Sloped roof snow load at a roof valley

[sundale]

If one has a steeply sloped warm roof with metal deck roofing, and one wanted to calculate Ps (not Pf), would a roof valley be considered an "obstruction"? Namely, would it be appropriate to take the valley angle to compute Cs, not the rafter angle?

I recall that the old UBC used to have a Cv boogie factor that implied additional drifting in a roof valley...

I normally just apply Pf to a roof and ignore any sliding implied with Cs, as the roof may be re-roofed with asphalt shingles in the future by another owner.

Thanks for any opinions offered.

 

[RARWOOD]

I would not considered the valley as an obstruction. I would use Ps for design of members away from the valley.

I genmerally will design for additional snow in the valleys using the 91 UBC as a guide. I am not sure how ASCE-7 deals with snow accumalating in valleys.

I know from looking at buildings around Minneapolis, that snow does accumalate in valleys so I generally consider it in my design.

I have noticed that the unbalanced load requirements are much higher in the ASCE-7 than what was required under older Minnesota State codes. Currently ASCE-7 treats unbalanced snow as a drift condition on the roof. That my be the reason ASCE-7 does not deal with accumalation in the valleys, I don't know.

I design strictly in wood, and will reduce for sliding snow when I am able to. I am not a big believer in designing to account for what the owner may do in the future. With wood the major area where failures occur is at the connections.
As a general rule the smaller the number and the smaller the diameter of the fastners, the better the joint performs.

I recently dealt with a job where the reactions were provided for use in the wood connection design. All the connections involved a larger number of bolts with top flange bearing. I found out latter that the reactions including a 50 p.s.f. loading to account for the possiblity that at some time in the future the owner might decide to add roof top units.

As a result of that 18" wide and 20" wide glulams were specified to meet the bearing stress requirements, in applications where 8 3/4" wide and 12 3/4" wide glulams of the same depth were capable of carrying the gravity loads.

The situation could have been improved if the architect had allowed some changes which would have increased the bearing area for the beams.

However there would have been large savings to the owner if the 50 p.s.f. load had not been added, just in case the owner wanted to add roof top units later.