American Society of Civil Engineers.

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

. (page 90 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 90 of 167)
Font size
QR-code for this ebook

was made of co-ordinate paper with twenty divisions to the inch, and

the scales adopted were 1 000 sec. to 5 in. horizontally, and 0.10 ft. to

2.5 in. vertically.

A point, E, is then found by trial on the horizontal line, GE, that
is, for H = 0.60 ft., such that for the value of t for this point, E,

dH . ^

the straight line drawn through E and having a slope, —tt , obtained

from Equation (6) for these values of H and t, intersects the previous
tangent line, DF, in a point, F', making F'D equal to F'E. We have
thus determined very closely a new point, E, of our curve and the
tangent line, F' EJ, at this point. A similar procedure, involving the
tangent, E J, and the horizontal line for H = 0.7 ft., gives another
point on the curve and also its tangent line ; and so on.






/ F





2000 3000

Time, in Seconds

Fig. 37.

The five points thus obtained in addition to the starting point, D,
are plotted in Fig. 38 and give the curve there shown. The figures
annexed to the points give the abscissas (time), in seconds.

Investigations involving a change of origin and the plotting of
the logarithms of the new co-ordinates, brought out the fact that
the relation

II = 0.30+ ( — ^-^^ — ^ (2 000 + tf-"^ (7)

V 10 000 000 / ^ ' ^ '

is practically the equation to this curve, within the limits shown;
giving results not more than 1% in error when K is solved for. The
t of this equation is reckoned from 9 A. M., as before.



For the point, d, in the author's Fig. 12, the value of ^ is 5 400 sec. Mr.
and this value, in Equation (7), gives H = 1.04 ft. For this point ^*'"''^''-
the author has 137.0 ft. as elevation of the surface; that is, H = just
1.00 ft. ; and, under the circumstances, this may be considered a very
fair agreement of the two methods.




















2000 3000 4000

Time, in Seconds,from 9 A.M.

Fig. 38.


M. M. O'Shaughnessy,^ M. A:^r. Soc. C. E. (by letter).— The _ Mr.
author is to be commended for the thorough manner in which the nessy.
reconstruction of this dam was imdertaken. The complete way in
which the details of the construction and the reasons therefor have
been described are also worthy of commendation.

To a Western engineer, the extraordinary flood flow per square
mile, of the region which drains into the dam, appears to be alarming.
It would seem that the new spillway capacity, more than six times
the original size, should surely be ample to underwrite the structure
against flood damage in the future. It wo\ild also seem that this
hollow, reinforced concrete type of dam was well adapted for this
particular site, and that its original failure was due, not to the type
of structure, but, in a large way, to lack of adequate care with the
foundation portion of the work. Neglect of this nature is likely to
cripple any dam, and one of this type perhaps less than any other, as
a fracture of a couple of panels will release the stored waters and
safeguard the rest of the structure.

* San Prancisco, Cal.

1082 discussion: reconstruction of stony river dam

Mr. One of the cardinal principles to be observed in any dam foundation

nessy. is to remove the uplift pressure and make a perfect contact between
the water-tight skin or face of the dam and the solid reliable foundation
material. The results of tests on the friction resistance of clay soil
and the analysis of the sliding factors furnish data of original labor,
which are creditable.

The method of pressure grouting the footings is a matter of
interest, and apparently this work was very well done in the recon-

The observation that "great care is necessary even in foundation
work, where it is too often assumed that any kind of work will suffice",
should be impressed on the minds of all young engineers as the most
important of all the elements that need attention. No matter how
perfect the superstructure, if the foundation is inadequate, final failure
or consequent trouble will result. If the leakage on May 25th, and
November 1st, 1915, was actually measured, the facts would be of

The reason for the construction of a curtain-wall and roof to avert
the freezing of the top of the weep-holes and deep drains, is a matter
of much interest to engineers working in a country which is practically

The author's comments on cracks in the original structure from
lack of proper contraction joints is illuminating. The writer's expe-
rience has been that where large masses of concrete were exposed to
rapid temperature changes, there should be joints at least 40 ft.
apart, and, in ditch lining, he has used joints as close as 12 ft. to

With reference to asphalt in expansion joints, there is a tendency
of this material to flow in hot weather. Such a condition was studied
in the reconstruction of the Twin Peaks Reservoir, in 1912, and by
adding a percentage (I7i% by weight) of diatomaceous, kieselguhr,
or infusorial (different names for the same substance, which is almost
pure silica), to the asphalt, this tendency to flow was removed, and
the asphalt still retained its plastic condition.

In stating the cost per cubic yard, the details of the cement and
iron costs are missing, and would be interesting if added by the author.

The writer has seen no paper, for some time, which has been so
candid in every manner in describing all the engineering features,
so that brother engineers can profit by the experiments and experience
gained in this structure.
Mr. Joel D. Justin,* M. Am. Soc. C. E. (by letter). — This paper

teaches a much needed lesson; a lesson not so much needed by the
Engineering Profession as by the promoters and owners of projects
in connection with which the construction of dams is a feature.

* Chippewa Falls, Wis.



If an owner is considering the advisability of constructing a dam, Mr.
there is one way of no regrets for him to go about the inception of
such a project. If he is not already in contact with engineers versed
in that particular line of work, he can readily meet them through
other owners who have built dams. Having chosen his consulting
engineer, the owner should authorize him to make a complete pre-
liminary report on the project, including the economic features. The
extent of such preliminary report and investigation should be left
largely to the judgment of the engineer, just as the number of visits
that a doctor makes his patient should be left to the judgment of the
doctor, and not to that of the patient.

Of course, such an engineer should have nothing to sell except
service. If the preliminary report is favorable, this ideal owner will
say to the engineer :

"Yery well, then, we will build this structure as you recommend,
and you will consider yourself retained to design the dam and
appurtenant work, and to supervise the construction. We will expect
you to have on the ground at all times competent assistants to see
that your orders are executed. In this matter, Mr. Engineer, we put
ourselves entirely in your hands."

It is true that the best of engineers sometimes make mistakes. So
do doctors and lawyers. The difference is that the two last named
professions have an excellent opportunity of "getting away with it",
whereas the mistakes of the engineer stare the whole world in the
face, for they can be seen and photographed. It is probably for this
reason that the business man, who accepts as infallible the advice
of his lawyer or doctor, questions and often ignores the advice of
his engineer. It is certain, however, that if a larger proportion of
owners were to follow the course of action just outlined, there would
be fewer failures charged to the Engineering Profession.

In the case of the Stony River Dam, a competent consulting
engineer was retained, apparently merely to investigate the feasibility
of the site. His connection with the project ceased when the contract
for the construction was let, and he apparently had nothing to do
with the design. Had he also been retained to design and supervise
the construction of the dam, it is improbable that Mr. Scheidenhelm
would now be writing about its reconstruction.

Hollow dams of the type used at Stony River have manifest advan-
tages over other types for the conditions existing at the chosen site.
In many such dams insufficient attention has been given to foundation
problems. It woidd almost appear that it has been quite customary
to use what may be designated as stock designs, without much regard
for local conditions, and without very thorough sub-surface inves-

1084 DISCUSSION : reconstruction of stony river dam

Mr. The paper proves that the author went into this matter in a most

' thorough and complete manner. He realized that a reconstruction
of this nature required as much if not more engineering investigation
than an entirely new project. In spite of his convincing hydrological
study, Mr. Scheidenhelm's provision for a spillway capacity of 1 840
sec-ft. per sq. mile seems rather high. Much of this, it is true, is
obtained by merely allowing spilling over the bulkhead section or
"Intermediate Spillway." It appears probable, from the author's study,
that the duration of a flood when the capacity of the original spillway
would be exceeded would be brief. Consequently, it would seem
that the author might have saved the cost of the down-stream apron
on the new spillway. The abutments at the east and west ends of
the masonry structure might have been raised several feet to prevent
water flowing over the earth, and then, when the capacity of the
old spillway section was exceeded, discharge would take place over
practically the entire vndth of the dam. It would have been necessary,
of course, to extend the tumbling hearth for the entire distance. It
would probably also have been desirable to provide an additional
apron to protect the soil just down stream from the dam.

The author shows most conclusively that the resistance to sliding
of the structure as originally built was insufficient. This condition
is no doubt quite often the case with existing hollow reinforced
concrete dams built on soil foundations. In the design of such
structures there has often been far too little attention given to making
adequate provision against sliding. The methods adopted by the
author for increasing the resistance to sliding are most ingenious
and effective. At the same time their execution must have been

After investigation the author decided that the safe soil pressure
had already been reached, and therefore he did not consider it safe
to obtain the desired additional resistance to sliding by filling in with
earth on the floor of the dam. This objection might have been
overcome by extending the reinforced concrete floor or footings down
stream from the buttresses until the desired maximum soil pressure
was attained. To take the reaction from the soil pressure it would
have been necessary to cantilever the buttresses out in a down-stream
direction over the additional floor or apron. This cantilever would
probably take the form of a triangular addition to the buttresses at
their down-stream ends. The earth filling, to produce the necessary
weight, would be placed inside the dam, and the bulk of it as far
up stream as possible, thus obtaining a lower maximum soil pressure
per unit of area at the toe than if the fill were evenly distributed.

The two problems of securing proper resistance to sliding and
adequate cut-off should be handled as one, and this the author has



done. The construction of the cantilever heel, the underpinning, and Mr.
the extension of the original cut-off were clearly very expensive. This "^*'°
work, aside from the mere grouting of leaky joints and of voids along
the cut-off, would be omitted in the method the writer has in mind.
In the case of a dam on a soil foundation, the construction of an
impervious cut-off to rock is frequently out of the question. The
problem is more generally one of introducing sufficient frictional
resistance to the flow of water under the dam, so that the velocity
of the water issuing from the down-stream toe is insufficient to move
the foundation soil. In fact, the desirability, in the case of a yielding
soil, of a deep cut-off at the heel to rock is open to question, as it
in-oduces a tendency to increase the maximum soil pressure at the toe.
The necessary frictional resistance to the flow of water under the

Xew concrete to be exceptionally
/well bouJed to old buttress.

Extension to old buttresses.

|l^3 Wrought-iVon pipe, perforated
'] with four ^i holes per lin. ft. one
1 1 on e.-ich quadrant.

Fig. 39.

dam is obtained by increasing the distance through which the water
must flow before it can escape at the do\vn-stream toe. This may
be obtained by (1) increasing the depth of the cut-off, (2) by increasing
the width of the base, (3) by a combination of these two methods.

The additional down-stream floor or apron, suggested by the writer
as a method of obtaining decreased soil pressure, would, by increasing
the frictional resistance, decrease the velocity of the water flowing
through the soil. This, however, would probably not be sufficient.
Therefore, the writer would use an up-stream apron. The thickness
of this layer need be only 6 in. or indeed it might consist of a few
layers of "Gunite" on a light-mesh reinforcement, there being slight
stresses to be resisted. For the t.vpical section, the width of this
apron should probably be about 100 ft., as it would not be wise to


1086 DISCUSSION : eeconsteuction" of stony river dam

Mr. place great dependence on the original cut-off. At the junction of
' the apron and the heel there should be a special flexible joint which
would remain tight even if there should be a slight movement in
the dam. A small fill of selected clay should also be placed at this
point. The alternative for providing against seepage and sliding
here suggested is shown in Fig. 39, the dimensions being merely

In outlining this method of increasing the resistance to sliding
and to seepage, the writer realizes that it was doubtless considered
by Mr. Scheidenhelm. The only reason for suggesting it is, that,
from the author's description, it would appear that it would have
been cheaper than the equally efficacious one adopted.
Mr. Eoss M. EiEGEL,* Assoc. M. Am. Soo. C. E. (by letter). — The
writer desires to commend the practice of using a hypothetical storm
with a graph of the run-off, in determining spillway characteristics.
His observation has been that, in the design of most dams, this deter-
mination has been largely a matter of conjecture, as is attested by
numerous failures. The author's practice is similar to that of the
Miami Conservancy District in the design of flood-control dams for
the Miami Valley in Ohio. In the latter case, recorded gauge heights
enabled a graph of the run-off to be constructed for the great flood of
March, 1913. From this, by the addition of a margin of safety of
about 40%, another graph was developed for the determination of the
spillway and outlet capacities for the dams.

Although the spillway capacity of the Stony Eiver Dam is high, as
compared with the usual practice, the circumstances of high altitude,
heavy rainfall, and small drainage area, and the fact that the dam as
reconstructed must not fail, combine to justify an attitude of great

The essentials of dam construction, aside from spillway capacity,
are a tight face to retain the water, a stable structure to maintain the
integrity of the face, and satisfactory drainage facilities to carry away
any water which may pass the face or the foundation. In all these essen-
tials the original dam appears to have been deficient. The tight face
was wanting by reason of an insvifficient and faulty cut-off; the drain-
age provision was such that it was subject to interference by freezing
during a great part of the year ; and the resistance to sliding seems to
have been markedly deficient.

In the reconstruction, these faults have been remedied with a
degree of thoroughness not often seen. The improvements at the
heel, as well as the various extensions and underpinning of the cut-off,
have corrected the faults in the original face. In his discussion, Mr.
Eutenberg seems to have overlooked this point entirely. Had nothing
been done to improve the cut-off, the drainage provisions of the recon-

* Dayton, Ohio.


structed dam might indeed have been insufficient, and Mr. Rutenberg's Mr.
points would have been well taken. The fact that such leakage as did '^^^ '
occur after the dam was refilled was readily checked, or checked itself
after a short time, is ample testimony to the effectiveness of the
measures taken.

The provision made to protect the drainage outlets from freezing,
namely, housing in the base of the dam, savors of the heroic, but seems
to be justifiable. The author was repairing an existing structure, and
to have developed an ideal system of sub-drainage with protected outlets
below the dam was evidently prohibitive as, to cost. Under the system
adopted, any seepage water will find its readiest avenue of escape
through the weep-holes or pipes. Its escape immediately under the foot-
ing will be practically prevented by the new "toe" wall. Moreover, after
passing the improved cut-off, the motion of seepage water through any
strata below this wall will be so slow that no erosion need be feared.

The provision of increased resistance against sliding seems to have
been necessary and to have been well worked out. The writer fails to
see any better method of remedying this defect, and feels that the
device is worthy of imitation. Although the criticism has been made
that the heel- and toe-walls are too light to carry the entire horizontal
thrust, it seems to the writer that such a requirement would be exces-
sive. Their function is to add to the resistance normally offered along
the top surface of the foundation itself, and, therefore, their strength
is doubtless sufficient.

On the whole, the writer has been impressed by the care and thor-
oughness of the author's methods. He wholly fails to comprehend Mr.
Rutenberg's criticism, which seems to have been based on rather hasty
consideration. For instance, his remark that the dam is less safe now
than before reconstruction seems particularly unsoimd, inasmuch as the
original dam failed after being in ordinary service for 65 days, and the
reconstructed one has now been in service for 2 years.

F. W. ScHEiDENHELM,* M. Am. Soc. C. E. (by letter).— The dis- g^j^^.M^Jij^^,^
cussion of the paper has been very interesting to the writer, as
presumably it has also been to other engineers concerned with the
problems treated. Among other things, it serves to emphasize the
fact that there is rarely agreement among engineers as to the exact
measure, or margin, of safety which should be given to, or is possessed
by, a structure. On the whole, the writer finds much cause for
satisfaction in the comments contained in the discussions.

Frequently, papers presented before the Society and the discussions
thereon result in a symposium of data and opinion on certain subjects.
Such is the case with regard, in particular, to two problems which
were involved in the reconstruction of the Stony River Dam, and
are treated in the paper.

* New York City.


1088 discussion: keconsteuction of stony kivp:r dam

Mr. The first of tliese problems concerns the determination of the

maximum flood for which spillway capacity is to be provided, and
the consequent determination and design of the spillway provision
itself. Of the ten members who have discussed this matter, only
one, Mr. Gregory, feels that the writer has "surely gone beyond the
necessary limit." The great majority, however, apparently agree on
the proposition that a drainage area as small as that of Stony River
at the dam site requires entirely different treatment than do larger
areas, and that the exceptionally large maximum run-off adopted,
although surprising, is warranted.

Referring to the comment of Mr. Gregory (page 1060) the wi'iter
feels that it is certainly necessary to take into account, not merely
rainfalls of 2 or 3 hours' duration, but those of considerably longer
periods, and more especially their consequent floods. This is for the
reason that from such rainfalls the run-oflF (into the reservoir) which
immediately precedes the maximum or critical periods of, say, 2 or
3 hours, necessarily affects the absorption or equalization capacity
of the reservoir during the vital period.

It is true, as pointed out in the paper, that a record or graph
of run-off never truly represents the corresponding rainfall; that is,
the maximum rate of run-off is always less than the maximum inten-
sity of rainfall. Yet, even making full allowance for this fact, as
applied to actual Stony River conditions, it is reasonably evident,
from the studies depicted in Figs. 12 and 13, that the length of the
period of rainfall which should be considered is at least 6 hours; and
it must be remembered that rainfall does not persist at a uniform
intensity through any 24-hour period. Thus, were there an automatic
gauge record available for the 22-in. rainfall during 24 hours at
Altapass, N. C., on June 15th-16th, 1916 (page 930), the record for
the maximum 6 hours would probably be astounding.

Again, looking at the matter from Mr. Gregory's point of view,
and making the reasonable assumption, size and topography of drain-
age area considered, that "the peak of the run-off at the dam is equal
to the greatest average rate of rainfall during any 45-min. period"
(page 940), it is to be noted from Fig. 15 that "the most severe
run-off conditions reasonably conceivable", there shown, indicate a
rainfall with an intensity averaging only about 4 in. per hour for
the maximum 45 niin. This intensity is somewhat less than that given
for the same period by the Talbot formula, quoted by Mr. Gregory,
and, of course, is less, by an even greater amount, than the cor-
responding "Maximum for Greater New York" (Fig. 32). Incidentally,
the Talbot and the Greater New York curves of Fig. 32, when plotted
on Fig. 9, fall outside of the group of curves shown thereon; in other
words, the curves of Fig. 32 show the greater intensities of rainfall.

discussion: reconstruction' of stony river dam 1089

It is interesting to note that apparently the Cane Creek and , .Mr.
Elkhorn Creek floods, which formed the basis for the studies pre-
sented in the paper (page 926) were of such, intensity as not to be
included within the formula proposed by Weston E. Fuller, M. Am.
Soc. C. E.* The same is true also of a run-ofp of 1 302 cu. ft. per
sq. mile, reportedf for the drainage area of 143 sq. miles of Devil's
Creek, near Yiele, Iowa.

The method of determining or checking spillway capacity by means
of graphs of hypothetical maximum floods to be provided against,
seems to meet with approval. Similar methods are shown to have
been used in the design work of the Miami Conservancy District at
Dayton, Ohio, as referred to by Mr. Riegel (page 1086), and dis-
cussed in considerable detail by Mr. Grant (page 1067). A remark-
ably close agreement of results is evident from comparison of Eigs.
12 and 36, the former being based on the method used by the writer,
whereas the latter is based on the flood routing method applied on
the work of the Miami Conservancy District.

The second feature provoking considerable discussion is the utiliza-
tion of the underlying foundation soil in setting up resistance to
sliding. On this subject the discussion of Professor Cain is a
contribution of exceptional value.

One may fairly question whether present experimental information
regarding the behavior of soils is sufficient to warrant the acceptance
of Professor Cain's application of Coulomb's laws to cases, such as
that of the Stony River Dam, where there is involved the question of
sliding, and more particularly of impending sliding. In fact, the
writer still feels constrained to distinguish between the conditions

Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 90 of 167)