Silicone vs. elastomeric roof coating and some other flat roof questions
I am insulating on top of an existing low-slope built up roof in climate zone 2A. The insulation will consist of 4 inches of recycled polyiso topped with 4-7 inches of EPS. (I’m getting some tapered EPS panels cut to add a little more slope to the roof). I’m cutting and cobbling around the supports for an existing solar panel installation, which means I’ll have numerous penetrations to seal.
I’m planning to use polyester fabric and a roof coating directly on top of the EPS. I also would prefer to rely on adhesives rather than try to screw down the foam. This portion of my house is an addition used as guest quarters and laundry, and it’s only 500 square feet, so I’m willing to be a little experimental – and I’d be interested to hear if anyone thinks any of these ideas are a poor bet. (I already know what Martin thinks about roofing around the solar panels supports, but I wasn’t able to come up with a reasonable alternative at the time of the solar install, and a least a few local folks have done something similar).
Finally, I am wondering if it’s worth paying 4 times as much for a silicon roof coating (looking at Koolseal Tundra, which has a “limited lifetime warranty”). The cheaper alternative is a higher-end siliconized elastomeric product, which has a 12 year guarantee. Anyone used a silicon coating? I’ve used elastomeric products before and they require maintenance/recoating more often than I’d like, but the more expensive ones do seem to last longer. Since I’m already committed to the EPS being the top layer, I have to stay away from any solvent-based products that would melt the EPS.
P.S. Is my recycled polyiso still green since I had to drive 300 miles roundtrip to get it? At least it’s on my roof and not in a landfill.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Nick,
I think you are painting yourself into a corner. A flat roof is one of the most potentially problematic building assemblies I can think of - and that's without adding pre-conditions which increase the risk.
On new construction done by experienced contractors, you don't see them working around poorly detailed penetrations, or relying on coatings as the waterproofing layer. With what you are proposing there is a good chance you roof will leak. If not now, a lot sooner than you would want.
Thank you Malcolm. The alternatives I am aware of are as follows
1) lay some rolls of peel and stick roofing and detail the best I can around the penetrations.
2) Put a top layer of polyiso with the kind of paper that allows for hot roofing. I did find a roofer that was willing to work around the panels, and someone who works for the solar installer did the same thing at his house.
Given the corner I'm already in, I'm curious what you would suggest.
Nick,
I would get a quote to remove and re-install the solar array. I would consider using one of the flat roof systems that are commonly used, which have good track records and that installers are familiar with. https://www.iko.com/comm/complete-guide-commercial-flat-roofing-systems-materials/
My own preference would be to0 use a two-ply modified bitumen membrane - either torch down or self-adhesive. I would consider changing the mounts on the solar array to ballasted ones that didn't require fasteners, and changing the location of any electrical connections to exterior walls, rather than the roof surface.
The consequences of flat roof leaks are considerable and hard to remediate. They often don't become visible until considerable damage has already occurred, and the fixes usually require removing more than the water-proof layer - something that is difficult to do in the sort of inclement weather that caused the leak in the first place.
If you have the structural support for this change, why not move to a ballasted system for the PV, eliminating or greatly reducing penetrations?
Peter
Originally I tried to talk to the installers into a ballasted system. They discouraged the idea and said I would need a civil engineer to evaluate whether the support was sufficient. Given that I have a 70 year old roof that is built with 2 x 6 joists/rafters spanning 12 feet and already has an inch or 2 of sag, I assumed that it would mean tearing down the roof to the walls, beefing up the rafters and starting over. I guess that's where I'll be if this roof leaks, so perhaps that would have been the best long term plan. At the time we all thought we had a very small window to get the system grandfathered in under Arizona's soon-to-be-extinct net metering, so I made the choice to just bolt the system to the rafters.
I think the best I can do with what I have now is use some high quality collars on the penetrations. That's what the installers did 2 years ago on the original install, and it's held up fine so far. If it works, great; if not, lesson learned. Light duty roofs with frequent maintenance (annual inspection and recoating every 3-4 years) is common practice for older homes around here.
Nick,
Fair enough. The answers you get here tend to reflect our feelings about situations in general, and often don't adequately reflect the individual circumstances of the posters. If that's what you need to do, that's what you should do. Sorry I don't know enough about the difference in coatings to usefully answer your original question.
A flat roof with an inch or two of sag sounds dicey to start with.
You mentioned polyiso, but AFAIK polyiso isn't useful anywhere that it stays wet - it's too open-celled to deal with water and retain any R value. That makes subgrade a problem, but also a poorly drained flat roof, I would expect.
The other solution is to install pitch pockets around all of the penetrations and flash them into a standard membrane roof. At that point, modified bitumen torch down, EPDM or PVC would all work fine. In my experience, relying on a liquid-applied coating to provide durable waterproofing for the long term is a high-risk approach.
As long as your solar panel structural supports are round pipes, just go with a TPO roof using the prefab round bellows for penetrations. Seal them up with the TPO mfg recommended sealant, then apply a metal pipe clamp, then more sealant. Reseal every 10-15 years. That is how commercial projects do it. Most commercial projects try to avoid any penetration that is not round.
https://www.arcat.com/cad/carlisle/075423_tpo-u---pre-molded-pipe-flashing.pdf
A rather belated thank you for all the suggestions. For now this section of roof is coated with 3 coats of elastomeric - it was all I could manage before the monsoons set in. When I insulate the main roof, I'll plan further in advance and budget for a longer term solution for both sections of roof. I like the idea of a manufactured membrane and standard, widely used components to seal around the supports (and yes, they are round).
This project has taught me a couple of valuable lessons. One is that the solutions I come up with are skewed by the limitations of what I think I can handle as a DIY project. The second is that I can handle more than I think. With 20-20 hindsight, I now think it would have been indeed been better (and maybe even cheaper) to tear off the old roof, putting in 2 x12 rafters and filling the bays with cellulose. Then I could have put a lot less foam on the top of the roof deck and used a ballasted system for the solar. But I couldn't see that at the time, so perhaps it took doing it my way (which was mainly cutting and glueing foam) to get to the confidence that I might be able to handle doing a job that required learning more roofing and carpentry skills.
And to Malcolm: I do appreciate that GBA is a source for best practices, and I took your advice in that spirit. It made me stop and think about how I got into this situation and the pattern of choices that led me there.
Nick,
If I sound discouraging in a response it's because I've been there, made more work for myself, and some bad mistakes. Working for other people, and being responsible for things over their lifetime, has made me much more risk adverse and conservative in what I'm happy doing.
I agree. What's nice about being involved in these types of projects is how empowering and satisfying acquiring the necessary skills is, and finding out what you are capable of doing.