Dead Level Roofs

Archive of Older Topics Archive - Specifications Discussions #5
1

More and more protected membrane roofs are being placed on older building, many of which have conventional roofs over dead level decks which are no longer acceptable in most of the newer building codes. How do the different regions address this situation?

Does one apply with AHJ’s for exceptions; place a tapering compound or product on the existing decks providing a slope to new roof drains; just place the new membrane and hope the water eventually reaches the drains; add a filtering drainage panel product over the new dead level membrane; or another scenario?

After consulting with NRCA and several mfrs, they just do not address new roofs over dead level decks. After consulting with several roofers, they say they just install new roofs on dead level decks and claim it is an existing condition.

As specifiers, in most cases, we are acting as roof designers in behalf of the Architects. But when architects are questioned, they have a ‘bambi in the headlights stare.’ Should specifiers assume the role of designers?

2

Specifiers are every bit designers as the rest of the architectural staff. More so, in many cases (this being one).

My tack with this particular issue would be to specify hot rubberized asphalt waterproofing - which is warranted by nearly every manufacturer over dead level deck. I am also familar with one modified asphalt system that also warrants their system over dead level deck (APP-modified).

I suspect that some of the better SBS modified waterproofing systems out there would also warrant their systems over dead flat deck.

We have successfully argued with building code officials and local jurisdictions that slope for these systems is not necessary, based on manufacturer’s warranties and long term in-service histories, with pages and pages of examples of projects built in the same way.

Good luck, Mr. Designer!

3

Technically it can pass as a waterproofing system which does not require slope to drain.

4

Thank you, Ken for your obversation. I have suggested to many people that the inherent difference between roofing and waterproofing membranes is exposure to the sun (and UV radiation). Roofing membranes must “stand on their own” in keep water out of the building while being directly exposed to damage from the elements (sun, rain, snow, hail, etc.). Waterproof membranes must only keep water out of the building although usually these systems are subjected to a higher hydrostatic head.

“Green roof” systems are usually protected by protection board, insulation, plant growth medium, and plantings. Sounds like waterproofing to me.

Incidentally, I always thought the hot-applied asphalt systemswolud be more expensive than the mod-bit systems, but a recent conversation with a waterproofing sub indicated that the hot-applied systems were significantly less expensive.

5

I am currently with the Pennsylvania Department of General Services, Bureau of Engineering and Architecture. We are responsible for all construction in the Commonwealth. That said we do lots or re-roofing as part of our maintenance programs. I have often worked on a building with a dead level roof structure. Since we are also often trying to improve the roof energy efficiency we do complete tear offs (due to deferred maintenance issues) and install tapered insulation systems to ensure positive drainage.

6

I’ve never had an issue with any of the multiple of choices of membranes. I do ‘wonder’ what really happens to the water. We all agree that the water eventually makes its’ own way to a roof drain. There are products that facilitate the drainage, however.

There are ribbed rigid insulation (recommended for use under precast ballast) and boards with kerfed or channeled edges always along the long edge. These channels almost never go the the drains in a direct route. These boards are much more expensive than a square edged board but they do permit a better flow of water through the system. Also, there are composite drain panels (installed on top of the membrane) which gives a 360 degree flow pattern to drains.

What I ‘wonder’ about is: (i) how much does the presence of water in the system reduce the thermal value of the system? …and does the speed at which water is removed reduce any loss of R-value?, and, (ii) does the use of a drainage panel create a thermal break in the system? …or does the ‘dead’ air space improve the thermal value of the system?

7

Ron brings up some interesting questions. I have to believe that a drainage panel would not reduce the system R-value but the R-value of the drainage panel would vary depending on the amount of water in that component at any given time.

It could also be argued that without a drainage panel to allow for easier movement of the water there could be an increase in the moisture content of the insulation which could amount to reduced R-value in that part of the system. Of course the type of insulation could influence how much moisture will be able to affect the insulation’s R-value.

In general our office does not use a protected roof membrane system in our projects so we would typically deal with a dead level roof with tapered insulation and then a membrane on top of that to provide slope to drain.

8

If you have a level roof then the water will pond. The framing needs to be checked for the additional weight of this water.

9

There are a couple issues here which seem to need to be clarified.

First though, some of my experience. I have been doing dead level roofs on new construction since the 70s when Uniroyal came down out of Canada to the US - ultimately, this became what today is American Hydrotech. In the early 1980s, one of the highest profile projects was the expansion of the National Geographic Society with a new building. It was done as a dead level roof, it is an American Hydrotech system, and it has never leaked or had any problems. In any given year, 90% of our projects at WDG are dead level roofs as new construction, here its more common to see dead level roofs as new construction.

The first issue that seems adrift here relates to the use of the term ‘waterproofing’. These are roof membranes, not waterproofing membranes. Those that make membranes for dead level roofs often have a twin for a waterproofing membrane relating to the second issue below. Terminology is specifications is paramount. And though a roofing membrane is technically protecting one from water, they should never be called waterproofing. Some installers and some manufacturers will gladly install their waterproofing membrane and give you the lower warranty. Or, since most of the differences are in flashings, give you the non-UV resistant flashings simply because they are mostly covered up anyway. If the membrane is on a slab elevated above the first floor slab, technically its a roof membrane and should be shown and specified as such. Also, it used to be, though not as common anymore, that the various manufacturer’s had a different certification program for waterproofing vs roofing membrane installers, roofing was always more rigorous due to the longer warranty periods. Several of the 1970s/80s era major manufacturers had this, right now its not as common, but, this was another way that calling it on the drawings and in the specifications as a waterproofing membrane could get you the lower quality installer subs. Some subs that did extensive installs of a manufacture’s waterproofing membrane were not even qualified to install that same manufacturer’s ‘identical’ roofing membrane.

The second issue regards the heritage of dead level membranes and why all the manufacturer’s that offer them have a simple letter that can be sent to any jurisdiction. No code variant or request needs to be gained, most major building areas the code reviewers are well acquainted with the issue and the systems. The letter will almost always relate that what is being installed is truly a waterproofing membrane on a roofing condition, and that the products are certified for impounded water. Done. They still are marketed as, and should be shown and specified as, roofing membranes.

Other thoughts:

Older buildings do need care, can the roof and structure carry the load. This is not so much an issue for new construction due to the scupper overflow requirements as the required second drainage system. Some projects do have a secondary drainage system for the roofs, but many go for the scupper system. Then in some buildings coming of age for roof replacement came out of the 80s and early 90s when ‘slow drains’ were a requirement. This results in the requirement to impound water on the roof so as not to impact the local storm water drainage system. Many of these were done as dead level. We have a project right now where a university is stripping down what an existing building that was once a 1960s hotel to its structural shell and want to provide a vegetated roofing system on it. Previously the membrane was a single ply over sloped insulation. First thing structural did was run calculations for the allowable weight of the new system including maximum saturation with water. It will work, but maybe because the entire building, including its facade, is cast in place concrete. Its still close though, and it has to be a thin system.

Comments are about where does the water go. Ponding will result, but it is of no harm. These membranes are capable of sustaining impounded water. Some are even used as pool and fountain liners which have more harsh chemicals than rain water.

The best design for a dead level roofing system is an IRMA, inverted roofing membrane assembly. The insulation does not absorb water, typically it is extruded polystyrene. Its insulation value is not lowered whether it is submerged or not, its closed cell.

The most commonly thought of membrane for a dead level roof is hot fluid applied rubberized asphalt (Hydrotech, Henry, Barrett are the most common, Suprema has a system for this now as well, there are others like Carlisle.

There are a number of single ply membranes that are also certified as waterproofing membranes for impounded water, many of the PVC and KEE systems are certifiable for dead level roofing. Though frankly in an inverted membrane assembly all those seams freak me out. There are some other generic examples of single ply that can certify for dead level, but dealing with seaming systems that are not welded seams is somewhere I can’t being myself go - weld seams I can imagine but other seaming systems cause me nightmares.

So a lot of systems out there. I myself try to stay with the hot fluid rubberized asphalt, and limit myself to a very small number of manufactures that I have followed or worked with over the years, who have great details and offer great technical support during design (I can call my local rep for these manufacturers and have an answer that may relate to a project on the other side of the country that the local rep out there verifies with a third rep following up with a visit to the production group which may be in yet a different part of the country a few days later).

And every one I deal with has to be able to provide a single source warranty (everything from the slab up to whatever is ballast (crushed stone or pavers) or to the top of the plants for vegetated roofing systems, transferable, no limitation on moving and replacing any overburden, ballast or insulation of any kind for the duration of the warranty period.

Some of these systems (notably hot fluid applied rubberized asphalt) are performing so well that if you have one of these roof systems, and its at the end of its warranty or out of warranty even, they will come in, inspect the system, give the owner a report about how many panels of insulation may need replacement (if any) and if some details for flashings need to be revised or upgraded and quote a roof renovation that does not touch the membrane at all - and give you a new 20 year or greater warranty as though you just installed a totally brand new roof. And that because examining these 20 and 30 year old systems, the membrane itself is responding and testing out as though its only a few years old.

A comment about additional drainage panels. Of the manufacture’s that I have worked with for an IRMA type hot fluid applied rubberized asphalt, none of them require a drainage panel. Some will put up with its being installed though they do not encourage it (waterproofing consultants like to keep adding things for some reason). A few will actually refuse to install them stating it will void their warranty. So, for non-vegetated systems from the slab up its membrane, separation material, extruded polystyrene insulation, ballast (crushed stone or pavers and sometimes filter fabric under the ballast. All insulation is required to have drainage channels - use of the ribbed bottom insulation with many manufacturers will void the warranty. For a vegetated system, remove the pavers/ballast and continue up with root barrier, various elements for water retention depending on the depth of growing medium/plant types, soil erosion control embedded if permanent and plants. Get a single source warranty for from the top of the deck to the top of the plants.

I specified (with my good friend and fellow specifier Joe Winters for HOK DC - our landscape consultant) the second largest vegetated roofing membrane installation in the country and it is built this way, much of it on dead level roof slabs - US Coast Guard Headquarters in Washington, DC. The only real difference is that being a design/build project where the bridging documents did not require a single source warranty, the GC declined to provide one and movement of overburden above the insulation is not included in the warranty for the roof system.

William

10

I would second William’s comments about successfully using hot fluid applied rubberized asphalt membranes in protected membrane roof systems on dead flat roofs for many years. The projects have been mainly CIP concrete structures in the DC area.

11

Happens all the time with Hydrotech 6125 FR (215 mils). Hydrotech will warrant applications in dead flat slabs, but AHJ may object just because.

12

American Hydrotech (as well as the others I have worked with) has a nice very short letter that they provide, and that’s that. Its never failed to have been accepted the few times it was required. We have one jurisdiction that requires it to be sent every time regardless of it being the same reviewer.

13

What I was hoping to get from this thread is anything specific on the use of composite drain panels. Or, are they just a unnecessary product proposed by some mfrs?

Being from the “old” school, I still think in terms of “the quicker water gets off the roof the better” or in the case of protected membrane systems “the quicker water gets OUT of the system the better.”

14

Ron, I think the bottom line with drainage panels is that you don’t want them on a dead flat surface since they just become reservoirs and provide no benefit but probably add cost and risk.

If you’ve got a sloped deck and are using a sheet membrane, you may want to speak with the manufacturer. The panels may be seen as providing protection to the membrane but I doubt that any of the better roof systems require them except in vegetated assemblies. You’re not looking at sandwich slabs are you? No matter what, remember you need a drain at the membrane.

15

Ken:

By design protected roof membrane systems permit water to seep below the insulation where it is protected from freezing. Eventually it works its way by gravity and capillary action to the waterproofing membrane, and in a properly sloped deck, to roof drains - but dead level decks do not permit such a flow to drains. A drainage panel with a thickness of approximately 1/4" limits the reservoir to which you make reference to only that thickness. Continued incoming water forces the water into the deck drains.

In response to my inquiry to Carlisle about the subject of this thread their response included the following statements:

  1. As long as the thermal plane is not broken continuous vertically through the insulation, there should be no loss of R-Value. So a drainage board underneath the insulation should not reduce the R-Value but potentially increase as dead air space.
  2. Typically we see it simple as two options, either a drainage board and flat insulation or pre-channeled insulation with flat on top. In the case of a flat deck, [we] would recommend the drainage board over the pre- channeled insulation, because there is a good chance that water will always be found at this layer. A pre-channel board would have the potential of heaving in a freezing condition. In a drainage board anything that freezes would just spread out through the board.
  3. We do include our drainage components under our warranty when our products (insulation, pavers, or vegetated components) are used above as overburden. If products are supplied by others, that manufacturer should recommend what drainage method would be acceptable for their products.
  4. [We] do offer one cautionary note: In the respect of vegetated materials, it is not recommended to install this type of overburden on a dead level surface. The reason is that plants love water and will send roots out to look for it. If ponding is occurring at the membrane level, there is a high probability that the roots will search out this water and damage the primary waterproofing membrane. We recommend putting positive drainage under the membrane so that ponding does not occur and attract the roots.[1]

[1] Email of 13Aug2014 from Chad Buhrman, CCM Insulation Product Manager, Carlisle Construction Materials who forwarded the comments from Carlisles Project Design and Review Department.

16

I believe item #4 from Ron’s post is vitally important. Vegetated roofs are just one of many potential forms of protected membrane roofing systems but they have some unique considerations.