Snow Drifting

[jwags]

In ASCE7-05 the drift height is calculated by a graph or equation involving the ground snow load and the length of roof upwind of the drift. My question is "Is there a reasonable maximum length of roof upwind to consider in the drift?" This is a result of a maufacturing facility that is adding a addition on the side of their exsitng building. The change if roof height is around 10'. The problem though is the length of the existing factory building 1000'. I have calculated the drift using lu=1000 but I am wondering if I can reduce this value.

 

[jike]

I have also wondered about this. I assume that the code provision is assuming that a portion of the snow on the flat roof area is blown to create the drift. Would it then be logical to use the distance to the first parapet, assuming that the parapet would interupt a clear path of wind blown snow?

 

[JAE]

No limit that I'm aware of.

They do allow you to use 3/4 of the calculated drift height for windward drifting. (See 7.7.1)

 

[msquared48]

However, in the no drift condition, the higher, unreduced uniform load would apply anyway for the structure of that roof area. So it's really a moot point.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[jwags]

Thanks for the comments. I didnt think there was a limit and I could not find anything in the code but I wanted to know if I was missing anything.

 

[tolchijb]

I have seen a similar situation in which a 500' long building was built 10' higher than an adjacent existing building. I was call out when the existing pre-engineered metal building failed due to drifting snow. The drift was every bit of 9 feet tall possibly even more. When the new building was built no one considered the new drift condition on the existing building. The first bay of the existing building adjacent to the new building had the roof purlins go from staight I shaped members to a big V. The purlins buckled midspan. All the workers could do was try to poke holes in the bottom of the V and try to heat the underside of the roof to reduce the load. Luckily no one was hurt. That being said, I would not try to reduce any drift loads. It maybe an extreme condition (we had heavy snow and high winds), but it is a possible condition over the design life of the structure.

 

[paddingtongreen]

I had a case where we were putting in a new extension about 10' higher than the existing building. We looked at putting in a false roof over the existing one, sloped in the shape of the drift, but it was a shade more expensive than strengthening the existing roof.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[NITTANYRAY]

I would never use less than the graph shows because you are making assumptions without any research data to back you up.
A lawyer would eat you up if you were ever involved in a lawsuit due to a roof collapse were you used a lesser snow load than the graph shows.
I once investigated a low roof collapse similar to the building you described but only about 300 feet long. There was only about 8" of snow on the ground but the lower roof had about 8' of snow on it when it collapsed.
The snow density could not be determined but it was probably less than the max assumed by the graph but the snow height was also considerably higher than the graphs indicated.
When dealing with snow drifts Murphy's Law should always be used.