Firewall sealant and Fire Safety

[bkthomps]

i eagerly await your results

[rocketbob]

Originally Posted by dpansier

Many years ago I used Cotronics Resbond 907GF in a OEM application and recently tested it for sealing firewall penetrations and holes.
http://www.cotronics.com/vo/cotr/pdf/907gf.pdf

Hi Don, where does one get this stuff?

[flyingriki]

Order from them:
https://www.cotronics.com/vo/cotr/order.htm

I wonder about the small tube fo 907. Seems to be very high temp too and smaller, maybe cheaper...?

2300°F Resbond™ 907GF Adhesive & Sealant
This moist, creamy putty is easily applied from standard caulking cartridges for use from -300°F to 2300°F.

Just apply to steel, stainless, iron, most metals, ceramics, ceramic cloth, tapes, gaskets, tadpole gaskets, etc.

907GF air dries in 4-12 hours at room temp. (curing may be accelerated with mild heat) and is resistant to most chemicals, solvents, oxidizing and reducing atmospheres, aging, thermal cycling and electricity.

Use for bonding, sealing and repairing exhaust systems, fire brick, gas turbines, brazing fixtures, equipment, stacks, etc. Ideal for use in any high temperature Assembly, Production, Repair or Maintenance Application.

[civengpe]

Originally Posted by flyingriki

Order from them:
https://www.cotronics.com/vo/cotr/order.htm

$20.75 for the 11 oz tube. Salesman said that it has a 6 month shelf life from the shipping date.

Shannon

[dpansier]

I sent a sample of the Resbond 907GF Adhesive & Sealant to Dan Horton so I suggest holding off purchase. I believe it will perform well in this application but Dan's testing will confirm.

[Jamie Aust]

Originally Posted by DanH

A variety of popular sealants have arrived here, with one more on the way. I will give them **** (literally) and report, hopefully in a week or so.

Hi Dan, will you be testing RED RTV 736 ?
This is what my friend at Boeing said they still use. I just had a read on the back of the tube and it said "When heated to temps above 150C (300F) can form formaldehyde vapor" This is a potential cancer hazard.

What can I expect the temps to get up to under my cowl in normal flight ?

Thanks

[DanH]

Originally Posted by Jamie Aust

Hi Dan, will you be testing RED RTV 736 ?

I wasn't planning on that particular product (Dow Corning, yes?), but I do intend to stop by the local auto parts store and pick up a tube of "Permatex Ultra Copper". It seems to be the common-as-dirt choice for a high temp silicone sealant here in the States.

What can I expect the temps to get up to under my cowl in normal flight ?

We're not yet into the really hot time of year here in Alabama, but I've been seeing inflight air temperatures around 250F exiting the bottom of a cylinder head and around 160F exiting the oil cooler. I've not measured temps after shutdown, but I'd bet money on 200-250F. Obviously radiant heating could go higher in proximity to an exhaust pipe.

[DanH]

Ok, here we go. Sorry to take so long. Had to assemble a new fire rig as the old one was long ago dismantled and recycled, and I decided to allow plenty of cure time for the test materials.

First, the candidates:



Cotronics Resbond 907GF (courtesy of Don Pansier)
Rectorseal Biotherm 100 (courtesy Dan Langhout)
Flamemaster CS1900 (courtesy Dan Langhout)
3M FireBarrier 2000
Permatex Ultra Copper

Practical notes, not fire related:

All are single-component except for CS1900, which requires mixing.

Resbond doesn't tool very well. The three silicones (Biotherm, FireBarrier, Ultra Copper) tool, well, like silicones; FireBarrier and Ultra Copper do best in this group. CS1900 is nothing like proseal and doesn't really tool at all. It is more like a blob of uncured rubber or playdough.

The silicones stick well, the Resbond a little less so, and the CS1900 is not very adhesive.

When cured, Resbond is hard. The silicones are what you would expect. The CS1900 feels like soft tire rubber.

The setup:



18 x 24 sheet of stainless firewall material. At the top are simple strips of sealant on the front and back of the panel (staggered). At the bottom we have actual firewall electrical penetrations, very basic; 6-gauge tefzel wire through plastic snap bushings, then covered with sealant, without SS shields. I selected this sort of pass-through because they are commonly seen on experimental aircraft. On the left is a personal experiment, the pass-through style on my own airplane, a 2"x2" .032 plate with a 1.5" long .035 tube, welded 4130. The cable, wires, or whatever is inserted, then the tube is pumped full of sealant.

Here's the back of the panel. I made a point of equally filling each snap bushing with sealant

[DonFromTX]

Great testing Dan, I can hardly wait for the results.

[DanH]

Bring on the heat.



First the burn rig was calibrated to approx 2000F using copper sheet. Note that (compared to previous firewall tests) I have reworked the torch nozzle to obtain a tighter, more horizontal hot spot pattern. Also note that none of the sealant samples here are actually in the 2000F hot spot.

Let's go closer:



Approx 30 seconds into the burn. The CS1900 has burst into flame. All three silicones are also outgassing flammable material, but not like the CS. The Ultra Copper sample is the worst of the silicones in this regard. The burn exhibited by these four samples is not entirely a bad thing, as the probable operating scheme is to form an insulating char. The Resbond appears to be inert. Compare the silicones to the previous setup photo and you'll see the Biotherm and FireBarrier samples are exhibiting some degree of intumescence, swelling and expanding. Now look at the 6-gauge wires; so far all the sealant samples are holding up ok (even if outgassing flammable material), but the wire insulation is burning away. I point this out because sometimes I bump into a builder who assumes Tefzel insulation to be fireproof.....it ain't so.

Back of the panel at about 45 seconds:



Hmmm...the Ultra Copper strip sample has burst into flame. The Biotherm sample had flamed slightly and then fallen away leaving some adhered residue. The CS1900 has peeled and fallen away cleanly, not even leaving a mark. Resbond and FireBarrier are doing fine.

Looking at the steel pass-through, we see intumescent FireBarrier squeezing out of the gaps in the joint. Same is true at the plastic bushing (center). So far all the plastic bushings are holding up ok, which means all the sealants are doing a decent job of slowing heat transfer; in thick sections they are serving as insulators.

Front of the panel at about 1 minute:



All the pass-through samples have remained intact and formed an insulating char. Of the strip samples, only the CS1900, FireBarrier, and Resbond remain adhered. The difference is mechanical; the 6-gauge wire is keeping the sample in place over each of the bushings. There's a lesson here; give your sealant a mechanical advantage!

Back of the panel at about 1-1/2 minutes:



The snap bushings are now melting. The copper wire cores are very hot, so the tefzel insulation is beginning to soften and swell. All the strip samples have peeled away, except for the Resbond.

More later.