Unbalanced Snow Condition

[pioneer09]

Is the unbalanced condition a function of the windward or leeward roof slope?

I have an existing 4/12 pitched gable roof. An addition to the structure will require re-roofing over the entire leeward side of the existing roof. This will make the leeward roof have a pitch of 1.75/12 (8.3 deg). In WI, an alternate in the building code allows use of the unbalanced roof load design from Canada that states roofs of less than 15 degrees require only balanced snow loading.

I tend to think that the unbalanced snow load is a function of the leeward roof pitch, so is it correct to assume that only the 4/12 pitch side of the roof will experience unbalance snow per the alternate mentioned?

 

[dcarr82775]

The ASCE7 unbalanced snow load provisions are from windward to leeward. So the snow is picked up from the windward side of the ridge, and deposited on the leeward. So if the leeward side is flat I can see a good argument for not needing unbalanced snow there.

 

[BadgerPE]

Pioneer,

From you handle, I assume you are a UWP grad in 2009. Would I be correct assuming that?

 

[pioneer09]

BadgerPE,

You hit that one right on the end.

 

[BadgerPE]

Cool! UWP BSCE Dec '09 here!

 

[dhengr]

Pioneer09:
Sounds like a large shed dormer is in the offing, on the back of a house. Don’t forget that the thrust at the front exterior wall and at the ridge still has to be dealt with on the remaining 4/12 rafters on the front of the bldg., that doesn’t go away. And, a ridge board won’t cut it any longer, plus you have the reaction from the new shed roof rafters. Do you really mean unbalanced loading or do you mean non-uniform (as in drifting)? Drifting snow can occur on either side of a change in plane or an elevation change, but I don’t have any of the latest codes, so I can’t comment on that aspect of the issue. I would expect the 4/12 roof to show more/heavier drifting however. You can’t go to far wrong following the Canadian code on snow loading, their early research and codes was really the impetus for the U.S. finally getting up to speed on snow loading issues.

 

[KootK]

My understanding of the aerodynamics of unbalanced snow loading is the opposite of yours Pioneer. I believe that the windward roof slope is what matters. Or, more precisely speaking, the angular change between windward and leeward roof slopes.

As wind passes over the windward roof slope, it is forced to speed up, similar to what happens with an airplane wing. When this happens, more snow is picked up like sediment in a fast moving river. When the wind passes the ridge, it starts to slow down and deposit that same snow onto the leeward roof.

As a mental experiment, imagine a roof that is 6:12 on the windward side and flat on the leeward side. I'd still definitely expect to see unbalanced snow accumulate on the flat portion of the roof. It would probably be a bit less than if the leeward roof were also 6:12 but it would still be significant. And trying to accurately estimate the difference would be a dicey proposition in my opinion.

For what it's worth, I'm a Canadian who learned to engineer in Madison. I didn't know about the option to use NBCC. Good to know...



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[woodman88]

Oh for the "good old days" of the UBC
"Snow loads full or unbalanced shall be considered in place of loads set forth in Table No. 23-C, where such loading will result in large members or connections.
Potential accumulation of snow at valleys, parapets, roof structures and offsets in the roofs of uneven configuration shall be considered..."

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[JLNJ]

It pays to be conservative when it comes to snow drifting. Plus, you don't want to do $5000 of non-exact science engineering to save $5000 in construction costs. The Owner is probably better served having a stronger roof.

 

[medeek]

I would just run the unbalanced snow load for the 4/12 pitch and this way enveloping the design.

A confused student is a good student.

 

[dcarr82775]

KootK, I have to disagree in part.

I agree 100% the windward side is where the snow comes from, and on the mechanism the snow is moved/deposited. But, the leeward side is rather flat. The area is exposed, and while certainly there will be a disturbance in the wind, which releases snow, I do not see that it can accumulate given the lack of a shadow. It will continue to move on down the roof. If you could get unbalanced snow of significance I would think it would happen farther downslope in a location not predicted by the Code (ASCE7 anyway).

 

[pioneer09]

dcarr82775,

I agree with your last response with regards to the snow not wanting to accumulate on the flatter leeward roof. Yes the snow travels from the windward roof and moves over the leeward roof. My visualization leads me to think, the flatter the leeward roof the easier it is to keep moving and not deposit on the this roof This leads back to my initial post of unbalanced snow being a function of the leeward roof pitch.

 

[KootK]

@dcarr/pioneer,

Well, truly, I'm just fudging my way through this using my intuition and knowledge of fluid dynamics. It's an interesting conversation however.

It sounds as though we agree on everything other than the conclusion:

1) There will be pickup on the windward side. Check.

2) There will be drop off on the leeward side. Check.

3) There is potential for accumulated snow to blow off of the the leeward side. Check.

The trick with #3 is that it is true whether the leeward side is flat or sloping. It may well be more true when the leeward side is sloping but, again, how do you quantify that benefit in a way you can rely upon it? On a normal gable roof, the snow could also blow off but we don't account for that benefit in that situation.

Another thing to consider is that, when the wind comes off of the roof peak, it likely is not travelling horizontally. In that sense, there is "shade" on the leeward side.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dhengr]

ASCE7...., What an inferior code or std. You mean they don’t make you account for every 6" wide, 18" long and 4" deep driftlet behind every vent stack. There once were plenty of roof failures because engineers were completely ignoring (at least not paying much attention to) the fact that snow can accumulate in non-uniform ways caused by roof elevation, shape and slope changes, etc. The fact that we are now required to spend pages of calcs. and hours of computer time to figure drift length, width and depth to the inch is a great waste of time and gives false sense of confidence that we are doing a better job. Just because we can do something with enough time and computer power, doesn’t always make it much better. The real issue is, just don’t forget that it happens, and use a little engineering judgement in applying loads adequately. Remember, our structures are supposed to be designed to carry all the loads they may be subjected to during their lifetime. But, arguing about a few inches of drift depth is usually not a failure mechanism.

 

[KootK]

@dhengr: were I to read you last post in isolation, I would interpret it to mean "just design for the unbalanced load and move on", which is my stance.

Unbalanced snow is really more of a macroscopic phenomenon significantly affecting the need for extra capacity over the entire roof. I wouldn't put it in the same category as snow drift behind a whirlybird.

I've heard that Norway has a pretty good -- and liberal -- snow drift code.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dcarr82775]

KootK,

I just really wanted to debate something other than strut and tie with you, and sadly this is really just a gut feel debate. The snow is dropped off in the area of turbulent flow. Certainly there is a little area of that right at the ridge, but the wind would re-attach itself much sooner to the remaining flat roof thus greatly reducing the amount of snow that could be deposited. I have seen unbalanced snow that follows no pattern in ASCE7 (and it changes often). I just can't see a mound of snow forming as you drew it.

 

[KootK]

I've been on a self imposed strut and tie moratorium since last fall. Once the playoff's are over, however, I'll be back with a vengeance!

Back in my metal plate connectied wood truss days, I did a bunch of ski lodges up at Whistler-Blackcomb, the local ski hill near Vancouver BC. I've actually seen over 6' of valley drift in real life. Terrifying.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.