Flat roof deflection causing ponding 1

[demayeng]

Hi,

I have to do an analysis on a flat roof that is deflecting and has a pond at the lower end full of water rusting through the roof sheet.

The purlins are 200x47 oregon spanning up to 7.5m (so they are undersized already). The roof sheet is deep profile (Kliplok here in Australia) with 41mm ribs. The manufacturer recommends 1 degree min, preferable 2 degrees min slope. I'm not sure what the slope of the roof actually is, but looks very flat.

I've been told that a build-up of hail has caused the ponding in the first place. I haven't been to the site so can't tell you if the gutters and downpipes are undersized and a blockage has caused a build-up of hail.

My question is: if there is now a permanent pond in the roof holding water, my assumption is that it is caused by general deflection of the undersized rafters under the amount of water that it is currently holding and not by a specific hail event.. is that correct or do you think an overloading case for max a couple of hours may have induced a permanent deflection greater than it already had?
Maybe people who live in snow prone areas can tell me more about this..

I guess I'll need to work out the depth of water that would build up during a storm event over the length of roof also..

Thanks

 

[rowingengineer]

I would say the rafter and purlins are undersized for dead load deflection and the roof has a flat zone, the hail wouldn’t have helped. I would do a model and check you deflection ratio’s I think for 1 deg you should be aiming for L/500 or something like that.

An expert is a man who has made all the mistakes which can be made in a very narrow field

 

[hokie66]

Lysaght recommends 2 degrees (1/30) as the minimum pitch for the lighter .42mm roofing, and 1 degree for the heavier grades. I think those recommendations are based on the use of steel purlins. With timber framing, the control of the roof plane is much less certain, and I think even 2 degrees is too flat. There could have been differential deflections in the timber, some from ponding and some built in.

What is the purlin spacing? Hail buildup can certainly cause yielding of steel roofing, so permanent deflections between the purlins. Hail is a load case not dealt with well by the Australian loading standard, but after every major hail event, there are reports of leaking roofs or roof collapse of flat roofs and/or roofs with box gutters.

 

[paddingtongreen]

Just on the basis of your post, the roof could have overloaded by the hail, which might then have melted, forming a pond of the same load. If the pond hasn't dried up, that pond has become a long term load.

I've seen something like this on a steel framed roof, a heavy tree branch fell on the roof during a storm, causing enough deflection for a pond to form. The pond was still there when the tree was removed until the water evaporated, allowing the roof to go back to normal. The members were still in the elastic range.

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

 

[MiketheEngineer]

NEVER - EVER build a flat roof. That said:

It may have permantently deflected or it may have been built wrong. The problem is how to fix it. More info would be required to even give a wild guess.

Good luck.

 

[ToadJones]

Mike-
Flat roofs work very well in places like The Gobi, The Sahara, The Mojavi....I kid, of course.

 

[jike]

Getting a permenent sag out of a flat roof is next to impossible without rebuilding it.

Sounds like a good location for a roof drain!

 

[msquared48]

"Sounds like a good location for a roof drain!" Good comment Jike.

There is your best solution, temporarily. Still should look at the rest of the roof structure though and possibly beef up the framing.

If there is any capacity in the structure, you could also consider installing tapered foam insulation over the structure with a 1/4" per foot minimum slope, and roofing over. Cost may be prohibitive though.

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

 

[dhengr]

Of course, ponding is a vicious cycle; initial load causes deflections, which begets larger loading over a greater area, which causes more deflection, etc. etc. approaching structural collapse. And, Mike has it right, while fixing the roof may be cost prohibitive, collapse can be catastrophic and pretty expensive too. There is ample discussion about this problem in codes and structural design books, it should not have been neglected, and I would get the original designers and builders involved in fixing this problem, if for no other reason to save their own butts. Again, I think this is paraphrasing Mike, from another thread; they can pay now to make it right, or pay later, big time.

You can’t figure this problem out sitting at your desk or chatting on a forum, you must get out to the building to determine what’s actually going on, and the owner should have no qualms about paying for this investigation work. Stretch a string (laser) on some of the various roof framing members to assess their current deflections, over the areas of concern. What is the actual roof slope, if any? This should be a meaningful average of conditions for your own edification. Then do an analysis of the various framing members to determine how much of the total deflection each contributes. Maybe a purlin needs to be added every other purlin space or some such. Stiffening which system gives you the most bang for the buck, and brings the entire roof back into code complience? You say the roof sheathing is already being damaged by this condition, and that would seem to indicate a new roof, which would all be part of your argument for doing it right this time. Look more closely at the original design loads (both DL & LL) maybe you can find a few lbs./sq.ft. to help compensate for the new sloped insulation.

 

[demayeng]

Thanks for the replies.

Here's a photo attached for your viewing pleasure

I think a pond on a roof is unlikely to not dry out long enough for the roof to rust, so this indicates that there is permanent dead load deflection causing the pond to refill every time it rains.

The roof sheet is rusted, so no problems asking for that to be replaced with an appropriate profile for the slope. . The purlins (at 1m centres) are undersized for dead load deflection regardless, so they are going to be replaced or supplemented with new ones. I will recommend they go to 2 degrees minimum, regardless of the manufacturer's recommendation, as part of the replacement process.

Noone is in doubt that the roof needs replacing. It's a cheap roof on a shed with no internal lining, just roof sheet and purlins. I'm doing the calcs remotely, so can't visit the site

I believe calcs showing that the purlins are undersized for dead load and an explanation of the dangers of flat roofs will be sufficient for the owners of this roof.

I like rowingengineers suggestion of upping the deflection criteria to deal with the flatness aspect.

 

[apsix]

I'd be careful reusing the existing purlins.
If they are only supporting the cladding, even halving the spacing won't reduce the dead load significantly.

At least turn them around if reusing so that the camber is up.

 

[demayeng]

Yes I agree. They'll only stay if it's easier to leave them in place than take them down.