Activate Javascript or update your browser for the full Digital Library experience.
Previous Page
–
Next Page
OCR
sectional measurements were'ma.de at the centres of the long members: but
Fig. 10 indicates the typical distribution near the quarter point.
STRESS DISTRIBUTION AT TRANSVERSE DI.iPIjIiuGMs.fAs several of the
nickel steel columns in the series of 1910 had failed at dlaphmgmsy ff“ 93%”
was made to secure some data as to the effect of these in the present series. F. 11
member TC, three sets of measurements wereyymade at 21dJ3Ce“V 59Ft‘01;15 l ‘K’
11 (a)]. Readings at section A1 ‘fstraddled . the d13P11T3flmI VIVh1%9, t i’i9t:E
sections BB and CC covered the adjacent portions of the I'll.()iS.b V Inn 113- CS and
stresses given by the readings over the diaphragm are denote 1) lu liiih leach
the mean of those at adjacent sections;the latter agreeing c os:);lti1N mater
other-by dotted lines. The stresses at the dlffphragmdalge Cgnsflsllfj vii gTab1C
than at the adjacent sections, as indicated by Fig 11 3“ Y t e 0 ‘ g ’
the difference being from 10 to 14%.
M5051” ed J'I7'er: - 7' bawmo’ $9 ” '
0 m 4.6’ 30 3 5
‘V .1 30 I .5 ll 5 M,
$5 9
t ‘ -'
s I r
" ,1 A sum
FUMIZ 7"z7al[a1xe
, V, szzm-,;.,.z;.;2i,-amzm.
.1 in
x V
0') 0‘
M’ $3 .- Sfn-Mr off 1:775” ,4].
'5 .' .' --- mpg?! o;’.!‘7fI.r.rtJ‘ J‘n77iinJ 55MdCC.
3 i 7
V) I
9
L 5 M 75 90 2" 3”
Meaxored Srre.rs- Tjaomxva’ ff! ' ' [ J
”g EA,”fo,wl%j,,,y,” Mffrmrex 7 I
. ' Sections
, t Diaphragm and Adjacent I
TABLE 7sCom()3“5On of Strellliesiiiliier T631
MEASURED STRESSES
ed] d Difference
Appli oa M
lbs. per sq. in- Mean of adjacent
At diaphragm sections
9310 935
9,475 19233 15-42% ill?
15,790 24'400 21.57.? '
20,200 I gm.-
.
207
On member TC.,2 readings were taken at corresponding section with a
20-inch gauge, but on the edges of the outstanding legs of the upper flange angles
only. The results are as follows re-
TABLIE 8sCoinparison of Stresses at Diaphragm and Adjacent Sections,
Member TC.2
MEASURED STRESSES
Applied ]0".d Difference
s. per sq. in.
At diaphragm Mean of adjacent
sections
9,500 10,300 9,450 850
15,800 19,400 18,100 1,300
20,200 28,100 ‘ 25,100 3,000
The differences run from 7 to 12% and are thus a trifle less in the member
with lighter lacing. The discrepancy between the applied load and measured
stress is, of course, due to the fact that the measure-
ments were made on the upper flange only in the 77"+'57 if
vicinity of the quarter point of the member, where the LOW“ 50'5-
bending stresses are considerable.
In the case of TC41 a set of measurements
was made “straddling" the diaphragm, but not at
adjacent sections. Comparison can, therefore, be
made only with the applied loads. The same action a
as in the former members is indicated. In member L
rivals.
E mu;
TC6, on the other hand, which has an II section,
there.is practically no evidence of greater stre$ at 1'
the diaphragms than elsewhere. Fig. 12
The stresses just discussed can, of course, be computed approximately by
treating the member as a framed structure and making the usual calculations
for secondary stresses. Such calculations give considerably higher stresses
than those indicated above at the outer edge of the flange angles, but lower
stresses in the webs. A consideration of the relative positions of the diaphragms
and lattice bars makes it seem probable that the web at the diaphragms is bent
on both horizontal and vertical planes, as shown in Fig. 12, which action would
‘ lessen the amount of bending in the flanges. This assumption is borne out by
the fact that a few vertical readings taken on the web at diaphragm points show
distinctly less vertical extension of the web, or lower values of Poisson's ratio,
than at points where there
are no diaphragms. I 24.‘
INTERACTION or-‘ Ln-
TICE BARSAND TIE-PLATES
-As the effect of lattice ‘sf ;
bars in a member under N
compression is to spread the rm,re,?;i" TC 2
Tib5y it is 0bVl0“5 that 507710 ,>‘gw1’ogx?xA: -ffazmrtlrzrnzr laitz73l;9ameJ:’z7'2d47VZ'x91x27lJ'I)'rJ5.
bending must take place . Fig, 13
5 v a
gut o"il>'DoV'i n‘“‘n"’i' all 12”" iiliiti l