wire rope as tension member in a truss

huangyhg

wire rope as tension member in a truss
we are using wire ropes as tension members in a 120 ft long, 6 ft deep trusses. the trusses are to support a gym roof. we sized the wire ropes so that the factored tension is about 85% of its breaking strength. the wire rope is connected to gusset plates using open swage sockets. the trusses have only tension diagonal members and are not subject to uplift. any comments on this would be appreciated. thanks
wire ropes have a lower modulus of elasticity than steel. it can be as low as 50% of the steel e-modulus. so one has to accept higher deflection, but i guess you (and the analysing software) knew this already.  
not may terminations reach 85% of the cable strength. especially not in fatigue.
cheers
greg locock
dear greg and alimar,
i've done something similar to this before on a major structure (domed stadium) with bowstring trusses but using bridge strand.  attached is a photo of the bottom chord/diagonal/vertical connection.
i don't recall to what level we designed the cables but i seem to recall the weak link was the clevis connection and not the strand itself.  with this also, i think the modulus of elasticity is only slightly smaller (28,000 ksi??) but since the strength is so high, the area comes in low and then you do get deflection issues entering into the solution.

iliconsult
there is not such a thing as a yield strenght on a wire rope, breaking strength is when the rope breaks
many connection devices are only designed to achieve a % of the breaking strength of the rope. 80% / 90% ... usually that is specified by the manufacturer. there are connections that achieve 100%, but not all.
ailmar, i am with greg, the 85% seems high. another comment, which i am sure you already have considered is that the wire rope bracing would only be effective once tensioned (if installed slack it would take quite a bit of deflection go get it working) so you have two problems when you instal it pre-tensioned. ensure that it has the right amount of pre-tension, and consider the pre-tension loads on the truss.
the infinite life fatigue strength can be as low as 25% of the breaking strength in steels depending on the dynamic loads expected. 85% design may be dangerous.

fe
it seems you would need to use a lo-lax strand to prevent geometry and truss load changes due to strand relaxation, especially at the high stress levels you mention. the e for typical 270 ksi lo-lax strands used in bridges (referred to by jae)is 28,600 ksi. you should note that these strands are only taken to 0.70guts final stress and to 0.80guts max during stressing operations.  
very important inputs. thanks to all.
iliconsult: yes we considered a low e for the cables
jae: the connections are same as your
kelowna: although we asked for no slack in the cable, we considered the possibility of some slack by designing the truss as a viedrendeel truss to support dead loads and checked the max. deflection. so, the truss will be ok even with a slack equal to the deflection due to dl. local cable suppliers said we can't prestress.
  
i've got a pretty good technical data manual from wireco that you may want to see if you can get your hands on.  in addition to the lower e for wire rope it also talks about "metallic area" which is used in calculating the stretch of the rope and is lower than the nominal area.
i've got to pile on about the 85% of breaking stength thing.  unless you used really high load factors (4 or 5 maybe) that seems dangerously high to me.
dozer,
a factor of safety of 4-5 seems to be very high for this application. i understand the need for such high sf for lifting equipments where dynamics could be unpredictable, but for a truss supporting roof where the major load will be from gradually applied snow load, do we really need sf=4 to 5?