1 AllOrl IIIfl1/8 G" SlcI I

in lI'ashillllto/! .••. . . . • • 6.7

2 Questions atu! ...h1311'erS
11/, the AISC Spec
and Codt ..•........•.. 14·1:,
8 Steel Geodesir [)",ne
('uts Cost and
COIlSirllrti(JU Tinlf' ..... 8·[1
 •
Published by
CONTENTS
American Institute How Quiet Are Steel-Framed Floors 3-5
of Steel Construction Apartments Go Steel in Washington 6-7
101 Park Avenue, New York 17. N. Y. Steel Geodesic Dome Cuts Costs and
Construction Time 8-9
O"ICE .. a: Experimental Design for Short-Span Bridges 10-12
Harold G. Lewis, President Low-Cost School 13
R. C. Palmer, First Vice President Questions and Answers on the AISC
J. Philip Murphy, Second Vice President Spec and Code 14-15
W. R. Jackson, Treasurer Research Cuts Stiffeners on Thin-Web
John K. Edmonds,
Plate Girders 16-18
Executive Vice President Research Progress in Bridge Diaphragms 18
M. Harvey Smedley,
Counsel and Secretary Coach House Motor Inn 19
Stacking the Deck 20

EDITORIAL aTAPp
William C. Brooks, Editor
Olindo Grossi, FAIl\, Architectural
Editor
E. E, Hanks, Technical Editor
AN INVITATION

Economy and vel'satility combined with safety and

strength are the major advantage8 of strllcturalsteel, and
the quickening pace and results of research are producing
•
better ways to design in steel, Whether the results come
from the laboratory, the experimental and probing mind
of a designer, or the actual construction of a bridge or
building, the American Institute of Steel Construction is
RE.IONAL OP:PIC ••
attempting to dessiminate useful design material as
Atlanta, Georgia quickly as possible,
Birmingham, Alabama For future issues of MODERN STEEL CONSTRUCTION,
Boston, Massachusetts we would like to invite a,'chitects and engineers to send
Chicago, Illinois in manuscripts, photographs and .~ketches describing their
Columbus, Ohio structures, methods or just ideas that they feel would be
Dallas, Texas of interest to ow' 120,000 ,'eadel's, It is 0111' hope that we
Detroit, Michigan can provide a forum for new thinking and useful inforrlla-
Greensboro, North Carolina tion, The design professions, clients and the public u:ill
Houston, Texas benefit through more economical and better structures,
los Angeles, California
Minneapolis, Minnesota A REMINDER
New York, New York
Omaha, Nebraska
Architectuml A wal'ds of E:rcellence Entries for build-
Philadelphia, Pennsylvania
ings completed 01' occupied during 1962 are due in the New
York office of the AISC by May 1,5, 1969, Announcement

•
Pittsburgh, Pennsylvania
St. louis, Missouri
of the winning architects will be made about July I, A
San FrancIsco, California
brochure containing entry rules was mailed the end of
Seattle, Washington
March, and additional copies are available through Ollr re-
gional offices 0" fl'om New York,
Washington, Oistrict of Columbia
• HOW QUIET ARE

STEEL-FRAMED FLOORS
John G. Hotchkiss,
Senior Regional Engineer

Impact noise-control is a major prob- strike the floor, it vibrates and in turn help control the transmission of air-
lem in a multi-family dwellings due to radiates sound into the air in the apart· borne noise. The converse case is very
the increased noisiness of our present- ment below. unlikely to occur.

•
day environment. Impact noises are the The problem of insulating airborne The United States is one of the few
most annoying kinds of noise. Such un- noise is minimized by using structural countries in the world which up to reo
wanted sounds as thumping footsteps, assemblies that are massive or are com- cently has had no noise·control crite·
a fallen object, a neighbor's TV set, or a posed of heterogeneous materials. rion. Recognition of such a standard
noisy dishwasher overhead can readily Changes in the floor surface, as men- has been established by England,
be transmitted by vibration through the tioned above for impact sound isolation, Sweden, Germany, Austria, Bulgaria,
floor and ceiling if adequate measures have little effect. By contrast, impact Canada, Denmark, Finland, France,
are not taken to reduce the impact sounds are generated by objects striking Netherlands, Norway, Czechoslovakia,
sound waves. or scraping across the floor. Noises such and the USSR. Due to the growing num-
Perhaps because little information as a dishwasher, an object dropped on ber of complaints of inadequate sound
about impact noise is currently avail- the floor, a bathroom shower, furniture isolation in multi-family dwellings, the
able in this country, certain misconcep- being moved, a rotating machine, or Federal Housing Adm inistration in Jan-
tions and fallacies have arisen. One of footsteps, are sounds that are passed on uary of this year prepared the first Amer-
these involves the use of steel bar joists to the floor by direct mechanical con- ican impact noise-level criterion, along
in a floor assembly. Such construction tact. Of importance is the fact that the with a compilation of noise performance
is often erroneously alleged to be noisy. sound thus radiated within the floor data. This publication, FHA NO. 750,
This can be readily disproved. Test re- continues on radiating waves from both i mport .voise COlltrol in ft1ulti./amily
sults on floor assemblies both here and the top and bottom surfaces of the Dll'el/i/l g. was the result of interviews
abroad show that it is only the surface floor. Thus certain floor assemblies de- with thousands of tenants as well as a
of the floor that is critical for noises due signed to control impact noise will also review of many foreign codes.
to impact and not the structural ele-
ments. Floor treatment is independent
of the structural assembly.
Since sounds in a room may be trans-
mitted by two different means, they
accordingly have been classified as
impact noise and airborne noise. (Fig.

•
1) The more familiar airborne noise is
produced by the TV or hi-fi set, the
human voice, or a musical instrument
radiating sound waves through the air
in the room above. When these waves

Air borne Figure 1 Impact

 Any criterion for noise-control must
acknowledge the possible existence of
where ambient noises produce a mask-
ing effect. This criterion. as suggested.
included in this Table to indicate how
transparent such floor construction is
•
ambient noises: the quiet background of takes into account an ambient noise for to the transmission of impact noises.
urban environment. the background typical apartment construction as given The one test shown for concrete joists
noise of flowing traffic in the cities or in the guide published by the American was included to show that an elaborate
simply an air-conditioning machine. So:iety for Heating. Refrigeration and construction - a floating wood raft on
These ambient noises may help mask Air Conditioning Engineers. a glass fiber blanket - only provided an
the sporadic. unwanted sounds coming The Table. "Impact Noise Control INR = + 2. Such construction would be
from a neighbor's apartment. One must Tests." shows various assemblies with ruled out as too costly. Furthermore.
also consider how much isolation the structural elements. floor finishes. this assembly does not meet the require-
can be provided without greatly increas- ceiling finishes. and in the boxes. the ments for a three-hour fire-resistance
ing the cost of construction. Since years INR rating. The testing authorities are rating.
of construction without code restric- also given. Obviously this listing is far Thus far only impact noise-control
tions for noise-control produced in- from complete. However. the few se- has been discussed. Insofar as floor
adequate isolation. new practices will lected serve to demonstrate the noise- construction is concerned, impact iso-
cause some increase in construction control attainable by careful choice of lation is of far more importance. How-
cost. Significantly. however. the amount floor finish. It should also be observed ever. there is test data currently avail-
of this increase is small if noise-control that the softer, resilient materials are able covering sound transmission of air-
is considered in the early design stage. more effective in the isolation of im- borne noise through floors. Disregarding
Addition at a later date often is costly. pact noise. Take the case of the steel any floor treatment required for impact
For multi-family dwellings the FHA bar joist floor. The highest rating (INR noise-control. the typical steel bar joist
has recommended values for the max- = + 26 tested at the Bureau of Stand- floor with a plastered ceiling has sound

•
imum impact sound-pressure level in ards). published by FHA. requires a :;B- reduction ratings in the order of 55-
floor assemblies. It is a criterion of ac- in. nylon carpet on a %-in. foam rubber 65 db.· Good American practice accord-
ceptable, but not excessive, isolation pad. The structural assembly is typical ing to Bolt & Newman (well-known
against impact noise. By comparison construction used in multi-dwellings acoustic consultants). suggests 4D-5D
with this criterion. impact performance where a fire-resistance rating of three db. A bare concrete slab has a 41-43db
characteristics have been classified by hours is required. For this fire rating. rating.
type of floor construction and the ability gypsum perlite plaster would be used in Some further comments regarding
to provide. or not provide. the recom- place of the sanded gypsum plaster noisy dwellings should be mentioned.
mended isolation. A si ngle number tested. Since the composition of the because noise-control involves many
impact noise-rating (lNR) is used. A ceiling finish has little or no affect on more problems than that of the floor
floor assembly rated INR = D or above the INR rating. such a substitution is alone. A common problem concerns the
is acceptable. The larger the INR the valid. The second highest INR rating control of transmitted sound over and
better the floor. It is sometimes advan- published for bar joists (lNR = -2) around partial-height partitions. In such
tageous to know the ranking of a floor shows Y4-in. cork tile cemented to the cases. particular attention must be di-
construction; viz .• does it barely exceed concrete with linoleum paste. A change rected to the elimination of serious air
the minimum or does it greatly exceed from INR = + 26 to INR = - 2 is very leaks. Air-borne sounds originating in
the requisite? A negative INR rating. significant. yet no change was made in one apartment may pass through a sus-
for example INR = - 4, means that an the structural assembly. pended ceiling into the adjacent room
improvement of 4db would be required Now consider the INR ratings for the where the partition extends only to the
to reach acceptability (INR = D). A rat- flat concrete slab construction. which suspended ceiling. Other typical leaks
ing of + 26 - the bar joist assembly is a typical framing system used for between apartments very often come
shown in the Table - means that the multi-family dwellings. The highest rat- from plumbing fixtures. radiators. light-
impact isolation rating has so much ing (lNR = + 12) published shows a ing conduits and back-to-back outlets.
tolerance that a reduction of 26db cocomat floor cover or three-eIghth-inch In such a short discussion it is im-
would be required to bring it back to carpet laid over a three-quarter-inch possible to cover the complex problems
zero. Such a design would be far more plaster screed. If one-eighth-inch lino- confronting the architect or acoustical
than required for even a luxury apart- leum is cemented to a two-inch con- engineer who attempts to control noise
ment. crete screed which in turn rests on a within cost limitations. Nevertheless. it
According to FHA. an apartment in
a quiet area should have a floor con-
struction better than normal. INR =
=
+ 5. A poorer construction (INR - 5)
may be tolerable in noisy urban districts
one-inch glass fiber blanket atop a five-
inch concrete slab. the INR = + 1.
Again this is a significant change
amounting to a reduction of lldb. The
bare concrete slab (lNR = -17) was
is recommended that such allegations
as "noisy" bar-joist floors should be re-
evaluated in light of the new test results.
'Such
Bureau
t~sU
0/
hUlr bun Nmduclrd by the Natio nal
SlandlJfds Qlld the Electrical Research
Products . 'nc. ( Wr.rlt',n Electric Company ).
•
,.
':.>
,.

• IMPACT NOISE CONTROL TESTS

Source: Impact Noise warm/ FilA ~ 75tJ
STEEL BAR JOISTS FLA T CONCRETE SLAB
r 2"concrete slafJ 017 %" 10M ..-Floor fil7/sh
.f - A .'._ .. D ., ..' .r _.(' '. -:.~ 1A · · •f & - ,,' . • ', \
.~ -' '''''' ' :A,. ' :.' .. ',' .• : • ../ . ...
~4 .. • ~ • ., .: .. .. ~ . ,~:.. I

.~ " .... - ' .'. '4,' .:. . . , .~ :.-

.. I' . ' ;' ' . ~ ',' 1";-
I- 7"bar jai.5/s . . . "" .. . - .
.... . .... : ..
. • • • ..
. . ..... c ...r~
A
!',
2 rae.
; .' - . ...... ......
~ ~ ," ':'
L Ceiling fin/s/1
'-~" 9YjJSllm /Joan:!
'- ?1~" sdntiedp/dsr~r Cocomatfloorcoveroro/a" +12
l'l(/' lime putt, finish carpet on 3M "plaster scrt!'ed
017 4Y2"concre/"eslafJ. 1-_-1
3/a"plaster cel/loq.
318"Nylon carpet ( -!i/JI!e)on +'))/'" Authorify.' Reseorch Inst for
14" rubber foam pad L 0 Publk Health Enqrq IYet/7er/cmds
Authority.' Nat-1. !3ureau 195tJ.
of StC7/ldards.

• Remarks: WIf/7)//' qyp5um perlite

pIaSTer ond I"POqa. mesh tIlis
assembly hC7s <7 3

14"Cork tile cementC'dfo

fire rotloq. or
-2
lJa"U17oleum cemented fo 2"
c017crt?tescreedon I"glass
/
fiber blankel"o17 5 "C017crete L - - - l
slab. 12" plaster cel/inq.
Autnoril"y.- Parkin,Purkis$ Schol!:s.
+

concrete wit"h linoleum L017do17, /960

pasl"e.
A utIJorify: Natl 8ureau or lJare coaCl'Pfes/obor s~" -17
Src:mdords. pitch mastic or -%''' compo-
sif/onon GYz"-9%" conc;p"tJ.---I
fa" Vinyl asbestos tIle I'() slob. Yz"plaster ceilinq or bare.
cemenred toco/Jcrelc!w/tlJ -(4 A uthor/t)'.' pprK/~ PurKist/5cIJoles,
hfiokumpas~. , ~-~I London, 19GO
AuthoriTY: Not! 8ureauof'
Standards.

CONCRETE JOISTS
rFloatinqra/t"or /3"Tt!6rlr.bot7rds17aikdlobolhls +Z
-S" ~:T.::".:,.', ..~:·~:.,:·::~,~~WS;::;e:e:::::;:r~
fi't·(/ ~;:r,'> Precast concrete cM/)mls, Z"s/ofJ, ribs 15a.c.3,{d~'p

L..JAI"plo.sk>rDoardwil'lJplostersK!i17CcKlt: AutlJorify:Pa/"Jl'ln,PtlrhsISckl:-s

5
APARTMENTS GO STEEL IN WASHINGTON . I

One of the most significant trends in

the construction industry is taking place
in the nation's capital. Architects and
engineers in the area are providing their
clients with probably the most econom-
ical yet handsome apartment buildings
anywhere in the country through welded,
continuously-designed steel frames.
The chart on the opposite page tabu-
lates some statistics for nine projects
which are typical of the more than two
dozen, seven to twelve-story apartment
houses which have been built in and
around the District of Columbia in the
past several years. As a result of the
economy inherent in welded continuity,
FigltT6 !
other framing schemes, widely used for
many years, are becoming obsolete for
multiple dwellings in the area. Con- Whitthall Apartment.,

•
struction is less expensive, and the steel Bathe.da, Md., hat'€'
frames are being erected as fast as .pan, 0/ ~8 to 30 ft.
other types of framing. Balconie. are framed
By designing steel as a continuous u'ith ('xpo,rcl
frame (as most engineers design con- H -,cetion •.
crete), savings up to 20 per cent in
weight are possible. In addition, the
detail material is reduced to about one
per cent. The combined weight of col-
umns and beams averages about 5.5 Figure 3 ...
pounds per square foot with an average
cost of 15¢ per pound. The actual unit- column design. K-trusses were used as rise apartment building, constructed of
weight depends on the particular span the most economical method of carrying the finest materials and of a design
and bay area. The absolute minimum in wind loads to the foundations. forward-looking in its approach yet con-
weight and floor depth can be achieved Most of these welded structures have temporary in its appeaL"
by using cover plates in the negative been erected with the aid of Saxe clips. The higher unit weight was due to
moment areas and designing the beam These specially forged connection units longer spans (28 to 30 ft.) and a con-
for the usually smaller positive moment. are shop-welded to the beams and col- siderable number of balconies.
This approach was used by the struc- umns and hold the connections rigidly The Horizon Apartments (Figure 5)
tural engineer for the Congressional in place while the field-weld is made. is another handsome but very econom-
Apartments (Project 9). Temporary seat angles and erection i:al structure. Floor beams are ten-inch
Washington designers are also taking bolts are therefore eliminated. Welding wide flange sections fully welded for
full advantage of the new AISC Specifi- is done by certified welders and is in- continuity which serve as wind beams
cation to achieve light steel weight. A36 spected by independent inspection and main-load carrying members.
steel is used for basic steel framing, but service agencies. Cost for this welding The Park Maple Apartment House
A441 and other high-strength steels are inspection service runs about two to (Figure 6) was designed with an unusual
also specified for lower columns. A six- three dollars per ton. balcony structure. Rectangular steel

•
story building is presently being de- Architects in the capital area are not tubes run the full height of the building
signed entirely with 50 ksi-yield steel. sacrificing beauty for economy. The and support the balconies at each floor .
For Crystal House, Arlington, Va. (Proj- Whitehall Apartments in Bethesda This luxury building, equipped with air-
ect 8), the engineer designed with 50,- (Figure 2) recently won for its builders conditioning, swimming pool and park-
000 psi steel in the lower columns to the Oliver Owen Kuhn Cup for "an out- ing facilities, was occupied nine months
save weight and maintain minimum standing example of luxury-type, high- after ground was broken.

6
I •

...
• Project
l. Park Maple Apts. To.om. Por'. Md.
No. Stories
12
FIGURE 1

No. of
Units
135
Floor Area
Square feet
157,000
Steel
Wl psf
6.55
Joist
Wt. psf
1.91
2. Park East Apts. W"hlnaton, O. C. 8 88 60,000 5,75 2.67
3. Whitehall Apts, Bethesda , Md. 11 209 265,100 7,32 3.08
4. Horizon Apts. Arlln"on, v•. 9 77 105,400 5.2 2,5
5. Abingdon House W"hlnaton, O. C. 7 82 88,700 4.65 2,1
6. Suburban Tower Silver Springs, Md. 12 174 202,800 6,3 2,8
7. Summit Hill Apts. SIlYer Sprlnas, Md. 11 123 189,000 6.4 2,85
8. Crysta I House Arlin~on, Va. 12 426 435,000 Total 9.8
9. Congressiona I Apts. Rockville, Md. 8 300 358,000 5.56 3,60

Archilect, Structural En,in •• rs

aerla Abel, Washington, D. C.: Project 1 Michael K. Johns, washington , 0 C.: Projects 1. 2
Werner-Dyer Associates, Washington, O. C.: Project 2 H. l. Keller & Associates. Silver Springs, Md .: Projects 3,6,7,9
Cohen, Haft AlA & Associates , Silver Springs. Md.: Projects 3, 6, 9 HoratiO Allison. Rockville, Md.: Projects 4,5
Sheridan Behm, AlA, Arlington, Va .: PrOject 4, 5 Heinzman & Clifton, 'Nashlnston, D. C.: Project 8
Edmund W. Dreyfuss & Associates. Washington, D. C.: Project 7
Edwin Weihe, Washington, O. C.: Project 8

• ,....,"'-{1Vp.

~£
... " TY?
CL,ps.
IOW=ZI
I" (1'1p.

Figure .4 Figure 5
Builders of apartment houses (both Horizon Aparlmr1l ts, Arlingtou, Va., allra hOt! r(wtiift'rrrd ball'lmir•. Figurr .4
public and speculative) have been ex· .hows the simplc, incrpcusit·c dctaiiuscci by the r"uiuar fflr tllr balcony framing,
perimenting for years to achieve low
cost yet attractive and serviceable de·
signs. The significance of the experi·
ence in Washington is that examination
by engineers of continuity design,
utilization of economies outlined in the
new AISC Specification, and framing
where applicable with high strength

• steels meet these requirements .

Steelwork for the Park Mapl.

Apartments, Tokoma Park, Mel., was
erected in 7lh week8. Building was
occupied nine month. after
Fillure 6

ground was broken.

 STEEL GEODESIC DOME
•

•
by E. E. Hanks
Technical Editor

The John T. Williams High School The frame is composed of 378 trihe-
Gymnasium, recently completed in Char· drons made of 8 B 10 steel sections with
lotte, N. C., is a geodesic dome covering arched fascia members of the overhang
13,000 sq ft and costing $154,000, or of 15 [33.9. The tension ring is made of
about $12 per sq ft. two 8 [18.75, first welded together
The steel frame projects over a brick back-to·back and then butt-welded
wall which encloses the gymnasium. In· around the circumference of the closure
side diameter of the closure wall is wall. Members of the trihedrons were
128 ft, with an area of 13,000 sq ft. Out· sl ightly curved to conform to the curva-
side diameter of the frame at minimum ture of the dome. Control radius of the
overhang is 132 ft, and 154 ft at maxi· sphere is 105 ft. Vertical component of
mum overhang at the six supporting reaction is carried by the bearing (clo-
concrete piers. Height above tension sure) wall.
ring is 22 ft·2 in., and 32 ft·2 in. above Weight of the structural steel (grade
the finished floor. A-7) was about 57 tons or eight pounds
The design of the dome combines per square foot including overhang area.
structural steel framing with poured This was the first geodesic dome for the
gypsum deck on bulb tees, with the steel fabricators, Southern Engineering Com·
exposed on the underside for a pleasing pany, Charlotte, N. C., and personnel had
architectural effect and to provide noise· some misgivings and apprehension
damping coffers. about ta ckling this job. However, they
The designer estimates that the report that no appreciable difficulties
dome design and construction methods were encountered in the preparation of
saved about 17 per cent over the can· the shop drawings or with the fabrica-

•
ventional rectangular building totaling tion and erection of the steel. Field con-
several thousand dollars in construction nections were made with high strength
costs. Cost of erection was low, as a bolts. No falsework was needed.
steel geodesic dome-frame supports it- The geodesic dome is an excellent
self as it is being erected, so no false- example of how structural steel, due to
work is needed. its ductility and pliancy, can be shaped

8
I~

Skctt"h sholn a ttution at su pport

of t {, ~f.ion r in g. S ix stn tctural

• z-
t{'(' 8Il PPo." -aBBe mb lic8 u:cr c
/abric at rd (1 1/(1 bolted to the
f ounda tion 11.;a llB. Tension ring t('a 8

h 4·' TREATED WITH

in tU rll bolted to the assemb lies

PENTACHLOROPHENOL
6' x 2 ' CARRIAGe.
~OLT5 @ 2'0' o.C.

METAL FLA5!1ING

I (

r CANT (POUReD
GYP.5lIM

~:J::3§~::::3=t'LU~RICATt.
: ~LIP
I J OINT bETWEEN TE.E.
FLA NGE e t>45E It

• eS'-1/ ' TO CENTER OF t:>lDG.

to fit any form or shape. Designers can

take advantage of steel's inherent
strength , economy and compactness re-
(MINIMIIM OVeRHANG

gardless of the contour of the structure.

Arch itects were Biberstein, Bowles,
Meacham & Reed, Charlotte, N. C. Robert
H. Pinn ix, Inc., Gastonia, was the general Skrtch at Irlt ,hOII ', typiC'ol
contra ctor. 01 dIU/If frame
(,()IHlt'Cti()Jt
nlrmbas. Photo bduw
The design was by Synergetics, Inc.,
sholf's the C'ompfrt('(l geodnic
consulting engineers, Dr. M. E. Uyanik,
dOlnf' be/ore 'roofin(J wa.
structural consultant, of Raleigh, North 1)I.ialled .
Carolina. The firm specializes in the de-
sign of structural steel geodesic domes.
The geodesic dome has all its members
as arcs of greater circles. The center of
all the greater circles is the center of
th e sphere. It is a cQmparatively new
concept in steel design, and the firm
has designed several structures.
In addition to the Charlotte dome,
they include the Union Tank Car Com-
pany Plant (384-ft diameter), Baton
Rouge, La. - the first one bu ilt from

•
plans developed by R. Buckminster
Fuller: the Wa lt Whitman High School
Fie ld House (1S7-ft clear span), Bethes-
da, Md .: and the American Society of
Metal Research Center dome, Cleve-
land, Ohio.
 ~'" """,0 Bum "1
6teel ~;I.
Hi" IZ·tJ.
•
..l /eICOL £P0Rtt'Y SECTION

EXPERIMENTAL DESIGN FOR

by R. S. Delameter
Consulting Engineer

•
Our office has recently developed considerably reduced weight of steel what we believe is maximum economy
a type of composite·beam design which without materially increasing the cost of and simplicity in forming. The deck can
we believe will prove to be practical as constru ction. This method of design has be formed by setting full -width, four-
well as economical for constructing sin ce been adopted as a standard by the foot plywood panels which will just fit
short-span, continuous-beam bridges. Secondary Road Department of the Kan- between the beam webs and which can
The design uses a large structural tee sas Highway Commission. be supported on transverse four-by-
with the flange serving as the lower Our objective on shorter spans was fours blocked up off the bottom flange,
flange of the composite beam, and a to design a bridge suitable for use on thus eliminating any cutting of form
concrete slab anchored to and partly local roads where traffic generally is boards or the use of any complicated
encasing the upturned edge of the tee quite light, mostly local in nature, but form -hangers. It should be possible to
web to form the entire top flange. The where occasional heavy loads may be reuse the form panels several times, and
design, we think, may be of some inter- expected and must be carried without both we and the county engineer feel
est to anyone having occasion to design damage to the bridge. It was desirable that this is about the simplest forming
short-span bridges. to keep the total depth of deck to a system that could be devised for this
Since most county highway depart- minimum and to use a continuous~ beam type of bri dge.
ments traditionally have minimum funds design, not only for economy but also The width of the roadway is really an
to maintain the kind of county road sys- to eliminate the maintenance problem outgrowth of the beam spacing. We
tems which the public has come to which usually accompanies joints in the started with a roadway width of 24 ft,
expect, maximum economy in construc- deck. We decided to try a fully com- pretty much a minimum for county roads
tion becomes almost a primary require- posite beam with the concrete slab in this state; however, using seven
ment. Adequate structural design is, of forming the top flange throughout the beams to allow for good anchorage, we
course. the first requirement of any length of the bridge. We adopted for wind up with a 26-ft slab, and, in effect,
bridge-design system, and the cost of stringers a structural tee-section with figure that we have a foot and a half of
maintenance certainly must be taken the web turned up, and the concrete extra roadway at very small cost, which
into account. To achieve maximum slab encasing the top inch of the web is more or less a bonus and certainly
economy we have tried to take advan- and anchored to it by connectors con - won 't be any disadvantage. If it were
tage of every design method which of- sisting of stud-welded rods bent up desirable to build a narrower roadway

•
fered a possibility of reducing construc- enough to be fully embedded in the one or more beams could be eliminated,
tion and maintenance costs without slab. longitudinal steel in the slab will reducing the deck width by four-foot
sacrificing the load-carrying capacity of provide tensile reinforcement required increments.
the bridge. Some years ago we started to carry the entire negative moment at For the short spans involved (30-ft
using composite design on our rolled- the piers. main spans) we decided to see if we
beam steel bridges and found that it We set the stringer spacing to give could achieve the "ultimate" in com-

10
 J!

--- -

•
-r
T
~
-~ .- - i ~-.

I
- -n
I !I

I
remp F./~_r' !

U
~
~
~pporl C.mbt:rJ4·

T
!
I
I
I
I
i
!
I
I
I
I

- I
I
I

I
- --

PA.Ii'T .s,DE VIEW

U
SHORT·SPAN BRIDGES

•
posite construction; to have no steel
compression flange whatever. We found
that this wasn't too much of a problem
in the positive moment section. A large
structural tee would carry the dead-load
moment without much trouble; the only
question was what kind of connectors
to use to develop bond between the
upper edge of the tee section and
--+-1f¥Idr"111o~1
the concrete. "'H. ....,,,,,,
We considered several alternatives ~.~" ,,'IoW) .,~ ""~_I,.-.d f~

and finally concluded that the simplest {f!f'I9'"'' ,""""" ~ be ct!J'"b~d'/

''W' . ' " or I~ cor,t,«IIN, Oc.If
would be stud·welded connectors, as we ,,.,....
~. "n.,J bt6 ""'3f .f IN powJ/ ..,

have used for several years on flanged

sections. However, the ordinary headed·
stud would barely be covered by the
slab. We decided to use a stud·welded
rod long enough so it could be bent up
and fully embedded in the slab. The
manual operation of bending up the
stud after welding doesn't seem to be a
major problem, and it does provide prob-
ably the simplest and fastest means of
putting a connector on the stringer, and
do so without excessive welding which
might distort the comparatively light
members we are using.

•
In order both to decrease costs and
to help keep the deck clean we are using "I
a deep-beam-type steel guardrail to
serve as combination curb and handrail. q~ '~.$TlJ~I'~ '7'1-
The rail is supported on posts bolted di- Top-'tbl7iim --
_S(,l.4/l INf.vl!C'\lO PLAN
rectly to the edge of the slab and to

11
 ~.
'4$' £"Id"P'"
' 4"4 Ct:nler~"

•
Abul P/~r Brtdge

L ~.5/X 4.5/X /I SPC.IJ/Z'-/I ',O' • {; SDC . ~/2" • -'SpC.@ ~ (J .5pC.<f/2" ~ '-q:...

€~",j'O @~"3'( , - (; '-0' 12 ' -fj"O'
~~ ~~

HALF PLAN OF STRINIS~R

stru ts we lded to the bottom fla nge of falsework to be placed at the center of

• <l
the outside stringer. This arrangement
is not only very economical but should
allow fast and economical replacement
in case the rail is damaged.
Design loading on the bridge is
H15-S12, which is heavier than that
used on most township bridges in this
each span to carry the dead-load of the
slab during construction. Fa lsework
piling can be easily driven at the same
time as the pier piles, and the single
bent of shoring will not offer a serious
obstruction to the stream during con-
•
struction. We bel ieve that th is wi II cost
area, and which should insure that the less than strengthening the tees enough
bridge will not be damaged by occa- to avoid the use of falsework.
sional overloading. Th is design was prepared at the re-
One question which arose in this de- quest of John Maddox, county engineer
sign concerned the bearing of concrete of Stafford County, Kansas. Mr. Maddox
on the narrow top edge of the web. We tells us that he expects to start con-
concluded, however, that th is is not
struction soon on the first bridge using
critical, since the shear connectors,
this design, and, if the design proves
6.1v D~ep a~8m acting vertically, will help transfer the
satisfactory, to use it as a standard for
~'~I Q8d, I Z"(Ja slab-load to the web. In addition, the
the constru ction of several bridges In
bottom layer of tra nsverse steel will rest
d'x d "113.3 7 the county.
or L'Jf4".< 5,," directly on th e top of the web and help
We do not. of course, have any actual
to distribute the load. There will cer-
tainly be a tendency for the slab to costs, but indi cations are that the total
4 ·x J '..0. Mach. Bol f crack over the stringers, but, if neces- cost of the bridge should run in the
sa ry, th e t ra nsverse steel can be in- neighborhood of $125 or less per linear
creased to preve nt th is wi thout affect- foot, or $4.00 to $5.00 per square foot,
ing the economy of the design. which we would consider to be quite
We found that the tees had sufficient satisfactory for this type of bridge. This
stiffness to carry the dead-load in the compare s with $150 to $175 per foot

•
H A NQ QA IL D € T A I, posi tive moment sections without shor- which is the usual cost for bridges de-
ing; they could not, however, carry the signed to Secondary Highway Stand-
negative moment over the piers without ards. This IS by no means the final
consi derable stiffening which would answer to the problem of low-cost bridge
involve ext ra steel and extra we lding. construction, but we do believe that it
We designed instead for a single bent of is an interesting step in that direction.

12
 The prime appeal of the Amite, La.,
Elementary School to the townspeople
is a savings in taxpayers' dollars - only
$7.50 per square foot. Second, is its

• attractive, carefully designed appear-

ance - clean and neat, yet highly func-
tional and adaptable.
An analysis of the most economical
framing system led to the use of a rigid
bent based on lightweight 12-in. steel
beams using continuous design and
supported by 4WFl3 columns. These
support a wood joist roof system. The
structural steel beams and columns are
exposed throughout providing the uni-
• Tilt' t'la.,room 1!"itlg. oprn dircdlJ/ to tilt ploJlIlTOIHtd at tltt'
fied architectural discipline for the
Trar of tltt> .("hool atld to tilt' moj" (,(,Jltra' corridor at tlu
entire building. Only 35 tons of steel Iront, u'hick in tllMI leatl. to Oll' ("a/rtcria, ,,,,,dlroom.
were needed; shop and field connec- admini.tration oDic,., and bll.foadi,lg.
tions were all welded.
The building design is based on the .. Extrrior 11'01/ mattrial. OTt' briel..· Il'illt 'Onlt' t'J'pn.td .olar
separation of grades into 3 wings of 8 tile. h,tcrior material. orr terrazzo floor. ottd plo,tie u.'all.
classrooms each, primary (1-2) in the lOT corridor•.
first wing, grades 3-4 in the central wing Cl4.SIl< ''''. CL AUll OO "'S C,-".,_ooon
and grades 5-6 in the last wing. If nec-
essary, these wings can be extended
to meet expansion requirements.
In the classrooms the floors are vinyl
asbestos over concrete, the walls are
corkboard and chalkboard over plywood.
The kitchens have quarry tile floors and
glazed tile walls.

•
The total cost of the entire structure
was $308,085. for 40,965 sq ft.
Oesmond-M "emont & Associates of
Baton Rouge and Hammond, Louisiana
were architects and engineers; John
Belcher was the general contractor.

fLOOR PLAN
 on the
UESTIONS AISC
•
& Spec
NSWERS and
Code
Q. The alignment chart given in the cooling when there is no preheat. The not exceed a total displacement
Commentary is a function of the column cooling rate of tack welds is likely to be of two inches toward, nor three
and girder stiffnesses, and from it, it is severa l times as severe as the Cooling inches away from, the building
possible to determine the factor K to rate of normal wel ding. line."
compute the effective length of the col·
A. The diagram on the opposi te page

•
umn. Is the factor K thus obtained the Q. How are the provisions in the Speci-
shows the column erect ion tolerances.
same for vertical loads as well as for fication and the Code of Standard Prac-
The envelope defined by the green area
vertical plus lateral loads? tice relating to column erection toler-
represents the maxim um variation of
A. The K·factor, which is used to deter· ances applied? These provisions are:
the column center line. The broken line
mine allowable axial stress F.. is the Spec Section 1.23.S.1: " Compression represents the error limitat ion of 1:500
same whether the column does or does members shall not deviate from straight- described in the Code.
not resist concurrent bending due to ness by more than 1/ 1000 of the axial The reasons for allowing a smaller
lateral forces. F, is defined in Sect. length between points which are to be displacement toward the outside of the
1.6.1 as the stress that woul d be per· laterally supported." building (the building line) area are:
mitted if axial stress alone existed. Of Code Section 7(h): " In the erection of 1. Encroachment on th e adjoining
course, as bending stress is increased multi-story buildings individual pieces property or over th e property line
due to heavier lateral loading (the ver· are considered plumb, level and aligned woul d be more serious th an hav-
tical loads remaining the same), the im· if the error does not exceed 1:500, pro· ing a column displaced towa rd
portance of the term f, / F.. and hence vided that: the center of the buildi ng.
the importance of K, diminishes. 2. Experience by major multi-story
(1) The displacement of the center-
line of the columns adjacent to building fabricato rs in New York,
Correction
elevator shafts, from the estab· who fo rmed the Subcomm ittee
A more proper answer to th e quest ion which made th is code rev ision
lished column line, is no more
on tack welding concern ing Secti on in 1959, indicates a tendency
than one inch at any point in the
1.23.6, which appea red in the first qua r- towards a shortening of beams in
first 20 stories. Above this level,
ter issue of MO OERN STEEL CON- upper stories. The cumUlative
the displacement may be in·
STR UCTION, woul d be: effect of many column-web half-
creased 1/ 32-in . for each add i·
Q. Why is tack welding included in the tional story up to a maximum thicknesses' being deta iled on
preheating requirements in Table of two inches. the "thick side" according to the
1.23.6? (2) Th~ displacement of the center- Manual, as well as the effect of
A. Preheati ng before tack wel di ng line of exterior columns from the beam lengths being detai led
can be considered to be at least as im- established column line is no 1/ 16-in. short, makes exterior

•
portant as preheating before final weld- more than one inch toward, nor columns tend to be displaced
ing, whether the tack wel d is incorpo- two inches away from, the build· away from the "building line."
rated in the final weld or not. This is ing line at any point in the f irst The limi ts given in the Code and Spec
necessary to prevent undue hardeni ng 20 stories. Above this level these were agreed upon by a Subcommittee of
and possi ble cracki ng of base metal limits may be increased 1/16-in. architect s, elevator manufactu rers and
which ca n develop as a result of rap id for each additional story, but may fa brica tors.

14
•, .

• , 'l
....,
=f! , : Cl
-t-B_UILOIN6 PLAN ~~ 8UILOING PLAN
.~-
I
y Ii 2' 2'
"
IlIi rt
,r
I I
~
-.. JI .5LOPE I.
~
~
PER 3TO.
-.J IrE5 TA8L/SHED "7.
\..... ,~ COLUMN LINe";
<:'bI \
4J
~~
~
I,
;i
PLUM8 LINE

.5LOPE JfG"
,
I

~::5
""
Q8 ,I PeR 5TORY

~\ I

• ..
Cll;)
@ !i I
~l
I
,rr II
'\
LIN
I

I
,"
" r-CENTt.R LINe. AXIS
8e.TWeEN INTER ..
"~t [,- 't-F'ELD ~PLICc.
i't')
~ SECTlON.5 OF FLOOR.
" ~- 'f IO ~ 8EAMS OR OTHER
~C( . '?-"-'I
1
LATERAL .5UPPORT.5
~~
I

,~Il)
t:::\
':<:4Jv
,~ .....
IIJ-.J

0 II
II
"-
\~ .,
L /\\
/ I
"
-H~
f. ~-Icb:J
J Af30UT c/THER AX-IS
~4..it
1;)11)
)-MAX.'".5a:J . ~I'
FIEl..O [;SPLICE
~::r:Cl
\.!)K-.J
<:\t)1IJ ® " j'

"-J<:i;: I
"IIJ
7'<: -.J I~ I
-..J

U\IJ~
.....
0 l- iL- ACTUAL COL . i uI
•
~ )....

~~IIJ ~ \ \
I-...Q.IQ
~ .6.4.5£ "
£L£V. EXT COL. ELcV. COL.
15
 RESEARCH
Steel has long been a reliable and
economical structural material espe-
cially adaptable and useful in the con-
struction of bridges. It has been with
us for so many years that there has been
little questioning of the design con-
approach is in contrast to elaborate
analyses based upon criteria of incom-
plete strength.
A good example of the merits of
proven strength may be found in the
case of the thin-web plate-girder inves-
•
CUTS cepts and criteria of its use. With the
increased popularity, availability and
tigation which has been under way for
some time. The results of this investiga-
economy of some of the newer higher- tion have been incorporated in the new
STIFFENERS strength steels and extensive efforts to
achieve maximum economy together
AISC Specification for buildings. Previ-
ously the simplifying assumption had
with safety and serviceability, many of been made that membrane action was
ON THIN-WEB these long-accepted concepts of steel
design have been brought under close
inconsequential and should be ignored
in the design of plate-girder webs. Ex-
scrutiny and restudy. tensive tests on full -sized girders proved

PLATE Perhaps the most notable develop-

ment has been the highly successful
conclusively that rather than being of
minor importance, membrane action
plastic method of design. At the present accounts for a large portion of the ulti-
time this concept has little applicability mate strength of plate-girder webs.
GIRDERS in the design of bridge structures; there-
fore it is not appropriate for discussion
Rather than being on the verge of col-
lapse when the beam shear-strength has
now. However, it has keynoted the ques- been realized and web buckling com-
tioning, research and restudy of many mences, a "thin" girder web accepts
by William A. Milek, of the long·accepted design criteria by larger shears without distress. In the
Research Engineer, AISC AISC. Greater emphasis has been interest of economy this strength should
placed on a frank recognition of proven not be ignored.
strength as an engineering criteria. This The tests were completed on static-

. .4.-
r
·ft·
..
•
u
l
Girder FI .,., About l. Eacept
Beanng of ShfltMrl
j • ~
1;--11", ," /
t •
...
12· . I·~ ~

~
t ...
~wu ;J....
I ' 3"' f ~wu 0 ~ Ii ."00
i I / L... , /-+-
12",,'J
I I II·.I · ~
I
~
I
n

1O"f;t
J!r I
~
I
20' -0'
_I

I
----- -
l~"_O '
-
I I
I
Girder F2
12',,'_ ...,..
Sym. About tEat.
of Stlfl.,..
j.
"+----., r"r- II "" !

r. ... , ~
vi~ I~"oo
~t... I,,,
•
.; !"',f
~ I,
4'-2 " ... 1
4-
-~ r*-.-
1. 1·-"---
i
,,-. ,-J I
/

3~ _ "4'_ _ "'4'
I~'·-'·
3'-4" I 19"
- -=---
3'-0' 4'·z· 4 '- 2~
,.... ,"
4' .e-
- 4'-Z' ,
"I "-0' _ _3'-4' _
13'-,'
;"-4'
5"·'1
Figure 1 Test girder. and let up
16
• ally loaded girders at the time of the
writing of the AISC Specification; how·
ever, it was anticipated that girders of
similar proportions would prove their
strength under fatigue·type loading in
an equally convincing manner. The pro-
gram is continuing to assemble the data
shown in Figure 1.
The procedure used to determine the
test loads is best explained by a modi·
fied Goodman diagram. As shown in
Figure 2, plotting maximum loads as
ordinates and minimum loads as ab-
scissa, point A is located as the static
of reasoning, a range of loads from 70
to 130 percent of P (46.5 k to 93.0 k)
was adopted for the tests. This should
give the same fatigue life as that of a
girder subject alternately to full static
design load and no load.
It was estimated that the girders
to corroborate this assumption. ultimate load, that is, the load at which might sustain 2,000,000 cycles of load-
To date two full·scale fatigue tests tension·field stress reaches yield point. ing, and in such event the range woul d
have been completed and reported. The Point B is located on the vertical axis be increased to the capacity of the test-
girders were 41 ft long, 50 in. deep with indicating the maximum load when the ing machine after 2,000,000 cycles in
webs only 3 16·in. thick. These propor· load is cycled from zero to maximum. order to speed up failure.
tions reduce to a web depth·thickness Its position vertically was determined Figure 3 shows the history of the
ratio of 265 to 1. They differed from by assuming the working fatigue·load loading. The load was cycled between
each other only in the spacing of inter· to be In proportion to the ultimate load 46.5 k and 93.0 k at 250 cycles per
mediate transverse stiffeners employed as the static working stress is to the minute with a shutdown for stat ic test-
in the test section. For the test, one yield point stress. Thus, p. = 18 33 ing and close inspection at 1,000,000
girder had stiffeners spaced at intervals (P. ) = 0.545 P". By connecting point A cycles. At 2,000,000 cycles the girder
equal to the girder-depth. The second to the origin with a straight line, and was static-tested to 110 k, and, upon in-
had stiffeners spaced at 1V2 times the point A to point B with a straight line, spection under load, two hair cracks
depth . One girder was in accordance we have a complete diagram. Load were discovered adjacent to one stiff-
with the provisions of the November ranges from the lower to the upper line ener. Since the cracks did not penetrate
1961 AISC Specification, while the sec· at any load-level would be expected to through the 3 16-in. web, an attempt
ond was beyond the limits of the recom· have equal lives. Under actual condi- was made to repair by welding over the
mendations. Loading was arranged to tions, dead-load might reasonably be crack without gouging. However, the
produce high uniform shear throughout expected to equal about one-half the cracks reappeared after 70,000 cycles.
the test section. These specimens are full dead plus live·load. With this line Loading continued for 500,000 cycles

•
p
erae\
Max.
oad
I GIRDER F2
r rCrOCk rCrOCk

OOI~----------------~-- ••-- 0 OL-______- L________ ~ __~ __~ __~----~

Pu Min. o 2 t 3x cycl ••

• Load LWlldWl4 Wlld"'4

Figure ~ Ttlt load. Figure 8 Stqlltnce 0/ tuting

17
at which time the cracks had penetrated survive more than 4,000,000 cycles of specimen would require a transverse

•
through the web and started to propa- loading which perhaps is a more con- stiffener spacing of 17 in. in addition to
gate upward and downward. Before too vincing bit of evidence that fatigue is a horizontal stiffener at the web thick-
much damage had occurred the area not a serious problem in thin-webbed ness of 3 16-inch. If the spacing was
was isolated by the welding of addi- plate girders. the same as spacing used in the test
tional stiffeners either side of the What is the significance of these girders, the web thickness would be re-
stiffener along which the cracks had tests? Certainly, successful tests of two quired to be 'l8 inch. According to the
developed. The new stiffeners per- girders cannot be considered as con- new AISC rules a 'l8-in.-web would re-
formed the function of taking that clusive, but it is important that these quire no stiffeners at all. Thus We can
portion of the web and the original two girders did perform under severe safely draw the conclusion that all stif-
stiffener out of action. Now the test fatigue-loading without exhibiting more feners can be eliminated on many plate
section consisted of two panels in which severe fatigue problems than have ex- girders with a significant cut in fabrica-
the space between stiffeners was 0.85 isted in the past. It does demonstrate tion costs. On very deep girders, two-
times the depth and one panel in which that the AISC rules for plate girders in th irds of the stiffeners can be elimi-
the space was equal to the depth. building construction, suitably modi- nated with perfect safety.
Loading was resumed at a higher load fied to the lower stress-level which Further tests are planned to broaden
range of 55 k to 110 k. After 3,080,000 bridge engineers habitually employ, may the range of conditions tested and to
cycles cracks appeared in the larger of quite possibly be used in the design of provide an accumUlation of data suffi-
the three panels. The cracks did not plate girders for bridges. cient to be considered conclusive evi-
penetrate through the web and loading The employment of such rules may dence of the serviceability of thinner
continued under close observation up or may not result in a great reduction in girder webs under fatigue loading.
to 3,277,000 cycles at which time the the thickness of plate-girder webs, but
test was terminated. the elimination of some or all of the in- Tlli~ article i. adapled fro111 an
As was to be expected, the second termediate stiffeners is an opportunity addr... delivered by Mr. Milek al Ille
girder did not perform quite as well as for economy that is waiting to be Hig/nt-IllY Re!(>aTclt Board in Janu ary.
the first, since it was beyond the limits plucked. He describes how Te8earc/r. ha5 con-
of the Specification even for a static- For example, designed in accordance tributed 6;gui{rcant savings by the
load condition. On the other hand, with with AASHO Specifications for com- pliminution 0/ ,til/ener! on plate

•
repairs made as cracks occurred, it did parable geometry and loading, the test girders.

RESEARCH PROGRESS IN BRIDGE DIAPHRAGMS

A research project to study the effec- four different sizes (l4WF 61, 14 WF the test bridge and provision for H-15
tiveness of diaphragms in steel girder 34, 10 WF 25, 8 [11.5) located at either truckloading. No interaction between
bridges is being conducted at North quarter or third points of the span, with slab and stringer has been provided in
Carolina State College, Raleigh, N. C. variable end connections from full re- order to maximize the deformations.
Objectives of the bridge project are straint to practically hinged ends. In the third or final phase, a stand-
to establish the effectiveness of dia- The Carolina Steel Corporation, in ard bridge with a six-inch concrete slab
phragms in lateral load distribution and cooperation with AISC, contributed five in composite action with the stringers
to develop criteria for the design of 60-ft stringers, diaphragm material and will be duplicated for experiments. To
diaphragms, including design data for all necessary high strength bolts for this determine maximum load, the bridge
end connections of diaphragms to fundamental research effort. The con- will carry to a point of failure.
stringers and efficient diaphragm spac- tribution amounted to 40 tons of steel. The bridge project is co-directed by
ing for load distribution. Prof. C. R. Bramer, acting head of the
The study, conducted by members of Tests on the diaphragm action in the Department of Civil Engineering, and
the Staff of the Department of Civil bare steel stringer grid have just been M. E. Uyanik, civil engineering profes-
Engineering, is one of sixteen projects completed and data is now under an- sor, with the assistance of civil engi-
in an intensive Highway Research Pro- alysis. In this phase of the investigation, neering graduate students.
gram administered through the Depart- the concrete slab was omitted on the The aim of the Highway Research
ment of Engineering Research in coop- test bridge and loading with hydraulic Program, coordinated by Prof. C. M.
eration with the North Carolina Highway jacks was restricted to each individual McCullough of the civil engineering
Commission and the United States stringer. Variables were the load posi- faculty, is to find the most efficient and
Bureau of Public Roads. tion, the number and position of dia- economical methods for construction
phragms, and the size of the diaphragms.

•
The 60-ft full-scale model bridge is and maintenance of North Carolina
located on the grounds of the Highway Strain gages were used in the measure- highways.
Bridge Maintenance Department. It con- ments of the load distribution. After all tests are completed and
sists of 5-24 WF 100stringers (designed A second phase of the project has analyzed, a complete report on the find-
for an overstress of 50 per cent for H-15 begun, with the same variables, but in- ings and the criteria developed will be
loading) and four sets of diaphragms of cluding a three-inch concrete slab on published in this magazine.

18
 • COACH HOUSE MOTOR INN
The new Coach House Motor Inn,
recently completed in Milwaukee, Wis.,
is a single-bay, seven-story cantilevered
building that demonstrates the eco-
nomies which can be effected by a
structural steel frame fireproofed with
vermiculite plaster.
The rigid frame design reduced the
framing cost about 15 per cent com-
pared with prestressed, precast, or
poured-in-place concrete. It also made
a 15 per cent saving in building height.
Beams cantilevered from the columns
and flange-to-flange welding achieved
a closely ba lanced moment design for
the specified floor loads and a wind
load of 70 mph. The simplified full-
section flange-to-flange welding detail
used was considered to be 25 per cent
more economical than rivet ing or high
strength bolts for the moment connec-
tions requ ired, and allowed half of the

•
welding to be done in the shop.
I Contact ceilings of vermiculite plaster
on metal lath tied to bar joists provided
the fireproofing and the finished ceil -
ings and made a saving of 12 per cent,
compared with a suspended ceiling
system. Cantilevers ore erpr(,88cd arC'hitrcturally by curtai~f wails brtturn the five upper
The Coach House contains 102,500 /lOOTSand by a belt 0/ precast Iwit8 01 the floor line. The (Ollflll}1 bay is
sq ft, including indoor parking. Its cost emphasized by a tJcrt ieal pa~t('l 0/ precast nnits ill which the aggregate is exposed.
was $1 ,350,000 or $13.17 per sq ft. The
building is 77 It high . The five upper welded to th e colum ns, which were fab- varie ty of accommodations in a limited
stories are 46 It wide and 232 It long, ri cated in three le ngth s. Th e onl y space. The five upper hotel f loors have
with floor-to-floor spacing of ten feet. braci ng used were temporary cable ties 105 sleeping rooms, including f ive lux-
The two lower stories are 70 It wide and and bolted erecti on angles. As soon as ury suites consisting of two rooms, two
250 ft long. The first story is 15 It high; an assembly was in place on each side, baths, and an outside balcony.
the second, 12 ft. th e ce nter bea ms were la id between and The basement and the first and sec-
The building is carried on two rows fiel d-welded to th e col umns flange-to- ond floors are given over to parking
of 14-in. columns with a maximum flange. In addition to visual inspection, space, meeting and dining rooms, ki t-
eight-inCh flange width in order to save th e we lds were exam ined wi th radio- chen and bakery, and a large banquet
as much floor space as possi ble. Col- graph s and magnetic pa rticle testi ng. room. The Coach House will cater to in-
umns are spaced 18 ft on center the The column fi reproofi ng is vermicu- dustrial conventions that present new
length of the building to accommodate lite plaster, 1% in. thick on sel f-furri ng products, and there is an over-sized
lower level parRing. The center bay is metal lath. Theceil ing th ick ness is ;,.-i n. door to the second floor banquet room
30 ft wide. The cantilevered section s on rib metal lath. Both assemblies have that permits displays seven feet high
project from the columns seven feet on three-hour f ire ra tings. and eight feet wide to be trucked intact
one side, nine feet on the other. The columns are concealed in 12-in. into the room.
The typical center beams are 16 in. partition s to in c rease the so und- Architect for the project was Sheldon
deep to provide the desirable ten foot reduction valu e of the wall between oc- Segel, AlA. The struelural engineers

• floor-to-f loor height. S-6 bar joists, 12

in. deep, span the beams two feet on
center. The decking is corrugated steel
with a 2¥.!-in. concrete topping.
The cantilevered beams were shop-
cupancies and to keep the colum ns
from project ing into th e rooms.
Corridors are off-ce ntered to permi t
rooms 19 It deep on one side, and 20
ft-6 in. on the oth er, which provides a
were Collings Engineers. Steel fabrica-
tion and erection were by the Milwaukee
Bridge Company, and general contractor
was Drobac & Associates. All firms are
in Milwaukee.

19
 •

•
STACKING THE DECK
by Robert F. Lorenz, use of composite action. More impor- typical 15-ft dimension. Shear con-
Regional Engineer, AISC tant, the cost was within the budget nectors are shop-installed five-inch
figure, and construction of the project channels placed at equal spacings along
could proceed. each girder's flange. A haunch varying
The ramp, covering a full acre, con- between one inch and three inches is
The Ryan Parking Deck in st. Paul, sists of two levels of attendant parking used to meet the sloping slab's grade
Minn., had originally been planned as and will have a capacity of 265 cars. requirements as well as to vary section
an IS-in. reinforced concrete flat slab The lower parking level is below grade, properties when required. Standard con-
using waffle-slab construction with 24- with the next level at street grade. The nections are made to the columns with
in. drop-panels. The scheme was re- design includes the possibility of adding high strength bolts.
jected when the bids exceeded the four more levels at a future date when The structural steel is A36 material
project budget limitation. conversion to a self-parking ramp is designed for 24,000 psi bending utiliz-
James Tillit, structural engineer, with contemplated. The lower level is serv- ing the compact section principle which
Walter H. Wheeler, architect-engineer, ioed by a circular ramp. AISC also introduced in the new Speci-
Minneapolis, then decided to implement The structural steel composite girder fication. This feature alone reduced the
the new AISC design criteria for com- system supporting the slab consists of weight of a typical girder by as much as

•
posite design of buildings, combining a two bays of 56 ft with two additional 10001bs.
six-inch light weight concrete slab com- narrow bays of 39 ft and 19 ft respec- Jack Geller, St. Paul, Minnesota, is
positely with structural steel girders. tively. For the typical 56-ft span, beams the owner-contractor of the project.
Although the projected height of the are composed of 33-in. shapes with 36- Structural steel was fabricated and sup-
building would be slightly greater with in. sections utilized where clearances plied by St. Paul Foundry and Manufac-
steel framing, this was minimized by are not critical. Beam spacing has a turing Company. St. Paul, Minnesota.

20