American Society of Civil Engineers.

Transactions of the American Society of Civil Engineers (Volume 81) online

. (page 91 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 91 of 167)
Font size
QR-code for this ebook

existing before a body of clay has failed in shear along a given plane
or surface and the conditions existing after the resistance to shear has
been overcome (that is, when, in his terminology, cohesion has given
way to friction, plus any adhesion). Such distinction does not appear
to be inconsistent with Coulomb's laws, especially as expressed by
Professor Cain somewhat more broadly in his paper on "Cohesion in
Earth.":}: In fact, Professor Cain's more frequent use of the term
"shear" in his analyses based on CovQomb's laws, as applied in his
discussion (page 1045), is significant. Even Professor Cain now
considers that possibly not all "cohesion" (Cain) is destroyed on the
beginning of the actual motion or sliding.§

However that may be, all must be in full agreement with Professor
Cain in his emphasis of the "need for comprehensive experimentation

* Flood Flows," Transactions, Am. Soc. C. E., Vol. LXXVI, p. 564.

t Water Supply Paper No. 162, U. S. Geological Survey.

t Transactions, Am. Soc. C. E., Vol. L.XXX, p. 1316.

§ This appreciable part of "cohesion" everted in addition to friction during the
actual sliding of clay on clay (p. 1045), the writer would term "adhesion", as more
accurately describing its nature.


1U90 discussion: eeconstkuction of stoxy river dam

Mr. to determine the coefficients of cohesion." Moreover, the writer be-

lieves that Professor Cain has performed an important service in
presenting in such detail the several analyses of the Stony River
problem on the basis of coherent earth. It is not improbable that,
as the result of such work as is now being carried on by the Society's
Special Committee on Soils, the treatment advanced by Professor
Cain may be adopted, at least in part, and especially as applied to
cases of actual sliding.

The writer agrees entirely with Professor Cain that probably the
effect of a longer time of application of a given shearing load on a
body of clay under test would be to afford lower values of unit shear-
ing resistance (page 1046). It should be noted, however, that, in the
Stony River design, no reliance was placed on the shearing resistance
of the foundation soil.

In the interesting analyses (pages 1046-48) of the original struc-
ture before failure, based on the theory of coherent earth. Professor
Cain unwittingly made several erroneous assumptions.

The mistake apparently was caused by his treatment of the section
of dam shown on Plate XII as being taken at the maximum height por-
tion of the dam. This is not the case, for Plate XII shows the
"Typical Section at Bay 35" referred to in Columns (3), (5), (7), and
(9) of Table 3. The resulting error in the value of Q, as derived by
Professor Cain (page 1047), is not serious, but, unfortunately, the
same mistake made later in his discussion robs the results of their
significance, though, of course, it does not deprive the theoretical
analysis of its value as such.

Thus, in assuming anchoring walls applied to the original struc-
ture, as shown on Plate XII, Bay 35, Professor Cain perforce used loads
(page 1048) as given in the writer's Cases I and II (pages 956-958)
and applying to the sections of maximum height, instead of to the
section at Bay 35. The difference in height of sections will be appar-
ent on reference to Plate IX. Moreover, the loads used by Professor
Cain apply to the structure in its original condition, whereas both
concrete and water weights were greatly increased in the reconstruc-
tion, without any increase in the horizontal water thrust (Table 3).
The horizontal water thrust for Case I, for instance, is but 53 820 lb.
per lin. ft., instead of 81 000 lb., as assumed by Professor Cain.

In consequence, the conclusions as to resistance to sliding, fur-
nished under the respective hypotheses set up under the theory of
coherent earth, are not applicable. The corrected figures show much
larger margins of safety. Nevertheless, it appears to the writer that
the appropriateness of the methods used by Professor Cain has in
no wise been affected; and, for the purpose of theoretical study, those
interested may well disregard the inapplicability of the figures.

discussion: reconstruction of stony river dam 1091

Professor Cain neglects the resistance along BG and IJ of his Mr..
Fig. 28, on the ground that there is "but little weight over parts of Scheidenhelm.
these surfaces" (page 1053). It should be evident, however, on study
of Plate XII and the description of the toe- wall and its effect (page 964),
that the new toe-wall in essence extends the base of the dam as far
as F (Pig. 28), and that, in fact, the greatest vertical loads on the
soil occur at the very down-stream edge of the toe-wall, namely, just
to the left of the plane represented by the line, FC. For this reason
the writer believes that one is not warranted in omitting from consid-
eration the resistance along BC and IJ.

The graphical solution for the passive resistance of the down-
stream foundation soil is exceedingly interesting, and here again those
interested incur a debt to Professor Cain. The point made by him,
that the writer was too conservative in assuming zero load at the
free surface of the earth at F (Fig. 28), is well taken, as is evident
from the analyses (pages 1057-58).

On the subject of the safety of the structure, as repaired and
strengthened, the measures taken appear to have general approval, ex-
cept by Mr. Eutenberg. In view of the nature of the discussion sub-
mitted by him (pages 1029-35), the writer is loath to make a detailed
reply. However, a number of the premises adopted by him diverge
so widely from the facts that it appears advisable to examine some
of the points which he has raised, in order properly to co-ordinate
the paper and the discussion thereof. Except in special instances,
the writer will refrain from discussing mere expressions of opinion
appearing in Mr. Rutenberg's contribution, for, of course, each engi-
neer is entitled to his own opinion, and it would be futile to attempt
to obtain agreement on all points :

1. — With reference to the discussion on remedying the faulty con-
dition of the foundation soil (page 1030), the writer desires to
emphasize that reliance against leakage was placed solely on the exten-
sion and perfecting of the cut-off wall, and not on the compacting of the
foundation soil by grouting. The object of the grouting, as stated, it is
believed clearly, on pages 987-988, was to compact the soil for
bearing purposes. In other words, it was to remedy conditions caused
by leakage which occurred prior to the failure of the dam; or, in
the words of the desideratum expressed by Mr. Rutenberg himself,
"to fill, if possible, all the holes found in the foundation soil" (page
1033). It will be noted that the subject of pressure grouting was
treated by the writer primarily imder the heading of "Footings", and
not under that of "Leakage and Drainage."

In passing, it may not be out of place to remark that it is the
writer's experience that pressure grouting in porous soils and gravels is
not a sure means of causing water-tightness, but that, on the other hand,
such grouting is generally effective in rock having pronounced seams.


1092 DISCUSSION : reconstruction of stony river dam

Mr. 2. — The writer agrees in essence with Mr. Rutenberg's statements

in the last paragraph on page 1030, but cannot agree with his con-
clusion as expressed in the paragraph following. The writer must
still insist that ''the smaller the coefficient of frictional resistance,
the less the net effect of uplift pressure." Perhaps this anomaly of
agreement on certain premises and disagreement as to the conclusion
is explicable on the ground that Mr. Rutenberg has not noted that the
writer's statement applies solely to the effect of uplift pressure in
so far as resistance to sliding is concerned. This is manifest from
the context (page 947).

The meaning of the writer's statement can perhaps be made more
clear by using a numerical example. Thus, if over a given area of
base of dam there should exist an uplift pressure of 100 000 lb., there
would result the equivalent of a net reduction in weight by the same
amount. The net reduction in weight, however, is a different thing
from reduction in resistance to sliding. For instance, if in this hypo-
thetical case the coefficient of frictional resistance were 0.75, the
resistance to sliding would, by reason of the assumed uplift, be
decreased by 75 000 lb. In the case of the Stony River foundation
soil, however, the coefficient of frictional resistance probably averages
about 0.33. This coefficient, applied to the uplift condition just assumed,
would result in decreasing the resistance to sliding by only 33 000 lb.

3. — The expression of opinion that leakage and percolation in the
dam should have been intercepted "directly and immediately on the
up-stream face of its cut-off wall, and discharged down stream by
special channels without any pressure" (page 1031) is so unique and,
in the writer's opinion, illogical, that he hesitates to believe that it
was intended to make such a statement. The writer insists that one
need be concerned only with such water (leakage) as passes through
the diaphragm (intended to be water-tight) of which the cut-off wall
is a part. Of course, the primary effort was devoted to making the
cut-off wall as tight as possible. In this regard, the results of the
reconstruction speak for themselves, namely, that at present the
merest trickles of water pass through or under those portions of
the dam founded on earth over-burden.

4. — Although desiring to avoid useless quibbling over terms, it
seems necessary to point out that the statement imputed to the writer,
"that water under pressure in the foundation soil is never harmless"
(page 1031), is not warranted by the statement that, in the case of
the Stony River foundation, deep-lying pockets, if any, of water under
pressure are "relatively harmless" (page 997). By "deep-lying pockets
of water" the writer meant accumulations of water, under pressure,
existing at elevations below the bottom of the new anchoring wall at the
heel ; in other words, below the probable location of the main "planes of
least resistance." Such deep-lying pockets of water are relatively


harmless, partly because of their depth, and partly because of tlie Mr.

comparatively small coefficient of frictional resistance obtaining in
the Stony River overburden, the eifect and relation of which coefficient
to uplift pressure and resistance to sliding have been explained in
Paragraph 2.

5. — The evidently unintentional misstatement is made that the
writer "admits the presence of considerable hydrostatic pressures at
a distance of 20 ft. from the down-stream side of the cut-off wall"
(page 1031). The writer's actual statement, as given on page 998,
is believed to be self-explanatory to a careful reader. The writer
does not admit the presence of such considerable hydrostatic pres-
sures; on the contrary, he desires again to point out that no pressure
head of more than 2 ft. has been detected at the top of any weep-
holes or drain pipes since the reconstruction was completed (pages

6. — The premises being erroneous, as shown above, Mr. Rutenberg
arrives at a conclusion manifestly erroneous, namely, that "the
up-stream half of the base is exposed to dangerous uplift pressures"
(page 1031). This erroneous conclusion leads to a likewise erroneously
applied train of "previously stated consequences" (for which see
page 1031).

7. — Again, there is imputed to the writer cognizance of the actual
existence of the rather strained hyoothetical phenomena which Mr.
Rutenberg sets up as to the possibility that "the drainage pipes
will be filled through the higher perforations from the higher strata
of the foundation soil, etc." (page 1032). It is conceded that such
conditions might exist, but, if so, it is far from certain that the
results would be harmful. Be that as it may, however, the phenomenon
referred to by the writer (page 1000) clearly lends no weight what-
soever to the hypothetical conditions proposed.

Relative to this matter, it should be noted that the fact that so
many of the weep-holes and drain pipes remain just full of water, but
not over-flowing, is sufficient proof that the source of such water is
ground-water, and that it is not the result of leakage from the reser-
voir. The writer is inclining more and more to a belief in the
correctness of the explanation of the phenomenon on the basis of

8. — From a casual inspection of Plate XII, Mr. Rutenberg ventures
to pass judgment as to the strength of the anchoring wall at the
heel. That the inspection must have been casual is shown by the fact
that, contrary to his statement (page 1032), Plate XII does give the
essential dimensions, and even the details, of the reinforcement of
•the anchoring wall at Bay 35. Furthermore, data as to the anchoring
walls at Bays 30 and 31 are shown on Plate XIII.

1094 DISCUSSION : eeconstkuction of stony rivee dam

Mr. All exposition of the detailed calculations involved in the design

■ of the anchoring walls might have been included in the paper. How-
ever, even as it stands, the paper has been characterized as exhaustive.
The addition of such an exposition, therefore, did not seem advisable,
and at this date the military duties of the writer preclude such
diversion on his part. Hence it must suffice to state that the anchoring
walls at heel and toe were designed, with standard factors of safety,
so as to carry, jointly and safely, all the horizontal thrust, excepting
only an amount equal to the product of the weight of the super-
structure (plus water load) multiplied, in the case of the clayey soil,
by a coefficient of frictional resistance of 0.2 (page 978). The latter
amount of horizontal thrust can certainly be counted on to be trans-
ferred into the foundation soil directly by the frictional resistance
existing between the soil and the base of the superstructure. Essen-
tially the same method has been used by Professor Cain, in the
admirable analysis contained in his discussion (page 1050-51). The
only difference is that Professor Cain there considers the writer's
"frictional resistance" as subdivided into friction and "cohesion."

As a matter of fact, the results obtained by Professor Cain for
the load carried directly from the base of the dam into the earth,
and thence down to the "planes of least resistance", without being
transferred through the anchoring walls at the heel or toe, are very
much greater than those obtained by the writer by the use of the
coefficient, 0.2. It will be noted that the coefficient assumed by the
writer for this particular purpose is only 80% of the "probable
minimum value" shown in Table 1; and, as such, it is believed to be
thoroughly conservative, with the purpose and result that the anchoring
walls were correspondingly built so much the stronger.

9. — The fear is expressed that the toe-wall "will aid in collecting
the leakage under the dam" (page 1033). Inasmuch as weep-holes
were left througli the footings (see Plate XII), it is not believed that
the hydrostatic pressure in the foundation soil at the toe-wall can be
sufficient (the depth of the wall considered) to require relief through
the toe-wall. An exception exists in the case of the old spillway,
where the toe-wall is especially deep (17 ft.). There, drainage openings
were provided through the vertical member of the toe-wall, the details
of which are shown in Fig. 27.

10. — With reference to the effect of the heel-wall and its "under-cut
slope" (page 1033), the attention of Mr. Rutenberg is respectfully
called to the physical fact that the heel-wall cannot possibly present
additional surface for the application of uplift pressure without at
the same time presenting additional siarface for the application of
downward water pressure, which, of course, resists overturning. Even
granting that the up-stream portion of the heel-wall is subjected to*
full hydrostatic pressure on the "under-cut slope", the net effect of


the heel-wall would still be to resist overturning, for the reason that Mr.

the specific gravity of the heel-wall concrete is much greater than that *^ eidenhelm.
of the water which it displaces. In this respect the location of the
heel-wall so far up stream, and its correspondingly great lever arm,
are of no small consequence in actually increasing the stability of
the structure against overturning. Manifestly, the result is diametri-
cally opposite to that stated in the discussion.

11. — The writer quite agrees with the implication regarding the
tests of frictional resistance of various Stony River soils, to the
extent that it would be dangerous to rely on these as giving truly
representative results (page 1033). However, it by no means follows
that the tests are "without practical value." On what is our knowl-
edge of the frictional resistance of soils based, if not on observation
and tests, as well as on theory ? It should have been unnecessary again
to point out what was evidently overlooked, namely, that the vsTiter
discounted the test results to such a degree that, instead of assuming
an average coefficient of 0.61, he based the design on a coefficient of
0.33 (page 953). Moreover, the structure is shown actually to be
safe with a coefficient of 0.25.

12. — Similarly, it should not have been necessary to remind any
one that no pretense has been made by the writer that the weight
of the underlying foundation soil aids, in the least, to resist over-
turning. The weight of the underlying foundation soil is utilized
solely to resist failure by sliding. Of course, once such soil has been
washed away, it can no longer resist sliding; but failure resulting
from such conditions would have been due primarily to leakage or
undermining; and against failure of that sort entirely different
precautions have been taken.

13. — Mr. Eutenberg is under a misapprehension as to the writer's
"adding 3^ ft. of water to the original level in the reservoir" (page
1034). As a matter of fact, the writer did not add to the water level,
but merely made provision for taking care of floods, which would have
come with equal probability in the cases of the original and the
strengthened structure. True, a parapet was added to prevent over-
flow at the bulkhead sections and consequent undermining at the
down-stream edges of those sections. However, that the additional
horizontal water pressure caused thereby is negligible should be patent
to every one with a moment's computation. On the other hand, the
new spillway serves to hold down the water level in the reservoir
for a given flood discharge, as compared with the water level which
would have existed in the case of the original structure for the same
flood discharge. It is pertinent to note that, for instance, even on the
basis of a run-ofl of 1 386 sec-ft. per sq. mile (as reported for Cane
Creek, page 926), the bulkliead section of the original structure would
have been overtopped by 2^ ft. That is, such a discharge could have

1096 discussion: reconsteuction of stony rivee dam

Mr. taken place only after the water level had risen to an elevation of

e m. g^i^Q^^ 141.5. The corresponding water level under the new conditions
for a spillway discharge of 1 386 sec-ft. per sq. mile will be approxi-
mately at Elevation 141.65. Evidently, the difference is negligible.

Again, reference to Table 3 will show that the same elevation of
head-water, namely 140.5, was used for the stability analyses for both
the original and the strengthened structures under ''normal maximum

14. — It is true that the superstructure of the new spillway section
is heavier than that of the original portions of the dam. However,
it does not follow, by any means, that either "the original structure
is insufficient or the excess of the new is useless" (page 1034). As to
the original structure, the writer was forced to choose, on the one
hand, between recommending the expenditure of a relatively enormous
sum of money in order to strengthen the decks and buttresses, and,
on the other hand, accepting somewhat higher working stresses than
would have been allowed had the structure been designed initially
for the water loading for which it should have been designed. As to
the new spillway, the increased cost of using normal working stresses,
as compared with the cost for the slightly higher stresses obtaining
in the original portions of the dam, for the same loading, was not
great, and there would have been no justification for holding down the
new spillway dimensions to correspond with those of the original
structure (page 1005). A situation of this sort requires the use
of common sense as well as the application of theory.

15. — The unit bearing stresses for the soil and the base of the
dam, the necessary data for the computation of which Mr. Rutenberg
misses (page 1034), are set forth in considerable detail on pages
978-982. Further details might have been added had it been desir-
able to extend indefinitely the limits of the paper.

16. — The writer is unable to agree that the reservoir or ground-
water leakage, mentioned on page 1000 and discussed on page 1035,
has found other opportiuiities for "harmful erosion." Mere differences
of opinion are immaterial, but it is to the point to know that since
about a month after the reservoir had completely filled, there have
been no further appearances of muddy water at any of the many
drainage openings in the dam. This fact would seem to bear out the
writer's opinion (page 1000) that the muddy nature of the water
which appeared was due to adjustments in the foundation soil to the
new conditions of loading.

The foregoing comments have been essentially on divergences from
the facts. However, certain mere expressions of opinion on the part of
Mr. Rutenberg aeem to warrant passing mention. For instance, he
"believes the reconstructed dam is less safe than the original structure"
(page 1035) and "is firmly convinced that the reconstructed dam will

discussion: reconstruction of stony river dam 1097

also be broken, perhaps soon, if adequate repairs are not made in Mr.

time" (page 1035). It is unfortunate that opinions of so grave a
nature and predictions so dire should be advanced, based, as they
are, on erroneous premises, and certainly not substantiated by the
facts. Probably the most important considerations in this regard are
the results of the reconstruction, which speak for themselves :

a. — In the first place, the reconstructed dam has now stood for more
than 2 years under full water load. The original structure stood only
65 days under full water load prior to failure.

h. — Shortly after the reconstruction was completed, the dam was
subjected to a flood which caused the head-water to rise about 3J ft.
higher than the highest water level which had occurred during the
short life of the original structure.

c. — Finally, an inspection made recently by the writer showed that
the dam is in excellent condition.

Certain criticisms have been offered by Mr. Moore (page 1035),
Two of these in effect duplicate criticisms made by Mr. Eutenberg,
namely, as to the value of the writer's tests of frictional resistance and
as to the structural strength of the new anchoring walls at the heel
of the dam. The writer's comments in regard to these criticisms may
be found in the paragraphs numbered 11 and 8, respectively.

The contingency of the fractiire of the anchoring wall at the heel,
to which Mr. Moore refers, is not believed by the writer to be a serious
one. The extent to which such considerations are pertinent is a
matter of judgment, as to which it is not surprising to find differences
of opinion. Had it been considered necessary by the writer, the heel-
wall and toe-wall could, and would, have been made even deeper, thus
further reducing the unit soil pressures caused by such horizontal
thrust as is transferred by the anchoring walls into the foundation
soil. The writer desires, further, to point out that it would be im-
possible to have uplift pressures from both clay and water acting
on the same areas of the base of the dam. Inasmuch, therefore, as
certain allowances were made for the uplift pressure of water (Table
3), it is evident that considerable provision has already been made
for flotation effect, if any, on part of the foundation soil.

Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 91 of 167)