asce valley drift load accumulations

huangyhg

asce valley drift load accumulations
i work in the state of wisconsin.  we are in the process of switching from our own building code (with some vague guidance for valley drift load) to the ibc.  as such, loading criteria will now fall under the umbrella of asce-98 (loads for buildings & stuff) for the most part.
my question is this: how does a guy (or gal) use the drift load provisions of asce to create drift loads in valley sections of roofs (eg. where two large sections of gable roof intersect perpendicularly)?  there are some provisions for saw toothed roofs that could be applied at an angle to the valley ridge line i suppose.  however, i don't get the sense that that is what the saw tooth stuff was intended for.
what are other people doing with respect to drift/accumulation loads at valley regions?  based on the plans that i've seen here, a lot a architects/truss designers don't bother to consider them at all.  suggestions??
adam p
check out our whitepaper library.
i agree that this is a problem for us engineers.  i live in canada, and our nbcc (even the commentary) is not entirely helpful when it comes to valley snow accumulations, although they make a stab on page 64, fig. h-3 of commentary h ("snow loads") of our national building code 1995 "structural commentaries". in frustration with this very problem of how to provide for valley accumulation (which can be substantial!), i wrote away for some old research that was done by the national research council of canada, division of building research.  they (for a fee of something around $30 (c$) will send you two photocopies of case histories:  "snow accumulation in canada: case histories i and ii"  compiled by schriever, lutes, et al.  [technical papers no. 237 [case i - 1967]  & 339 [case ii - 1971].  although far from satisfactory (in comparison with a nice table of coefficients, or even an iso-bar map!), it is a start, at least.  if you wish to pursue this further, the address of the nrc (for these case histories) is  
thanks aton.  i'll see if i can get my hands on that nrc report.  i worked as a roof truss designer in bc for several years while i was a student.  during that time, it was basically company policy to ignore the phenomenon unless the loading was directly spec'd by the a/e.  i see the same thing happening here in wi and, from the engineers perspective this time, i find it kind of unnerving. i want to find a good algorithm for specifying valley drift loading on our plans.  maybe even as a standard detail for the usual 90 degree roof intersection if possible.
adam
i also am somewhat shocked that no provision was made in your past work experience for valley snow.  when you get those old nrc reseaarch reports, you will feel justified in pursuing this concern.  good luck. if you have trouble getting them, let me know, and i can photocopy my own and send them to you.  at least it is a start...  let us all know how you eventually handle the problem!  this sharing of knowledge is invaluable.
sustainable, solar, environmental, and structural engineering: appropriate technologies for a planet in stress.
anton,
it would be great it you could send me those reports.  also, could you send me that page page 64 of the canadian code.  i don't want to buy the whole thing for one page but i'm curious about what it says.  here's my contact info:
fax: 15092756364
adam, i'm sending some relevant excerpts (about 10 pages), and you can see how useful they are.  be in touch...
-aton
sustainable, solar, environmental, and structural engineering: appropriate technologies for a planet in stress.
thanks anton!
the information that you faxed me was very helpful.  it's surprisingly similar to the information in the wi building code (although a good deal more explicit).  when i come up with an algorithm or standard detail for my plans, i'll be sure to come back and post it here.  thanks again.
adam
i suspect you find little info on snow load acccumulation in roof valleys because valleys for the most part occur in small buildings (residential) wood framed, very redundant, canadian building code part 9 permits you to ignore all snow load coeff. if wood framed, members 24" oc or less, spans less than 40', this is based on past performance record of these systems, and strong lobby of wood industry, cwc etc..., we have a problem with this minimum standard, especially in light of much of the past performance based on poorly insulated steeper roofs were the accumulated snow often melted and ran off the roof.
having said that we do alot of agricultural buildings, wood framed, 25,000 to 40,000 sq.ft. with significant valleys. we ignore the canadian farm building code which basically allows much reduced snow load coeff's. based on the fact that 'collapse won't result in loss of life', or some such gibberish, lsd importance factor = 0.8, load sharing wood members on 4' centers, slippery roofs, (slippery slope if you ask me), etc., and use the following; part 4 snow loads, cb=0.8, cw=1.0, cs=1.0, ca=1.25.  after we have done this we divide this load by the avg. denisty of snow to get its depth and depending on the geometry, height difference of the two intersecting ridge lines adjust this load up, never down. yes we also find the the truss manufacturer has little interest in all this, prefering to design the truss for minimum load, unless the 'engineer of record' provides design loads, all the plate manufactures state this in their disclaimers ...'truss designed for loads shown it is the responsibity of others to verify' ... fair enough, their manufacturers, suppliers, subtrades to a job primarily interested in profitability ...