Liu2006 2

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

Y. Liu et al.

/ Journal of Constructional Steel Research 62 (2006) 132–143 133

of two parallel HSS 254 254 9 5 frame members spaced


Nomenclature at 1240 mm on centre. Details of the loading beam and
load actuator attachment are shown in Fig. 2 while details
Bk width of knife plate of the reaction beam and flat load cell mounting are shown
bc width of Charpy V-notch coupon in Fig. 3. The horizontal test frame was bolted to the strong
E modulus of elasticity of material floor by means of three 50 mm diameter A490 high strength
Fy specified yield strength of material bolts through the web of the loading and the reaction beam.
Fu specified tensile strength of material Elements of the test frame were designed to be capable of
hc height of Charpy V-notch coupon applying a maximum axial load of 1500 kN to specimens.
hc n notched height of Charpy V-notch coupon
hn notch depth of Charpy V-notch coupon
2.3. Test specimens
Lk length of knife plate
L length of longitudinal fillet weld
As shown in Fig. 4, each test specimen consisted of a
l weld leg size of longitudinal fillet weld
hollow structural section with a knife plate connection at
R radius of gyration
either end. The HSS members were fabricated from G40.21
th wall thickness of HSS
350W [2] steel while the knife plates were fabricated from
tk thickness of knife plate
ASTM A36 steel [6]. [ To mount a specimen for testing,
gross cross-sectional stress
one knife plate was bolted to a clevis plate (Fig. 5) at the
AVE average gross cross-sectional stress
load application end while the other end was bolted in a
i strain at i th location
similar manner to a load cell attached to the reaction beam.
u ultimate strain
The length of the HSS, 1700 mm, was chosen to eliminate
y yield strain
interaction of boundary effects. Measured dimensions of
the twelve specimens are presented in Table 1. Values
shown represent averages of three to seven readings in
each case. Specimens KPC01–KPC10 were fabricated using
HSS 102 102 6 4 while specimens KPC11 and KPC12
were fabricated using HSS 102 102 4 8. Specimens
KPC01–KPC04 were designed to be identical with only
slight variations in the fabrication. KPC05–KPC07 differed
only by the length of gap between the ends of the knife plate
and the slot. Specimens KPC08 and KPC09 were fabricated
using an 18 mm plate as opposed to the other specimens
which had 12 mm plates. To evaluate the effect of weld
length, KPC10 was designed for 200 mm length welds in
contrast to 150 mm length welds for all other specimens.
Fig. 1. Typical configuration for a knife plate connection.
2.4. Test apparatus
to investigate the strength and behaviour of knife plate
connections under the action of cyclic loading between zero The test set-up for all specimens is illustrated in Fig. 6.
and a prescribed upper level of tension stress. The computer controlled load actuator was used to apply
both static and cyclic loads to specimens. The knife plates
2. Test program at either end of a specimen were bolted to the clevis plates
using nine 25.4 mm A325 bolts designed and installed to
2.1. General prevent slip during application of test loads.
Test loads were applied axially along a specimen using
The program included testing twelve full-scale specimens a 1960 kN capacity load actuator mounted in series with a
under cyclic tension stress ranges. Two specimens were 900 kN capacity flat load cell. As shown in Fig. 6, the load
selected for a strain gauge study of stress distribution in the actuator was anchored to the loading beam by means of a
elastic stress range under static loading. Associated tension 25.4 mm collar plate and two 150 150 32 side plates
coupon tests [4] on HSS and plate specimens and Charpy [5] which were welded to the collar plate. Each side plate was
V-notch impact tests on HSS material were also conducted. welded to an end plate bolted to the loading beam flange by
four 25.4 mm A325 bolts. The web of the loading beam was
2.2. Test frame stiffened using two pairs of web stiffeners as indicated.
The load cell was monitored by a computer that
The test frame had a simple rectangular configuration continuously recorded the applied loads during tests. The
consisting of one W360 91 beam connected at either end load cell, made of stress-proof steel with a minimum yield

You might also like