than the
surplus sale agreement for online times up to 60%. As with section 4.3.1 Total Solids
Concentration, this is because as the digester electrical output decreases, the system is

offsetting less on-farm electricity.

Figure 15. The Effect of Online Time on NPV

 

NPV vs Online Time

 

$0
-$200,000 +~ , w W J - ~ ., , ..-._-_.ﬁ_._,__._______ , *,__
$400,000 --_ ——-~~—« or, , w av ~ , 77 g .7- _,_ -. _,-__, ,

 

$600,000 «u—r-rm :k‘ﬁ ~ —-‘—~— - — e—k————~v e~ *m— +smpius Sale
«$800,000 .
$1,000,000 ~

—-— Buy-All Sell-All

NPV

 

 

 

 

4w Net Metering

 

-$1,200,000 -

 

-$1,400,000 ~

 

 

-$1,600,000
.99 9° <29 «Q «9 «.9 v9 (5° '9
Online Time (%)

 

 

 

 

On-farm electricity is valued at the commercial retail rate which is roughly double the
average monthly LMP. in addition, standby charges, which are a constant and determined
by the farms peak usage, represent an increasing percentage of the electricity revenues.
For example, at an online time of 80%, the associated standby charges use up 45% of the
electricity revenue on average with a surplus sale contract. If this level decreases to 40%,
standby charges use up 78% of the electricity revenues. At a level of 40%, a buy-all sell-
all agreement would be preferable to offsetting on-farm electricity with a surplus sale

agreement. At a level above 60%, however, the reverse is true. On average, every 1%

120

increase in online time, increases the return on investment by $8,5 55 with a surplus sale

agreement, $6,591 with a buy-all sell-all agreement and $6,227 under net metering.

Lastly, net metering would be the preferred agreement throughout the range of online
times tested. As mentioned in previous analyses, this benefit is speciﬁc to a 1,000 cow

dairy and should not be taken out of context.

4.4 Section Three - Policy

4.4.1 Part I - Current Policy

The third component of this chapter analyzes how Michigan energy policy affects the
return on investment of anaerobic digesters over a range of herd sizes. While the previous
two sections dealt speciﬁcally with a 1,000 cow dairy, examining a range of herd sizes
more effectively highlights the differences between the three electricity purchase
agreements. Unless otherwise indicated, assumptions used in all sensitivity analyses

performed in this section are listed in Tables 23-24.

Table 23. Section Three Assumptions- Policy

 

 

 

lnﬂuent
Herd Size: 1,000
Daily Flow Rate (gpd) 29,950

 

 

Biogas Production

 

Methane concentration (%) 60

 

Biogas Yield (ft3/lb VS) 4.3

 

 

Electricity Generation

 

Online Time (%) 9o

 

Engine Efficiency (%) 35

 

Engine Generator Size (kW)

 

Surplus Sale and Buy-A ll Sell-All 180

 

Net Metering 95

 

Heat Recovery Efficiency (%) ' 4O

 

 

Parasitic Energy Requirement (%) 2

 

 

 

121

Table 23. Section Three Assgptions- Policy (Continued

 

Digester Tank and Heating

 

Heat Loss (%)

5

 

Volatile Solids Loss (%)

0

 

Total Solids (%)

Base Case Variable

 

 

 

 

 

 

 

 

 

 

 

 

 

Design Temp (°F) 95
Hydraulic Retention Time (days) 20
Boiler Efﬁciency (%) 80
Pricing

Carbon credits ($) 2
REC's ($) 26.5
Propane gas (S/gallon) 2.31
Retail Electricity (S/kWh) 0.0988
Financial Inputs

Return on Equity (%) 10
Tax Rate (%) 33.45

 

 

180 kW Engine-Generator

Total Project Cost

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$1,038,040
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $519,020 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $259,510 1 25%
105 kW Engine-Generator Total Project Cost

$909,893
USDA REAP Funding Total Principal Term APR Percent of Investment
Loan Guarantee $454,946 15 6% 50%

Amount Duration (yr) Percent of Investment
Grants $227,473 1 25%
Table 24. Current Policy Summary
Policy Summary Surplus Buy-All Net

Sale Sell-All Metering

Standby Charge Threshold (kW) 100 100 150
Average Monthly Value of Excess Electricity ($/kWh) 0.0435 0.0435 0.0612
AdministrativeCharges (S/kWh purchased)“ 0.0010 0.0010 N/A
System Access Charge 100 100 N/A

 

 

 

 

 

‘5 Details on administrative charges included in Chapter 3

122

 

 

Figure 13 shows the NPV for digester investments with herd sizes ranging from 500 to
4,000 lactating cows“. The “saw tooth” effect observed in the graph comes from the fact
that investment costs only come in discreet units. In particular, larger components such
as digester tanks and engine-generators are primarily responsible for the variation. Since
this scenario is based upon current policies, it is also used as a baseline of comparison for

other analyses in Section Three.

Clear economies of scale are present under the surplus sale and buy-all sell-all
agreements with larger herd sizes exhibiting greater returns on investment. Despite the
signiﬁcant differences in business models between the surplus sale and buy-all sell-all
agreements, their returns on investment over the range of herd sizes are extremely close.
Buy-all sell-all agreements do not offset on-farm retail rate electricity, but instead receive
compensation at the LMP which is less than half of the retail rate. In contrast, surplus
sale agreements offset on-farm electricity use at the higher retail rate and only sell the
excess to the utility at the LMP. While common intuition would suggest a surplus sale
agreement to be the superior choice, standby charges paid under a surplus sale agreement
reduce net revenues from electricity offsets and sales by 42% on average. This makes the

NPV of both electricity purchase agreements almost equal.

This would suggest that a farmer may need to rely on other factors to make a decision
between electricity purchase agreements. One factor may be the anticipation of higher

retail electricity prices. As depicted in Figure 5 from the 1,000 cow example, higher retail

 

‘6 The assumption was made that dry cows are kept in separate barns and do not contribute manure to the
digester.

123

electricity prices make a surplus sale agreement more favorable. Another factor could be
the involvement of a third party (e.g., energy project developer) who negotiates the
electricity purchase agreements on behalf of the farm. For example, a third party47
energy project developer may have a ﬁnancial interest in the electricity sales and may not
realize a beneﬁt from offsetting the farm’s electricity use. If this is the case, the farm may

select a buy-all sell-all agreement.

A net metering agreement shows higher returns for herd sizes ranging from 700 to 1,450
cows. As mentioned in the previous two sections, there are three main reasons for this
result. First, under category three net metering, customers only pay standby charges if
their engine-generator has a nameplate capacity greater than 150 kW. Second, they
receive the power supply component of the utility bill (average of $0.0612/kWh) for
electricity produced. The other two agreements receive the LMP which is a lower value
(average of $0.043 5/ kWh). Lastly, they do not pay administrative charges or system
access charges. The other two agreements must pay both of these charges (see Table 23).
Also note, however, that for herd sizes from 500-600 cows, surplus sale is clearly the
preferable agreement. This is because these farms sizes would not be subject to standby
charges under either a surplus sale agreement or net metering. Since a surplus sale
agreement involves a larger engine-generator and increased electricity production,

however, the return on investment is greater.

For herd sizes greater than 1,450 cows, however, net metering exhibits the lowest returns

of the three agreements. This is due primarily to reduced electricity revenues from the

 

‘7 Third party is deﬁned as an entity that is not the farm or the utility company.

124

smaller engine-generator which is sized based upon the farm’s average yearly electricity
usage. At the same time, standby charges do not decrease since they are based upon peak
on-farm energy usage measured in kilowatts (kW) and are unrelated to the average yearly
electricity usage which is measured in kilowatt hours (kWh). Therefore, standby charges
comprise a larger portion of the farm’s electricity revenues under net metering. For
example, at a herd size of 2,000 lactating cows, standby charges represent 60% of
electricity revenues. For the same herd size under a surplus sale agreement, however,
standby charges represent 42% of electricity revenues. As a result, the NPV of a 2,000

cow farm under net metering is 37% less than it would be with a surplus sale agreement.

Figure 13. NPV Compared to Herd Sizes (500 to 4,000 cows)

 

NPV vs Held Size

 

 

" ______ . V' . -w __ - - z +Surplus Sale
+ Buy-All Sell-All
”M - -- ____ ~— —O—Net Metering

 

 

 

 

 

 

 

 

 

 

 

Herd Size

 

 

 

4. 4. 1A Value of Odor Reduction
While Figure 13 shows negative NPV’s for herd sizes ranging from 500 to 4,000 cows, a

farm may still wish to install a digester for the odor reduction beneﬁts that it provides. In

125

order to quantify of the value of the digester’s odor reducing beneﬁts in more
meaningful units, the NPV of the investment was broken down into a cost/cow per day
with the results diSplayed in Figure 14. The assumption was that the difference between
a negative net present value and zero represents the value of the odor to the farmer.
Throughout the range of 500 to 4,000 cows, the value of odor control was valued at $0.10
per cow/day with a surplus sale agreement, $0.11 per cow/day with a buy-all sell-all
agreement and $0.12 per cow/day with net metering. If examining only the larger dairies
with herds over 2,000 cows, the cost/cow per day decreases to an average of $0.08 per
cow/day due to economies of scale. When put in these terms, it would appear plausible
that a farmer would still invest in an anaerobic digester despite a having a negative net

present value.

Figure 14. The Cost per Cow/Day of a Digester Across a Range of Herd Sizes

 

Herd Size vs Cost per Cow/Day

$0.00 11171

 

~$0.05~~~ ~ 2» —~ -- . ,C

 

'30-10 T ‘ +Surplus Sale

—l—- Buy-A11 Sell-A ll
- __ #2.. - _- . e - -v .. .. .. -O—NetMetering

Cost

-$0.15 —

 

 

 

 

-3020» » ~ , we»

 

 

 

-$0.25

Herd Size

 

 

 

126

4.4.2 Part II - Recommendations

This section brings together the results of the previous analyses in this chapter and uses
that insight to make recommendations for energy policy which is more favorable to
anaerobic digesters. The most recent purchase agreement option for digester owners is
net metering which was ﬁnalized in July, 2009. In Figure 13 of this chapter, the model
results showed that net metering only showed an advantage over the other existing
agreements for herd sizes ranging from 500 to 1,450 cows. From a previous analysis in
Section One, it was determined that this advantage was due in part to the higher threshold
for standby charges (150 kW), the lack of administrative and system access charges and a
higher value ($/kWh) for excess electricity produced by the system (power supply
component of customer’s electric bill). Since these elements were identiﬁed to be
beneﬁcial aspects of the new metering law, a series of sensitivity analyses were run to
determine the effect of applying these speciﬁc policy elements to the other two
agreements. An additional policy scenario was tested in which the current net metering

policy for digesters was compared to “true” net metering which is currently only offered

to small wind and solar technologies.

4. 4. 2A Standby C harge_S

In order to better understand how the standby charge threshold relates to herd size, Figure
15 shows the average yearly electricity production (kWh/hr) for a range of sizes.48 For
example, under a surplus sale agreement with a standby charge threshold of 100 kW, a

1,200 cow dairy would be subject to charges. This is because based upon the average

 

‘8 As with section one, the average yearly electricity production is used to predict the appropriate nameplate
capacity of the engine generator.

127

yearly electrical output potential of the 1,200 cow dairy, a digester would require an
engine-generator nameplate capacity of approximately 195 kW. Since 195 kW is greater
than 100 kW, the farm would be subject to standby charges if they wished to received
service form the utility company when the digester engine-generator is down. The same
size dairy with a net metering agreement and a threshold of 150 kW”, however, would be
limited to a nameplate capacity less than 125 kW. In this circumstance, the farm would

not be subject to standby charges.

Figure 15 can also be used to measure the effect of either raising or lowering the standby
charge threshold on digesters with a range of herd sizes. For example, consider an 800
cow farm with an estimated engine-generator size of 130 kW and a standby charge
threshold of 100 kW. In this scenario, the farm would likely be subject to standby
charges since the digester would require an engine-generator with a nameplate capacity
greater than 100 kW. If a herd size of 500 cows was considered, then standby charges

would not be required.50

The suggested nameplate capacities used here are considered close approximations, but in
reality a farmer will be limited by the engine-generator offerings which are commercially
available. In addition, a farm may wish to install a larger size generator than needed if
planning to add additional feedstock or may decide on a smaller size to save costs and

simply ﬂare the extra biogas. The estimated engine-generator nameplate capacities in

 

49 Category three net metering sets the standby charge threshold at 150 kW.
A farm is not required to purchase standby service, but during generator downtime the utility is not
obligated to provide service.

128

Figure 15 incorporate a 10% down time51 and a 2% parasitic energy re uirement and
q

therefore are considered to be a realistic estimation of the appropriate size needed.

Figure 15. Estimated Engine-Generator Nameplate Capacities and Herd Size

 

Estimated Engine-Generator Nameplate Capacities vs Herd Size

 

 

+ Surplus Sale and Buy-All
Sell-All FJigine-Generator
Size (kW)

~+- Net Metering Engine-
Generator Size (kW)

Nameplate Capacities (kW,

 

 

 

 

 

 

 

OITIIIIIIIITTIIITTTTITITIIFITPﬁTIHIII

o o o o o o o o o
o o No 6° 6 6 6° 0° Op o ,9 o

Herd Size

 

 

 

4. 4. 28 “Net Metering Components ”

Scenario 1

In Figure 16, all beneﬁcial components of the net metering law (a standby charge
threshold of 150 kW, produced electricity valued at power supply component of the
customer’s bill, and no administrative or system access charges) were applied to the other
two electricity purchase agreements. Since this scenario does not include any changes to
the current net metering policy, the results for the net metering agreement will not change
from Figure 13. Table 25 summarizes the purchase agreement assumptions used in this

analysis.

 

5| - . . . . . .
This implies an operational online time assumption of 90%.

129

The model shows that the returns achieved through both the buy-all sell-all and surplus
sales agreements are affected by herd size. For example, between a range of 500 to 950
cows, the surplus sale is clearly the preferred agreement. In this range, the beneﬁcial
components of net metering increases the NPV of the digester by 30.5% with a surplus
sale agreement compared to 20.0% with a buy-all sell-all agreement. This is because a
farm with a herd size under 950 cows would not be subject to standby charges. The buy-
all sell-all agreement does not pay standby charges under any circumstance, since it does
not involve offsetting on-farm electricity use. For herd sizes over 950 cows, however, a
buy-all sell-all agreement becomes the preferred agreement. In this range, the digester
NPV increases by an average of 34% compared to 18% with a surplus sale agreement.
These values represent increases from the current policy depicted in Figure 13. Despite
the increase in returns from the proposed scenario in Figure 16 “NPV vs Herd Size (Net
Metering Components)”, the digester does not achieve a positive NPV for the herd sizes

tested.

Table 25. Scenario 1 Policy Summary

 

 

 

 

 

 

 

 

 

 

Policy Summary Surplus Buy-All Net
Sale Sell-All Metering_
Standby Charge Threshold (kW) 150' 150 150
Average Value of Excess Electricity ($/kWh) 0.0612 0.0612 0.0612
Administrative Charges ($/kWh) N/A N/A N/A
System Access Charge N/A N/A N/A

 

130

Figure 16. NPV vs Herd Size Using Net Metering Components

 

 

 

 

 

 

 

 

 

 

NPV vs Herd Size (Net Metering Components)
$0 lIlTlTj—ﬁllllllIIITTIIIITTIIIIIII
-$20,000 _, — — —— — '———~;——w 1-
l i 11' .l \ I:
-$40.000 _--_._..-_ l; -— , - —~-—
> -$60,000 TN ‘1‘- - I:_~___H___ ‘ - - _ +Surplus Sale
A. u —I-— Buy-All Sell-All
2 '$80'000 "’ EL 7 _- “'T—_‘ ”a“ ﬂhw” ' i —O—Net Metering
-$100,000 “ii/T” — * ———— ———— ~ —~ — — — 4 ~——
-s120,000j ——— — — — —~~ — ——~ ~— ———
-$140,000
0 Q Q Q Q Q Q Q Q Q Q
o e e o o 6 6° 6
9° 9» c tr (1 $439.5.» ,9 ,2» 4:9,?»
HerdSize

 

 

The next two scenarios maintain the same assumptions from Table 25, but increase the
standby charge threshold to 400 kW (Figure 17) and 800 kW (Figure 18). As discussed
in Part One (Figure 15), increasing the standby charge threshold essentially exempts

larger farms from paying charges for standby service from the utility company.

Scenario 2

In Figure 17, the standby charge threshold was raised from 150 kW to 400 kW. Under
this scenario, the buy-all sell-all agreement shows the same results as in Figure 16 and is
unaffected by the threshold increase. With the surplus sale agreement, the only difference
pertains to the range of herd sizes that are able to beneﬁt from not paying standby
charges. In this scenario, herd sizes ranging from 500 to 2,500 would beneﬁt from this
sort of policy change compared to only up to 950 in Figure 16. The NPV increases by
56% with a surplus sale agreement and 28% with a buy-all sell-all agreement on average

for herd sizes ranging from 500 to 2,500 cows. Between this range, the surplus sale is

131

 

clearly the preferred purchase agreement. By not being subject to standby charges, the

farm is able to reduce expenses and increase cash ﬂows.

Figure 17. NPV vs Herd Size with Standby Charge Threshold of 400 kW

 

NPV vs Herd Size (400 kW Threshold)

 

 

 

$0
-$20,000 -
-$40,000 a
> -$60 000 -, _ +Surplus Sale
2 ' -I—- Buy-All Sell-All
$80900 ‘ —O—Net Metering

 

 

-3100000
-3120000 .
-$140, 000

We§@@§§§§§§

’1; W '1’ ‘L “or" (5’? W6
HerdSize

 

 

 

 

 

 

 

For herd sizes ranging from 2,600 to 4,000 cows, the NPV remains unchanged from the
scenario in Figure 16 under a surplus sale agreement. In this range, a buy-all sell-all
option becomes the preferred purchase agreement. For net metering, however, raising the
threshold to 400 kW essentially eliminates standby charges for almost all herd sizes
tested. Only dairies with a herd greater than 3,900 would still be subject to the charges.
When compared to the current policy (Figure 13), the NPV increases by an average of
28% across herd sizes ranging from 500 to 3,800 cows. Despite the increase, however, it

is the least favorable of the three agreements.

132

529113.093

Figure 18. NPV vs. Herd Size with Increased Standby Charge Threshold of 800 kW

 

NPV vs Herd Size (800 kW Threshold)

$20,000

$0 IIIIIIIITIrrrfFfrrrrIII'W

-szo,ooo -—~— ___4,__-___ -- I.

3 i.“ .‘
‘34°'°°° ' ” $7 '- + Surplus Sale
-$60,000 — a -

 

 

 

 

 

 

 

 

 

 

 

 

 

E l ., -‘—ﬂ—————-i«——--—- +Buy-AllSell-All
-$80,000 q 7 , ____...-4_____. _4m ## #7 m _ﬁ_ ,__“-- —0—Net Metering
-$100.000 - A*—ﬁ- ,__-_-_.___ r" *‘W’ rw __..
-$120,000 a _-__-_.___ - ~- —*— ' *-- W’ _#____ A
-$140,000

Q \N63 0 63 § Q
or e° 39,3 W§Wr§w§ (ff 699$

Herd Size

 

 

 

In Figure 18, the higher standby charge threshold essentially eliminates the charges for
the entire range of herd sizes. In addition, for farms with more than 2,600 cows, the
returns start to become positive under a surplus sale agreement. On average, the NPV of
the surplus sale agreement increases by 60% across the herd sizes tested. This scenario
points out the fact that a digester investment could be a marginally proﬁtable investment

for surplus sale agreements with an increase in the standby charge threshold from 100

kW to 800 kW.

4. 4. 2C Modiﬁed Net Metering vs. “T rue ” Net Meterilg
The ﬁnal scenario in this chapter addresses the question of whether “true” net metering
would be beneﬁcial for anaerobic digesters. “True” net metering means that a customer

receives payment for their electricity at the same price at which they purchase from the

133

utility company. Currently, only small wind and solar systems with nameplate capacities
less than or equal to 20 kW can beneﬁt from this policy. In Figure 19, all assumptions
from Tables 22 and 23 are the same expect for the value of the electricity produced,

which was assumed to be the same as the retail value ($0.0988/ kWh).

Figure 19. “Modiﬁed” vs “True” Net Metering Across a Range of Herd Sizes

 

NPV of Modiﬁed vs "True" Net Metering

 

 

 

-*—A-“ Modiﬁed Net Metering

 

NPV
3
.0
o
o
o

—l— True Net Metering

 

 

-$100,000 1

$120,000 J

 

 

 

 

-$140,000

0 ° 0° 0° c9 0°

«5° «9 .5 a .1 ,§,2§9,@°°,9P°,29°.,29°,e
HerdSize

 

 

 

 

The results show that “true” net metering would not produce a signiﬁcant increase in the
return on investment of a digester system. This is because the smaller sized generator
produces less excess electricity to be credited at the higher price. Therefore, an increase
in the price credited has little effect on the overall returns from the system. This result

suggests that pursuing “true” net metering for digester systems would not be worthwhile.

134

Chapter 5: Conclusions

Anaerobic Digestion is receiving a great deal of attention as a viable alternative in
supporting residuals management for livestock operations. In contrast to conventional
liquid and slurry management systems, anaerobic digesters provide multiple
environmental beneﬁts such as odor control, improved air and water quality, improved
nutrient management ﬂexibility and the opportunity to capture biogas for heat and
electricity production. (US. EPA, 2002). “Without the environmental beneﬁts provided
by AD technology, some farmers might be forced out of livestock production and a
digester is sometimes the only technology that allows growth in the livestock production

business” (Lusk, 1998, p.1-2).

The digester system is a process which includes collection and handling, anaerobic
digestion, by-product recovery and efﬂuent use, biogas recovery and biogas use. There is
signiﬁcant variability in digesters from one farm to another and it is difﬁcult to make
generalizations and comparisons. Proper maintenance and monitoring of equipment and
the microbiological conditions inside the tank itself are crucial to the success of the

digester.

Although energy production alone has not been cited as the primary motivation for the
installation of anaerobic digesters, state polices on distributed power pricing can greatly
affect the economic viability of digesters (Lazarus, 2008). In order to analysis the
situation, a multi-purpose model was developed with the capability to research the

economic effects of the three electricity purchase agreements available to digester owners

135

in Michigan. In addition to a research tool, the model can be used for outreach purposes
to examine speciﬁc systems and assist engineers in making design decisions. In Chapter
4, a series of analyses were performed to demonstrate its use and ﬂexibility. In order to
effectively summarize the key ﬁndings of this research, conclusions have been broken

down by section.

Section One

In section one, the results suggest that the business model of each electricity purchase
agreement will determine its response to price increases. For example, although all three
agreements show an increased return on investment from higher retail electricity prices, a
surplus sale agreement beneﬁts the most. This is because it is based primarily upon
offsetting on-farm electricity at the retail rate and only selling the excess production at
the locational marginal price. In addition, future energy legislation such as feed-in-tariffs
would have the most signiﬁcant effect on a buy-all sell-all agreement although lesser

beneﬁts were also observed with the other two agreements.

With the 1,000 cow example, net metering was shown to be the most preferable
agreement under the analyses tested given the prices assumed in the model. This is due
to the fact that the farm would not pay standby, administrative or system access charges
based upon the engine-generator nameplate capacity required from predicted average
yearly electricity production. As the prices were increased, however, net metering was

shown to be an inferior agreement compared to the other two options.

136

Overall, the breakeven prices calculated by the model appear to be feasible given the
trend for higher electricity prices, pending cap and trade legislation and a demand for
utility companies to comply with Renewable Portfolio Standards. Furthermore, it is
likely that increases in prices will occur simultaneously which would lower the breakeven

prices calculated in the model.

Section Two

High levels of VS loss and low TS concentrations both lead to digester heating deﬁcits
and decreased electricity production. According to the model, however, low total solids
concentrations have a more signiﬁcant impact of the NPV of the system. This is primarily
due to the fact that lower TS concentrations increase the capital costs of the digester
system with higher levels of water in the digester inﬂuent requiring larger digester tanks
and more heat. With net metering, the smaller 105 kW engine-generator made the
digester system less sensitive to these changes. This is because the extra biogas allows

the engine generator to run at full capacity despite decreases in biogas production.

In terms of online time, it was shown to have a linear relationship with average yearly
electricity production. For example, every 1% increase in online time increased the NPV
by $8,555 with a surplus sale agreement. In general, the information from this section
will allow an engineer to more effectively predict digester performance and quantify the

effects of engineering design decisions.

137

Section Three
Over a range of heard sizes, a digester investment does not achieve a positive NPV under
the current policies and assumptions. When the costs are considered on a per cow/day
basis, however, the costs appear to be low enough to justify the investment for certain
farmers. It is assumed that this cost represents the value of odor reduction to the farm

OWI'ICI'.

When considering policy recommendations, the model suggests that applying the
beneﬁcial components of net metering to the other two purchase agreements would not
be sufﬁcient to produce a positive after-tax NPV. Subsequent scenarios, however, show
an increasing beneﬁt to larger dairy farms (with surplus sale and net metering
agreements) as the standby charge threshold is increased to 400 kW and 800 kW. At a
threshold of 800 kW, a digester system begins to show positive returns on investment.
This suggests that a change in the standby charge policy of the major utility company
examined would produce signiﬁcant results for digester owners under a surplus sale or

net metering agreements.

An additional policy recommendation would be to pursue “true” net metering for digester
systems which currently operate under “modiﬁed” net metering arrangements. The model
shows, however, that this effort would not be worthwhile as only slight increases in NPV

are achieved through this policy change.

138

5.1 Areas for Future Research

Since multiple levels of detail are built into this model, future analysis could center on
further exploring tradeoffs between engineering design decisions, energy uses and
additional feedstocks. To achieve this purpose, new components to the model could also
be added. For example, programming the model to predict biogas production outside of
the target temperature range would be a valuable tool for engineers. In terms of
ﬁnancing, new mechanisms could be explored (e. g., federal investment tax credits) and

the model could be used to evaluate their effect on digester systems.

Another area of research would be to determine the optimal price of retail electricity,
carbon credits, RECS, etc. when considered in a single scenario. The current analysis
examines each element in isolation, when in reality price changes are often connected.
This would give a more accurate evaluation of the breakeven prices required for a

digester system.

139

REFERENCES

American Society of Agricultural Engineers (ASAE). 2005. “Manure Production and
Characteristics.” Publication No. D3 84.2, ASAE, St. Joseph, MI.

Beddoes, Jenifer C., Bracmort, Kelsi S., Burns, Robert T., and Lazarus, William F. 2007.
“An Analysis of Energy Production Costs from Anaerobic Digestion Systems on
US. Livestock Production Facilities.” United States Department of Agriculture
and Natural Resources Conservation Service. Technical Note No. 1.

Bracmort, Kelsi , Burns, Robert T., Beddoes, Jenifer, and Lazarus, William. 2008. “An
Analysis of Anaerobic Digestion System Costs on US Livestock Production
Systems.” American Society of A gricultural and Biological Engineers, An
ASABE Meeting Presentation: 084643.

Burke, Dennis A. 2001. “Dairy Waste Anaerobic Digestion Handbook: Options for
Recovering Beneﬁcial Products From Dairy Manure.” Environmental Energy
Company, Olympia, WA.

Chicago Climate Exchange (CCXa). 2009 “Agricultural Methane Collection and
Combustion Offset Project Protocol.” http://www.chicagoclimatex.com
/docs/offsets/CCX_ Agricultural_Methane_Final.pdf. Accessed May 15, 2009.

Chicago Climate Exchange (CCXb). 2009 “ FrequentlyAskedQuestions.” http://www.
chic agoclimatex.com/docs/offsets/CCX_Agricultural_Methane_Offsets.pdf.
Accessed May 15, 2009.

Ciolkosz, Daniel, Topper Pat, and Graves, Robert. 2009 “Biogas Digesters-Engine
Lubricating Oil for Your Digester’s Engine-Generator.” Agricultural and
Biological Engineering Cooperative Extension. Pennsylvania State University,
State College, PA.

Consumers Energy. 2009 “Rates & Rules: Electric Rate Book.” http://www.consumer
senergy.com/welcome.htm. Accessed August 8, 2009.

Crook, Michelle F. and Gould, Charles M. 2009. “On-Farm Anaerobic Digester Operator
Handbook.” Michigan State University Extension. East Lansing, MI.

140

Database of State incentives for Renewables & Efﬁciency (DSIRE). 2009. “Biomass
Gasiﬁcation and Methane Digester Property Tax Exemption.” http://www.dsire
.org/incentives/incentive.cfm?1ncentive_Code=M122F&re=1&ee=1 . Accessed
July 20, 2009.

F ulhage, Charles D., Sievers, Dennis, and Fischer, James R. 1993. “Generating Methane
Gas From Manure.” Department of A gricultural Engineering, University of
Missouri, Columbia, MO.

Gebremedhin, Kiﬂe G. 2006. “Biogas Production Model for Plug-Flow Anaerobic
Digesters.” American Society of Agricultural and Biological Engineers, An
ASABE Meeting Presentation Paper: 064172.

Gebremedhin, K.G., Wu, B., Gooch, C., Wright, P. and Inglis, S. 2005. “Heat Transfer
Model for Plug Flow Anaerobic Digesters.” American Society of Agricultural and
Biological Engineers. 48(2):777-785.

Gloy, Brent A., Enahoro, Dolapo K. 2008.“Economic Analysis of Anaerobic Digestion
systems and the Financial Incentives provided by the New York State Renewable
Portfolio Standard (RPS) Customer-Sited Tier (CST) Anaerobic Digester Gas
(ADG)-to-Electricity Program,” Department of Applied Economics and
Management, Cornell University Extension Bulletin 2008-04.

Gloy, Brent A. 2008. “Inputs for Biogas Economic Assessment.” Department of Applied
Economics and Management, Cornell University Extension Bulletin 2008-03

Jewell, W.J., P.E. Wright, N.P. Fleszar, G. Green, A. Saﬁnski, and A. Saﬁnski, and A.
Zucker. “Evaluation of Anaerobic Digestion Options for Groups of Dairy Farms
in Upstate New York.” 1997. Department of Agricultural and Biological
Engineering, Cornell University, Ithaca, NY.

Jones, D.D., J .C. Nye, and A.C. Dale. 1980 “Methane Generation From Livestock
Waste.” Publication #AE-105. Purdue University Cooperative Extension
Service, West Lafayette, IN.

141

Kaplan, Jonathan D., Johansson, Robert C. and Mark Peters. 2004 "The Manure Hits the
Land: Economic and Environmental Implications When Land Application of
Nutrients is Constrained." American Journal of A gricultural Economics 86.3:
688-700.

Kramer, J. 2004. Agricultural Biogas Casebook: 2004 Update. Madison, WI: Resource
Madison, WI: Resource Strategies, Inc.

Krich, K., D. Augenstein, J .P. Batrnale, J. Benemann, B. Rutledge, and D. Salour. 2005.
“ Biomethane from Dairy Waste. A Sourcebook for the Production and Use of
Renewable Natural Gas in California.” Report prepared for Western United
Dairymen, Modesto, CA.

Lazarus, William F. 2008. “Farm-Based Anaerobic Digesters as an Energy and Odor
Control Technology: Background and Policy Issues.” United States Department
of Agriculture. Agricultural Economic Report Number 843.

Lazarus, W. and M. Rudstrom. 2007 “The Economics of Methane Digester Operations on
Minnesota Dairy Farms.” Review of Agricultural Economics: 349-364.

Lazarus, W., and M. Rudstrom. 2003. “Financial Feasibility of Dairy Digester Systems
Under Alternative Policy Scenarios, Valuations of Beneﬁts, and Production
Efﬁciencies : A Minnesota Case Study.” Paper presented at the conference
“Anaerobic Digester Technology Applications in Animal Agriculture: A National
Summit,” Raleigh, NC (June 2-4).

Lazarus, W. University of Minnesota (UM). https://tinyurl.com/digester-xls. Accessed
October 20, 2009

Leuer, Elizabeth R., Hyde, Jeffrey and Richard, Tom L. 2008. “Investing in Methane
Digesters on Pennsylvania Dairy Farms: Implications of Scale Economics and

Environmental Programs.” Agricultural and Resources Economics Review 37/2:
188-203.

Liu, Yan, Miller, Steve A., Safferman, Steven 1. 2009. “Screening co-digestion of food
waste water with manure for biogas production.” Biojpr: Biofuels, Bioproducts &
Bioreﬁning. 3:11-19.

142

Lusk, P. 1998. “Methane Recovery from Animal Manures: The Current Opportunities
Casebook.” Report prepared for National Renewable Energy Laboratory, Golden,
CO.

Martin, J .H. and K.F. Roos. 2006. “An Evaluation of a Mesophilic, Modiﬁed Plug Flow
Anaerobic Digester for Dairy Cattle Manure.” Eastern Research Group, Inc. EPA
Contract No. GS 10F-003 6K Work Assignment/T ask Order No. 9.

Mehta, A. 2002. “Economics and Feasibility of Electricity Generation Using Manure
Digesters on Small and Mid-Size Dairy Farms.” Energy Analysis and Policy
Program, Department of Agricultural Economics, University of Wisconsin,
Madison, WI.

Michigan Department of Agriculture (MDA). 2009. “Frequently Asked Questions about
Anaerobic Digesters (ADs). http://www.michigan.gov/documents/mda/MDA
AnaerobicDigesterFAQ_1895l9_7.pdf. Accessed May 22, 2009.

Michigan Department of Environmental Quality (MDEQ). 2009. “Organic Residuals
Exemption On-F arm (Anaerobic Digestion)” http://www.michigan.gov
/documents/deq/DEQ-WHMD-On-Farm-Anaerobic—Digestion_266497_7.pdf.
Accessed June 10, 2009.

Michigan Public Service Commission (MPSC) 2009. “Direct Testimony of Hubert W.
Miller III on Behalf of Consumers Energy Company.” http://eﬁle.mpsc.state.mi.us
/eﬁle/docs/15805/0024.pdf. Accessed May 15, 2009.

Miner, Ronald J. 1997 "Nuisance Concerns and Odor Control." Dairy Science Journal
80:2667-72.

Natural Resources Conservation Service (N RCSa). Undated “Anaerobic Digester Fact
Sheet.” http://www.mi.nrcs.usda.gov/teehnical/engineering/animal_waste.html.
Accessed July 12, 2009.

Natural Resources Conservation Service (N RCSb). 2005 “Conservation Practice
Standard Anaerobic Di gester—Controlled Temperature No. Code 366.”

143

www.wcasfmra.org/biogas_docs/NRCS%20Biogas.pdf. Accessed April 10,
2009.

National Renewable Energy Laboratory (N REL). 2009.“Geothermal Heat Pumps.”
http://www.nrel.gov/learning/re_geo_heat_pumps.html. Accessed June 20, 2009.

Ontario Ministry of Agriculture Food & Rural Affairs. 2009. “Energy Yields from a
Farm-Based Anaerobic Digestion System.” http://www.omafra.gov.on.ca/english/
engineer/facts/enyields.htrn. Accessed June 20, 2009.

Persson, S.P.E., H.D. Bartlett, A.E. Branding, and R.W. Regan. 1979. “Agricultural
Anaerobic Di gesters Design and Operation.” Pennsylvania State University
Agricultural Experiment Station Bulletin 827, November.

Pennsylvania State University Department of Agricultural and Biological Engineering.
2007. “Penn England Farm Case Study.” http://www.biogas.psu.edu/casestudies
/pennengland.pdf. Accessed July 10, 2009.

Pronto, Jennifer and Gooch, Curt. 2008. “Anaerobic Digestion at Sheland Farms, Inc.:
Case Study.” Department of Biological and Environmental Engineering, Cornell
University. http://www.manuremanagementcomell.edu/Docs/Sheland_case%20
study_original.pdf. Accessed July 10, 2009.

Rosentrater, Kurt A., Heath R. Hall, and Conly L. Hansen. 2006. "Anaerobic Digestion
Potential for Ethanol Processing.” American Society of Agricultural and
Biological Engineers. An ASABE Meeting Presentation Paper Number: 066167.

Safferman, Steve and F aivor, Louis. 2008. “Anaerobic Digestion Biogas Potential
Assay.” Michigan Dairy Review. https://www.msu.edu/~mdr/voll3no4/
safferman.html. Accessed January 15, 2009.

Scott, Norman, and Jianguo Ma. 2004. “A Guideline for Co-Digestion of Food Wastes in
Farm-based Anaerobic Digesters.” Cornell Manure Management Program, Fact
Sheet FW-2. www.manuremanagement.comell.edu/.../Food%20Waste%20fact
%20$heet%202.pdf. Accessed July 16, 2009.

State of Louisiana Division of Administration Ofﬁce of Community Deve10pment. 2007
“Engineering Fee Schedules and Policies for the LCDBG Program.” P. 263-271

144

Steffen, R., Szolar, O., and Braun, R. 1998. “Feedstocks for Anaerobic Digestion.”
Institute for Agrobiotechnology Tulin. University of A gricultural Sciences
Vienna.

US. EPA. 2006. Winter AgSTAR Digest. AgSTAR Program. http://www.epa.gov/agstar/
. Accessed June 16, 2009.

US. Energy Information Administration (U .S. DOE). 2009 “Electricity Prices: US. Data
Projections.” http://www.eia.doe.gov/oiaf/aeo/excel/aeotab_8.xls. Accessed May

1, 2009

US. EPAa. 2009. “Anaerobic Digestion Capital Costs for Dairy Farms.”
http://www.epa.gov/agstar/pdf/digester_cost_fs.pdf. Accessed March 20, 2009.

US. EPAb. 2009. “Operational Anaerobic Digesters, Sorted by State.” AgSTAR
Program. http://www.epa.gov/agstar/pdf/digesters_operational.xls. Accessed October

9, 2009.

US. EPAc. Undated. “FarmWare 3.1” AgSTAR Program. Washington, DC. Available
at http://www.epa.gov/agstar/ resources/handbook.html. Accessed April 24, 2009.

US. EPA. 2002. “Managing Manure with Biogas Recovery Systems: Improved
Performance at Competitive Costs.” AgSTAR Program, Washington, DC.

US. EPAd. 2009 “High Global Warming Potential (GWP) Gases.” http://www.epa.gov/
highgwp/. Accessed September 5, 2009.

Wilkie, Ann C. 2005. “Anaerobic Digestion of Dairy Manure: Design and Process
Considerations.” In Dairy Manure Management: Treatment, Handling, and
Community Relations. Natural Resource, Agriculture, and Engineering Service,
Cornell University, Ithaca, NY.

Gipe, Paul. 2009. “Wind Works- What are Feed in Tariffs?” http://www.wind-
works.org/artic
les/feed_laws.htm1. Accessed April 10, 2009.

Wright, P. 2001. “Overview of Anaerobic Digestion Systems for Dairy Farms,” Natural
Resource, Agriculture and Engineering Service ﬂV RA 1554 43).

145

11111111

03063 5225

H“
Sjljj
R"
av...“

N”
U"
E"
Tjjt

3 293

1111111111