Effect of spray foam on AC refrigerant lines & electrical wiring
Finally we are ready to make our attic a conditioned space. R38 on 2x6s and R21 gable walls. The new closed-cell Demilec HFO looks very promising. After all the research, I’m done to one major concern:
The effect of the installation heat upon my Romex, alarm system wiring and AC refrigerant line. About five feet of the latter would be buried in the foam; and quite a few runs of Romex and low voltage alarm wire would also be buried. The concern is not only the initial, temporary 200ªF temps of the foam as it cures but the lack of ability for the heat of Romex wires when they are in use to dissipate due to the effectiveness of the insulation. I can’t imagine what the initial heat would do to the rubber tubing on one of the AC lines or the copper tubing underneath.
I’m attaching some article links below. Generally Romex is rated at 90ºC (194F). Some foam manufacturers correctly say that there will be no air to start a fire if the wires overheat, but if the wires short out from melting insulation, other parts of the system could be affected and cause a fire elsewhere. Multiple Romex wires put together through bored holes in the rafters increases the risk. I have scoured the internet and have found only a few reported cases where the intense heat has left the sheathing intact but melted the wire insulation beneath. But admittedly not many cases are reported. Common sense tells me not to do it.
My solution to this may be:
1-Remove the wiring first and put it in conduit over the sheetrock when I install that over the foam.
2-Or I could spray foam three inches instead of five—thus not touching the wires– and then use cellulose for two inches.
3-Or de-rate the amperage for the breaker that covers the attic circuit and/or change the standard breaker to a combo AFCI-GFCI one.
4-Some installers report the using the trick of spraying a quick inch or so on the wires, let it cure 15 minutes and then spray the rest of the cavity. That might solve the initial overheating problem but not the long term inability of line heat to dissipate. In ordinary use –mostly for lighting–not much of the 15 amp capacity would be used, but it’s hard to predict what certain outlets might get plugged into them in a house. This is an old house with some updated wiring and fairly new main panel 200 amp service, but this one line covers two bedrooms and the hall, outlets and lights.
5-For the AC line I can reposition it so it only goes through four inches of the gable wall, or better yet since my AC is already over 30 years old, rip it all out and install a new, smaller, more efficient unit away from the foam.
I’d appreciate any feedback on this.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
David,
Your links don't work for me. I'm guessing that these are the two articles that you are trying to link to:
Thermal Effects of Type NM-B Cable Installed in a Residence Encased in Spray-Foam Insulation
Effects of Polyurethane Foam Systems on Wiring Ampacity
Martin Holladay
David,
Code officials have accepted the use of spray foam insulation in residences for a long time, and it's common to encapsulate non-metalic cable (Romex) in spray foam. I find it hard to believe that building officials are ignoring a fire hazard.
Right off the bat, I'd like to point out that when spray foam is installed too thickly, the exothermic reaction that produces the foam can result in very high temperatures -- certainly high enough to scorch the jacket on Romex wiring. It seems to me that this risk is greater than any risk from properly installed spray foam.
I look forward to reading the comments of other GBA readers.
-- Martin Holladay
HFO blown 2lb foam is not nearly as exothermic during installation as the HFC blown goods. It is usually OK to install in at least 4" lifts whereas HFC blown goods are limited to 2".
Demilec claims HFO High Lift is safe to in install 6.5" at a time (R49 in a single pass according to the marketing fluff):
http://www.demilec.com/documents/Heatlok-HFO-High-Lift-Flyer.pdf
The HFO blowing agent itself is somewhat flammable, but the diffusion/leakage rates aren't high enough to be of concern. They indicate in attics & crawl spaces at thicknesses less than 6.5" it doesn't need an ignition barrier on vertical installations (walls), or thicknesses under 11.5" if horizontal:
http://www.demilec.com/documents/Heatlok-HFO-High-Lift-CCRR.PDF
I'm skeptical of a long term R7.5/inch though, but there isn't a sufficient track record of third party tested fully aged HFO blown stuff- it's too new to tell what it's performance will be in 50 years.
Thanks Dana and Martin, and for the corrected links. I spoke to a Demilec rep who confirmed their HFO formula comes out of the nozzle around 105ºF but the reaction raises the temp to 180-200º which dissipates a bit after fifteen minutes or so. Hard to imagine that not affecting at least the insulation on the wires under the sheathing of Romex or the alarm wire or, due to the complete encasement, the wire's long term ability to dissipate heat from current. Also wondering about the effect on the few feet of refrigerant line that will encased--see photos. Anyway thanks for your advice--the lack of a longer list of reported problems with this does seem to agree with your point.