H_ :::_______ ___:__::__:_:_ _ __,

 

a .
"r‘
‘l’\'\ I

\P‘
3!! I D
.. TI.

3 1.
5'0” .
3.. k

 

.I
\
I
_
.
.I
.
.
.
I
.
. V
I vs
..
I
_ .
. .
It .
I: ~
I.
I
I u
.
.
I
. .
. .
I
o
. I . I
~ I
. I I
.
a I
..

.‘..’Lgul-__..

o
' v
I
I
I
I
O
' v
m

,.
.

.

\

.

I
I
3'-

nah

 

\ I . l . I . r
I I
I I
l v I I
I l 0 w i
— r u I
' I
I
n u
u 4 I I
. 4
VI ~ I
I _
. I ;
I. ’ — I r f 4
I I. .
I . .
II . .
I I I . I a
I . .
. . . N . . I . av A . .
. , I . . \
3 a
I ‘ I 4‘ .
.
. . I .~ . . I

.
I

.

\

l

l
a... ....

J
.
I
J.
I
...:IS

(43 ‘h

.I
“A.

a
t
—/
O
.
' J-
g
Wanna-m

1
”I

‘ I
I
“Io-C"

r:
7393.;

I
l

I
r
I
I
I
I
IU’PP

 

. .. I
I a I . I I I.
\ \ - l
1
. . . C.‘
v a I I a .
. . . . . a.
. I x.
. 4 I .
o I u '-
. . .
I. 4 I I
. . I. I ‘ . II. I , I
-_ ., , .I . ....JlK .19.} .a F -fi.:a..v.4...¢‘ .l:.s..H.-rfr .
I. . . A . - 7 . a .II ......lr‘ I Iv . t0 ,. . . I. .. ... ...
. I . . x . . \o I . l1: 35 I I .
. . . I . I n p. I. I > '3 III. I I r . I
. . I I t . I . ..1I JP. . . I , a . . . . . _
I . I h I I I \I I
I \
I . , I . I k . < o
. I ~ . I .
. . ._ I . - . n .
. . . . .. I; I
' I. I I . n
u I I a .
. . . . .. . t I
I. . I I . . . . ’ .
. (I T- I . I II . I
n .. I II 3 A o.
I . . , I I It. . .
.. . .v
I . . I . I I 4 a I I .
I I. I I I I I I .
. y D I a I 1 6
4 J A I. . . . . . I I
I I : _
- . _ a n- 1 I
. .0 I I1 ' y. . .. I . ...
I .I . I
r . l‘ I . . I
. I . . I . .
. I . . I , I \ . 4 _a
. . . :
. . .I . ~ . v I I .
. . A
. . I . . . . _
I I. l I I I Q . l v
. . . . . .
I . V . P
I I I I I a
4 . 1
. . _ .. . . . . ..I. . .
I ~ .I .
I I . II I I . .
I . . I . .
. . I .
\ . a . .
. . - . .
. . . . . . I g
. \ . II. I, . . .. I .
I . I . . . . . . I
I I I 4 II .
I
. . . . . . \ . . I
v . . I
r n . . I
. . . . . I, \ . . _
I _
. I . . . II I
.q - . . . I . .
I I I I I _ . . I
I .
I I . .
. . I I
. A l o

 

 

 

....) [lb .N

 

‘ .
.. A: .... f

.. ..vv.’.0..:f.rr. .Ju..:w....v1..t .. .. ....i... 3...: ... . ..,.‘Y.£ ....:.I..n...¢s.~..v?.. ..
a . .. .. .‘ . . ... .. v,

 

AN EXPERIMENTAL STUDY IN PISE'DE
TERRB CONS'HUCTION FOR WALLS
OF DWELLING HOUSLS

A Thesis Submitted TO

The Faculty Of

2m; MCKIGAN STATE COLLEGE
OF
AGRICULTURE Azm APPLIED SCIENCE
BY
PHILLIP L. $111415

Candidate For Degree Of Bachelor
Of Science

June 1927.

“T <31: \ \
"P \‘\; Q

Q (1‘le
\

 

Introduction.

Shortly after the War (1920-22), considerable
interest was aroused, in the European Countries. in the
building of pise'de terre houses. Many welcomed this
idea with all the enthusiasm of revivalists. and claimed.
in it. we would find the solution of our housing problem.
They pointed to the houses of our early ancestors and
claimed that with a few changes we might expect to adapt
that type of construction to modern day needs.

Because of this interest. the Committee of the
Privy Council for Scientific and Industrial Research, in
England, appointed a Building. hesearch Board, in 1920;
its chairman being ”The Most Hon. the harquess of Salisbury.
The director of building research was Mr. H.O. Waller
B. Sc.h.lnst.C.E. The purpose of this board was ”to con-
sider and direct the conduct of research on building
materials and methods of construction."

About the same time as this board was carrying
on its investigations. there were a number of investiga-
tions and eXperiments carried on in Germany, France,
Spain. and New South Wales. However. the above mentioned
board. together with "AXperimental Cottages," a report
on the work of the Department of Scientific and Industrial
Research at Amesbury,Wiltshire. by w. h. Jaggard, seems
to be the most complete investigations carried on in

pise'de terre in Burcpe at the present time.

191737,:- I.)
.- U.‘d -rd

In the United States. interest in pise'de
terre was not aroused until some time later. In 1926
Profs. Memfree and Franklin. both of the University of
hichigan took up the study of pise'de terre. Articles
appeared in the Lng. News Record (Vol 26-page 267) and
also in the Scientific American (Vol—l30-p—233) on
“Rammed Earth." and ”houses of dud." respectively.
Others interested in the same subject being - J.D.Long,
Ag. Eng.Dept.,Davis Cal. and h. C. Killer. Ag.Eng.Dept.
Fargoe N.Dakota. About the most complete article. in
regard to both the type of house suitable for pise'de
terre construction and the construction itself is one
in pamphlets form gotten out by the 0.8. Department of
Agriculture and is known as Farmers' Bulletin No. 1500
The article is entitled "hammed Barth Walls For
Buildings."

Pise'de terre construction is not a new idea.
It has been gradually develOped through out the ages.
Probably the earliest form of habitation were built out
of wattle work, consisting simply of interwoven branches
and twigs with the leaves on. This was improved in
pro-historic times by covering in earth and clay, or
some other adhesive material. This process in England
became known as daubing and the actual construction
was spoken of as "wattle and daub." Such primative

construction probably entailed no end of repairs.

Banksrt. in his "Art of the Plasterer”
describes wattle and daub as follows;~

”___W_'_ Clay mixed with short straw and
chalk was then worked in between the main timbers.
thus covering the hurdle-like lathing with a rough
‘daub' cement.

“Whilst still the clay was wet it was
scored over to give keying to the finishing coat of
lime plaster, which was itself colored and often
painted with decorative designs on the inner surface."

Later. wattle and daub passed out of
existence and cob toca its place.

In a letter to "Country Life” (Nov. 22,1913),
Kr. Norman sewscn described the traditional way of
making and using the cob as follows:-

” _____ The cob is made by mixing clay
and straw with suffiCIent water to get it to the right
consistency. It is generally mixed at the edge of a
pond, a horse being used to tread the clay and straw
together. A course about a foot high is then laid on
top of the base wall and wellatrodden down, beginning
at one corner of the building and following right
right around, so that by the time this is finished, the
cob first laid is hard enough to receive the next course
and so on r a ‘ a after the wall is brought up to

the desired height, the surface is pared off and

generally plastered. * a a a,“

He also remarks that "* * * *With reference
to cob cottages. which are well known to be comfortable
to live in. and the materials for building which are
generally abundant I believe there are few men to be
found nowadays who will build them as cob-building
is hard work and messy. ******."

Mr. Tuycross in writing to the same paper
at about the same time (Mar.l4.l9l4) stated that
until the year 1850 the entire town of Naseby con-
sisted almost entirely of mud buildings and at the
present time garden walls are still built in the same
way. He also lamented the fact that there are few
men left who can do this "mud-walling." and remarked
that "houses can only be built where you have the
right sort of tenacious mud."

An analysis of a piece of typical Devon
cob for Mr. A. Alban H. Scott gave the composition

as follows:— .
Per.cent.

Stones (residue on 7x7 mesh sieve) 24.40
Sand.fioarse (residue on 50x50 mesh sieve) 13.70
Sand.Fine (through 50x50 mesh sieve) 3 .50
Clay 20.60
Water 1.

, 100.00

Pise'de terre or the system of building
with earth rammed between boards, has deservedly
received more attention than Cob in recent years as
[being afitter method for revival and adaption to

modern conditions. It must not be confused with

’ l

either of the two foregoing discussions of daub and
wattle. or ooh. Although pise'de terre is not intended
to replace cob it is far more adaptable to our needs
than any other “Earth Houses.”

Some peonle have thought that pise'de terre
is the modern solution of our housing problem. Others.
refusing to admit any good points about it. Jestingly
refer to it as “Castles of Mud". From the scientific
standpoint. this type of construction can be said to
be in its infancy yet. From past experiences and
experiments however. indications are such that might
lead one to think that with more time and effort on
the part of the engineering world. something of commer-
cial value to the present methods of building may be added.

There are a number advantages already proven
for pise'de terre construction. There is ample proof
that thoroughly sound and durable walls were constructed
on the most ancient of pise'de terre systems.

Plivy in his ”Natural History". comments on
the strength and lasting qualities of ”formocion“ walls.
"moulded. rather than built. by enclosing earth within
a frame of boards."

Failures were no doubt common. but were
probably due in most cases to causes that could be

avoided in modern work.

Those who have followed the lead of hr. St.
Loe Strachey. the revivalist of the method in England,
and whose work is described in “Cottage Building in
Cob. Pise'Chalk and Clay." by Mr.Clough Williams-Ellis.
have progressed a long way towards the elimination of
such causes. With the more efficient types of shutter-
ing now available. there is no real restriction of plan.
There are perhaps more limitations to such a method
than to one employing a small unit of construction. but
any good straight forward plan should be workable.

For maintaining an equable temperature within.
pise‘ walls are hard to best; they name the most
comfortable of dwellings in summer or winter. and
excellent stores for produce or materials affected by
excessive heat or cold.

Costs depend so largely on local conditions
that it is practically useless to state a relationship
between costs of pise'de terre and other building
materials. However it is quite safe to say that whatever
the conditions are. pise' construction will cost about
(2/3) two thirds of the cost of a similar building in
brick or stone.

As one readily sees. in reading that which
has been done in pise' de terre construction. there has
been no method by which a person. wishing to use this

method of construction. can pre-determine the strength

he can reasonably expect the soil on his preperty to
attain. A person of limited means might very well
hesitate then to expend much time and money on a thing
that he could not be reasonably sure of standing after
it was built. The purpose of this thesis.therefors.

is to study the strength developed in pise'de terre
walls and to determine if possible some simple tests to

pro-determine this strength.

Outline of Experiments.

In this study of pise'de terre construction
there are four results hOped to be attained by ex—
periment. First. to determine whether the strength of
pise'de terre is due to Just the ramming of the earth.
to the clay acting only as a binder. or is due partly to
the actual strength of the clay. Also to derive a curve
showing the compressive strengths that can be developed
by variable amounts of sand and clay. hereafter this
test will be known as test number one.

Second. taxing a sample of soil. run the
tests developed in test J1. mate test cylinders and
breah the same in a compression machine and check the
actual strength with the strength from the curve of
test fl. This experiment will hereafter be listed and

spoken of as test number two.

Third, maxing two sets of three cylinders
each. both sets having the snme composition. to dry
one set with artificial neat and the other set to
dry naturally. determine whether any strength is lost
due to rapid drying. this test will be known as test
number three.

Fourth, to build a small model house. to
use both for exhibition purposes and also to expose to
the elements in the study of waterproofing and paint:

to be used in this construction.

Kethod of Procedure.

Test no.1.

 

In order to mate a comparisan between two
substances of different composition, we munt have none
standard of comparison. Ere standard in this case
was a ratio exoreescd between the voids in a sample
and its clay content. In other words a voids—clny
ratio was the standard by means of which all samples of
earth were to be compared.

This particular ratio was chosen to see
whether or not the ole in tre pise'de terre construction
does not have the same function as the cement has in
concrete construction. That is does it not serve merely

as a binder?

Talbot's method of proportioning concrete
has data showing that the strength of concrete is
dependent upon the ratio of the volume of voids to the
volume of cement.

If then. the clay in pise‘de terre construction.
plays the same part as cement in concrete construction.
the rammed earth will have strength dependent on the
ratio of the volume of voids to the volume of clay in
the soil.

Talbot'e method of proportioning concrete is
identical with the water ratio theory of preportioning
concrete when the voids in the concrete are entirely
filled with water. To-day. the water ratio theory of
prOportioning concrete is thought to be the most scien-
tific way of mixing concrete. Therefore, since the
water has nothing to do with the strength of rammed
earth. and since the amount of water is the only
difference between Talbot's method and the water ratio
method of proportioning concrete, why is it not logical
to assume that the voids-clay ratio will determine the
strength of pise' de terre walls? If such an assumption
is correct. a curve. drawn with the compressive strength
as ordinate and the voids-clay ratio as absisss. should
drop off or at least not rise any higher, when the voids-
clay ratio reaches 100$.

Therefore in determining whether the clay in

pise‘de terre construction acts only as a binder or

whether it has an actual strength of its own. test
cylinders were made of pise' de terre in which the
voids-clay ratio was varied. These cylinders after
being thoroughly dried, were broken in a compression
machine. Their actual strength was recorded. This
value was plotted against the voids-clay ratio of the
cylinder.

The cylinders were considered thoroughly
dry when they assumed a constant weight.

he test cylinders were made of sand in
which some gravel was mixed.(gravel varied up to the
size of a walnut). The sand and gravel was mixed
mechanically with clay from a Lansing brick yard clay
pit. The clay was pulverized before using.

The percentage of voids in the sand and
gravel was determined first. The following method was
used. A known volume of sand and gravel was weighed.
(Height A). The container partly filled with sand
was then filled to overflowing with water and weighed.
(Weight B). The container filled to overflowing with
water alone was weighed. (Weight C). in each case the
net weights were used. Using the letters to represent
the weights. the following formula will give the absolute
volume of the sand and gravel.

B-(A x C) - absolute vol. of sand and gravel.
02.4

Then letting x a apparent volume of sand.
and Y. the absolute volume of sand. the fommula.

$90 ( x - I) n % of voids in the sand and gravel.
x

In order to proportion the various mixtures
of sand and clay by weight instead of volume. a ratio
between equal volumes of dry and rcdded sand. and dry
and rodded clay was needed. In this particular sample
the sand was 1.43 times as heavy as the clay. Therefore.
if a voids-clay ratio of 50% was desired. when there
was 20 pounds of sand which had a volume having 30%
voids. the following computation would give the correct
weight of clay to be used. ‘

30:: of 2019s“?e clay to fill 1007. of the voids.

50 x 6i . 3i (clay to fill 50% of the voids.
00 (if the clay and sand were equal

(in weight.

3} a 2.28} clay needed to fill 50% of
7;“ the voids in the sand when the
‘ 3 sand-clay weight ratio - 1.43.

The foregoing is an example of the steps
necessary to proportion a set of test cylinders. Each
set was comprised of three cylinders. In this test
eleven sets of cylinders were made having the following
voids-clay ratios:

4.1%; 12%; 2o.45z; 40%; 50.8% ,
75%; 102-4%3 125%; 150%; 175k; 200%.

The materials {sand and clay) that went to
make up a set of test cylinders were thoroughly mixed
while dry. Then water was added gradually until the
mdxture was damp enough to hold together'when a handful
of it was compressed. The least water used to obtain
a workable mixture. the easier it is to ram.the earth
solidly in the molds. If too much water is used. instead
of compacting. the earth will Just squeese up around
the rammer. If such a condition should arise. add a
little more earth or else allow the mixture to stand
long enough to dry a little before placing in the molds.

The molds used were made of three-ply cardboard.
well paraffined. They were cylinders having a three
and one quarter inch diameter. six and one half inches
in height.

The dampened soil was placed in the molds and
rammed with the peen end of an ordinary foundry rammer
until the earth sounded solid. Another layer of earth
was added and rammed and so on until the mold was filled
level full. In the last few layers the ramrod was
replaced by a two and one half inch wooden mallet which
was held in the mold and struck with a heavy hammer.

Due to the fact that the molds had to be torn
from the rammed earth. the cylinder was allowed to
stand twenty-four hours before its mold was stripped
from it. In this way there was not as much danger in

breaking the cylinder due to the tension in pulling the

The cylinders were then allowed to dry until
the sets containing the most clay assumed a constant
weight. They were then ready to be broken in a com-
pression machine.

Before breaking. however they were capped in
plaster-cf-paris to evenly distribute the pressure
over the entire area of the cylinder. Some of the
cylinders which were low in clay could not be capped
due to the fact that the moisture from the plaster-of-
paris would get into the cylinder and cause it to
crumble. They were broken without being capped. In
the latter cases soft card board was used in place of
the plaster-of-parie cape.

The data collected. together with the curve
of the compressive strength plotted against the voids-
clay ratio. will be found in another part of this
report listed under ”Data” and also “Results and Con-

clusions.”

method of Procedure.

Tegt no.2.

As was stated in the outline of tests. this
test was to develcpe some experiments to determine ,
the voids-clay ratio of a sample of earth. It was
also to take a sample of earth of unknown composition.
to run these tests on the sample and then determine
its strenght from the curve of Test No. I. Also. it
was to take this earth and make test cylinders. and
break them after being cured as in Test No.1. This
procedure gave two values. which should check fairly
close. for the strength of the earth.

This test resolves itself into two sections.
first. to determine the voids-clay ratio and second.
to make the cylinders. and break them.to find their
actual strength.

The first section has two parts vis..getting
the sand-clay~weight ratio and second actually de-
temmining the voids-clay ratio.

To determine the sand-clay-weight ratio.
lg_.‘Weight a sample of earth. and measure its volume.
222. Wash the clay from the earth my the following test:-

7 Break up all sand and clay lumps with a mortar
and pestle. being careful not to break the rock fragments.
Pour the mixture into a one liter beaker in which stands

8 con. of ammoniated water of a concentration of 1:500.

Allow the mixture to stand for eight minutes. and then
pour off the supernatant liquid. Repeat this process
until the liquid. after standing eight minutes is
clear. The clay has now been washed out and the sand
and silt alone remains in the beaker.

352, Weigh the contents of the beaker.and measure
its volume. when dry.

43h. ' The difference in the two weights is the
weight of clay. '

Egg. The difference in volumes is the volume of
clay in excess of enough to fill the voids
in the washed sand and silt.

63h, Determine the percentage of voids of the
sand and silt without the olqy.

zgg. nultiply the volume of sand and silt (3;!)
by this % of voids. which gives the volume .
of clay necessary to fill 100% of the voids.

83h. Add this volume (7th) to the volume of the
(552) step which will give the total volume
of clay in the sample.

23;. Divide the weight of clay (41h) by the
total volume of clay (83h) to.give the unit
weight of clay.

1022' The weight of sand and silt (3rd) divided
by its volume. also the(3£g) gI?.. the unit
weight of the sand.

lith. Divide the unit weight of the sand (1953)
by the unit weight of clay (ath)to give the,
sand-claydwsight ratio.

The second part. to determine the voids clay

ratio. is as follows:-

1’1 O

gag.

Multiply the total volume of sand and silt by
the percentage of voids as determined in the
6th. step of the first part.

Hultiply this volume by the unit wt. of sand.
giving the wt. of sand necessary to fill the
'Qid.e

Divide the above weight by the sand-clay-
weight ratio. which will give the weight of
the clay necessary to fill l00fl of the voids.

Divide the weight of clay in the sample as
derived in the 4th. step of the first part.
This value multiplied my 100 is the voids-clay
ratio.

In the second section .2 Test No. 2 the

cylinders are prepared in the same manner as for Test

Ho.

1. and are broken in the same manner. See the

data sheet for computations and results.

Method of Procedugg.
22.2m-

This test. which was to determine if the
rapid drying out of the pies' de terre would cause
it to lose some of its strength. was carried out in
the following manner. ’

In test No. I.there was a set of cylinders
having a voids-clay ratio of 102.4}. These were
duplicated by another set of three cylinders of the
same composition. the same amount of ramming and the
same moisture €33¥235Z The first three cylinders
were drying naturally for Test No. I. The second
set were placed close together in a single row. A
bunson burner was placed on each side of the row and
a reflector placed back of each burner to cast as
much of the heat toward the cylinders as poesible.
The idea of this arrangement being to get a condition
as near as possible to what a person would have if
he wanted to hurry up the drying of his walls of
pise'de terre by building a fire on each side of the
wall. The purpose of this test being. as was previously
stated. to determine whether or not the rapid drying
process was detrimental to the strength of the pise'de
terre construction.

The data and results of this test will be
found listed as such in a later part of this report.

Method of Prgoedurg.

O 0 .

As was stated in the outline of tests. this
test is to build a model house of pise'de terre. The
purpose of this is two-fold. First. the primary
reason is to leave something tangible. besides a mere
report. to interest some student in the further study
of pise'de terre construction. Second the house was
~ painted with ordinary paint and is to be exposed to
the rigors of a Kichigan winter. Naturally. with two
such purposes in view. no results or conclusions will
be found in this report. The report will only include
a method of building the house along with a simple
drawing or plan of the same.

In the first place the house is not a
“Model” in the sense of being perfect as to design or
construction. The word "Model“ as used in this report
only means the building of a house on a small scale.
All sises and dimensions can be found in the blueprints
found with the “Results and Conclusions” of the previous
tests. in the final pages of this report.

The first step in the building of the house
was to build a foundation. Due to the small sise of
the structure. neat cement was used instead of concrete.
The foundation was anchored to the supporting platform
by nails having been driven into the platform and the

cement hardening in place around them.

As soon as the cement had begun to harden
a keyway was cut into it and then nails were driven
into the keyway. This was to bond the earth to the
foundation.

The forms for the walls were then built.
Ordinarily the forms for this type of construction
are only three feet high and eight feet long. The
walls are generally built by ramming this section
and then moving the forms forward and ramming a second
such section and so on until a course three feet high
is completed around the entire house. Then the forms
are raised and a second course is built in the same
way on top of the first and so on until the entire wall
is built. A very good detailed account of this con-
struction can be found in the pamphlete “0.8. Department
of Agriculture - Farmers' Bulletin Ho.l§00%under the
title “Rammed Earth‘Walls For Buildings.“

Because of the lack of rigidity in forms
built on a very small scale. the forms for the model
house were built in the form of the entire house.

The forms were placed on the concrete founda-
tion and the earth rammed between them. The window
and door caseing had a precast layer of cement around
theme This was to protect the earth around the windows
and doors from getting wet due to water soaking thru
the wooden sills.

The window casings and doors were rammed
into place and the walls built up to the required
height. On top of the walls. a layer of neat cement
was placed for protection as in the case of the windows
and doors. Before the concrete set. a wooden nailing
strip or sleeper was put into the cement. the sleeper
being flueh with the tap of the cement. The roof was
nailed to this sleeper. The roof was made of 'homosote.'

No attempt was made to design the house for
anything except to make it water tight and weather
resisting. A 'Hodel' house as to design and construc-
tion would be enough alone. for another thesis.

The houee was given three coats of ordinary
house paint after it had the appearance of being dry.

A detailed plan will be found at the end of
this report showing all the details that were introduced

in the construction of this model.

ata Sheet.

1

III! NO: I e

of cylinder 3 8.3 eg.;n.
:#e/eq.in.:TORS 1 on:

Greg! gectional area
CTLIRDER5VOIDS-CLAY:TOTAL

 

 

N0._ 1 RATIO :PRESSURB: :e ft. :APPEARANCB
: : #s. : : (Fell to
1 x 4.1 m : : : )pieoes when
2 : 4.1 t z : : (molds were
3 : 4.1 2 : : 3 )taken off.
4 § 20.45” E 280 E 33.8 i 2.44 (
5 : " : 220 : 20.5 : 1.91 )Very granular.
6 : ' : 230 1 27.7 2 2.0
: : : 2
7 : 50.8 5 z 675 : 81.4 1 5.86 )Still quite
8 : ” 3 775 : 93.4 i 6.73 (sandy but sub-
9 : ' : 615 : 74.2 1 5.35 )etantial look-
: g 2 3 1113s
10 : 102.4 4 : 1297 x 156 : 11.25 (Looks like a
11 z ' : 1242 : 150 a 10.81 )poor grade of
12 : ' a x 3 (concrete.
: : z z
13 : 75 7 z 1030 : 131 : 3.45 Quite compact.
14 z “ ‘- 1 : 118 e .5 send does not
15 x ' x 910 z 110 : 7.94 )1ooeen when
: z : : (handled.
a z : z
16 : 12 g x 100 : 12 z 0.86 {Hardly held
1 : “ : 95 : 11. 5 : 0.82 together
1 : ‘ : 110 : 13.25 : 0.9
: : z a
19 z 40 5 : 540 z 65 1 4.7 (More compact
20 z ' : 665 : 80 1 5.76 )but still quite
21 : ” z 420 : 150.5 : 3.64 (granular.
: : : s
22 : 125 5 : 1280 : 154 : 11.1 2Abcut the same
23 z ' z 785 : 94.5 : 6.8 as the 102. 4%
24 z ' : 1395 : 108 : 12.1 )cylindere
25 : 150 5 z 1100 x 132 E 9.5 (Very compact.
26 : ' 3 845 : 101 : 7.27 )slight
27 : ' z 1125 1 136 : 9.8 (cracks.
28 : 175 z 1 1255 E 151 E 10.8 )Full of hair
29 z ' : 1140 1 137 : 9.87 (like cracks.
0 t ' : 1095 : 132 z 9.3
1 z 200 5 3 1250 : 151 : 10. (The excess clay
2 : ' : 1220 a 147 1 10.6 )eeeme to have
I 3 . 2 122‘ 2 1417. K 1 10 - 6‘ (gnu-Ion an ind

Data Sheet.
1..t HOe lo

GTLINDRR velnsaCLxr:TorAL :is/sq.in.:TON8 EEK

 

 

N0. ; RATIO :PHE08URE: :sq.ft. 1

3 1 _¢_, : : z

1 E 102.45 3 1297 § 156 "E 11.25 2

2 : '_ : 1242 : 150 x 10.31 Iatural

3 : “ : :Cylinder was brokenz Dr e .
: : :in stripping mold.

4 E ' 3 1090 E 90.4 : 6.5 )

5 : " : 80 : 48.2 : 3.5 ( Heated.

6 : ' : 40 : 69.7 : 5.02 )

heated c linder develc ed

the wall; clogegt to the fireg. They also had a meg: ggggular

‘2 2.3r‘n62 e

   

 

8 LT & CONCLU ION
TEST NO. I.

The outstanding result of test No. 1. is
that the clay apparently has no strength of its own.
It only acts as a binder for the sand and gravel in
the mixture. This is shown by the voids-clay ratio
curve dropping off after it passes the 100% point on
the curve.

Another outstanding feature taken from the
curve. is that an excess of clay seems to cause very
eradic results. The value of the strength at the

751.3 Value /: [yam/abort»?
200% point was not shown on the curve9qbut the data
sheet shows that its average was higher than the 175%
point. The appearance of the 175% and 200% cylinders
may throw some light on this point. Although the sand
and clay was thoroughly mixed while dry. after moisten-
ing and ramming. the particles of excess clny seemed
to have greater cohesion than adhesion. causing a se-
gregation of the sand and clry particles. This condition
evidently caused weak spots in the test cylinders. The
cylinders having the greatest segregation. evidently.
would be the weakest cylinders.

The cylinders of this test did not stand up
under as great pressures as some made at the University
of Michigan. There might be two reasons for this;
either the amount of ramming. or the tenacious qualities

of all clays. light not be the same. The cylinders
at the university of Michigan were rammed with a '
pneumatic hammer until they sounded solid. The
cylinders of this test were rammed by hand until they
sounded solid. Two individuals might not agree on
this solid sound. The other reason would call for a
test of all clays possible. and then classifying
them by the use of some standard test such as the
“Capillary Potential Test”.

The above two reasons suggest two lines
along which further study might be of benefit to
this type of construction. First to determine Just
what part. the ramming of the earth plays in this
construction. To do this. make a set of cylinders
in which the composition was held constant. Strike
the cylinders with blows of equal intensity but vary
the number of blows. Allow the cylinders to dry and
break them in a compression machine.

Second - Get as many samples of clay from
all parts of the country as possible. Make test
cylinders holding the quantities of ingredients constant
but using different kinds or qualities of clay. Plot
the compressive strengths developed against the capillary
potential of the clay sample.

If the above tests proved successful. their

use along with the voids-clay ratio curve. would give

worlds of information to the timid and conservative
home-builder. who did not want to risk his money on
a venture in which he had to guess at the results he

could get from.the soil on his particular property.

—‘~v u—- w»— f—>‘— -—
.'.__ _'V_.'_ f , ' -__
? 4
- ..-...- -. “... _,._».... a.
T fl _ , ,
—-~-J.._i.--_'__‘.._.. ,- h
. o ' ‘ ‘

 

(

-—~.—-,~~»— ,L—.»

 

3 ' 5
.~.I'w hw— ..
- .. _.;._..._._.41

1 ,1... V -t—.— wt— ‘1’“ Y"
.-'. -.._.,_-... t.
..2,_..._,-_.. 2.- ~-

__, ' ,. 7 ,-r....1.._‘ 1....
, l 1 e V

_. .1". -. ‘ ' . ' L :
_ _ ‘,._,..a.._... -, .....- ..- 71,,“
. . g ' . 1 . .
I‘. w— » ‘-— —- .- - ‘ . ' ‘
.7 .1 .11 . . ‘ . , ...---..1-.+....L-+._1. . .. ... p--. .
r Y I ‘

‘ ‘ u 1 1

‘ ~ .
,_*.._.-.. ”—f“ b—

y I
.-_.. ;. -.

.',-_....__.__',_... . . . ' . . , . , . .',....._._
, _ n
u ‘ , ' ‘
.$—’~—r-e—-.-. . . , 7 . , , 11-. .
. .
. . _
r..._\.. .. .

7,_f:;:;‘:::1 .. .f 4 ,6 Ca/mrez/ (mi:

D

-... , 7 L ygggJ-CLALYEflflfik

Y.“ 2——~-----.—.~T—-m .

>-— ~._ ...,

I W“: . ' f , 1 a EVE';;:;::LL:_

3-4.11-1 " ' , ' ;. , ~‘ [‘- '. ‘2 ' MMJMTE cot/.255 ,
4...: .   . . a _ , vi /927"'-— -91...“

. . .
- . ifie- .
' 1

a _ .....A__._g....._...

1 ‘ V
_.4~4—- A..—
..o. -——..ev——J—»

~ep—__.—w—._.?..**—.A ,_ ,,

 

wunueeet-tmmwulm

Minoan»...- . A. .

 

RESULTS & CONCLUSIONS.
IE8: N03 113

The curve gives a value of 148 lbs. per
sq. in. for a soil having a voids-clay ratio of 170%.
The lowest cylinder actually broke at 159 1bs.per sq.in.

It was rather unfortunate to have tested
a soil above 100%. as the curve has not as even a slope
after passing the 100% point.

The values of the original test cylinders
were very eradic between 125% and 175% so the result
of this test is quite satisfactory. however before
any real conclusions could be drawn. a number of check
tests should be run checking the curve at various
points.

This test checks with the first test in
showing that there is a relationship between the con-
prsssive strength of earth and its voids-clay ratio.

It tends to further prove that the clay acts only
as a binder and does not lend any of its own strength
to the mixture .

HESULIS & CONCLUSIONS.
TEST N0, 111:

 

The data for test No. 3 shows that when
cylinders are dried hy artificial heating. eradic
results are liable to be attained. Therefore unless
a person could constantly guard the fires he was
drying his walls with. he might better allow them
the time necessary to dry naturally. By careful.
slow. heating good results probably could be attained.
but it evidently would not do to Just build the fires
and allow them to run their own courses.

Fast drying caused cracks to appear in the
outer‘walls of the cylinders*which‘would cause both
weakness and deterioration due to climatic conditions.
The cracks were due to shrinkage caused by the rapid

loss of moisture.

RESQLQS & CQQCLQSIONg.
2E8? 30, IV.

As has already been stated. no results
can be stated for this experiment at this time. The
condition of the model house at the end of a winter
season will be a convincing report of this test.

The plans and details of the house are

shown in the accompanying drawings.

mag Hex/,2: HO
Wowrwcw/g ,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

        

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fill/[f ”gamma/0N w/r/z 4 ‘R
[He's/.3 We a Q. ,2 ‘ '
fl: 2;: I? MM we era/P:
- _. , [M55050 m ND? 7
MfiME‘JV’YE ’9 2 7 affifflr WH/CH IJ‘ /"77wce
£2: El. PIP/IVE mm “P”
7'
£02 0
'7 1’ TOP OF 14/41; 5
g '%'fl¥narro 7‘ ~ in
- 0L0 lIETH . i_ ' \ .2 e?" 7
i _j 74 Fol/Noam " :
a ~
’ t 1‘ N
i .* 1.5065 70 '- . ‘
i .‘ JuPPaerFLooe‘ ». .1 . a
a \IYl/LS‘ To Bow . ' S
l' ‘ $3553.95?”
if r..._ 41 5' ' 5" ~ 4",
JI". — :3“ e i?‘:. 4 24"
L , .
£1" / ' (L20
f ’7 2 2. .5/0f [Iii/€770”
3o .
g {grieve/near 64.13/09
am una’doam'a n/mdono
r ( ' I § ,,
1'1 1 III 5
y
a g
‘h —s ‘r
‘\ 49 L ] {1 P r I fi
3. "_
3701" 1525 V67 W/Y
Fen/Yr [Ail/4 7/0”

 

 

[Eons—fl; L A/oop wen [5 flf /z ' Zw‘lfiff’
u MMDOMS 5365 54/175 3725- fl

 

 

Summary of Th.l3l°

Pise' De Torre construction. although an
old idea in itself._needs to have engineering principles.
and practices added to it in order to be of much use
to the present building situation. ‘

The houses are decidedly comfortable to live
in. both in summer and winter. The labor in building
constitutes their biggest item of expense. The materials
muet be found on the property. or site of erection.
in order to make the finished house cheaper than any
other house. When materials have to be hauled in to
get a soil of the right consistency. an ordinary frame
house would be Just as cheap.

Here-to-fore there have been no tests derived
to pre-determine the strengths of various soils. The
object of this thesis. primarily. was to try and develops
some such tests.

Starting with the same idea as is used in
proportioning concrete and assuming that the clay per-
formed the same function in piee' de terre construction
as the cement does in concrete. tests were run to check
both the idea and the assumption.

or course. no set rules or empirical formula.
should be accepted without their results being checked
by a number of tests. The time element limits this

experiment to a single check. This check seems to be

 

. . .Iiilv .a"?! I‘ll-nit

‘jr! (Hill

.l: 11‘ ¢ili| v

Quite satisfactory. the curve being on the conserva-
tive or safe side. However the curve should be checked
in a number of places before it is either accepted or
rejected.

The tests themselves suggested the following
lines of further study in Pise'De Terre construction.

I. . How much of the strength of construction
is due to ramming.

11. Will different clays give different results?

111. Does the sise of gravel or stones make any
difference in the results. .

IV. How long does it take to gain its full
strength. Show the gain in strength from
day to day by a curve.

v. Devise some way to make paint adhere to the
surface of the walls.

Vl. Design a model house which is a model both
as to design and construction..

The results of the tests performed were of
such nature as to need considerable more study but at
the same time they tended to prove that it is possible
to predetermine_the strength that soil can develope
without actually making test cylinders and waiting for
them to dry for four or five weeks.
If the soil is of the right composition. a
aubstantial and cheap house can be built of pise'de M
terre. If materials have to be brought in pise'de terre
loses its advantage. Piee'de terre of neeeesity must

be simple in design and constructed of local materiale.

iii 51‘: . I.

 

.% w

to

O

a
.a
u.“
l. n
.. . 4
e . .. ...
.. M
l . 0M

.

.w ..
o .h
a .

m .
. I 1&5

.

..
...
.
H
.
Ha
.
x h

 

 

‘tV

 

 

  

 

,uv—e—v - -. ~4v r

 

 

 

 

     

NNQN. -. .. prong} \
.a i aN % Q ”n §lQ§§

\anbwkuq N \h.

 

\~l
\

  

 

 

 

 

 

 

 

 

 

 

 

 

e >04”.
yin-“n; war
.3.

2; ..

.8

«

C-O-oL"
rm v» - 1.
1
Q
r
I
1
n
i
I
d
R
i
O
I
9
. L. . ~ -
...—“I- e-g A- ‘r-f—o*zrfi «qr-sewer -| .. e a

... moan—w r.
f4 ”or- V

....... ~
'9 I u. I. I .i e i.
Ir a! el a I .4 r 30!. st. .- IO..." 0. C .. e i.
urknei'h u. in..- I'FI 0. a... II la! a. I l I I ...!!! .0

 

3‘..‘ F.‘ _ u .' 'I... i. .u'.# - 51.1-, e'e‘b... I. 9...." -3!.-fli.‘.‘*.“.i\i » n. i b!

".....ws - --. - .. hr.-- n. ....HMU
a . . 3...”. K a. . m m
H . daub/1%? C ............ Lam; ...) W... w M
a . . ....
..wm .. ...}: 4...} F50..J5fle/...V a r
W”. ....v.......a.u..rl .. ......J .4) . w... v ..Aftjtede’..§fv w... .n. . . . m

... a

 

 

 

 

     

1‘M‘r‘.‘lw~‘

 

axexéi fingeohmtboxe
[Lieu testing .w

 

 

 

 

. Iii: all! '...t.e.§ec.,§ «venue... to 5 i

..s. we...” km. Syria...” 4.1]
J . .r i :1) . “.... ...

II. . . . eMD;a.IeI| PQngflhwnIm.» ...-l. .5 «I... u

._ v. . . and. .w. - ... I. o... .
OJ a IV... ... .0 ; ifismjldflungw. lie d .1; ..u .o .5”:

J ..efa a a
. . I; .. .. . I . . i . . .
0.. .- O of , M ... ... . i4.& .Dt... . a . .. ,n ’ ... é... V .. J
-... if .1! ...fl. 501cc)?» data. ,C in... 44.. .l'... ... EX." .5" .I.‘
‘. ‘7 Fe. n . I
...!‘e.’ O 9' 0 n I , -. a A I ‘ . a...
f v o .l u ... a . . r a“ It...
.‘wvb‘ Ie e I... he .' a 1. .VI I. as . v. D se t .9 Q... . as. r!
. a I. , . ...v. . . ... a {C
... r... A... .r. .. L ... . . ... . .. .
m. . ..- a . .. . tea (a... - .I
lies-...... at. ea -Gahle‘i . 5.. ...v I: . 1 v .11! "their. «29.31.... a. .03.
. ..s . a
I
‘ I 9H ... D.
. at ..
n v. r u t v .4 o e I. ..
595.”-..1001 ..e . . .- Vo. l M. .n‘ . o. 1 F. M ._ .V
e ..4
. ... .. .. N. 1 I n. . . .
.- I:..I..f§.“ ‘. .Il'. .9311. 'fl .I .I I I a.“ I re: . b I '3!
’ .( hi I .a
l D. .
d J mpg”! -r
.- see. (we ite..1 ..
...}. I . .I 1' ’1':
. r. ’l ' ft... \V P ’l, 3’
I ..m ... . . 1v: ”eh.0 .I. I
elute- v. ..ao. ... . , h i all
.. s r ... ...... .. v.- I. .. if.
e. .. . . .
v . fl . a. . .7 . I . ..k
.. . . I I .0 .....s. l. x. .. J
, .. ..l t... e! .u t . FY. . e
5". t .l. '51.! ......et. m H... v . .n . u. J?- . .41 . . ...... . ..

y . r In. .. v a .e I t.‘..
.. M. .. .. .H‘... r“ .....e ’r‘: ......Iavflw are .i- 623‘}. .1" an H
. H ... .v. .nh. {Iii-1L n is -‘L’I e. Exile-out" ..

a .
. n. .a is- L.. ..E .- .... ..ir ..II' .31... age-fr! a . e. ..
... a: .. u ....5'. ., . v. .... .Ifl...v..u.. ..u r. 1.....1‘

     

 

 

 

’9—
-.4b"

ww—

 

 

 

 

 

 

 

 

mumyuulfltmuIguflmlmujuWinn