which were used to determine the structural
weight of the truss system is supported just be-
integrity of the system.
low the level 3 collar, the lowest collar level (see
The derivation of the axial load on each col-
Fig. 5). Each column supports on average about
umn is straightforward. The weight of the build-
18 tonnes of truss at this collar level. When tak-
ing, the weight of the portion of column to the
ing sway bolt measurements at the level 3 collar,
point of measurement (the sway bolt location),
and the weight of the truss all contribute to the
tween the trusses and the supporting channel
axial load. Each column theoretically assumes
below the collar must be considered. The static
one-eighth of the building load at the second
coefficient of friction for dry steel on steel is 0.78,
floor level as well as one-eighth of the truss load
resulting in a break-out force of about 140 kN for
a 180-kN normal force. After the Teflon/stainless
at the level 3 collar location. Building loads were
steel bearing pads were inserted at the interface,
supplied by the contractor after the 1983 lift and
the static coefficient theoretically should have
loads at each level were calculated in 1985. Val-
dropped to about 0.04, with a resulting breakout
ues are given in Table 2. These loads remained
force of 7.2 kN for the same normal force. How-
constant for calculation purposes throughout the
ever, a drop in lateral resistance due to friction of
period covered by this report.
The horizontal loads experienced by the
this magnitude, approximately a factor of 20, did
not materialize after installation of the bearing
building and collars are far more complex. The
pads. This may have been due to the high loads
Table 2. Column vertical loads.
on the components or the measurement protocol.
Therefore, level 3 sway bolt measurements can
P/A*= fa†
Load (P)
only be assumed to be approximate.
Level
Column
(kN)
(kPa)
fa/Fa**
Another frictional effect to be taken into con-
1
A1
1.79
16.55
0.12
sideration when determining the horizontal
A2
1.79
16.55
0.12
component of the stress is the influence of fric-
A3
1.94
17.93
0.13
A4
1.35
12.41
0.09
N1
1.35
12.41
0.09
perpendicular to the sway bolt being measured.
N2
2.09
19.31
0.14
These bolts reduce the accuracy of the measure-
N3
1.79
16.55
0.12
ments from the constraints they impose on the
N4
1.79
16.55
0.12
free movement between the truss and collar. As
2
A1
1.94
17.93
0.13
resistance to movement due to these side loads is
A2
1.94
17.93
0.13
experienced when force is applied between the
A3
1.94
17.93
0.13
collar and truss during a measurement, the mea-
A4
1.35
12.41
0.09
N1
1.35
12.41
0.09
sured loads differ from the actual loads. Thus,
N2
2.09
19.31
0.14
sway bolt measurements should be used only as
N3
1.79
16.55
0.12
a guide in making an engineering decision on the
N4
1.79
16.55
0.12
condition of the structure rather than as an abso-
3
A1
1.94
17.93
0.13
lute quantity. It is in this manner that CRREL has
A2
1.94
17.93
0.13
used the data obtained over the years from sway
A3
1.94
17.93
0.13
bolt measurements.
A4
1.35
12.41
0.09
N1
1.49
13.79
0.10
N2
2.24
20.69
0.15
STRUCTURAL STRESSES
N3
1.94
17.93
0.13
N4
1.79
16.55
0.12
This report includes data and results for com-
Base
A1
2.09
19.31
0.14
prehensive stress measurements done by CRREL
A2
2.09
19.31
0.14
personnel for the years 1983 through 1988. Yearly
A3
2.09
19.31
0.14
combined stress factor graphs are included in
A4
1.49
13.79
0.10
N1
1.64
15.71
0.11
Appendix B. A complete set of data for the years
N2
2.39
22.06
0.16
19831988 can be found in Walsh (1992). Data for
N3
2.09
19.31
0.14
1983 are not complete because the lift operation
N4
1.94
17.93
0.13
was not finished at the time measurements were
* A is axial cross section of column (0.108 m2).
taken. A list of personnel involved in each series
† fa is the axial stress.
of measurements is given in Appendix C. The
** Fa is the allowable axial stress with no bending stress
methodology for obtaining the stress measure-
(137.9 kPa).
7