American Society of Civil Engineers.

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

. (page 102 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 102 of 167)
Font size
QR-code for this ebook

panel is approximately - W l, and not - — W I.

b 16

The matter may be clarified still further, per-
haps, by considering an entire row of panels across
a building. Suppose, for example, that the floor
is four panels wide. In Fig. 17 is shown a row of
four such panels, AF and EG being the exposed
edges of the floor; AE and FG are sections through the floor. The
columns are 2it A, B, C, D, etc. There are no bending moments on the
edges, AF and EG. The total load is 4 W, and the sum of the shears at
Columns A, B, C, D, and ^ is 2 W. The shears at the several columns
are no longer exactly equal, on account of the discontinuity at A
and E, but the sum is the only quantity needed. Now, take a section
through JK, and consider moments on the left. The external moment

is 2 TF X I ^^ - W I. This is equal to the sum of the internal

' No Moment

Fig. 17.


Mr. moments on the lines, JK and AE, acting perpendicularly to JK.
rneaure. '=■ tr r ^

The average moment per panel will be one-fourth of this, or ~ W I,

as previously found. This result seems very clear, as, in this case,
there are no internal moments along AF and EO to confuse the issue.

This analysis leads to the conclusion that the sum of the moments
on JK and AE is precisely the same (2iyZ), no matter what the
number and spacing of the columns along AE and FG may be, as
long as the arrangement is the same on each side, so that there is
no shear on the line, JK. If the supports consist of continuous
walls along the lines, AE and FG, the same result will be reached.

This leads to the general conclusion that, with symmetrical con-
ditions with respect to the axis, JK, the sum of the internal moments
along AE and JK is precisely the same whether the supports be walls,
or columns spaced in any manner, a result which is not incompatible
with the fact that the structure is statically indeterminate. Professor
Eddy seems to agree that the bending moment for wall supports in

the case considered would be — W I, but maintains, in effect, that if
we take away most of the wall, leaving only small columns, the

moment will be immediately reduced to — W I. As a result of this,


one-half of the reinforcing rods may be taken out and placed in the
other direction and the slab be made considerably thinner. This is
an astonishing result, and obviously incorrect. The interior moments
in the slab acting parallel to the axis, JK, cannot help out the moments
transverse thereto. They have their own duty to perform, namely,
to carry the load from column to column in the longitudinal direction,
a result brought about by changing walls into columns.
Mr. A. N. Talbot,* M. Am. See. C. E. (by letter).— Professor Eddy's

Talbot, ^igc^ggion contains so many fallacies and so many violations of
accepted principles of mechanics that it is hardly believable that it
is presented seriously by one who is familiar with the analysis of the
action of structural parts. Since reasoning presented in mathematical
form and coming from one of high standing may appear plausible,
the erroneous statements and conclusions may mislead, and a more
explicit answer is needed than will seem necessary to some readers.

Professor Eddy puts forward the astonishing statement that "any
assumption of the validity and applicability of statical analysis to
continuous flat slabs is incorrect, and leads to erroneous results, just
as much as in the case of continuous beams or of any other inde-
terminate structures." That statical principles are valid for all struc-
tures is generally accepted, and it is difficult to believe that Professor

» Urbana, IH.


Eddy intends to deny this generally accepted truth. Static analysis Mr.
does not reveal all the conditions of indeterminate structures. For
example, it cannot give the distribution of the intensity of the
resisting moment along sections of the flat slab. Nevertheless, the
indeterminate structure must be amenable to the statical laws of
forces and moments, even though Professor Eddy seeks to exempt
the flat slab from such laws.

On page 1183 he gives a mathematical derivation of what he
terms the total applied bending moment. In this derivation he uses
the accepted relation between vertical shear and rate of change of
bending moment, dM = S^dx. In determining /S^., he adds the
total vertical shear at a section, ^

AB (Fig. 18), and the total ver-
tical shear at a section, CD, and


Section of zero shear

Fig. 18.

equates this sum to ^ the

latter being the load between one
of these sections and a middle sec-
tion of the panel. It is evident

that -^j— is the value of the total

vertical shear on either one of the
two sections, and not that of the
sum of the total vertical shears
on the two sections. He does not
explain why he unites the shear
of these two ''free bodies" in the expression and yet considers
the external load on only one. By this violent assumption, at one
stroke the bending moment found in his analysis is cut in two. If
the analysis was in any way defensible, the resulting moment co-
efiicients would produce a wonderful saving in the dimensions and
costs of structures. As the analysis involves a gross violation of
the principles of mechanics, one must brand it as fallacious. That
the sum of the numerical value of the positive moment at a mid-
section and the numerical value of the negative moment at an

edge-section is — Wl, as a limiting value for very small capitals, and


not — Wl, as given on page 1184, rests on as firm a basis of logic

as do any of the accepted expressions for bending moment in beams,

girders, and other structures.

On page 1181 Professor Eddy states that the total vertical shear

in sections parallel to the sides of the panels of flat slabs is 50% less

than in a wide beam structure which rests on parallel walls, and


Mr. that the total vertical shear in a section parallel to one of the sides
Talbot. -1

of a panel is only that which arises from a uniform load of — W.

This means that the vertical shear along a line, AB (Fig. 19), close

to the supports, is only — TF, or less. It is plain that such a statement

violates the principles of equilibrium of forces acting on the portion
of the slab, ABFE, principles which are jus't as applicable with a
continuous slab having sections of zero shear along boundaries, except
at the supporting columns, as they would be for an isolated panel.
It seems too elementary to require pointing out that, for a uniform
load over the whole panel, the total vertical shear on the section, AB,
is once the load between AB and CD and not one-half of that amount.
The presentation made by Professor Eddy assumes that a series of
vertical forces along a side of a square which lies in a direction parallel
to one side of the panel (shear along one edge of a square column)
does not enter into the determination of equilibrium for sections at
right angles to this side. This method of cutting the shear in two
is not even ingenious, and it seems hardly necessary in a discussion
among engineers to add that
it has no logical basis. Of
course, the acceptance of Pro-
fessor Eddy's value of the shear
would result in dividing the
moment by two.

Attention may be called to
the different view put forth by
Professor Eddy in a paper* before the American Concrete Institute in

1916. In this paper he derives and accepts the value — WL as the

applied bending moment each way in a panel (the limiting value for

very small capitals), instead of — TFi, as given in his discussion.

Further on in this paper, starting with -:;- WL, he reduces the amount of

moment necessarily provided for by a series of steps; (a) he divides it
by two, because for two sets of reinforcing bars at right angles to each
other, embedded in a matrix of concrete, the effect is the same as if the
two sets of bars were a sheet of metal of the same weight, and hence a
set of bars will have the same resisting effect laterally as longitu-
dinally; (i) he divides it by two again, because as the bars do not
slip in the concrete, the concrete will do as much work in a given

• "A Further Discussion of the Steel Stresses in Flat Slab Floors," Proceedings,
Am. Concrete Inst., Vol. XII.

Fig. 19.


direction as the steel; and (c) he states that tests appear to show that Mr.
the effect of Poisson's ratio reduces the stresses in the reinforcing ^ ° '
bars of the slab to three-fourths of what they otherwise would be.
The combined effect is to divide the moment by 5J, and he gives

— — WL as a final value for the sum of the positive and negative re-
sisting moments, on which to base the design of the reinforcing bars,
instead of — WL before referred to. If these considerations were added

to the effect of halved shear given in his discussion already referred
to, the result (/^j of that given by static analysis) would produce
moment coefficients which would effect remarkable economies in struc-
tural design.

Professor Eddy favors bending all reinforcing bars downward at
a point fixed at the same distance from the panel edge. Although
the position of the point of dip may be expected to affect that of
the point of inflection somewhat, the differences produced for ordinary
conditions will be slight, especially as the stiffness of the concrete
in compression and tension has considerable influence. The recom-
mendation in the report provides for such changes in the position
of the point of inflection as may result from a change in the dis-
tribution of the load on the floor or from its concentration, and adds
to the diagonal tension resistance of the slab, as well as avoids a
possible plane of cleavage.

The actual distribution of the intensities of the resisting moments
along a section of the panel will be affected by variations in resisting
stiffness in different parts of the panel, but, as the concrete of the
slab in its compressive resistance, and to a certain extent in parts
of the slab in its tensile resistance, is a great factor in making up
the stiffness of the slab,' changes in the location or distribution of
the reinforcing bars over a panel-wide section will not affect the
distribution of intensities of resisting moment to any such extent
as is claimed by Professor Eddy. The report allows a considerable
leeway in the manner in which the reinforcing bars may be distributed
over a section. It may be added that experiments have not borne
out the claim for added strength in saddle-shaped portions.

Professor Eddy objects to the recommendations made for general
thickness of flat slabs on the ground that it has been indubitably
established that the ratio of steel reinforcement to concrete may be
made greater in beams of relatively large depth over that permitted
in beams of relatively small depth. The attempt, on page 1169, to
establish this statement by the relation between the sharpness of the
curvature and the horizontal shearing distortions in the concrete at
points distant from the sections where maximum moments occur, is

Mr. absurd. If true, the criticism would apply also to the principles of


design of all beams and girders. Besides, the relation of thickness
to length existing in the ordinary flat slab is that of relatively long
beams. In establishing values for the formula for slab thickness.
Professor Eddy starts with two assumed examples of thickness of
slab for definite panel lengths, considered to be kno^vn from experience,
and, later (page 1174), he naively suggests that it will be sufficient
to assume that I (given as the total panel length in the derivation of
formulas and Table 3) is the effective span (a dimension shorter than
the panel length) and may be so regarded in using Table 3. Professor
Eddy seems to have overlooked the fact, which a little study will
show, that the formulas given in the report for thickness of slabs
with and without dropped panels are devised to produce working
stresses, in both reinforcing bars and concrete, which are in close
agreement with the values of working stresses ordinarily used in
design when the moments and their distribution are taken as recom-
mended in the report.

On page 1191, Professor Eddy first uses L — 0, and then adopts

L — -7- C, as the span of the panel in determining the effect of the


size of capitals. The expression for moment given in the report
corresponds closely to the moment of the couple formed by the
resultant of the load on the half panel and the resultant of the shears
at the column supports. A more exact expression is found in the
paper by John E. Nichols, Assoc. M. Am. Soc. C. E., on "Statical
Limitations Upon the Steel Requirement in Reinforced Concrete Plat
Slab Floors."* That given in the report is a close approximation
to the more exact expression.

Any information or analysis which will furnish grounds for reduc-
ing coefficients and dimensions of engineering structures should be
welcomed. To be acceptable, such grounds must stand the tests of
the fundamental principles of mechanics and of the known action of
structures. Doubtless, the future will see many developments in
concrete and reinforced concrete. With increased knowledge and ex-
perience in construction, in the choice of materials, and in the methods
of work, and with the greater confidence which may develop with
continued use, it is not improbable that, in the future, it will be
foimd proper to increase allowable working stresses to more than
those now recommended, and even to encroach on the margin between
allowable loading and the critical load which a structure will stand —
the so-called factor of safety. Practice will change, and recommenda-
tions must be revised from time to time, but changes should not be
based on misapplication of basic principles.

* Transactions. Am. Soc. C. E., Vol. LXXVII, p. 1673.




This Society is not responsible for any statement made or opinion expressed
in its publications.

Paper No. 1399





With Discussion by Messrs. IT. S. Schick, Leavis A. Jones, and

E. S. Wise.

The American Society of Civil Engineers,
220 West 57th Street, N. Y. City.

Gentlemen: The Special Committee to Investigate the Conditions
of Employment of, and Compensation of, Civil Engineers submitted
its second report at the Annual Meeting held on January 20th, 1915.
With that report there was presented statistical information based on
4 796 replies to circulars sent out by the Committee, the information
being given in diagrammatic form and covering the following:

1. — Maximum, minimum, and average yearly compensation, together
with that of the middle man or the man occupying a position where
equal numbers of individuals receive a greater and a less annual
compensation. These figures were given for every year of experience
from 1 to 63 years, although the number of replies received from those
who have been in active practice for more than 47 years were so few that
the information with respect to them was considered of little value.

2. — The average yearly compensation according to the nature of
employment, classified in eight groups, namely, States and Counties,
National Governments, Municipalities, Technical Schools, Railroads,
Private Companies, Consulting Engineers, and Contractors.

3. — A geographical classification arranged in six groups, namely.
Southern States, Western States, New England States, Central States,
Middle Atlantic States, and Foreign Countries.

* Presented to the Annual Meeting, January 17th, ]917.


4. — A classification separating graduates of technical schools from
non-graduates, this being supplemented by a statement giving the
relative ages of graduates and non-graduates for those whose experience
covered from 2 to 15 years.

The report of the Committee was received and the Committee was
continued. The Committee has endeavored to secure information
concerning a larger number of engineers, and, with this in view, has
extended its inquiries outside the membership of the Society. While
additional requests for information were sent out and many replies
were received, the Committee was unable to present a formal report
at the Annual Meeting of 1916, and simply reported progress.

In accordance with the expressed wish of the Board of Direction,
that the work of Special Committees be completed as soon as possible,
the Committee now presents its final report.

During the summer of 1915 the Committee addressed the same
questions which were addressed to those whose replies formed the
basis of its last report to those who had become members of the
Society since the last preceding circular was sent out, the number
of inquiries so sent being 599, from which 263 replies were
received. Inquiries were also addressed to 6 393 non-members of the
Society, whose names were secured from the membership lists of
fifteen local engineering organizations in different parts of the country,
after having first eliminated from these lists all who were members
of this Society. From these non-members 1 319 usable replies were
received, which, together with the additional members of the Society
heard from, brings the total number of returns received by the Com-
mittee to 6 378.

The Committee has considered the advisability of addressing letters
of inquiry to members of the other National Engineering Societies,
but has concluded that this would be inexpedient, for the reason that
there are quite a number of members of these societies who are also
members of the American Society of Civil Engineers, and they are
probably those whose compensation or income from professional work
is above the average, so that, if those who are members of this Society
were excluded, the result would not indicate the average compensation
of members of the other societies, and might be misleading. The
Committee also made an effort to secure information from the different
engineering schools, but it was found that, with two exceptions, the
statistics collected by these institutions concerning their graduates,
with respect to the matters which have been the subject of the Com-
mittee's investigations, were so incomplete that this idea had to be
abandoned. An attempt was made to secure pertinent data from
officers of railway and other corporations, municipalities, States, and
Federal departments which employ large numbers of engineers, but

ISO 000




VOL. LXXXI, No. 1399.





Years of Experience 1
Number of Replies

10 11 12 13 14 15 16 17 18 19 20 21 23 23 24 25

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 « 45 46 47 48 49 50 51 52 53 M. 55 66 67

59 60 61 62

O 5! ^ o



they appeared to consider it impracticable to undertake the collection
of such data for the Committee, and, without their active co-operation,
it has not been possible to secure any information of value.

The Committee, therefore, is obliged to base its third report on
the information received from the 6 378 engineers who have replied;
and that these figures are worthy of serious consideration is obvious
from the fact that the 5 059 members of the Society represent a gross
annual professional income of not less than $20 952 952, while the 1 319
non-members represent a gross annual income of $4 467 709, or a
total for those whose replies form the basis of our report of $25 420 661.

The returns from the 263 additional members of the Society, the
great majority of whom were younger men, when combined with those
covered by the last report, result, as might have been expected, in
lowering the average compensation from $4 224 to $4 142, or $82.
The average annual compensation of the 1 319 non-members is $3 387,
or $755 less than for the members of the Society, including the latest
returns. The Committee realizes that in directing its inquiries to
individuals outside of the membership of the Society, but members
of local engineering organizations, it has addressed a number of men
who are not really practising engineers, as the local engineering organ-
izations almost always include in their membership men who may be
simply interested in engineering and some who have done little, if
any, professional work, and they have been frank enough to say so in
their returns, and their replies are not included in our statistics.

Twenty-two of the replies indicated that the men who made them
are employed as sales engineers or in other lines of commercial work,
although they are graduates of engineering schools. These men appear
to have turned their attention to this kind of work immediately after
graduation. The Committee has not included these returns in its com-
pilation, but they are so significant as to be deemed worthy of com-
ment. They include men whose active work covers periods of from
1 to 22 years, although in all but four instances their experience in
work of this character has been 10 years or less. A comparison of
their earning power with that of men engaged in what is generally
considered to be strictly engineering work indicates that it agrees
very closely with the averages shown by the diagrams, even though
the responsibilities which must . be assumed are less than those of
the practising engineer.

The Committee submits herewith five diagrams as follows:

1. — A diagram similar to that reproduced as Plate XL* in the last
reiM)rt, showing maximum and minimum average annual compensation
for professional work of 6 378 engineers, including 1 319 who are
not members of the Society.

» Proceedings, Am. Soc. C. E., for December, 1914.


2. — A diagram showing the average yearly compensation of 6 358
engineers, divided into members of the Society and non-members,
together with the combined average.

3. — A diagram showing the average compensation of graduates and
non-graduates, arranged according to 5-year periods.

4. — A diagram showing the average compensation of engineers
grouped according to nature of service.

5. — A diagram showing the average compensation of engineers
arranged in six geographical groups.

The first of the above diagrams includes all the returns which have
been received. The second and third include all returns covering
professional activity of 50 years or less, while the other two cover
all those which are capable of being classified with certainty and
which include 40 years or less of professional work.

Your Committee has considered the question as to whether
exceptionally large professional incomes, say those above $25 000 a
year, might not be due to some fortuitous circumstance or condition
entirely apart from professional ability, and whether they should not
be excluded from consideration in the preparation of the diagrams
by arbitrarily reducing them to $25 000. This was tried and, while
the effect was to modify very materially the curve showing maximum
professional incomes, its effect on the curve of average compensation
was very slight. The Committee believes that the data collected
include practically all the very large professional incomes, and that
an increase in the number of returns would tend to eliminate the
irregularities of the curve. The Committee's conclusion has been that
the returns should be shown as they were received.

A comparison of the diagrams with those presented with the last
report will show that the broader scope of the Committee's inquiry
has not produced any substantial changes in the results heretofore
submitted. The line indicating the average compensation of non-
graduate engineers falls below that showing the compensation of
graduates, except during the first part of the first 5-year period, where
the professional earnings of non-graduates appear to be slightly greater
than those of graduates. The Committee has communicated with
bodies which are collecting data concerning professional education,
in the hope that it might secure information relative to the com-

Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 102 of 167)