. _:_:____:__::__:_:_:_:_:______

 

K“

~de ,‘1

3“" I‘.'r‘
r

“xii;

I.

‘

TH ass

I ‘
{:J O \) s \ I

A STUDY OF THE WATERTIGHTNESS OF CONCRETE
WITH AN ADMIXTURE OF CELITE

Thesis Submitted to
The Faculty of
Michigan State College

by
Theodore fidenburg
Candidate for the Degree of

Bachelor of Science

June 1987.

I.
II.
III.
IV.
V.
VI.

C O PITT“? HTS

Introduction
Materials

Mixes

Apparatus

fcsts and Results

Conclusion

103303)

I. INTRODUCTION

watertightness or permeability cannot in any sense be
underestimated as to importance in concrete construction. The
ability to retain or rather the flexibility with which water
permeates through some concrete is indeed a big question. In
such structure as concrete pipes, concrete ships, drydocks, canals,
dams, swimming pools and many others, this then is an important
item.

Theoretically, and it has been proven practical, it
is possible to get a concrete which will be absolutely water-
tight and which furthernore will hold water under very high
pressures. If an exact design according to the Portland Cement
Comtany is made and used, a concrete with the minimum amount of
voids and maximum amount of strength and watertightness can be
obtained.

This theoretical concrete, however, does not, as most
engineers know, hold absolutely on every job, especially those
of a smaller nature. On a job where a good mix is not designed
the water can easily seep through. For such mixtures various
admixtures have been tried, one of which I have attempted to
exreriment with. I will then in the following consider only a
crude and rough mix, and with this mix, by adding the given
admixture, attempt to get the desired result of watertightness.

The admixture which I will consider goes under the
commercial name of Celite. It is claimed that by adding this

to the mix, a denser concrete giving greater strength, work-

ability, less segregation, and watertightness is
I will at this time confine all my attentions to
permeability and watertightness with a few minor
vations on some other points.

The prcblem then is to test a concrete
this admixture as to its watertightness. I will

discuss the materials used, then the mixes, then

obtained.
the

obser-

with
first
the

apparatus, and finally the tests and results obtained.

Finally, I will attempt to draw suitable conclusions

from them.

___.__.._..._.._._.__x._n _ __

II. MATERIALS

The cement used in all the mixes was Petoskey
Portland Cement. No tests regarding it were made. Rather,
it was used just as it came from the factory where it was
manufactures.

The fine aggregate which was used was obtained
from the bins in the concrete laboratory and was used
just as it came out of the bin without any tests being
made upon it.

The coarse aggregate was also obtained from the
bins in the concrete laboratory. It was very roughly
graded as to size and also as to cleanliness, the ex-
tremely large pieces being rejected and the dirty pieces
being also thrown out.

The Celite in composition is but a very finely
ground silica 3103. In the mixture it is inert and is
not acted on or does not act on any of the substances in
the mix. It was obtained from the Celite Products Company,
New York. It is through their courtesy and willinwness
to cocperate that we were able to obtain this admixture so
as to be able to experiment with it. No tests were per-
formed with the Celite other than using it in the various
experiments as it came from the company.

The water used in the mixes was obtained from

the college mains.

’1'! tr,

III. MIXES

 

hree mixes, only one being standard, were
used. A 1-3-4, 1-8-5, 1-8-6, mixes were used, the 1-9-4
mix being a rather common one, while the other mixes, the
1-3-5 and 1-8-6, are very rare and in being rare are ex-
tremely high in coarse aggregate. The Celite if acting
in its proner capacity is assumed to fill the voids in
the mix and in the case of the 1-2-5 and 1-2-6 mixes an
excellent opportunity to observe this could be had by
noticing their relative watertightness in the tests.
A mix of this nature could also then show us Just how
coarse a mix could be before it became pervious and also
through the addition of Celite it might be made impervious.

I first decided that two slumps, they being a
3' and 4', would determine the water content but this was
not followed out throughout the work. Discs were made for
both a 3-inch and 4-inch slump in the 1-8-4 mix, but due
to the coarseness of the 1-2-5 and 1-8-6 mixes, I decided
to use only the 4-inch slump in them, and attempt to get
the results with it. From each mix I made five discs with
varying reroentages of Celite 1n prorortion to the weight
of cement.

Number 1 with 0% Celite
agent).
- a 1:37
I 4 I 5% a
5

s 87: II

To obtain the required amount of Celite in
the mix I had to weigh the amount of concrete in each
successive disc and then find by proportion the weight
of cement in the disc or rather that remaining in the
mix and also the weight of Celite remaining in the
mix. This then gave me a figure for the weight of
cement remaining in the mix from which I could compute
the amount of Celite necessary to add each time.

A simple proportion is all that is necessary to de-
termine this and there is no need at this time to
demonstrate it. The following table will show the
quantities and system of numbering used as the discs

were made in the concrete laboratory.

Disc No.
l

COQQQU‘IIPOIN

a) t4 P‘ rd :4 #1.»: Id F‘ r4 r4
<3 a) co 7e c: tn 9- 0: ts F‘ C)

in;
1-2-4
I
I
I

1-8-6

mwomgmwom’gmwomm

H
'01»:

(I)

Date Poured

April 14

April 18

April 21

IV. APPARATUS

The suggestion or idea on apparatus was brought
to me through a series of tests conducted by C. F. Meansell
and C. J. McLean on their thesis work in 1930. A detailed
drawing and sketch of the apparatus may be obtained by
referring to the thesis by these two men. I accepted
their principles and have made a few slight revisions
which I thoughtwould better accomplish the results hoped
for. From the sketch which you see on the following pass,
the disc three inches thick and seven inches in diameter

is placed between two flanges and then firmly clamred
down by the bolts around the edge of the flan es. The
apparatus seemed to be very satisfactory except for one
detail which I think it best to explain; then if further
work is to be done along this line, a better and more
efficient method could probably be found.

The biggest question which I had to contend with
and which spoiled my results the most was the question of
making the joint between the iron flange and the concrete
watertight. The disc was made with an even surface on top
and with a rather rough surface on the bottom. To make a
watertight connection, it was first thought that just a
small rubber gasket would do but this failed miserably.

Larger gaskets were then used which worked much better but

 

 

 

not to satisfaction. Then the same larger gaskets were
used with a filler beneath them and with a thick layer of
putty around them which gave about the best results but
which still did not prevent leaking. It was found also
that the discs cracked because they were not brought up
even all around so plaster of Paris was placed on the under
side and in this manner an even surface was obtained all
around the face of the disc. This last combination of
gaskets and leak preventatives worked out the best but
they were far from perfect since even after they had been
put into use the water nevertheless leaked over the edges
between the flange and disc.

It was thought best at first to catch just the
water which came on down through the discs and then through
the two—inch hole in the bottom flange but since the
pressure of the water in the disc would be exerted equally
in all directions it would then be expected to find the
water seeping out of all sides of the disc as well as out
of the bottom. fhe discs were therefore placed horizontal
and all the water which permeated out of the sides as well
as through the bottom was caught in the pails placed on a
platform beneath the discs.

The water pressure was obtained from the college

and was registered by a gauge shown in the accompanying

photograph. Although I at first expected to use different
pressures ranging from approximately 10 to 70 pounds per
square inch, this was not carried out, and the maximum
pressure of 70 pounds per square inch was used throughout
every test. I thought that this maximum pressure would
give the most practical results, it being similar to those
in concrete pipe and other structures needing to be abso-
lutely watertight.

No apparatus was available to obtain a reduced
pressure so that seepage and other small pressure heads

similar to seepage could not be used.

V. TFSTS AND RESULTS

Each disc was mixed and formed in a metal mould
coated with oil on the inside to allow it to be easily
disconnected from the concrete. The discs were then
placed in a damp chamber and allowed to set for fourteen
days.

It was7 hoped that results of a very accurate
nature could have been obtained so that graphs and charts
of the flow through the discs might be made indicating
the time it took for the water to permeate through; the
amount through each disc; and the relative efficiency of
varying percentages of Celite as added to the mixtures in
the discs. Due to leakage in almost every case caused by
a bad joint, and cracks, it was found that accurate Quan-
tities of water could not be obtained. The combination
of bad joints and cracks in the concrete allowed an ex-
cessive amount of water to drip into the pails so that only
a rough account of quantities was obtained and written up
rather than being placed in.tabulated form or in graphical
form. Following is a list of the trials as they were
observed:

Disc 1. A 1-3-4 mix with 8 inch slump and 0% of
Celite. Cracked and considerable water'permeated through
cracks. A considerable amount also ran outside between

flange and disc through gasket. From close observation

the water was noticed oozing out the side and bottom and
more water came through from this disc than any of the
other three on at that time.

Disc 2. A 1-3-4 mix; 2 inch slump; 1% Celite.
Did not crack but water did get by the gasket. Water
also permeated through sides and bottom as in all of the
other discB by the beads noticeably forming- on the sides.

Disc 3. A 1-3-4 mix; 8 inch slump; 1%fi Celite.
Did not crack but gasket leaked. Considerable amount of
water came through.

Disc 4. A 1-3—4 mix; 3 inch slump; 6% Celite.
Cracked and no accurate results could be obtained.

Disc 5. A 1-3-4 mix; 3 inch slump; 8% Celite.
Did not crack but gasket leaked. Water permeated
through but it was not as outstanding as in the other
tests.

Disc 6. A 1-3-6 mix; 4 inch slump; 9% Celite.
Did not crack but was very porous. It was extremely
more porous than any other disc made. Indications
from the amount of water which came through the sides
and out of the bottom showed that this disc was very
porous.

Disc 7. A 1-3-6 mix; 4 inch slump; l? Celite.
This sample cracked but nevertheless considerable water

did come through other than the cracks.

Disc 8. A 1-3-6 mix; 4 inch slump; sfi Celite.

This sample did not crack but Quantities of water
similar to that through the others Game through this
one.

Discs 9-10. 1-8-6 mix; 4 inch slump; 1% and 8%
Celite. Both these discs cracked but water permeated
through them as much as through any of the others of
this series.

Disc 11. A 1-3-5 mix; 4 inch slump; 0% Celite.

Did not crack but very porous.

Discs 12-13. 1-2-5 mix; 4 inch slump; 1 and lhfi
Celite. Number 13 cracked but nevertheless it was very
porous in the uncracked part, similar to No. 14.

Discs 13—15. 1-8-5 mix; 4 inch slump; 6 and 8%
Celite. Neither of these cracked but water did permeate
through them shortly after the full force was turned on.
After allowing the water to remain on for four hours,
conside able water did permeate through but not of such
a large quantity as through those of a smaller percentage
of Celite or in any of those in the 1-3-6 mix.

Disc is. 1-3-4 mix; 4 inch slump; 0% Celite.
Cracked but very little water permeated through the
cracks. Some, however, did come through the sides and

bottom although in not such large quantities.

Discs 17-18. 1-2-4 nix; 4 inch slump; 1 and 15% Celite.
Number 1? cracked and water permeated through the cracks.
In Number 19, which did not crack, the water permeated
very slowly through the fissures and irregularities on
the sides of the disc. That which did appear, however,
seemed only to wet the surface, none getting through
due to evaporation.

Discs 18-20. 1-3-4 mix; 4 inch slump; s and 8% Celite.
Neither of these discs cracked. Due to their high
Content of Celite they were supposedly more watertight,
but did not however prove such. The water seemed to
permeate through these much faster after the pressure
was turned on than through the others of this same

as: 31880

VI. CONCLUSION

It is rather a hard proposition to draw con-
clueions from the data and results presented, due to the
great inaccuracy experienced on account of improper
gaskets to keep the water from flowing between the disc
and flange and also due to the cracking of some of the
discs. I had hoped to obtain graphs and more accurate
data from which proper results and conclusions could have
becneeasily obtained. If the time would have allowed
I would have deemed it more advisable to have disregarded
my tests and would have gone about making more which I am
sure would have been more accurate and safe to draw con-
clusions from.

However, from the results obtained I will attempt
to give my conclusions. It can very easily be seen that
the leaner mixes, as the 1-3-6 and 1-3—5, thrown together
as roughly as these were, are not satisfactory. In these
mixes it appeared to be rather an apparent fact, however,
that the more Celite added to the mix, the more impervious
the mix became. It is my contention that if the correct
amount of Celite be added to these mixes, they can then
be made somewhat impervious. I say somewhat impervious
because the leaner mixes would have the greatest amount
of voids and it is alsost an impossibility to fill all

these voids and do it properly.

My next point in conclusion is that to get an
absolutely watertight concrete the mix must be accurately
and thoroughly,designed and worked out. In the 1-3-4
mixes, both the 3 inch and 4 inch slump, it was found that
the water permeated through regardless of the varying
amounts of Celite added, although it was noticed that
' in the mix where only 1 and 15% of Celite was added, the
water did not appear to get through as quickly or in such
large quantities.

Celite as an admixture I would say should be
used in very small quantities principally as a safety
factor in a well-designed mix. It should not be relied
upon to make a mix which lacks thorough design meet the
requirements of a well-designed mix.

In summarizing I will quote from a chapter in
Hool's ”Reinforced Concrete“, Volume II, which will and

does explain my two points very thoroughly.

"The addition of so—called 'integral' water-
proofing compounds will not compensate for lean
mixtures, nor for poor materials, nor for poor
workmanship in the fabrication of the concrete.
Since in practice the inert integral compounds
(acting simply as void filling material) are added

in small quantities, they have very little or no

effect on the permeability of the concrete. If
the save care be taken in making the concrete
imperreable without the addition of waterproofing
materials as is ordinarily taken when waterproof-
ing materials are added, an impermeable concrete

can be obtained.‘

APPENDIX I

The followin: is a list of books and references
from which valuable information on concrete was obtained

in the preparation of this thesis:

I. Thesis of 1930 by C. F. Meanwell and C. J. McLean -
”Study of the Permeability of Concrete"

II. Hool - "Reinforced Concrete Construction"
Volume II, Chapter XXIII

"Waterproofing of Concrete"

III. H051 and Johnson -

Concrete Engineers' Handbook

APPIHUIX II

Watertightness or Pereeability of Concrete With

An Admixture of Celite
Outline

Materials:
1. Celite obtained from the Celite Company.
2. Gravels of various grades obtainable from the
college.
3. A slow and rapid setting cement to be used.

4. Sands of various grades.

5. Water to be obtained from the college mains.

Mixtures:
l. Mixes to be graded as to quality by
(a) Kind of materials used.
(b) Workability.
(c) Carefulness in preparation.
8. Three standard mixes will be used:
(a) 1-3-4
(b) 1-3-5
(0) 1-3-8
3. Percentage of additional Celite will be varied -
o-I-IL-a-s.
Tests:
1. All discs to be tested under water pressures of
varying amounts.

3. Mix to be tested by the slump test.

 

 

 

 

 

r
,
x
u
n y
b
a
v .

Erhlu'flfl .dl...r\l§n‘.l.t (I I. 911.1511i'1.1 until! Hu .1. «Ian-Una... 1.. .h.\V «II-31.: .INAII1II ..:!ufl.. IA. ..|. . i .. n I ,.,
.‘u . ‘5’: a,-vl.,r-.yl n . . , .

 

 

 
 

HIIHIIIIHIH

Ill

8