1 . . V 16‘ 79!. 5.7.41‘ at I 5...; .
Lafiti 1...: LI. 39...... v .
{ff-51 1v 7: fly ...

.. f. 73.8.. n v

. I . ‘

. 73‘ .5; I.

[lint-7'

21.1!!! \ovnv;\#n
u A.

.. o. ,1. I). . 1'.

5:9; ... .‘l

.3

.. v4
Iérprirni
r.?.l.|.n6..,

n p . u. I! v-

. 3%.):ih31.
a". 5.21. ..

s .90 rts.\:xr.

.. . . .. V

, ‘51:..451!
or V .III

1...

,QIOVSI VII-
.inaii .312...) 1.3....
....iv: .3\2...r|.

. 3| .r...
. .2. ' PQVII A,
. . I ,v n 1 . bV‘:-;."‘i V (In. 9’
hf V: innit... 4.4:.r.....,.....§§hn.:.x. 375“. 3.. v.3 ‘3.
13.. ,.a ‘ . ‘ i 3‘55; I! . I. e . .13.} 3 .:
.. \‘§:\u)bimlai LN“)? ,I. rib”. 3722a. . .25.... a A a
,p , i“. o: . . V . v ,
t. .::.‘\. .- 2A..» 9.1.1 u. a x .1 ‘

5.!51. .\ r.
F 33):; If:
.1... 5...! , .
z 3.2! .: . I 7 .I

, . finikliyri35l‘51‘x‘.‘ I :9».

3:: 1:11. sstltgtu; .. ~
altlaiii .. V1.3... :k!c....b:..l Z: .
.l . Ian. I. .2)...» .1 t-Qllxips. . it 1,315.»..3 :.
. . l {I}, V. .72.. . ...!!. ‘ . brat C.

fir...
4-é'fntfi‘ It I:— . ll.

 

,abbref
:. . 7.x... ‘
«x:

tn! 1.. ~I9
$3.0}... ~9-
.i. In. x

‘ 7,211.. i.
:2... -I... xtl’. :1.
V; .3- . 1

a. .: l:
2.1::

:5 ...
33.. 19-!!! It?

. V, t.’:. .1
t. f
‘ .£ :1 ” . .
O s i. Q. .n‘"\ ‘ .dop’
‘ v! . t. v) 3‘» .
IRS}. .34.... {A ‘ ,
Infra:

ERIE; it‘lrz. V
.fl.)l. A Ir... .1!
12s. :.\:~.y0)‘.rli;b
iii;
auhf‘Vlil . . :-
l.h‘v7 .l‘ (is: &{.c.
1!! 7.:
. 1... 1....
t: .

5 .1 \A :. i?v\£’
3419-! L.

A 9-. vzx
». 2:11... In
(platff,
at... '5

..v
I 5.312;; t)?! .v:
1...}.3

Z .
. :5...- ;

Ls $.35}. 1! .
..

.
A... 1.13...

 

..
I: 7375.. 7

 

ANSTATE UN VERS

IIIIIIIIIIIIIIII lIII IIIIIIIIIIIIIIIIIIIIII

31293 00881 2657

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

THE DESIGN, FABRICATION AND TESTING ISSUES OF
AN ELECTRORHEOLOGICAL FLUID-FILLED COMPOSITE
HELICOPTER ROTOR BLADE

presented by

DAVID MARK MOOREHOUSE

has been accepted towards fulfillment
of the requirements for

MASTERS degreein MECHANICAL ENGINEERING

 

 

C1.\/. gandh'

Major professor

Date 71/2- 71/? 3

07639 MS U is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY
Mtchlgan State
UnNersIty __

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

________________—————————-——"1___________————————————"——‘
I DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ ._ -- I. <_ _.__. _______"—__ __ 1
.a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
cmwnpma-pJ

,Efir...

 

THE DESIGN, FABRICATION AND TESTING ISSUES OP
AN ELECTROREEOLOGICAL PLUID-PILLED COMPOSITE
HELICOPTER ROTOR BLADE

BY

David Mark Moorehouse

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

MASTER OF SCIENCE

Department of Mechanical Engineering

1993

ABSTRACT
THE DESIGN, FABRICATION AND TESTING ISSUES OF

AN ELECTROREEOLOGICAL FLUID-FILLED COMPOSITE
HELICOPTER ROTOR BLADE

BY
David Mark Moorehouse

Major Professor: Dr. Mukesh Gandhi

Helicopter manufacturers continually strive to achieve
increased performance in the rotor blades. One area of
focus has been control of vibration. One proposed method of
achieving vibration control is through the use of
electrorheological (ER) fluids. This thesis describes the
design, fabrication and testing of an electrorheological
fluid-filled composite helicopter rotor blade which
successfully exhibited vibration damping via ER fluids. The
blade contained two parallel sheet aluminum electrodes
encapsulated in urethane rubber. The space between the
electrodes was filled with an ER fluid and the assembly was
wrapped in glass fiber/epoxy resin skins. The blade was
tested in a non-rotating environment on an exciter. The
test results showed the deflection of the tip to reduce by
14% and the velocity of the tip to reduce by 33% when
applying voltage to the electrodes and comparing the results
with the unpolarized case. Many design and manufacturing

issues are discussed.

TABLE OF CONTENTS

LIST OF FIGURES . . . .

INTRODUCTION . . . . .

DESIGN ISSUES . . . . .

Loading . . . . .

Electrorheolcgical

(ER) Mechanism

Design-for-Manufacturing . . . . .

Weight . . . . . .

Cost . . . . . . .

GENERAL PHYSICAL ATTRIBUTES . . . . . .

Leading Edge . . .
Skin Material . .
Filler Material .

Spar . . .

Airfoil . . . . .
Root End . . . . .
Blade Tip . . . .
Tip Weight . . . .

Aspect Ratio . . .

vi

IDWQUI

10

10

12

12

13

14

16

17

17

18

DESIGN CONCEPTS . . . . . . . .

Continuous Skin Blade/Urethane Filler/Integral ER

Cavity . . . . . . .

Continuous Skin Blade/Polymethacrylimide
Filler/Separable ER Cavity
Modular Blade With Composite Spar/Trailing

Skin/Separable ER Cavity

ELECTRORHEOLOGICAL ROTOR BLADE CONSTRUCTION

Electrorheological (ER) Cartridge

Silicon Carbide Leading Edge .

Glass/Epoxy Skins Composite Skins

Urethane Rubber Filler . .
Airfoil . . . . . . . . .

Aspect Ratio . . . . . . .

ER CARTRIDGE FABRICATION AND TESTING

Multi-Layer ER Cartridge .

Single Layer ER Cartridge

ROTOR BLADE TOOLING FABRICATION
Patte rn O O O O O O O O O

Molds . . . . . . . . . .

ROTOR BLADE FABRICATION . . . .
Cartridge Encapsulation .
Skin “yup O O O 0 O O O 0

iv

19

20

21

22

24
25
27
28
28
29

29

3O

31

35

4O
40

42

47

47

50

Silicone Carbide Surface

ROTOR BLADE TESTING . . . . .

MANUFACTURING ISSUES . . . .
Manufacturing Processes
Tooling and Equipment .

Quality Control . . . .
CONCLUSIONS . . . . . . . . .
APPENDICIES . . . . . . . . .

APPENDIX A:

APPENDIX B:

RESPONSE . . . . .

GENERAL REFERENCES . . . . .

Coat Application

NACA 0012 AIRFOIL DATA

MULTI-LAYER ER CARTRIDGE

51

53

57

57

61

64

66

68

69

7O

71

LIST OF FIGURES

Figure 1: Electrorheological Fluid Effect

Figure 2: Important Airfoil Geometric Quantities .

Figure 3: Aerodynamic Forces Developed by the Airfoil

Figure 4: Continuous Skin Blade/Urethane

Filler/Integral ER Cavity . . . . . .

Figure 5: Continuous Skin Blade/Polymethacrylimide

Filler/Separable ER Cavity . . . . . .

Figure 6: Modular Blade with Composite Spar, Trailing

Edge Skin and Separable ER Cartridge.

Figure 7: ER Fluid-Filled Rotor Blade Cross-Section

Figure 8: ER Cartridge Cross-Section . . .

Figure 9: Multi-Layer ER Cartridge . . . .

Figure 10: Bottom Side View of Multi-Layer ER

Cartridge Prior to Pouring . . . . . .

Figure 11: Multi-Layer Cartridge After Pouring .

Figure 12: Multi-Layer Cartridge Assembly Showing

Sheet Wax Spacers Prior to Pouring . .
IFigure 13: Single Layer Cartridge Response
Voltage . . . . . . . . . . . . . . .
‘Figure 14: Single Layer Cartridge Response
Volts . . . . . . . . . . . . . . . .

IFigure 15: Pattern Template . . . . . . .

vi

15

21

22

23

24

26

31

33

33

34

38

39

4O

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Pattern Assembly . . . . . . . . . . . . .
Pattern Assembly After First Spline . . . .

Pattern Box Prior to Surface Coat . . . . .

Pattern Box After Surface Coat . . . . . .

Casting Sand/Resin in Pattern Box . . . . .
Both Mold Halves After Demolding . . . . .
Cartridge After Encapsulation into Airfoil

Blade Droop and Flexibility After

Demolding . . . . . . . . . . . . . . . . . . . .

Figure
Figure
Figure
Figure
Figure

Figure

24:

25:

26:

27:

28:

29:

Volts

Plan View of Blade . . . . . . . . . . . .
Root End View of Blade . . . . . . . . . .
Tip View of Blade . . . . . . . . . . . . .
Blade Tip Velocity - No Volts, 3000 Volts .
Blade Tip Velocity - No Volts, 2000 Volts .

Blade Tip Deflection - No Volts, 3000

vii

41

42

43

44

45

46

49

50

51

52

52

54

55

56

INTRODUCTION

Helicopter manufacturers are constantly seeking new
design and manufacturing options that offer greater levels
of performance in the main rotor blades. The inherent
complexity of both designing and manufacturing rotor blades
yields many factors to consider when attempting to improve
performance. This multitude of factors makes improving
performance a difficult task. Some of these factors are
static loading, dynamic loading, vibration, fatigue,
temperature, moisture, hub attachment, blade flexibility,...
Of these issues, one area of focus has been control of
vibration due to flapping. This attention is because rotor
blade performance is limited by vibration.

The use of composite blades has improved the ability of
a rotor blade to dampen vibrations simply due to the
inherent structural damping nature of composite materials.
Additional advantages of composite blades are decreased
weight, greater resistance to crack propagation, lower
vulnerability to corrosion, improved adaptability to complex
geometries, improved aerodynamic performance, and reduced
manufacturing costs. Although composites offer all of the

above advantages and dampen vibrations better than metals,

2
rotor blade designers still desire to have improved damping
of composite helicopter rotor blades.

One proposed method of achieving vibration control is
through the use of electrorheological (ER) fluids.
Electrorheological fluids are suspensions of fine particles
in a non-conducting oil. A commonly used ER fluid is corn
starch mixed with silicone oil. These fluids have the
ability to change viscosity as a function of the electric
.field in which they exist. Upon application of an electric
field, the polarized particles align and the fluid becomes a
gel-like solid. Once the field is removed, the fluid
reverts to its original viscosity. Figure 1 below shows

this ER effect.

ND VOLTAGE VOLTAGE

 

 

‘—1:L_ _:EE-

 

 

 

 

Q

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1: Electrorheological Fluid Effect

3

Using ER fluids inside a rotor blade would create a
variable stiffness blade which would enable one to tailor
the overall or local stiffneSs to respond optimally to
varying operating environments.

Most work performed on ER fluids to date has been with
small parallel electrodes on a laboratory scale. There has
been little effort to advance these materials towards a
production environment by fabricating larger test specimens
and addressing some of the manufacturing issues.

This thesis describes a research effort which
successfully utilized ER fluids in a scaled-down rotor blade
to control vibration. Many of the design issues and
physical attributes of ER blades are explained. Some broad
design concepts are presented and a final design
construction is generated based on the concepts. All
aspects of tooling and actual blade fabrication are
detailed. The testing procedure and results are discussed.
Some of the manufacturing issues that surfaced through the
research are highlighted. Finally, conclusions are made

based on the research.

DESIGN ISSUES

Designing composite helicopter rotor blades is a very
complex task. The requirement of vibration control via ER
fluids adds an entirely new dimension to the design process.
As a result the design process must consider many new
issues. This section outlines the critical design issues of
an ER fluid-filled composite blade. The issues fall into
the categories of loading, ER mechanism, design-for-

manufacturing, weight, or cost.

Loading

The loads induced on rotor blades may be structural,

environmental, fatigue, or galvanic.

figzggggzgl; These types of loads include all static loads
such as bending moments in the hub area due to the weight of
the blade. Other types of loads induced include dynamic
loading due to the centrifugal forces, vibration and

aerodynamic loads.

5
fingi;gnmgg§gl; There are many types of issues resulting
from the environment which may cause degradation of the
blade. These types are as follows: ‘
1) Thermal 5) EMI Shielding
2) Moisture 6) Deicing

3) Ultraviolet Radiation 7) Lightning Strike
4) Corrosion

zatiggg; Due to the cyclic loading imparted on a rotor
blade, fatigue is a very important consideration. The mode.
of failure must also be considered (catastrophic versus non-

catastrophic).

figlgggigy This type of loading or degradation would exist
when two dissimilar materials such as aluminum and graphite
come in contact in the presence of an electrolyte, such as

salt water.
Electrorheological (ER) Mechanism

Fundamentally, the requirements for an electrorheological
effect to occur are simply two isolated electrodes made of a
highly conductive material separated by a constant gap of
.020" - .040". The gap is filled with the ER fluid. Some
of the new issues associated with an ER mechanism are

described below.

6
n1a2sr2g2_nas2rialLSeasiaszfiizezlnsnlafignl Typically.
sheet aluminum .015" - .030" has been used as the electrode
material due to its high electrical conductivity, low cost
and availability. Spacings of .020" - .040" have been used
with good success. Gaps larger than .040" begin to show
deterioration of the ER effect. The width of the electrodes
must also be selected. Electrode widths from 1" to 12"
square have been successfully used and it appears that the
only limitation of electrode width is the assurance that the
two electrodes do not come into contact and that the gap
size remains constant. Finally, the electrodes must be
insulated from each other as well as surrounding materials
which may be damaged by the presence of an electric field.
This is usually done with a silicone or urethane based

rubber.

Ilni§_nrain_£lngzzill_zlusl Once the rater blade is
completely fabricated, there must be provisions for filling

the ER cavity between the electrodes with fluid. It also
may be necessary to drain the fluid on occasion hence there

should be a drain plug.

2212:.S22211132g21r22222a1 Voltages up to 3000 volts have

been used to produce an ER effect. The voltage requirements
must be evaluated so that appropriate hardware such as

transformers, power supply, etc. can be determined.

7
11;g_ngzgggg; There must be wires from the transformer to
the electrodes. The wires and connections should not be
exposed to the environment yet should be accessible in the
event of a repair. The connections shall also remain secure

during all rotational speed and angles of the rotor blades.

Localiged Contrg; Vi; Separatg ER Zones. It may be desired

to only vary the stiffness of certain regions of the blade
for optimal performance. This could be done with separate
ER zones which would be insulated from adjacent zones.

There would be separate wires and controls for each zone.

Packaging; Adding all of these additional components
(electrodes, wires, etc.) to the blade may be a cumbersome
task. It is desired to have the blade packaged in such a
way that in attaches on to the hub similarly to existing
blades with one wire connector which would also need to be
connected. The more consolidated the wires, controls, etc.
are, the more appealing this concept would be to helicopter

manufacturers.

gon;rgl_§1g§gg; There would need to be additional controls
on the instrument cluster to vary the voltage in the
different ER zones. Also needed would be some sort of
sensing mechanism to inform the pilot when and how much to

vary the voltage. This sensing mechanism could utilize

8
optical fibers, however, addressing that issue would be a

study in itself.
Design-for-Manufacturing

These new issues involved in designing an ER fluid-
filled rotor blade also make manufacturing more difficult.
The design process must include evaluation of design-for-
manufacturing concepts such as processing and tooling. It
should be determined beforehand whether a research effort of

this nature will be either process or tooling prohibitive.

Processing; Typical composite blades are fabricated using
cocured or precured and bonded techniques. The cocured
process entails curing the composite skins and in the same
cure cycle, bonding them to a filler such as honeycomb or
foam. Precured techniques involve precuring the skins and
bonding them to the core in a subsequent operation. Adding
an ER mechanism will require a thorough process evaluation.
This evaluation will entail determining whether the process
will be cocured or precured and outlining the manufacturing

sequence to determine feasibility.

Tooling; The evaluation of tooling required will be done
concurrently with the process evaluation. When each

manufacturing step is defined, a list of tooling for that

9
step should be generated so that a continual feasibility

study can be performed.

Weight

Weight is an important factor in rotor blade design. The
blade shall have enough weight to provide sufficient
rotational inertia to lift the helicopter yet not be too

heavy to decrease fuel consumption or performance.

Cost

The issue of cost must be thoroughly evaluated via a cost
trade-off compared to existing composite blades. It should
be approximately determined how much the increase in
performance will cost on a production basis due to the

increase in material cost and processing time.

GENERAL PHYSICAL ATTRIBUTES

There are many attributes which are common to all rotor
blades and it is these attributes which the design will
define based on the design issues covered in the previous
section. This section discusses these attributes and some
of the issues pertaining to the applicability of these
attributes to an ER fluid-filled blade. The ER mechanism
discussed earlier is also a physical attribute of the blade,

however, it obviously is not common to all blades.
Leading Edge

Due to the operating environment of rotor blades, the
leading edge must be extremely erosion resistant. This is
primarily to protect wear caused by sand, rain or dust.
There are several materials and methods that have been or
can be used on the leading edge to provide erosion
resistance. One important characteristic of a leading edge
is that it should be repairable or replaceable. Eventually,
any of the materials will wear necessitating restoration of
the erosion resistant cap. Materials that can be used as an
erosion resistant leading edge include metals (titanium or

nickel), polyurethane tape, elastomeric sheet such as Viton,

10

11
kevlar/epoxy, thermoplastic film such as
polyetheretherketone or polyetherimide or a silicon
carbide/epoxy surface coat.

Although metallic leading edges are good erosion
protectors, the tooling costs are high and the blade must be
removed to replace the cap. Metallic leading edges require
surface preparation, NDT and bonding processes which add
cost and complexity as well.

The polyurethane tape typically is backed with an
acrylic adhesive and it can be replaced while still on the
aircraft.

Elastomeric sheets have been successful in this
application. They may either be cocured with the composite
skins or bonded on in a subsequent operation.

Kevlar/epoxy requires a separate mold in which the
leading edge can be premolded. It is then bonded and/or
fastened to the blade in a subsequent operation.

Thermoplastic films make excellent erosion barriers but
are a poor bonding substrate. This disadvantage has made
them unsuitable for most applications.

The silicon carbide/epoxy surface coat is applied to
the female airfoil shape mold prior to laminating.. During
the cure, the silicon carbide flows partially into the first
few plies and becomes an integral part of the laminate. An
alternative application technique is to abrade and solvent
clean the outside of the composite skins after the cure and

spray the coating on. This coating is also easily

12
repairable as it can be splined on the leading edge and will

cure at room temperature.
Skin Material

Rotor blade skins have been made out of kevlar,
graphite, glass and boron. The fiber orientations are
typically :0 or :45 degrees with 0 degrees being spanwise.
In many cases unidirectional graphite will be put in
localized areas requiring higher stiffness in the spanwise
direction. The most suitable fiber and orientation depends
on the structural, weight, and cost requirements. Epoxy is
generally the matrix system used. To properly select the
fiber type and orientation for an ER fluid-filled blade, the
stiffness of the ER mechanism with and without voltage

applied across will have to be characterized.
Filler Material

The filler material is the material used to fill the
empty space between the skins, spar and trailing edge. It
should be lightweight, have a high compressive strength, and
shall have processing requirements conducive to the overall
processing of the blade. Materials used by blade
manufacturers have typically been either honeycomb or foam.
Honeycomb must be machined prior to bonding. The honeycomb

can either be a fiber reinforced polymer such as Nomex or it

13
can be aluminum. An important consideration when selecting
the most suitable honeycomb is the potential for galvanic
corrosion. If it is aluminum honeycomb, it has to be coated
with a corrosion resistant primer prior to bonding. The
foam is usually an epoxy based, microballoon filled
compound. A unique foam sometimes used for this application
is Rohacell foam made by Rohm Tech, Inc. It is a
polymethacrylimide lightweight foam capable of exerting
pressure on the skin laminae while being cocured to them. A
scrim-supported film adhesive is placed between the foam
core mandrel and the skin plies.

The ER blade will be significantly different than
traditional blades from a filler standpoint because the ER
mechanism will occupy most of the space between the skins.
The material will have to be compatible with both the skins
and the insulation material on the outside of the
electrodes. A third possible filler material is the same
rubber used as insulation for the electrodes. Using rubber
rather than honeycomb or foam reduces the stiffness which
would further enhance the contribution of the ER mechanism

to the overall stiffness.
Spar
The spar is the spanwise structural portion of the

blade whose purpose is to counteract centrifugal forces as

well as bending and twisting moments. Spar materials have

14

typically been epoxy resins reinforced with glass, graphite
and kevlar fibers. Due to the purpose of the spar, the
fiber orientation is always 0 degrees (spanwise). The
configurations of the spars are C spar, D spar, and multi-
tubular spar. These spars are usually premolded separately
and post-bonded in the blade assembly. The necessity of a
spar in an ER fluid-filled blade needs evaluation. The
stiffness of the ER mechanism may be sufficient to eliminate

the need for a separate spar.

Airfoil

There are several different sources for airfoil shapes.
One popular source is the National Advisory Committee for
Aeronautics. They publish a vary complete catalog including
x-Y coordinates for numerous airfoils. Much of the data
generated by the helicopter manufacturers is proprietary and
more difficult to access. Figure 2 contains a diagram of an
airfoil showing the important geometric quantities. These
are the chord, chord line, camber line, and maximum
thickness. The chord line is the line connecting the most
extreme points at the leading and trailing edges. It is
considered to be the reference line of the airfoil. The
camber line is the line equidistant from the upper and lower
flow lines and is indicative of the amount of curvature in
the airfoil shape. Highly cambered airfoils generally
;produce more lift than slightly cambered ones. If an

15
airfoil is symmetric about the chord line, it has no camber.
Maximum thickness is the greatest thickness in the airfoil
measured perpendicular to the chord line. This term is

expressed in terms of percent chord.

t Maximum thickness

 

 

 

 

 

 

 

 

l Chord fine

1 Camber line

yh)

x
lyb) f
F
Leading edge Trailing edge
x:O x=t0

Figure 2: Important Airfoil Geometric Quantities

The purpose of an airfoil is to generate a mechanical
force as a result of relative motion between it and a fluid.
Relative motion implies that either the wing is stationary
(airplane) or the fluid is stationary (helicopter). The
primary force generated by an airfoil is called lift.
Secondary forces also generated by an airfoil are drag
forces and pitching moments. These forces are shown in
Figure 3. The point of application of the resultant force

is at the one-quarter chord point on the airfoil.

16

Lift

     

Pitching moment

 

 

 

 

 

\\~—d’
/ $014,]

 

 

(3 a»

Figure 3: Aerodynamic Forces Developed by the Airfoil

It is recommended to select a symmetric airfoil for an ER
blade to simplify tooling. Also, it may be beneficial to
select an airfoil with a relatively small thickness. This
would enhance the ER effect since the moment of inertia of
the blade would be less. One possible airfoil is NACA 0012.
This designation indicates that the maximum thickness of the
airfoil is 12% of the chord, which is the lowest of standard

airfoils.

Root End

The root end is the portion of the blade which

transitions from the constant cross-section region to the

hub interface hardware. There are several different methods

17
of attaching the blade to the hub. The most common in use
are the "wrap around lug" concept and the simple "bolt
through" concept. Regardless of the attachment method, the
transition from the blade airfoil and root should provide a
smooth and continuous path for the centrifugal forces, twist
forces, bending moments and drag forces. A root end for an
ER blade will require provisions for the lead wires to be
routed from the electrodes to the electrical circuit at the

hub.

Blade Tip

Blade tips are important because they sustain the
highest dynamic pressures, are at the origin of the
formation of the tip vortices, and generate most of the
rotor drag and noise. The concepts used for attaching the
tip caps all fall under the category of mechanical fastening
or adhesively bonding. Either of these techniques should be

suitable for an ER blade.

Tip Weight

Tip weights are needed to move the center of gravity of
the blade near the one-quarter chord which is needed for
stability and also to give the blade adequate rotary inertia
capability to lift the helicopter. For the non-rotating

tests that the ER blade went through, the requirements of a

18
tip weight do not exist hence the ER blade does not have a
tip weight. Generally, tip weights are machined out of a

metal such as brass.

Aspect Ratio

The aspect ratio of the blade is the radius of the
blade divided by the chord (r/c). This generally falls with
the range of 10-18. Many performance requirements influence

the optimal aspect ratio.

DESIGN CONCEPTS

Based on all of the design issues discussed and the
physical attributes described, three different
configurations of an ER fluid-filled composite rotor blade
have been conceptualized. The features of these blade
concepts may be interchanged resulting in a multitude of
design possibilities. Also, some features may be
eliminated. For example, a spar may not be needed for
spanwise stiffness depending on the stiffness of the ER
fluid mechanism. The concepts consist of cross-sections
only. It is not necessary to focus on the blade tip and
root end until a working cross section is discovered. The
blade only needs to be adequately fixtured to the
appropriate test stands. These concepts consist of the

following:

1. Continuous skin blade/urethane filler/integral ER
cavity.

2. Continuous skin blade/polymethacrylimide filler/
separable ER cavity.

3. Modular blade with composite spar/trailing edge skin/
separable ER cavity.

19

20

Continuous Skin Blade/Urethane Filler/Integral ER Cavity

Figure 4 shows the concept of a continuous skin wrapped
around a urethane filler with an integral ER cavity. The
skins are continuous around the airfoil and are fabricated
as a detail as is the composite spar. A tip weight is
machined to fit in between the composite C-spar and the
skin. The electrodes and spar are fixtured and suspended
within the airfoil and urethane rubber poured in around the
electrodes to serve as a filler. There is a filler material
of some sort in between the electrodes during this operation
to keep the ER cavity empty. The filler material could be
sheet metal or sheet wax. Once cured, the filler is removed
(or melted in the case of sheet wax) and the space between
the electrodes is filled with an ER fluid and sealed off.
The leading edge of the blade has a formed titanium detail
bonded on. Advantages of this configuration are that the
continuous skins do not require any mechanical fastening or
bonding, both of which add cost and uncertainty. The
integral ER cavity makes fabrication easier but does not

facilitate easy service as a modular unit would.

21

UP VEIGIT

ER FLUID CAVIYY
C’SPAR
ELECTRODES

CENTIWS
CDPOSITE SKINS

mm LEADIMS ENE

Figure 4: Continuous Skin Blade/Urethane Filler/Integral ER
Cavity

Continuous Skin Blade/Polymethacrylimide Filler/Separable SR
Cavity .

Figure 5 shows the concept of a continuous skin wrapped
around a polymethacrlyimide foam filler and a separable ER
cavity. The electrodes are encapsulated in an ER
"cartridge" made of urethane rubber with a rectangular cross
section. The cartridge is placed inside a molded cavity
within the polymethacrylimide foam. The solid composite D-
spar with molded-in tip weight and the foam are used as a
mandrel for the continuous skin plies. After curing the
skins, the ER cavity is installed. Finally, a coating of
epoxy-based silicon carbide is sprayed on the leading edge
for erosion protection. In addition to the advantages
achieved with continuous skins, the ER cartridge can be slid

out of the blade for servicing. This concept is more

22
difficult to manufacture than the previous concept due to

the premolding and bonding of the foam and spar.

AR

SP
oz cavmcs
nevus" .
//// ' '

ml j\>

 

  

 

 

 

 

 

 

 

 

 

 

 

 

E?* "
~ \-mnmn1rmm
- "macaw
cumulunnm: “wan5 L_
, uwmmcmnmn
_ corms": was a c ARTRIDGE)

Figure 5: Continuous Skin Blade/Polymethacrylimide
Filler/Separable ER Cavity

Modular Blade With Composite Spar/Trailing Edge
Skin/Separable ER Cavity

Figure 6 shows the modular blade concept a composite
spar, trailing edge skin and a separable ER cavity. This
concept is totally modular with the trailing edge composite
skins, kevlar/epoxy leading edge with tip weight, ER
cartridge and aft honeycomb or foam filler all being
separate pieces which are prefabricated and assembled
together. The primary advantage is serviceability. The
tooling and processing is more complex due to the separate
pieces and the need to mechanically fasten the skins to the

leading edge.

23

REM/EPOXY SPAR AND LEADING ENE

TIP HEIGHT [ER CAVITIES
////

DIET WIND! mucus

 
   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

' \_.

pm... L

CMSITE SKINS

 

 

FONS?“ or run:

mam RUBBER
(ER CARTRIND

BlTED AND ENDED

Figure 6: Modular Blade with Composite Spar, Trailing Edge
Skin and Separable ER Cartridge.

ELECTRORHEOLOGICAL ROTOR BLADE CONSTRUCTION

The design issues and concepts were evaluated and a
first iteration ER fluid-filled rotor blade was
conceptualized and defined in terms of its physical
attributes. These attributes are described in this section.
Figure 7 Below shows a_cross-section of the blade.

Urethane ER Cartridge
/ ,F Urethane Filler-‘02 Aluminum Electrodes
£4; *: 2”

LGlass/Epoxy Skins with Sir: SurFace Coat
.04 ER Cavity

 

 

 

Figure 7: ER Fluid-Filled Rotor Blade Cross-Section

For various reasons, some of the attributes were not
incorporated into the ER blade. These are the spar, root
end, tip cap and tip weight. The ER blade does not have a
spar because it was desired to maximize the effect of the ER
mechanism had on overall blade stiffness. The ER blade was
made to evaluate non-rotating vibration tests only, hence,
it did not need a root end. A root end for an ER beam being
used for rotating tests will require provisions for the lead

wires to be routed from the electrodes to the electrical

24

25
circuit at the hub. The existence of a tip cap would not
have had any bearing on achieving vibration control with ER
fluids. Tip weights are needed to move the center of
gravity of the blade near the one-quarter chord which is
needed for stability and also to give the blade adequate
inertia capability to lift the helicopter. For the non-
rotating tests that the ER blade went through, the
requirements of a tip weight do not exist hence the ER blade

does not have a tip weight.

Electrorheological (ER) Cartridge

The design presented in this paper utilizes an
electrorheological cartridge. The cartridge is the core of
the blade and is a separate assembly which is fabricated and
tested independently prior to encapsulating it into the
blade. Since a blade of this nature has never been designed
or fabricated before, it was desirable to have a somewhat
modular design which would facilitate easy design
verification and change implementation. The cartridge
consists of two electrodes made out of .020" 6061 aluminum
sheet and a gap of .040". The electrodes are separated by
silicone sheet strips around the periphery. The entire
assembly is encapsulated in castable urethane rubber and the
finished cartridge has a rectangular cross section. The

urethane is cast on the four lengthwise edges and one end

26
simultaneously. Figure 8 shows a cross-section of the ER

cartridge.

 

-<le £175 4-—

 

 

 

-1 2.25 ’-

 

 

 

 

 

h‘I——————— 11§7S -—I————4-

 

 

 

.1875 J

.080 Thick Electrodes
.040 Thick Gap

Figure B: ER Cartridge Cross-Section

Aluminum was selected for an electrode material because
of its cost, availability and electrical conductivity. The
electrodes have tabs at the root ends to allow for
attachment for lead wires. The gap of .040" was selected to
facilitate filling the cavity with the ER fluid and make it
more difficult for the electrodes to touch during testing.
Prior work has proven that gap dimensions up to .040" can
exhibit the ER effect and that this effect deteriorates with

larger gap sizes. Silicone sheet was used for a

27
separation/insulation material because of its insulation
properties, cost, availability, and consistent thickness.
The sheet was cut into .1875" strips and bonded to the
electrodes around the periphery with an epoxy-based glue.
For testing purposes, a fourth strip was placed between the
electrodes at the root end after filling and clamped in
place to act as a seal for the fluid cavity. The electrode
assembly was encapsulated in urethane to isolate the
cartridge from all other materials in the blade. The

urethane is a two-part, castable material.
Silicon Carbide Leading Edge

The coating selected for the rotor blade is a silicon
carbide surface coat. The surface coat is applied to the
female mold which forms the blades skins and leading edges
prior to laminating. During the cure, the silicon carbide
flows partially into the first few plies and becomes an
integral part of the laminate. The coating can also be
applied using a spray-on method. This coating is also
easily repairable as it can be splined on the leading edge
and will cure at room temperature. This type of coating
will be used as the erosion resistant leading edge due to
its easy application method, repairability, and cost. No
additional tooling (form dies, etc) will be needed. In

addition, it is currently being used in production military

28
helicopter parts as an erosion coating. The exact'

identification of this product is Ren Plastics RP 3260.
Glass/Epoxy Skins Composite Skins

Due to the fact that the stiffness of the ER cartridge
with and without voltage applied across it is basically
uncharacterized, there was no logical basis for choosing a
particular fiber type and orientation. The skins were
constructed of two plies of bidirectional glass cloth. At
this point, it was desired to have minimum stiffness
contribution from the skins so that the ER effect from the
cartridge could be maximized. To eliminate the problems
associated with the transition from the leading edge to the
airfoil, continuous skins around the entire airfoil are used
versus a bonded leading edge for instance. The matrix

selected was XR100-317 epoxy resin made by REN Plastics.
Urethane Rubber Filler

. Most of the discussion on traditional fillers such as
honeycomb and foam fillers above does not apply to the ER
blade because the ER cartridge occupies most of the space.
For compatibility and processing reasons, what little space
there is between the inside of the skins and outside of the
cartridge is filled with the same urethane rubber used in

the cartridge. Using rubber rather than honeycomb or foam

29
reduces the stiffness which further enhances the

contribution of the ER cartridge to the overall stiffness.
Airfoil

The airfoil selected for this blade was the NACA 0012.
The are two reasons for this selection. First, the number
0012 indicates that the maximum thickness of the airfoil is
12% of the chord, which is the lowest of standard airfoils.
Minimizing the thickness of the blade reduces the moment of
inertia hence the stiffness. Thus, the ER cartridge will
have more influence. Second, this airfoil is symmetric.
This is advantageous from a tooling and fabrication

standpoint. Data on the NACA 0012 airfoil is in Appendix A.’
Aspect Ratio

The aspect ratio of the blade is the radius of the
blade divided by the chord. This generally falls with the
range of 10-18. The ratio used for the ER blade was 12
simply for convenience. The chord is four inches and the

radius (length) is 48 inches.

ER CARTRIDGE PABRICATION AND TESTING

The overall sequence in fabricating the ER blade as
described consisted of first manufacturing an ER cartridge
which exhibits the ER effect by itself. Once that was
accomplished, the rotor blade tooling was made. The
cartridge was then incorporated into an actual blade.
Finally, testing of the finished blade was done.

The ER cartridge presented previously is the result of
two iterations of design, fabrication and testing. The
first cartridge consisted of multiple layers of electrodes
and ER fluid cavities. It was unsuccessful in that it was
so stiff by itself, applying voltage to the ER fluid had no
affect. This section describes the fabrication and testing
of the multi-layer design as well as the existing single
layer design. It is believed that with additional
development, the multi-layer cartridge may have potential
applications. The fluid used in all testing was made from
55% by weight Dow Corning Diffusion Pump Fluid #704

(Silicone Oil) and 45% corn starch.

3O

31

Multi-Layer ER Cartridge

The original concept for the ER cartridge was to make
it as thick as possible and still fit into a realistic
airfoil shape. The cartridge consists of five electrodes
made of .050" 6061 aluminum and four cavities as shown in
Figure 9. The total thickness was 1/2" and the airfoil was
NACA 0018 (maximum thickness equal to 18% of the chord - 50%
thicker than the NACA 0012). This was the desired airfoil
to provide room for an ER cartridge which could occupy as
much of the cross-sectional space as possible. A chord of
6" was chosen for the cartridge as dimensioned below to fill

most of the cross-sectional area.

 

 

‘1 325

' v ELECTRODES. 5 PLCS

i

b— .187 f
— .375 .055 two plcs

.035 ER CAVITY. 4 PLCS

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 9: Multi-Layer ER Cartridge

A rectangular mold was made to cast the electrodes in
the urethane to achieve the dimensions shown above. Figure
10 shows a bottom side view of the mold and electrodes prior

to pouring. For the first attempt, sheet aluminum was used

32
as the spacer material. The ER cavity area of the
electrodes was taped off and all bonding surfaces were
lightly sandblasted. The assemblage of electrodes and
spacers was lowered into the mold. While the sandwich of
sheet metal was held Centered in the mold, bolts were
tightened through the side walls of the mold against the
electrodes to hold the electrodes and spacers tightly
together to eliminate any seepage of the urethane in the ER
cavity. Conathane TU-65 urethane was then poured around the
electrodes to encapsulate the edges. Figure 11 shows the
electrodes suspended in the mold immediately after pouring.
After the urethane was cured at room temperature, the
cartridge was demolded. At this point, it was discovered to
be impossible to remove the sheet metal spacers. The second
attempt at fabricating this configuration utilized .035”
sheet wax as spacer material. The sheet wax was cut and
placed on the cavity area of the electrodes. Figure 12
shows the assembly of electrodes and wax prior to closing
the mold and pouring. After encapsulation in urethane, the
cartridge was placed vertical with the open end down in an
oven at 200°F. The wax melted and flowed out at this
temperature. This process produced a very cosmetically

appealing cartridge.

33

/r.035 SPACERS (4 PLCS)

 

 

r l BOLTS TO HOLD .
ELECTRODES TOGETHER

 

 

 

 

 

 

 

 

 

 

 

 

.257f

 

l ' J 50'!

Figure 10: Bottom side View of Multi-Layer ER Cartridge
Prior to Pouring

 

Figure 11: Multi-Layer Cartridge After Pouring

 

Figure 12: Multi-Layer Cartridge Assembly Showing Sheet Wax
Spacers Prior to Pouring

Once fabricated and filled with fluid, the cartridge
was fixtured to an exciter (shaker). The natural frequency
was found to be 20 Hz. Lead wires were than attached. The
first sequence of attaching the lead wires was +, -, +, -,
+. Excitations were applied at the natural frequency and a
voltage of 2000 volts was applied and the tip deflection and
velocity curves were observed. There was no change in
either parameter. The lead wires were then hooked up +, -,
skip, +, -. Again, there was no difference with or without
voltage. Appendix B shows plots of input versus output for
both combinations of lead wires and with and without
voltage.

The conclusions made from this experiment were that it
is now desired to have as flimsy a cartridge as possible to
maximize the ER effect. This desire also applies to the

rotor blade itself. For this reasoning, the cartridge was

35
changed to a single layer as shown in Figure 2 and the

airfoil was changed from a NACA 0018 to a NACA 0012.

Single Layer ER Cartridge

The single layer cartridge was fabricated slightly
differently than the multi-layer cartridge. The aluminum
electrodes were .020" 6061 aluminum sheared to 2.25” x 48 '.
The ER cavity area which was 1.875" wide centered on the
electrodes was taped off and all bonding surfaces of both
electrodes were lightly sandblasted. Strips of .040" thick
silicone rubber sheet were cut to .1875" wide. The strips
were adhered with epoxy around the edges of the cavity side
of one of the electrodes. One of the ends was left without
silicone to facilitate filling. Once cured, a bead of epoxy
was run along the opposite side of the silicone strips and
the other electrode was placed on top of the epoxy. The
assembly was lightly clamped together and the epoxy was
allowed to cure. Once cured, the exterior surfaces were
wiped with solvent to facilitate bonding.

A rectangular mold was made by sandwiching two four
foot long pieces of metal .1875” thick with a separation of
2.75" between two pieces of metal 4" x 48”. The two .187"
pieces were temporarily bonded with polyester filler to one
of the side plates. The opposite side plate was removed and
the mold was placed horizontally on the table. The mold was

filled with a .055" layer of Conathane TU-65 urethane

36 .
rubber. The electrode assembly was placed centered on top
of the bed of urethane. The assembly was then covered with
another layer of .055" urethane. The other side plate was
placed over the urethane and the entire mold was clamped
together. The mold was then positioned vertically and after
settling, additional urethane was poured in the end around
the electrode assembly to ensure a void free ER cartridge.
Once cured, the cartridge was carefully demolded.
The cartridge was extremely flimsy and was not self-
supportive. The stiffness was not great enough to hold the
entire cartridge up while constraining it at one end.

The cartridge was filled with ER fluid and a seal was
clamped between the electrodes at the fill end to keep the
fluid from leaking. The cartridge was placed on the shaker.
Mounting the cartridge consisted of fixturing it so that
only 12" of the 48" extended past the shaker. The other 36"
was supported on the opposite side. The natural frequency
was found to be 7 Hz. This natural frequency was then used
as the exciter frequency to yield maximum deflection. After
2000 volts were applied across the fluid, the amplitude of
the tip reduced by approximately 33%. Figure 13 shows input
versus output for no voltage. The peak on the bottom graph
(output) is at approximately 4.2 volts which is equal to 4.2
'mm of deflection (.165"). Figure 14 shows the same graphs
generated while applying 2000 volts. The peak on the output
graph is at approximately 2.8 volts equal to 2.8 mm
deflection. After the voltage was applied, the deflection

37

was 2.8/4.2 = 67% of the initial deflection with no voltage.

Once this ER effect worked, a spanwise elastic modulus
was determined. The cartridge was fixtured as a cantilever
beam with a length of 12". A load of .3249 lbs (147.5
grams) was applied and different voltages were induced. The
deflection for a 12" portion for any voltage was .242".
Using E = (PL3)/(3vI) yielded an elastic modulus of 516,000
psi. The cartridge was then fixtured for a 3-point bend
test with a span of 16". A load of .6498 lbs (295 grams)
produced a deflection of .047”. Using E = (PL3)/(48vI)
produced a spanwise elastic modulus of 787,200 psi. This
was done for no volts and 3000 volts. A conclusion was made
at this point that the ER effect is a dynamic phenomenon and

cannot be detected in static tests.

 

 

a WWI—hm '1 ' ”ILA—m"

2:5 /\ A A /\ d
..::.\j \/ \/ \ -

-a- .—
V

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9tert: O I Stop: 999.02 me
Time Chen 1

”FELL—2M4me V

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

...;I / /‘\ /"\ A \ /‘\j
\/ z \/ \~ \ V i

-.F . '
V

‘8tert: O I 9top: 999.02 me
Time Chen 2

 

Figure 13: Single Layer Cartridge Response - No Voltage

39

.Pc

 

 

 

 

 

MV
Idiv

\

\/

 

 

 

 

'3'"

 

 

 

“VFK/ E1

 

 

 

 

 

tat-rt: O I
Time Chen 1

9top: 993.02 a;

 

'= 1:?433V

 

T

 

H001

 

i
AIAfl

 

/

 

 

 

_
-6-

V

 

 

l l

 

 

vi V«

 

 

Start: 0 e

9:?ime Chen 2

Figure 14:

e . f\ a
\/% x.
l .

atop: 999.02 ee-

single Layer Cartridge Response - 2000 Volts

ROTOR BLADE TOOLING EASRICATION

Aside from the ER cartridge molds, the tooling required
to encapsulate the rectangular ER cartridge into an airfoil
shape consists simply of a pattern and two mold halves.
Since the airfoil is symmetrical, both mold halves are
identical. This section explains the procedure for

fabricating both the pattern and the mold halves.

Pattern

The pattern was basically half of the airfoil and

served as the form to cast the mold halves. The first step

consisted of machining 18 templates as shown in Figure 15

 

 

 

 

 

 

 

below. 600 _
- 100
/-NACA 0012 Airfoil
. \
.50 T
Q Lg) 1...
i
i
.50 -

Figure 15: Pattern Template

4O

41

The templates were made of .080" steel. All 18 pieces
were rough cut, stacked together and positioned on a CNC
mill. CNC code was written in SmartCAM software and
downloaded to the mill to machine the profile on the
templates.

Once the templates were machined and deburred, two
pieces of 54" long .500" diameter rod were slide through the
.501" holes in the templates. The templates were
distributed every 3" along the rods. A piece of 1" x 4"
wood was cut into 17 pieces, each piece being 3" long. The
wood pieces were placed between the templates. The
templates were then clamped tightly against the wood pieces
and hose clamps were tightened onto the ends of the rods to
hold everything tightly together. The entire assembly was
then fastened to a piece of 1" x 10" wood 5' long. This

assembly is shown in Figure 16.

 

Figure 16: Pattern Assembly

42

Next, glass microballons were added to room temperature
curing epoxy resin to make a paste—like, low density
sandable filler. This syntactic foam was splined between
all 18 templates with a straight edge using the templates as
guides. Once cured, the entire surface was covered with a
thin layer of polyester filler. The filler was rough sanded
to remove the high spots, followed by a wet sanding
operation to make a very smooth pattern surface. Once the
entire surface was wet sanded to the templates, the pattern

was complete. Refer to Figure 17.

 

Figure 17: Pattern Assembly After First Spline

Molds

The molds were made out of a mixture of sand and resin
with a surface coat on the mold surface. In order to cast a
mold half, the first step was to build a box around the
pattern. This was done by screwing 1" x 6" boards tightly
around the sides of the pattern. Once the box was made, the

molding surfaces of the side boards and the pattern were

43
sealed with paste wax and the same surfaces were coated with

a release agent.

 

Figure 18: Pattern Box Prior to Surface Coat

The first step in casting the mold was to coat the
insides of the pattern box with a graphite-filled, black,
high-temperature surface coat. The product used was REN
CGL1320 made by REN Plastics division of Ciba Giegy. The
surfaces included the wet sanded surface on the pattern and
the inside surfaces of the side rails. The thickness of the
surface coat was .050" - .100". The pattern was left alone
for several hours to allow the surface coat to become tack-

free.

44

 

Figure 19: Pattern Box After Surface Coat

Once tack-free, a mixture of three parts-by-weight sand
to one part high temperature laminating resin was used to
fill the pattern box. The resin system used was RP4005/1500
made by REN Plastics division of Ciba Giegy. Once the box
was half full with the sand mixture, two pieces of 1" square
tubing were placed in the sand. The tubing was covered with
the sand/resin mixture and the mixture was added until the
entire mold box was filled. Refer to Figure 20. The
purpose of the steel tubing was to provide additional.
stiffness in the mold. The mold was left to cure at room
temperature for 24 hours. Once cured, the box was

disassembled and the mold was removed from the pattern.

45

 

Figure 20: Casting Sand/Resin in Pattern Box

The boards and pattern were thoroughly cleaned and
reassembled. Paste wax and release agent were applied as
before. The entire process of applying surface coat and
filling with sand/resin was repeated and a second identical

mold half was produced. Figure 21 shows both mold halves.

46

 

ing

Both Mold Halves After Demold

Figure 21

ROTOR BLADE FABRICATION

At this point, the ER cartridge was fabricated and
proven to exhibit the ER effect by itself. Also, all
necessary tooling was complete. The next step was to
encapsulate the cartridge into an airfoil shape with fiber
reinforced skins on the outer surface which had an abrasion
resistant coating. The manufacturing sequence was to first
encapsulate the rectangular shaped cartridge solely into
urethane rubber so the end result has an airfoil shape.
Next, the rubber airfoil shape was wrapped with fiber-
reinforced skins. Finally, an abrasion resistant coating
was applied to the outer surface. This section describes in
detail these three processes performed to fabricate the

rotor blade.
Cartridge Encapsulation

The process used to encapsulate the cartridge into an
airfoil shape was similar to that used initially to
encapsulate the electrodes into a rectangular cartridge.

The main difference was a rectangular mold was used for the
initial encapsulation and the airfoil shape sand/resin molds

were used for the final encapsulation.

47

48

The first step consisted of applying sheet wax to the
mold surfaces to compensate for the skin thickness. The
airfoil of the rubber must be smaller than the final
airfoil, the difference being the thickness of the outer
skin. Since two plies of glass/epoxy at .010" per ply
thickness were used to wrap the blade, the sheet wax used on
the mold surface was .020". With the airfoil of the rubber
portion .020" smaller than the mold, upon wrapping it with
.020" of glass/epoxy skins, the mold will provide a tight
encapsulation of the entire assembly. The sheet wax was
carefully placed over the entire mold surface and trimmed
with a utility knife to the leading and trailing edges.

The mold halves were then aligned and clamped together.
Both mold halves were positioned vertically. A metal plate
was bonded with polyester filler to the bottom of the mold
segments to prohibit leakage of the urethane. The mold was
then filled half full with urethane by pouring it into the
open end on top. The outer surfaces of the cartridge were
lightly abraded with coarse sandpaper and cleaned with
acetone. A 1.5" wide x .035" piece of sheet aluminum was
slid into the ER cavity of the cartridge so that the gap
will be maintained during encapsulation. Without this
filler piece, there would be a strong possibility of the
walls of the cartridge being pushed in by the urethane
causing the electrodes to come in contact. Once the
cartridge was prepared, it was simply pushed down into the

mold. As it was forced down, the urethane in the mold was

49
gradually displaced around the cartridge and flowed towards
the top of the mold. The cartridge was pushed down until
only the ends of the sheet aluminum electrodes and a short
portion of the cartridge remained above the top of the mold.
Prior to this point, the urethane reached the top of the
mold, ensuring that the entire mold was filled. The
assembly was left overnight to allow the urethane to cure.
Once cured, the mold halves were separated and the part
removed. Figure 22 shows the cartridge after demolding.
The molds were then cleaned and stripped of the sheet wax.
The surfaces were coated with release agent and the mold was
ready for the next operation.

The part itself captured the airfoil as defined by the
molds and sheet wax very well. There was thin flashing on
the leading and trailing edges which was cut off with
scissors. The part was prepared as before with a light
abrasion using sandpaper followed by a thorough acetone

wipe.

 

Figure 22: Cartridge After Encapsulation into Airfoil

50

Skin Layup

The mold surfaces and outer surface of the part were
coated with a light application of REN Plastics XR100-317
laminating epoxy resin. Bidirectional glass fabric style
120 was spread out on cardboard and saturated with the
resin. The part was placed at one end of the fabric. The
end of the fabric was held against the part and the urethane
assembly was simply rolled into the fabric. The rolling
process was continued until the thickness of the skins was
equal to two plies (the blade cartridge was flipped four
times). Once wrapped, the assembly was placed onto one of
the mold halves and covered with the other. The molds were
then clamped together. The .035" piece of aluminum was
still inside the part to maintain the gap. The part was
left in the mold overnight to allow the resin to cure.

After the cure was complete, the part was demolded and
all of the flash was sanded off. The metal spacer was
pulled out of the blade. The blade seemed to have a
desirable amount of flexibility upon an intuitive

evaluation. Refer to Figure 23.

    

Figure 23: Blade Droop and Flexibility After Demolding

51

Silicone Carbide Surface Coat Application

The final step in fabricating the blade was applying
the exterior abrasion resistant surface coat. The blade was
thoroughly rinsed with acetone and the ends of the
electrodes were covered with masking tape. REN Plastics
RP3260 surface coat was mixed per manufacturer's
instructions. The surface coat was then put into an
agitating spray gun and thinned with methyl ethyl ketone
(MEK). Several light coats were sprayed onto the part until
a thickness of approximately .005" was achieved. The part
was placed in an oven at 175°F for 2 hours to cure the
surface coat. The part was removed from the oven and the
masking tape was taken off. This concluded the fabrication
of the blade. Figures 24, 25, and 26 show a plan view, root

end view and tip view of the finished blade.

 

Figure 24: Plan View of Blade

52

 

Figure 25: Root End View of Blade

 

Figure 26: Tip View of Blade

ROTOR BLADE TESTING

Before the blade could be tested on the exciter,
fixturing had to be fabricated to properly constrain the end
of the blade. The fixturing was made in two parts. First a
metal plate having a hole pattern matching that of the
exciter was covered with a thick (3/4") layer of polyester
filler. Before allowing the polyester to cure, the blade
was treated with a release agent and was pushed into the
polyester so that the polyester would capture one half of
the airfoil shape. Once cured, the fixture half was coated
with a release agent and a second metal plate was coated
with the polyester. The other side of the blade was then
pushed into the polyester on the second fixture plate and
bolts were tighten in each corner of the plates, drawing the
plates towards the blade. The polyester was then allowed to
cure. This fixuring was lightweight, captured the entire
airfoil on the blade for approximately three inches and
easily bolted to the exciter.

The blade was filled with ER fluid of the same
concentration as used on the cartridge (55% Silicone Oil,
45% Corn Starch). A piece of sheet rubber was cut to the
same width as the ER cavity, positioned in between the

electrodes at the opening and a spring clamp was used to

53

54
hold the electrodes tightly against the rubber, eliminating
the possibility of leakage. The blade was secured to the
exciter with the fixturing and the lead wires were attached
to the electrodes.

A random input was first used to find the natural
frequency. The natural frequency was found to be 13 Hz.
This natural frequency was then used as the exciter
frequency to yield maximum deflection. The first parameter
evaluated was the vertical velocity of the tip. Figure 27
shows the tip velocity with no voltage (top graph) and the
velocity after applying 3000 volts. The peak with no
voltage is approximately 160 mV or .160 v which equates to
.16 m/s. After applying 3000 volts, the peak reduced to 110
mV (bottom graph) or a velocity of .11 m/s. The velocity

was reduced by 31% by applying 3000 volts.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

'33 On A A ,,
egHIITHIH \
:;Ji/byigiug
égmflfinflnfl .
fEJHiVUl \VWV -

 

 

 

 

 

 

 

start: 0 e ‘. . ’
8:71am Chen 9 top. 999.09 no

Figure 27: Blade Tip Velocity - No Volts, 3000 Volts

55

Figure 28 shows the same graphs generated by applying

BOO
0V

Reel
60
mv

Idiv

-200

200
av

Reel
50
«iv

Idiv

-200

Figure 28:

A

 

 

2000 volts rather than 3000 volts.

from .16 m/s to .13 m/s for a reduction of 19%.

parameter evaluated was tip deflection.

A

 

 

 

 

 

The velocity decreased

The final

 

\
\HI

 

 

 

 

b

31/
wit,

 

ill \W

 

 

 

\J

 

 

stare:
9:?1me

0 e
Chen 2

'e

atop: 999.03 ..

 

 

 

-/

 

l\

\l

 

 

 

 

 

 

 

 

 

U" U’ 1' ‘v
start: 0 e

 

Blade Tip Velocity - No Volts, 2000 Volts

 

56

Figure 29 shows tip deflection with no voltage (top)

and after applying 3000 volts.

The deflection is

approximately 2.2 mm with no voltage and 1.9 mm after

applying 3000 volts.

The reduction of .3 mm (.012") equates

to a 14% reduction of tip deflection.

 

 

 

e

 

 

 

/\

   

 

 

 

 

 

 

 

i ii

 

 

 

U *

 

 

 

 

Start:
9:Time

o .
Chen 2

 

 

 

 

 

 

 

p-

 

 

 

 

 

 

 

 

 

Start:
S:T1me

Figure 29:

0 0
Chan 2

Stop: 999.02 me

Blade Tip Deflection - No Volts, 3000 Volts

 

MANUFACTURING ISSUES

The final segment of the research effort involved
highlighting some of the manufacturing issues that would
mandate further evaluation as this application of ER fluids
advances towards production. This section discusses the
manufacturing issues involved in the specific processes used
in fabricating an ER blade in a production environment, the
tooling and equipment needed and finally, the quality

control aspects.
Manufacturing Processes

The production processes used to fabricate a blade
similar to the laboratory blade would basically be the same
as those described. These processes are cartridge
fabrication, cartridge encapsulation, skin layup and
abrasion resistant coating application. Also described are
other components and processes that would be needed on a
production blade, but were not implemented on the model
blade. The materials used on the model blade were not
optimized as there has only been one blade fabricated and
tested. These manufacturing steps will require a re-

evaluation upon material or process substitution.

57

58

e c o The manual method of partially
filling the mold with urethane, positioning the electrode
assembly, filling with urethane and closing the mold worked
fine for a development blade. However, it was time
consuming and produced air bubbles. An alternative method
to encapsulate the electrode assembly in urethane is via a
Resin Transfer Molding (RTM) process. The electrode
assembly would be fixtured in a constrained position in the
center of the mold. Resin would ingress in one end and
would be forced through the mold around the electrode
assembly with a vacuum source at the other end. This
process would be more adaptable to production than the
pouring method. Also needed in production is a removable
cap for the ER cavity with provisions for lead wires to be
easily attached to the electrodes. The model blade has a

temporary cap which was clamped in place.

cartrisse_snsaesslasienl similar to encapsulating the

electrode assembly, RTM would be a viable replacement for a
production process. RTM for this process would work just as
explained in the previous section. Also, the sheet wax
procedure would not be needed in production since there
would be two sets of molds, one made to the outside of the
skin for the skin layup and the other made to the inside of

the skin to encapsulate the cartridge.

59
§313_Lgygp; Production processes such as filament winding
or automatic tape laying could perhaps replace the hand
layup procedure. Also, high performance prepregs may be
used for production blades. The layup process would be
similar but the curing process for these materials would

entail an autoclave rather than a room temperature cure.

W The process used

to manually spray the surface coat on was inexpensive and
easy. The uncertainties are longevity compared to
alternatives such as an elastomeric sheet and dimensional
conformance after spraying. If the coating is functionally
adequate, the method of applying the surface coat to the
mold surface rather than spraying it on the mold would make
a superior part since the flow surface would now be defined
by the mold and the surface coat would be allowed to
crosslink with the skin plies making it an integral part.

In addition to the processes described above, a
production blade would inherit additional processes. These
are due to attributes not used on the laboratory blade but
possibly necessary on production blades. These components
include the tip weight, spar, tip cap and root end. Some of
the considerations for fabricating these components are

described below.

1ip_ggighg& The model blade did not have a tip weight

because it would not have contributed to the objective of

60
evaluating the ability to control vibration with an ER
fluid. To incorporate a tip weight would involve fixturing
the piece of metal (machined or round) in the mold during
the cartridge encapsulation operation. If properly
fixtured, the weight will be cast in the correct location of
the blade and all of the surface contact between the weight
and the rubber should be sufficient to hold the weight in

place without any mechanical fasteners.

535;; It may be discovered that a spar is needed in
addition to the ER mechanism. If so, it could be fabricated
from unidirectional composite material and machined as a
separate detail. The spar would then be fixtured and
encapsulated (or bonded) at the leading edge in the same

manner as the tip weight.

Tip_gap; The tip cap could simply be a flat laminated
composite panel attached to the end of the blade. Inserts
could be cast in the urethane to allow for mechanical
fastening. An alternative approach is to mold the cap with
a flange that captured the end of the blade. Inserts could
be molded or potted in perpendicular to the flow surface and
fasteners put through the flange. This approach would be
less likely to fail since all of the axial load on the cap

is transferred through the fastener to the skin.

61
Rgg§_fipg& The root end consists of the hardware and
transition region in which the blade changes from a constant
cross-section to the hub. Many of the traditional methods
used on composite blades will work the same way on as ER
blade as long as there are provisions for lead wires to
unobtrusively be attached to the electrodes. One method is
the "wrap-around-lug" concept. The outer skin plies
gradually twist from the airfoil surface to wrap around
cylindrical bushings situated perpendicular to the plane of
the blade. The bushings are used to attach the blade to the
hub. Employing this method would require complex layup

molds with changing cross-sections.
Tooling and Equipment

Based on the construction of the blade and the process
as described, there are several facets of tooling and
equipment needed to fabricate a blade of this nature on a
production basis. The tooling and equipment needed for
fabricating the ER blade includes molds, fixtures, sheet
metal equipment, mixing equipment, and aluminum surface prep
equipment. Obviously, standard composite facility equipment
such as autoclaves, clean rooms, and machining centers is
needed. This additional tooling and equipment are described
in this section and the list is not an all inclusive as
there are always tools unanticipated at the concept phase in

any composite part program of this complexity.

62
figldg; A mold is needed anytime it is desired to have an
uncured material conform to a shape while curing. For this
blade, it is necessary to have at least three molds.

The first mold is the rectangular cross-section mold
needed to cast the ER cartridge. It is simply four
lengthwise pieces of metal bolted and doweled together with
a base plate. Fixturing is needed to hold the electrode
assembly centered in the mold during the urethane
ingression.

The second mold needed is a mold to encapsulate the
cartridge into an airfoil shape. This can be machined
aluminum or composite and the contour will need to be the
desired airfoil offset inboard the skin thickness.
Fixturing is necessary to constrain the cartridge in the
proper location during the urethane ingression.

The final required mold is the layup mold which the
urethane wrapped in composite is placed in to form the final
airfoil. It will be identical to the mold described above
only will contain the true airfoil. Both the second and
third molds will have the root end incorporated as specified
by the design.

If a composite tip cap and spar are to be used,
separate layup molds will be needed for each of these. The
spar mold can be machined metallic or composite and the tip
weight mold can be a flat metal plate with any necessary

steps or contours machined into it.

 

63
liztnggg; There will be several fixtures needed to
fabricate the ER blade. Aside from the fixtures needed to
constrain the cartridge and electrodes during encapsulation,
there may be electrode bonding fixtures, spar fixturing, tip
weight fixturing, tip cap fixturing, and root end fixturing.

There will need to be fixtures to hold the trimmed
metal electrodes directly over each other while bonding in
the silicone strip without closing the ER fluid cavity.

Spar fixturing and tip weight will include trim
fixtures to cut the pieces to length, machining fixtures as
needed, and constraining/locating fixturing to accurately
locate and hold the pieces while being encapsulated in
urethane.

Tip cap fixturing will consist of a periphery
router/machining fixture to machine the edges and drill and
holes needed for mounting.

The root end fixturing will involve fixturing to drill
bushing holes in either the composite or molded-in metal

bosses.

fihggt_fig§gl_§guipmgnt; Since the sheet metal components

used for the electrodes are not used on conventional
composite blades, equipment to fabricate these details is
needed. For flat electrodes, a sheet metal shear will be
adequate. For cambered blades requiring contoured

electrodes, either draw dies or roll form tooling will be

64
also required. A brake may be needed for any sections of

the electrodes requiring bends.

EB_z1uid_Eiainsznispsnsipg_sgsipseatl Electrorheological

fluids have two components which must be mixed within a
certain range of ratios to work properly. To perform this,
automated mixing and dispensing equipment is available and

needed to be efficient for a production application.

A1uminum_surfasa_zreparatien_£guiemsntl Prior to

encapsulation in rubber, the aluminum will undergo certain
processes which will require additional equipment. First,
the aluminum must be mechanically abraded in the areas that
get encapsulated. This can be done with conventional
sandblasting equipment. The aluminum will have to be free
of all contaminants before bonding. This may be done with
either a trichloroethane vapor degreaser and/or an etch or
anodic treatment followed by application of a corrosion

resistant primer.
Quality Control

I The aspects of quality control and inspection will have
added responsibility and complexity resulting from
incorporating ER fluids in rotor blades. These additional
responsibilities fall into the categories of Receiving/In-

Process Inspection and Non-Destructive Evaluation (NDE).

65
Reggiyiggzgp-grocess Inspection, Receiving inspection will
be mandated to trace certifications of all new materials
including the aluminum, ER fluid components (corn starch and
silicone oil), rubber encapsulant and all hardware
associated with the electrodes such as lead wires and
connectors. Certifications may include physical, chemical,
and mechanical test data. In-Process inspections will need
to be performed on the cartridge assembly or other
equivalent ER fluid-filled mechanism prior to integration
into the final assembly to ensure proper response prior to
integrating it into a rotor blade. This will require
sophisticated testing equipment such as customized rotating

shakers to simulate flight conditions.

Newman... Adding ER fluid to rotor blades

does impose additional non-destructive inspection (NDI)
requirements. These include surface inspection of the
aluminum with fluorescent penetrant techniques, inspection
of all bond lines for voids and other defects and a final
test inspection of the completed blade performed on the

rotating shaker mentioned above.

CONCLUSIONS

This paper presents one of the most advanced attempts
at controlling vibration in rotor blades by utilizing ER
fluids. A fiber-reinforced ER fluid-filled composite
ihelicopter rotor blade was designed, fabricated and tested.
Also, many design and manufacturing issues were explored.
Although there are many differences between a scaled down
rotor blade and an actual one, this study will provide some
ground work for future advancements in this area. Based on
the design, manufacturing and testing of the ER fluid-filled
helicopter rotor blade as described in this paper, the
following conclusions can be made on the feasibility of
utilizing ER fluids to control vibration in rotor blade

applications:

1. It is possible to achieve vibration control with ER
fluid-filled, fiber-reinforced composite helicopter
rotor blades on scaled-down blades with construction
representative of actual parts. The tip velocity of
the presented rotor blade was reduced 33% by applying
3000 volts and the deflection was reduced by 14%.
These reductions could be increased by optimizing the
design.

2. The blade was purposely made with some flexibility to
ensure realization of the ER effect. If the blade was
made with sufficient stiffness to meet the functional
requirements of an actual rotor blade, it may have been
too stiff to exhibit vibration control using ER fluids
as utilized in the design presented. To overcome this,

66

67

the ER cavity needs to be located off of the neutral
axis. Future work should focus on this idea.

An ER component made of layers of electrodes can be
easily and accurately manufactured. Although the
multi-layer cartridge did not work as attempted in this
study, additional development may make this idea more
feasible than single-layered components for some
applications.

Utilizing ER fluids for vibration control adds many new
design and manufacturing issues. Advancing ER fluids

*towards production applications will require extensive

evaluation of these issues.

Many new quality control issues will be required for ER
fluid-filled rotor blade fabrication and sale to an end
user. These issue will involve creation of many new
testing procedures and material and process
specifications.

An important issue in ER rotor blade design involves
characterizing the ER fluid mechanism. Many physical
and mechanical properties need to be determined at an
early stage in a research program. These properties
include optimal electrode material, thickness, size,
spacing, stiffness, etc.

appsnnrcss

69

APPENDIX A: NACA 0012 AIRPOIL DATA

2 I 3226:}
.00000 .00000
.01250 .01894
.02500 .02615
.05000 .03555
.07500 .04200
.10000 .04683
.15000 .05345
.20000 .05738
.25000 .05941
.30000 .06002
.40000 .05803
.50000 .05294
.60000 .04563
.70000 .03664
.80000 .02623
.90000 .01448
.95000 .00807
1.00000 .OOOOO

leading edge radius a 1.58
slope of radius = 0.00
airfoil is symmetrical - lower 8 upper

data is per unit chord

70

APPENDIX B: MULTI-LAYER ER CARTRIDGE RESPONSE

 

 

Iifiiiiiii Ii IiIi Ii Ii i Ii Ii Ii i Ii I i Ii Ii %

‘Stert: 0 9 Stop: 999.02 mm
S:Time Chen 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4 Y: 14 V
2
v

 

Reel
800,

 

 

 

 

mV
Idiv P

 

-2 b
V

 

Start: 0 e _ f i * 'Stop: 999.09 me
9:71me Chen 2

GENERAL REFERENCES

10.

11.

GENERAL REFERENCES

Immen, Frederick H. and Raymond L. Foye, "New Insights
in Structural Design of Composite Rotor Blades for

Helicopters," Proceedings of the Fourth International
Conference on Composite Materials, Tokyo, Japan, 1982.

Foye, R.L., et al, "Mechanics Problems in Composite
Main Rotor Design," 0.8. Army, Research, and Technology
Laboratories, 1983.

Hahn, M., "Design for Repairability of Helicopter
Composite Blades," DTIC report number AD-A178 691,
1987.

Holmes, R.D., "S-Glass Reinforced Plastic Adopted for
Helicopter Rotor Blades," SAMPE Quarterly, October
1975, pp 28-41.

MIL-S-8698

Goldstein, G., "Electrorheological Fluids:
Applications Begin to Gel," Mechanical Engineering,
October 1990, pp 48-52.

Marsh, G., "Plastic Tiger," Aerospace Composites and
Materials, Vol. 3, March/April 1991, pp. 4-7.

Adelman, H.M. and W.R. Mantay, "Integrated
Multidisciplinary Design Optimization of Rotorcraft,"
NASA Tech. Mem. #101642, 1989.

Law, M. and J. Williams, "The Influence of Helicopter
Operating Conditions on Rotor Noise Characteristics and
Measurement Repeatability," DTIC report number AD-
A121426, 1982.

Nixon, Mark W., "Preliminary Structural Design of
Composite Main Rotor Blades for Minimum Weight," NASA
Technical Paper 2730, 1987.

MW. AGARD report number 781.
DTIC report number AD-A230 577, 1991.

72

12.

13.

14.

15.

16.

17.

18.

19.

20.

73

Skinner, Gary L., et al., "Combined Preliminary
Airworthiness Evaluation and Airworthiness and Flight
Characteristics Evaluation of the UH-1H with
Preproduction Hub Spring and Composite Main Rotor
Blades Installed", AEFA Project #84-33 Final Report,
1988.

Miley, S.J., o ' e s

° s, Department of
Aerospace Engineering, Texas A&M University, February
1982.

Layton. Donald Mo. Helicopter_£srformanse. Matrix

Series in Mechanical and Aeronautical Engineering,
Matrix Publishers, Beaverton, Oregon, 1984.

Bazov, D-I-. "Helicopter_Aerodynamis§". NASA #N72-
23024, May 1972.

Barrie, Douglas, et al., "Advancing Helicopters",
Flight International, December, 1992, pp. 31-37.

Kolias, Alexander and Andrew Dimarogonas, "Rheology of
Zeolite and Corn Starch Based ER Fluids at High Strain
Rates," Proceedings of the Conference on Recent
Advances in Adaptive and Sensory Materials and their
Applications", Blacksburg, VA, 1992, pp. 548-557.

Morishita, Shin and Tamaki Ura, "ER Fluid Applications
to Vibration Control Devices and their Adaptive Neural-
Net Controller," Proceedings of the Conference on
Recent Advances in Adaptive and Sensory Materials and
their Applications", Blacksburg, VA, 1992, pp. 537-546.

Brooks, Douglas, "Applicability of Simplified
Expressions for Design with Electrorheological Fluids,"
Proceedings of the Conference on Recent Advances in
Adaptive and Sensory Materials and their Applications",
Blacksburg, VA, 1992, pp. 524-535.

Weiss, Kenneth, et al, "Electrorheological Materials
and their Usage in Intelligent Material Systems and
Structures, Parts I and II, Proceedings of the
Conference on Recent Advances in Adaptive and Sensory
Materials and their Applications", Blacksburg, VA,
1992, pp. 507-520 , 605-617.

 

HICHIGRN STATE UNIV. LIBRARIES
I“ll“NIWIIHIIWH”WINIWIHHIIHIIWIWI
31293008812657