Failed GRP Roof - Thoughts on replacement

[Diablo]

All,

 

We had flat roof installed about 4 years ago.  The handover from builder to roofer wasn't as seamless as I would have hoped, as the builder put the OSB3 deck down and did not cover it.  That night, the heavens opened up and by morning, the deck was drenched.  The roofer did explain the risk of continuing, but we took our chances and tried to torch the boards dry, before applying the GRP.  Long story short, the GRP did not fully bond to the deck and we have had several leaks over the years. 

 

We are looking to replace the roof with a 3-layer torch on felt.  The roofer is insisting that it would be fine to board over the top of the GRP and then lay the felt on the new boards, on the basis that the GRP will prevent any moisture in the old boards permeating into the new boards. He is prepared to guarantee it. Does that sound reasonable?  

 

Many thanks,

Rog

[epsilonGreedy]

I reckon the original roof failed due to swelling of the OSB3 and present leaks suggest this process continues.

 

If the current roof is prone to leaks why is the builder confident it won't release its existing captured moisture upwards into the new OSB overlay?

[Jeremy Harris]

Moisture seriously degrading the cure of polyester resin is a very well-known problem, and is why so much emphasis is placed on making sure everything is bone-dry when laying up a GRP roof.  Once cured, GRP is very weather resistant, and a properly done roof will last many decades.

 

Replacing the roof with a properly installed GRP one would be my preferred choice, as it will last a lot longer than a felt roof.  However, as you've found, laying up GRP outdoors does mean you're at the mercy of the weather, and any dampness that gets into the resin before it's cured will almost certainly result in curing defects and subsequent leaks.   At least with a torched on felt roof there's not too much of a problem with weather during the installation.  Such a roof will usually come with a 10 year guarantee, too.

[epsilonGreedy]

1 minute ago, JSHarris said:

Moisture seriously degrading the cure of polyester resin is a very well-known problem, and is why so much emphasis is placed on making sure everything is bone-dry when laying up a GRP roof.

 

 

The problem you refer to is osmosis in polyester resins and takes 10+ year to appear. It occurs even when the original grp layup was done correctly in dry conditions.

 

I suspect the OP suffered a mechanical failure and not a chemical one.

[Jeremy Harris]

1 minute ago, epsilonGreedy said:

 

The problem you refer to is osmosis in polyester resins and takes 10+ year to appear. It occurs even when the original grp layup was done correctly in dry conditions.

 

I suspect the OP suffered a mechanical failure and not a chemical one.

 

 

No, nothing at all to do with osmosis, that cannot affect a roof structure as there will be no continuous osmotic pressure - a roof does not sit immersed in water like a boat.  Polyester resin is simply very susceptible to moisture when curing.  It's a very well-known issue, and one that every laminator in the land will be intimately familiar with.

[epsilonGreedy]

6 minutes ago, JSHarris said:

No, nothing at all to do with osmosis, that cannot affect a roof structure as there will be no continuous osmotic pressure - a roof does not sit immersed in water like a boat.

 

 

A boat can suffer grp osmosis even when hauled out for 4 months of the year but yes must less of a problem on a roof which is why I doubted your theory.

 

8 minutes ago, JSHarris said:

Polyester resin is simply very susceptible to moisture when curing.  It's a very well-known issue, and one that every laminator in the land will be intimately familiar with.

 

 

Since the vast majority of grp fabrication takes place in a factory environment I doubt 1 in 10 grp laminators thinks about moisture at all. The vast majority of discussions on grp & moisture relate to osmotic degradation. 

[PeterW]

Just rip it all up and start again. Get the OSB off the roof, get it replaced with new OSB3 and start it off properly. The cost of a few sheets of material vs a new roof is not worth the hassle. 

[jack]

1 minute ago, epsilonGreedy said:

Since the vast majority of grp fabrication takes place in a factory environment I doubt 1 in 10 grp laminators thinks about moisture at all. The vast majority of discussions on grp & moisture relate to osmotic degradation. 

 

But we're talking about roofing here, surely? 

 

I GRPed our balcony floors and it was repeatedly hounded into me on every Youtube video I watched and article I read that everything needs to be kept bone dry to avoid curing problems. Nothing was mentioned about osmosis.

[Jeremy Harris]

5 minutes ago, epsilonGreedy said:

 

A boat can suffer grp osmosis even when hauled out for 4 months of the year but yes must less of a problem on a roof which is why I doubted your theory.

 

 

Since the vast majority of grp fabrication takes place in a factory environment I doubt 1 in 10 grp laminators thinks about moisture at all. The vast majority of discussions on grp & moisture relate to osmotic degradation. 

 

 

I can assure you that ALL laminators are acutely aware of the moisture problem - even condensation on a mould can inhibit a cure (ask @SteamyTea, he's done as much, or more, laminating than I have).  I once spent the night in an industrial unit, following the layup of the hull of a 36ft yacht, just monitoring the humidity and temperatures and making sure the resin surface didn't drop below dew point.  A helluva balancing act, as the heating was a large industrial diesel fired heater, that produced water vapour when running.

[SteamyTea]

Just a few drops of water is enough to stop the curing process.

What I suspect has happen is, as mentioned, there is a poor bond between the OSB and the initial layer, or layers of GRP.

What would concern me, and would need investigation, is where is the water getting in from.

Without knowing that, any replacement may just not work.

I developed an osmosis proof GRP system about 30 years ago. Some of the parts are still doing sterling service.

They system is for sale, but not cheap. £100k should cover it.

[epsilonGreedy]

8 minutes ago, JSHarris said:

I can assure you that ALL laminators are acutely aware of the moisture problem - even condensation on a mould can inhibit a cure

 

 

The curing of a resin is an exothermic process, laminators spend more time worrying about an excess of heat cooking the uncured laminate.

[Big Jimbo]

Ì built a boat from fibreglass about 40 years ago, and also did a large flat roof with fibreglass about 5 years ago for a friend. I hated the both doing, and the look of the fibreglass flat roof. Tons of edge pieces. Loads of fibreglass strips over the joints of the OSB3, before i could even start laying up the roof. I know other people love them, but i would not have a great deal of faith on a fibreglass roof. However, i'm not that keen on torch of felt either. I've done about 5 flat roofs over the last 5 years with EPDM, and just feel that it is a better system.

I would however, want all of the old OSB3 off and replaced, before i put on any new roof, regardless of what system i was going to use. Anything now just put over the top, comes accross as a bodge to me. If you are going to sell in the next few years, fine. If not sort it properly. You don't want to have to re-visit it again in a few years. No offence, but i doubt any builders G T will be worth the paper it's written on.

[jack]

1 minute ago, epsilonGreedy said:

The curing of a resin is an exothermic process, laminators spend more time worrying about an excess of heat cooking the uncured laminate.

 

Nice attempt at moving the goalposts.

[Jeremy Harris]

1 minute ago, epsilonGreedy said:

 

The curing of a resin is an exothermic process, laminators spend more time worrying about an excess of heat cooking the uncured laminate.

 

Yes, it is, but that is very section dependent (the elephant versus mouse problem), so thin, large, sections get cold, as their surface area is large compared to their volume.  Not sure how much laminating you've done, at a guess I'd say very little, as moisture cure inhibition, especially of MEKP catalysed polyester, is a very well understood issue.

[epsilonGreedy]

12 minutes ago, SteamyTea said:

I developed an osmosis proof GRP system about 30 years ago. Some of the parts are still doing sterling service.

They system is for sale, but not cheap. £100k should cover it.

 

 

The boating world now uses vinylester or epoxy resins to solve the osmosis problem.

 

The apparent randomness of the occurrence of osmosis in boat hulls perplexed the industry for decades, how did you solve the osmotic mystery all those years ago?

[epsilonGreedy]

7 minutes ago, jack said:

Nice attempt at moving the goalposts.

 

 

@JSHarris comprehended my point.

[SteamyTea]

1 minute ago, epsilonGreedy said:

 

The apparent randomness of the occurrence of osmosis in boat hulls perplexed the industry for decades, how did you solve the osmotic mystery all those years ago?

Was not random at all. I just happened to be in the industry as some new system came into place. This gave me the opportunity to compare them. Soon nailed the problem. It helped that I was dealing with high temperatures and humidity levels of 100%. So could quickly 'rapid age' materials.

If you want to know the solution I found, then get your cheque book out.

[Jeremy Harris]

Just now, epsilonGreedy said:

 

@JSHarris comprehended my point.

 

I did, but the concern laminators have over the exothermic reaction is really only related to pot life.  Once resin is rollered out into a laminate the heat from the curing reaction has little or no effect, curing becomes dominated by environmental factors, primarily the ambient temperature and humidity. 

[epsilonGreedy]

1 minute ago, SteamyTea said:

Was not random at all. I just happened to be in the industry as some new system came into place. This gave me the opportunity to compare them. Soon nailed the problem. It helped that I was dealing with high temperatures and humidity levels of 100%. So could quickly 'rapid age' materials.

If you want to know the solution I found, then get your cheque book out.

 

 

Let's summarize, you solved a £ billion manufacturing problem 30 years ago and the secret is still for sale. Hmm.

[epsilonGreedy]

1 minute ago, JSHarris said:

I did, but the concern laminators have over the exothermic reaction is really only related to pot life.  Once resin is rollered out into a laminate the heat from the curing reaction has little or no effect, curing becomes dominated by environmental factors, primarily the ambient temperature and humidity.

 

 

100% wrong. I once owned a new yacht with a curious crack in the deck, after much head scratching the problem was opened up with an angle grinder at which point the experts went ah ha, the laminate had cooked on the moulding joint where resin was laid on thicker. I was then advised there is a constant juggling act in hull lamination between production targets that favours more layers of laminate per cure compared to quality control who want to avoid cooked laminate.

 

 

[Jeremy Harris]

17 minutes ago, epsilonGreedy said:

 

100% wrong. I once owned a new yacht with a curious crack in the deck, after much head scratching the problem was opened up with an angle grinder at which point the experts went ah ha, the laminate had cooked on the moulding joint where resin was laid on thicker. I was then advised there is a constant juggling act in hull lamination between production targets that favours more layers of laminate per cure compared to quality control who want to avoid cooked laminate.

 

 

 

 

Not 100% wrong at all.  I've spent years making composite structures, including several boats and various aircraft parts.  From what you've posted so far I'd guess that you've never laid up a composite structure in your life.

 

The only time I've ever seen laminate overheating during construction was when incompetent people were laying up laminates that were far too thick in one go, usually using a chopper gun (should be banned, IMHO, as they produce horrible, resin-rich, lay-ups). 

[nod]

When did our GRP

A good friend advised us He works in a boat yard and works with GrP everyday 

He did say if your boarding gets wet you are stuffed You need to re board He advised a perfectly dry day Start early as we are south facing 

We started at 5am and finished at 10am just as it was getting hit 

I was about to start the top coat and he advised to wait til dusk 

He pointed out that direct sunlight can almost be as bad as the rain

[epsilonGreedy]

7 minutes ago, JSHarris said:

Not 100% wrong at all.  I've spent years making composite structures, including several boats and various aircraft parts.  From what you've posted so far I'd guess that you've never laid up a composite structure in your life.

 

The only time I've ever seen laminate overheating during construction was when incompetent people were laying up laminates that were far to thick, usually using a chopper gun (should be banned, IMHO, as they produce horrible, resin-rich, lay-ups). 

 

 

British yacht construction suffered much worse than comparable manufactures on the continent. The mainstream UK yacht manufacturers Westerly and Nicholson (think Austen/Morris and Jaguar) had a terrible osmosis record, it became an industry joke and their business failed. There is no volume mid market British manufacturer of sailing yachts left

 

My new yacht was manufactured in Germany by a company that is still the No.2 volume producer today. Defo no chopped strand there, infact the whole chopped strand debate is a bit of vintage discussion, I think a few budget brand powerboat companies in the US were the last to abandon that years ago.

[Jeremy Harris]

15 minutes ago, epsilonGreedy said:

 

British yacht construction suffered much worse than comparable manufactures on the continent. The mainstream UK yacht manufacturers Westerly and Nicholson (think Austen/Morris and Jaguar) had a terrible osmosis record, it became an industry joke and their business failed. There is no volume mid market British manufacturer of sailing yachts left

 

My new yacht was manufactured in Germany by a company that is still the No.2 volume producer today. Defo no chopped strand there, infact the whole chopped strand debate is a bit of vintage discussion, I think a few budget brand powerboat companies in the US were the last to abandon that years ago.

 

 

There's nothing wrong with using chopped strand mat, especially when a layup is needed that doesn't have unidirectional or orthogonal strength requirements.  It's pretty nigh impossible to make good hand layup without including layers of chopped strand mat, as woven cloth layers do not bond well to each other, unless the laminate is cured under pressure, either autoclaved or vacuum bagged.  Woven cloth or rovings  lay-ups aren't always better, in fact I avoid using them unless the structural requirements mandate their use.  Non-woven triax is probably my favourite general purpose material, as it drapes and forms almost as easily as chopped strand, but uses less resin.  Even so,  that really needs a layer of chopped strand between layers, much like rovings does, when hand-laid up, to ensure that the layers bond together well.

 

The major problem with mass-produced boats years ago was that they were being made by anyone with a tiny bit of knowledge, a shed and a chopper gun.  Any idiot can use a chopper gun, and as a consequence lots of boats were laid up by people who knew next to nothing about laminating properly.  Chopper guns use rolls of rovings, and can make very resin-rich laminates, especially if the operators don't put down the gun and spend time rollering the laminate out properly.

[SteamyTea]

@epsilonGreedy.

There are many reasons that you get delamination, star cracking and other problems.

I could write a book about it, but the trouble is, the industry is small, and to put it bluntly, not professional.

I had great trouble employing people from within the industry, opting in the end, to train people up.

Now when you consider that some of my products are still on the front of Aston Martins, after nearly 40 years, with some other building related products still spoofing people into believing they are genuine stone, I feel qualified to comment.

The British boat industry does not have a good record of mass production, but companies like Princess and Sunseeker made good products (got thrown off the Sunseeker stand at Earls Court boat show for pointing out there bad model making. Had the same problem with Mitsubishi too, but they were Colt Cars in them days).

Oh happy days they were, I still itch.