DIY Home Improvement, Remodeling & Repair Forum

DIY Home Improvement, Remodeling & Repair Forum (
-   Roofing and Siding (
-   -   Flat roof repair (

miamicanes 06-13-2009 11:03 PM

Flat roof repair
6 Attachment(s)
I have a conventional built-up flat roof on a concrete deck with dozens of blisters, nearly every single seam in bad condition, and water infiltration at at least one (probably more) penetrations including the bathroom exhaust fan duct shown in the last two pics. As badly as I'd love to have it torn off and completely replaced, I can't afford an $8,000 repair right now, period, and have no choice but to fix it up as best as I can so I can defer the full replacement for another 4-5 years.

I've gotten the impression that there's at least one repair process that's DIY-friendly, but will leave the roof in a state where it basically can't be repaired further, and nothing short of a full tear-out will be possible going forward (kind of like blowing emergency repair foam into a flat tire... it'll get you back on the road immediately, but a tire temporarily fixed with it has to be replaced because the moment you blow it into that tire, it becomes permanently irreparable). I *think* it involved a product with the word "cement" in its name, but I'm not really sure.

Keeping that in mind, is there any product available at Home Depot or Lowes that can be used to seal blisters that have been cut open and dried, and re-seal seams that are coming apart, that

a) won't render the roof irreparable going forward and mandate a complete tearout the next time the roof leaks

b) at worst, might need to be re-applied if it rains too soon after use (as opposed to, "If it rains too soon, you'll end up in worse shape than if you'd done nothing at all").

'B' in particular is vital... at this time of the year in Florida, there basically IS NO day when it truly doesn't rain at all, and you could confidently know that it wouldn't rain at all earlier in the day. A "dry" day is one with scattered light showers. A "rainy" day is one where there's a mid-afternoon mini-hurricane with 3+ inches of wind-driven torrential rain and the roads all flood...

Some roof pics taken this afternoon (1=facing west, 2=facing east, south side of roof, 3=facing east, 4=facing west, north side of roof; 5=master bathroom).

Nestor_Kelebay 06-14-2009 11:59 AM


I've gotten the impression that there's at least one repair process that's DIY-friendly, but will leave the roof in a state where it basically can't be repaired further, and nothing short of a full tear-out will be possible going forward (kind of like blowing emergency repair foam into a flat tire... it'll get you back on the road immediately, but a tire temporarily fixed with it has to be replaced because the moment you blow it into that tire, it becomes permanently irreparable). I *think* it involved a product with the word "cement" in its name, but I'm not really sure.

Whoever told you that was smoking whacky tabaky.

You don't have a "built up" roof, which is "built up" by mopping hot asphalt onto the roof and rolling successive plies of felt paper into that hot asphalt. I think you have something called "roll roofing". However, the following repair procedure should work equally well:

The standard repair procedure for flat built-up roofs is as follows, and it doesn't "ruin" your roof:

1. scrape away the pea gravel to expose the roof membrane so that you can find the tear, crack or opening where the water is getting in. Sweep away any dust or whatever to ensure you get a good bond between the existing roof membrane and the plastic cement you'll be applying.

2. apply "plastic cement" (which you can buy at any hardware store for roofing repairs) to fill the tear, crack, or opening. Also apply it several inches around the tear crack or opening. Normally you use a "pointing trowel" to do this, but any tool that allows you to spread a coat of plastic cement without getting it all over your hands will work.

3. apply fiberglass roofing mesh over the wet roofing cement. If you can't find the stuff made specifically for roofing, then use twice as much fiberglass mesh for drywall joints. The roofing stuff is much heavier grade of fiberglass mesh.

4. apply another coat of plastic cement over the fiberglass mesh

5. repeat steps 3 and 4.

6. broadcast the pea gravel back onto the wet roofing cement to protect the plastic cement from the UV light from the Sun.

The design of the repair is very simple to understand; the plastic cement forms a water proof seal over the tear or opening in the existing roof membrane and the fiberglass mesh embedded in that plastic cement carries any tensile stress (tension) that may be in the roof membrane on cold days so that there's no tension on the patched opening so that it doesn't open again. Anything you can do to improve that design (such as using more layers of fiberglass mesh across the tear or spreading the plastic cement for a larger distance around the tear improves the repair. So you want to clean the existing roof membrane well around the tear to ensure a good bond between the old membrane and the plastic cement, you want to spread that plastic cement for a good 6 inches around the tear and you want to use at least two layers of fiberglass mesh embedded in the plastic cement.

And, if it wuz me, I would buy some pea gravel at a garden center and cover your roof repairs with that stuff to protect the plastic cement from the UV light from the Sun.

miamicanes 06-14-2009 02:18 PM


You don't have a "built up" roof, which is "built up" by mopping hot asphalt onto the roof and rolling successive plies of felt paper into that hot asphalt. I think you have something called "roll roofing".
Hmmm. This is the first I've heard of it, but you might be right. I thought anything that involved a flat roof & asphalt-based products was a "Built Up Roof" (as opposed to SPF -- sprayed polyurethane foam, or single-layer PVC or rubber membranes). It makes sense, though... the neighbors' roofs look EXACTLY like a tile roof does before the tile goes on. It's probably just a "Dade County" thing, but down here roof tiles are regarded as little more than missiles ready to be launched by the next hurricane into your neighbor's house. Ever since Andrew, tiled roofs have to be designed so that they can be completely denuded of every last tile without losing their water resistance (based on the assumption that in a real hurricane, the roof WILL be stripped bare of its tiles before the eye even arrives at halftime.

One thing that's been bothering me ever since yesterday... is there any legal/engineering/technical reason why penetrations through concrete roof decks aren't themselves surrounded by a hollowed-out cone of low-slump concrete (kind of like a pipe emerging from the mouth of a volcano)?

To me, it seems like the best possible way to protect penetrations from water that works its way under the top layers of roofing material would be to make the top layers almost unnecessary. Form a steeply-sloping conical hilll around the entire penetration with a trowel and low-slump Xypex-enhanced concrete, and everything on top to provide further protection would practically be moot. Water flowing horizontally along the slab's surface would run into a steeply-sloping rounded surface (the cone's base) that's chemically monolithic with the slab itself (the surrounding slab would be Xypex-treated as well) and take the path of least resistance... around it. Just about the only way water could get in is if the hood were blown off, and falling rain entered from directly above.

Ditto, for the edges where the slab meets the parapet walls... trowel the same low-slump Xypex-treated concrete along the perimeter to form a sloping transition from the horizontal slab to vertical parapet walls, treat the adjacent surfaces with more Xypex, and even the complete failure of any additional roofing membrane would be of little immediate concern, because the whole roof would basically be like a shotcreted swimming pool sitting over a parking garage -- filled with water, but acting like a big pool because any water not flowing towards a drain would have no physical path into the home's interior.

The one possible problem I can think of is expansion joints. Would they be necessary at all on a 44x18 suspended slab roof deck, or is that small enough that merely isolating them from the roof slabs of neighboring units would be good enough to avoid failure due to cracks (particularly if the slab were protected by insulation and some kind of cladding, even if that cladding weren't strictly needed for water-resistance alone)?

miamicanes 06-16-2009 10:59 AM

Well, the first roofer was just here. I haven't gotten a price quote yet, but here's what I found out:

* My roof is absolutely, positively a 100% conventional oldschool hot-mopped built up roof.

* The roofer didn't want to do a complete tear-off... partly because there's no logistically good way to actually DO it. They can't park a truck alongside a corner unit, because I'd have to get permission from 3 neighbors to beat up THEIR roofs getting my own hauled away. I wouldn't want them to do it to me, and I assume they'll feel the exact same way.

- Removal via the rear is out of the question completely. The yard is walled in and covered with a screened enclosure.

- Removal via the front would entail disassembly of the pergola (not out of the question, but would likely kill an entire day if I did it myself doing it carefully so that I could put it back together again afterwards). It would also shred and destroy all the work I did back in January making the front look nice. Needless to say, I'm non-enthused about this option.

So... that basically leaves having the roofer do what I might otherwise do myself... cut, dry, and cement down the blisters and loose upper layer, prime it, cement down a layer of fiber reinforcement, then coat it with 3-5 layers of liquid elastomer.

If push came to shove, and I ended up doing the job myself, what order should I do the repairs in if there are simply too many to complete the whole job from start to finish in a single day, and it's likely to rain in between my repair days?

To help break it down, here are what I see as the main sub-areas:

a) The parapet wall tops and sides

b) The flashing between the parapet walls and roof deck

c) the front third (where ponding is the worst, there's no drain, and no positive drainage towards the rear -- where the actual gutter is)

d) the mid third (so many blisters, it's almost one big blister, plus two pairs of bath vents + stack vents (one on each side), and the air conditioner.

e) the rear third (puffy blisters along the perimeter, but all things considered, it's in visibly better shape than the mid and front thirds)

So, within those sub-areas a-e (and maybe some sub-sub areas within 'd'), what order would I:

1. slice, dry, and cement the blisters closed

2. prime

3. reinforce with fiber+elastomer

4. apply the first layer or two of heavy-duty elastomer

5. apply the finishing layer or two of white, surface elastomer


Put another way, would you do the steps roof-wide in order (fix all the blisters on the entire roof, then prime the entire roof, then do the fiber, etc), or would you do things a little bit out of order, like:

1. Slice, dry, and cement all the blisters along the roof's perimeter (say, within a foot of the edge)

2) Prime the parapet walls and the outer foot of the roof deck

3) apply one or more of the elastomer coats to the parapet walls ONLY


4) repeat steps 1-3 with the middle third of the roof

5) ditto, for the front third

6) ditto, for the rear third

My rationale is basically that doing the parapet walls first will get that whole sub-project out of the way, and won't substantially harm the unfixed areas of the roof in the meantime. I'd do the middle third next, because that's where I've observed outright water infiltration. This part alone will probably take me a few days to finish. Applying the elastomer to the middle first will probably worsen the ponding at the front third of the roof... but really, I don't see how it can get much worse than it already IS. Plus, I'll have a blower, so I can go up once or twice a day and blow the ponded water off. The rationale for finishing with the rear third is simply the fact that it's in the best condition of the entire roof (a few puffy blisters, but no ponding, and nowhere near as bad as the other 2/3 of the roof).

Also, keep in mind that if I do it myself, besides the rain, the amount of work I can do in any single day will be limited by the heat. Just going up on the roof this morning for 15 minutes with the roofer left me so hot and sweaty, it took nearly an hour to fully recover and feel normal again... and I dried off multiple times, changed clothes, and changed my shirt AGAIN 25 minutes later because I hadn't stopped perspiring... and I'm in fairly good shape, believe it or not. 90+ degrees at 100% humidity is brutal...


Nestor_Kelebay 06-16-2009 10:15 PM


I guess we're dealing with a difference in terminology in what is meant by "built up". It is clear to me that your roof is not made from felt paper glued together with hot asphalt because such a roof does need protection from the UV light from the Sun.

In roof shingles, the granules on the shingles serve that purpose by shading the shingle, and the amount of granules from the shingles that have collected in the evestrough is one thing to consider when assessing the condition of a shingle roof since you don't always see areas on the roof where the shingles are bare of granules.

Flat "tar and gravel" roofs, which are really asphalt and pea gravel use the pea gravel for the same purpose; to shade the roof from UV light.

On your roof, I don't see any pea gravel. "Roll roofing" is basically a large wide shingle that you roll onto the roof, but I'm not very familiar with it, or torch down roofing or any of the other single membrane roofs that are used on flat roofs. Since I don't know very much about these other kinds of flat roofs, I'm not in a position to give any sort of "expert" advice on your roof. But, in the following paragraphs I'll tell you how you can get reliable advice from a knowledgeable contractor and get the work done at a competitive rate.

I can tell you that blisters on a flat roof are an indication that the roof is nearing the end of it's life. I'm not sure it's a good idea to repair the existing roof. I think that might just be throwing good money after bad. If it were me, I'd probably try to replace the roof entirely so that the money you spend now won't commit you to continue repairing the roof so as not to lose the benefit of the money already spent. You might ask about RE-DOING the roof one third at a time, tho. That way, you're not committing yourself to an ongoing repair program.

Here's a pretty reliable way to find competent contractors that will give you both reliable advice and a reasonable price on whatever you decide to do:
Simply save your yellow pages phone book from year to year.
Anyone who's working for a roofing contractor now isn't going to start his own company until he's learned everything and made all his mistakes on his boss's customer's roofs. Only when he can see for himself that he could run a roofing company himself does he make the jump by quitting his job and start his own company. And, of course, it's not lost on him that he needs to have his new company in the yellow pages phone directory in order to get business.
So, keep the old yellow pages each year when you get new phone directories on your doorstep. By simply comparing the roofing contractors listed between this year and last year with a helper and jotting down the names and phone numbers of the companies that appear for the first time in the new directory and disappear since the last directory, you can get a pretty clear picture of who the new contractors are, who's retired, gone out of business or moved, and who's simply changed their business name for whatever reason.
Anyone starting a new company is going to play it safe. That is, they're simply going to do the best job they can for each customer they have at a competitive price. That way, if the business doesn't fly, they can't blame themselves for having charged too much, or done too little or gotten bad word-of-mouth advertising cuz of 1 and 2, or given bad advice to convince the customer to spend more or anything else within their control. They'd do exactly the same thing as you or I would; do an honest job for a competitive rate, and hope that's enough to make the business successful.

I would see if you can photocopy the Flat Roofing contractors in last year's phone directory at your local library. Certainly the new contractors who've just started into business will at least give you honest advice about what condition your roof is in and whether it's really wisest to repair it or replace it.

sorry I couldn't have been of more help in this case.

miamicanes 06-20-2009 08:19 AM

I talked to someone about the roof at a party last night. He said you're right, but not with respect to South Florida. He wasn't sure whether it was merely a norm, or an outright HVHZ (high-velocity hurricane zone) building code requirement, but down here built up roofs are usually(always?) topped with roll roofing instead of gravel. Basically, because in a hurricane, gravel on a roof turns into buckshot and causes lots of window damage to neighboring buildings. Apparently, the massive damage caused by Hurricane Wilma to some skyscrapers in downtown Miami (where Wilma's strength was nowhere near what it was in downtown Fort Lauderdale) was due to a single, relatively small building with lots of loose gravel on its roof (the window damage was almost all concentrated on a single wall of the affected buildings that faced towards the building in question). I feel sorry for that building's owner, because the businesses that had the worst damage were some of Miami's most elite & powerful law firms... he knew that at the very least, the topic of legal liability was going to come up a few hundred times at lunch over the next few weeks ;-)

So, as far as gluing down the loose flashing goes... any thoughts about Gardner Leak Stopper rubberized roof patch? Good, bad, or otherwise? Likewise, for gluing down the sliced & dried-out blisters, any thoughts on Wet-R-Dry roof cement? It's cheaper than their "Premium Roof Cement", but for some reason most of their OTHER products refer to Wet-R-Dry rather than its more expensive brother. The price difference is almost insignificant in light of the labor it's going to take to fix my roof and the importance of it not failing, so I'm trying to figure out which one might be better for this application (keeping in mind the fact that it will have at most a few hours to dry/cure before the next afternoon rainstorm).

Nestor_Kelebay 06-20-2009 12:30 PM

Hurricane force winds turning the pea gravel on a flat roof into buckshot is something I hadn't considered, and it does make sense. Millions of stones going 150 mph would be dangerous to anything in their path.

Here in Winnipeg, they also sell two kinds of roofing cements; standard and premium. The only difference is that the premium sticks to wet surfaces so that you can repair a leaking roof while it's still leaking. I think what you're dealing with there with the Wet & Dry and Premium is the manufacturer wanting to keep his market share of the contractor business (where the price is the primary consideration) but maximize his income from DIY'er business, where price is of lesser concern. That is, I expect the difference between the Wet&Dry and Premium plastic cements is much smaller than the price difference between them. It's just a manufacturer trying to make more money where he can; from the DIY'ers, who, almost to a man, will pay extra for the premium, presuming there is an important difference between them.

I wouldn't be concerned about a subsequent rain. Plastic cement is water proof from the minute it comes out of the cartridge or can. Once you get a repair down, then you can pour water on it. You'd have to use a fire hose on the repair to blast away the plastic cement and mesh to cause the repair to leak. A rain won't damage the repair.

I would talk to some people down there about fixing the blisters yourself. Just cut them open with a knife of some sort, smear some plastic cement under the roofing there to get the membrane to stick down to whatever is under it, then put down three more layers of plastic cement with a layer of roofing fiberglass mesh between each layer. And, based on what you're saying about hurricanes, I'd spread some plastic cement the size of a 15 year shingle over the repair and press a 15 year shingle down into that plastic cement to protect the underlying plastic cement from the UV light from the Sun.

(I say 15 year shingle because it's the granules on the shingles that protect the underlying asphalt from the Sun, and a 15 year shingle has as many and as large granules as a 30 year shingle, so from a UV perspective, they'd offer the same protection. 30 year shingles are made from thicker asphalt impregnated paper or from two layers of asphalt impregnated paper, and you're not relying on the paper to protect against water penetration; the plastic cement does that. You're only after UV protection.)

terryowood 06-20-2009 02:01 PM

Another option.I work at a hospital with a flat roof and I ve seen the roofs of several bussineses in my community,all roofed with 80 mil vinyl. we had a roof leak at church,some one had scraped the peagravel to try and repair the leak.I covered the are with a 10' by 40' piece of vinyl.No more leaks.I would lay it over exsisting roof.One negitive,vinyl
can be punchered but it can also be repaired easly .White vinyl will also deflect the heat.Just an opimion.good luck

miamicanes 06-21-2009 01:25 PM

Well, yesterday I spent a couple of hours up on the roof trying to patch the flashing with the obvious hole and two blisters. I came to the conclusion that there are SO MANY blisters up there, it would literally be easier to rip the entire upper layer off and replace it. Basically, every 3' x 3' area has at least one notable blister, and many of the blisters run into each other.

Then, I realized that the lower layers (hot-mopped asphalt + felt) are in nearly perfect condition (there was lots of water between the lower layers and the upper layer, but the lower layers themselves seemed to be almost flawless). It appears that my current problem is due to the penetrations being flashed to the upper layer. Water is getting under the topmost layer, running between the top layer and felt+asphalt, flowing under the flashing, then down into the hole.

It occurred to me: perhaps I could tear off the upper layer, then replace it with reinforced elastomer *instead*. Put another way, my roof would be (from top to bottom):

Ames Maximum Stretch white elastomer coating (2 coats)
Ames Super Elasto-Barrier (4-5 coats, with roof fabric embedded in layer #3)
Ames Super Primer
(existing felt + asphalt layers)
(existing insulation)
(existing vapor barrier)
(existing concrete roof deck)

I emailed Ames' tech support to ask whether it's a viable option, but don't expect to hear back from them until Monday. Intuitively, it seems like it shouldn't be merely "ok" -- really, it seems like it would be a superior option. Why? Because the elastomer would be applied to a foundation that's firmly in place and isn't going anywhere. If I apply it to a new layer of roll roofing, it will basically be resting on a layer of roofing material whose own service life is 5-10 years, max. If the roll roofing comes up, the elastomer will come up along with it. As I understand it, the asphalt + felt layer is basically what you'd end up with if you took a "conventional" built-up roof and removed 95%+ of the gravel, anyway. The lower layers would lack the physical protection of a hard "shingle-like" surface, but I'm under the impression that roof fabric embedded in Super Elasto-Barrier would be just as good, anyway.

If I do that, I'll probably just do the removal, primer, and two coats of Super Elasto-Barrier now, and embed the roof fabric and topcoat it this winter when it's not as brutally hot outside (and we can predictably have more than 12 rain-free hours in a row).

On the other hand, if they say that the roll roofing layer is absolutely essential, I'll probably just tear off the roll roofing, replace it with new roll roofing + cold-process roof cement from Home Depot or Lowes, and maybe put a coat of Super Elasto-Barrier on it for good measure. In any case, I don't think I'm going to even bother trying to cut, dry, and patch the individual blisters. There are just too many of them, with too much water trapped between the roll layer and felt+asphalt layers

The single hardest part I see is removing the old roof. I just might have to pay someone to do that, because I don't know whether I have the stamina to tear off the roll roofing from a 20x44 foot area, fix the flashing, prime it, and apply at least one layer of Super Elasto-Barrier in a single day... in 96-degree 99% humidity, no less. I know the big roofing companies won't touch a job that only involves tear-off, but maybe some of the 2-3 person small roofing companies might.

Nestor_Kelebay 06-21-2009 09:28 PM


If it wuz me, I would compare this year's yellow pages listing of "Commercial Roofing Contractors" to those listed in last year's phone book and ask those new companies about just replacing the deteriorated surface layer of your roof.

I'd be reluctant about puting down any elastomeric coating on your roof. Elastomeric coatings are more meant for masonary walls with active cracks that open and close with building movement. They're water proof, but I think replacing the surface of the roof you have with the same thing, only new, would be the better option.

I wouldn't try doing this work yourself. 90% of the problems that occur around a house can be done by a DIY'er with the resolve to at least try solving the problem before giving up. Re-doing a flat roof is one of the other 10%.

██████████████████████████████████████████████████ █
In solidarity with the freedom seeking people on the streets of Iran.

All times are GMT -6. The time now is 10:50 PM.