Thread: Poor roof design

The lower roof section slopes toward the wall. The main roof above does not extend over this lower section! This is a problem, installing a gutter above this lower roof section will help, not quite sure what else to say about this one.

Similar Threads:

• Design of Cathedral Ceilings
• Bad design idea?
• Dumb design
• Poor Cat
• Poor Man's Solar Collector

They either need a cricket in that corner or a really really big flashing behind those walls, and siding which will not deteriorate when it gets wet with as much water as that is going to get.

What's with the louvers missing on that vent to the left in the photo?

I sure hope no one opens that door when it is raining ...

That's what I was thinking Jerry, a cricket or monster sized corner flashing. Yes I seen the missing louvers on the vent cap, called it out.

I would recommend they install a little cricket, new corner flashing and a scupper box/downspout too.

The upper wall appears to line up with the lower, so why not cut the lower sloped section away, run the siding all the way up and install a gutter on the upper roof. A new roof over the doorway to complete the job. Send that "architect" back to school.

That would be one way of doing it.

Speaking of "Poor Roof Design is Us (TM)"... how they are going to fix that first one, I have NO idea.

In my opinion it doesn't look bad at all. Probably a 12" overhang and won't get much water at all.

Originally Posted by Wayne Carlisle

In my opinion it doesn't look bad at all. Probably a 12" overhang and won't get much water at all.

Serious question: are you serious?

Any water that gets into those window assemblies is going right into the wall/or and under the roof. The undersides of the overhangs are drywall, and are getting soaked. The metal cladding is a screed system that the builder attempted to install as a barrier system in some locations - they keep adding caulk to "repair" the wall leaks.

Yes. Just in case.... I am refering to the original post.

How much water can there be on such a small overhang. It would take a monsoon to even be a possibility to produce enough water to give it a problem at the base where the overhang abuts the wall.

It looks like it is flashed with at least 4 inches of exposure on the flashing.

A question for Trent. Is there any evidence of the intrusion of water inside?

Michael,

The fix is easy ... as long as one calls time consuming work which is not difficult "easy" (versus tearing the roof down and rebuilding it as being "not easy").

The windows must be removed, the roof covering removed, a proper flashing installed along the wall and under the windows, then a proper counter flashing installed over that (the flashing would need to extend up into the window openings as a pan, with the counter flashing simply serving as a joint to allow for movement), then the roof covering replaced.

What is that building structure? Frame? concrete columns and decks?

The repair also depends on the structure and what it is made of.

Easy? Yes, meaning it is not rocket science.

Easy? No, meaning that it will be a pain in the neck, require skilled workers, and involve removing the windows, etc.

Well... perhaps I'm missing something here, and if so of course I want to be the first to know, but as far as I can tell there is no flashing under those windows: it appears that the roofing was run to a point about three quarters of an inch below the lower edge of the window frame, so all the metal surface in the third picture is the window frame, and there is no flashing.

My opinion was pretty similar to Jerry P's; there has to be a flashing system on that window which provides a designed exit path for any water which enters the window assembly, either from the structure above, at the glazing seals, or the seals at the movable lights.

Currently, there is not enough height between the roof membrane and window to do that properly, best case you might get away with some kind of "better" arrangement where the top of a short counter flashing ends up just below the glazing. More likely (seems to me) the "right" arrangement will likely involve dropping level of the roof deck until you have a reasonable hold back to allow for reliable water control when eight or 10 inches of snow banks up against those windows and starts melting.

Otherwise any water that enters any of those points is currently being directed behind and/or under the window frame and either back into the wall behind and below the window an/or soaking out into the decking below the roof membrane. I don't see how anything else could be happening there, but if somebody can explain to me how this arrangement could work I'm ready to be enlightened.

............................

As for existing problems, that "soffit" (between the first floor balconies above the ground floor commercial space and extending out over the sidewalk) is the reason I was there - its underside consists of drywall panels laid into trays at the front and back of the soffit and supported who-knows-how perpendicular to the soffit (no access for inspection there), this drywall had been wet and failing and had recently been repaired at the time of inspection. (The explanation given by the company performing the repairs was the long dryer ducts passing through the soffit had not been insulated and condensation and dripped down onto the soffit floor - the problem was that I found similar deterioration on a floor above where the soffit was definitely not being used as a chase).

............................

As for the interface between the roofing membrane and the vertical surfaces below (and I deserve combat pay for getting those pictures, by the way) I think that's just nuts. For starters it's dependent on sealant at a junction between two radically dissimilar materials subject to regular thermal movement, it would have to be installed absolutely flawlessly to work even at the start, and while perhaps roofing in the Heavenly City and installed by angels is done this way, down here below roofs are installed by mere mortals, often to about the standard evident at those pictures of the roof 's corner.

Looks to me like the only way to fix that is to take off a strip of roofing at the entire perimeter and detail the replacement properly, which "properly" may mean entirely rethinking the roof projection flashing methods.

Again, I'm ready to be told I'm wrong, but if so what's keeping that arrangement water tight over time as the roof membrane creeps back-and-forth?

...............................

Jerry,

This building is what I call "CAD Gone Bad": the architects design it with a variety of materials - in this case some kind of rain screen metal panel over CMU, interfacing with stucco interfacing with wood and all interfacing with exposed CMUs.

Then some builder has to figure out some way to make all those interfaces "watertight" (usually their first instinct) or think things through and provide for proper water control - because as we know water inevitably gets in there. (Apparently, not a line of thought instinctively attractive to most humans if there is a caulking gun within reach).

I attempted to identify the metal cladding material via the architect, who said that what he thought was specified was "Omega Light Cladding", but that "the builder may have substituted a similar material". No luck with Google for this material . Where I can get a sense of what's going on behind the cladding it's installed on metal standoffs over Tyvek over the CMUs (insulation back there too?), but at many interfaces with horizontal material below I don't see any evidence of flashing or any other detail directing water out from behind the wall.

I do know that some things were not done per plan where the plans were available, for example both the architectural specifications and drawings called for flashing between caps and parapets which was absent (a lot of builders around here do it that way because they believe the flashing will "break the bond"). There also a lot of just plain ignorant mistakes, for example at some parapets wicks have been installed below the caps, but somebody then put a row of sealant along the junction, including carefully covering the wicks - presumably "so they wouldn't be an entry point for water".

...........................

I only had access to the interior of one unit (condo board member) but there had been recent water intrusion at a window (not on that front wall) and at an exterior corner IR indicated there might be some water in the ceiling, however the Tramex showed only a 2-3% elevation there so I can't be certain.

I do know that to fix the leak leak at the window the builder came in and caulked the interior window junction with the wall, so we know where that's headed.

......................

Basically unless somebody can convince me I'm completely off-base here my recommendation is going to be to get one of the Big Dogs in there to do a building envelope analysis while the builder is still talking to the association - hopefully whoever they select will any me to tag along to see what they find when they start removing metal cladding, etc., but knowing the big-name architectural engineering firms probably the best I can hope for is to see the report.

Originally Posted by Michael Thomas

of theWell... perhaps I'm missing something here, and if so of course I want to be the first to know, but as far as I can tell there is no flashing under those windows: it appears that the roofing was run to a point about three quarters of an inch below the lower edge of the window frame, so all the metal surface in the third picture is the window frame, and there is no flashing.

To me, 'there could be' a flashing installed, just not installed 'properly'.

Currently, there is not enough height between the roof membrane and window to do that properly,

I think there might be enough height, but the base flashing would need to extend in under the windows and turn up, creating a pan the counter flashing, then the window, sets in. That base flashing would also need to extend up behind the siding.

This building is what I call "CAD Gone Bad": the architects design it with a variety of materials - in this case some kind of rain screen metal panel over CMU, interfacing with stucco interfacing with wood and all interfacing with exposed CMUs.

...

I attempted to identify the metal cladding material via the architect, who said that what he thought was specified was "Omega Light Cladding", but that "the builder may have substituted a similar material". No luck with Google for this material . Where I can get a sense of what's going on behind the cladding it's installed on metal standoffs over Tyvek over the CMUs (insulation back there too?), but at many interfaces with horizontal material below I don't see any evidence of flashing or any other detail directing water out from behind the wall.

I do know that some things were not done per where the plans were available, for example both the architectural specifications and drawings called for flashing between caps and parapets which was absent (a lot of builders around here do it that way because they believe the flashing will "break the bond"). There also a lot of just plain ignorant mistakes, for example at some parapets wicks have been installed below the caps, but somebody then put a row of sealant along the junction, including carefully covering the wicks - presumably "so they wouldn't be an entry point for water".

...

Basically unless somebody can convince me I'm completely off-base here my recommendation is going to be to get one of the Big Dogs in there to do a building envelope analysis while the builder is still talking to the association - hopefully whoever they select will any me to tag along to see what they find when they start removing metal cladding, etc., but knowing the big-name architectural engineering firms probably the best I can hope for is to see the report.

Unless you convince the person(s) you are talking with to use you are their representatives to the Big Dogs, in which case the Big Dogs will HAVE to allow you to be there whenever you want to be and see whatever you want to see. Make the fee to the association low enough such that they have no reason not to use you as their eyes and ears and to photo document the investigation and repairs.

Without removing at least that one window, no one is going to know *HOW* it was built, regardless what the plans show.