What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Firewall Penetrations

DanH

Legacy Member
Mentor
From a previous thread...

"...Yes Mickey, I constantly come across RVs that have firewall penetrations made with nylon snap bushings covered with a blob of "high temperature" red RTV such as Permatex Silicon Gasket Maker or similar. This is crazy..."

I guess you ought to address this with Vans. Referring to RV-10 plans, section OP37-2 figure 1, electrical harness, Vans says drill a 1/2" hole for firewall penetration of the battery cable and use a SB500-6 snap bushing...

It may not be what YOU consider airworthy, quality, or whatever you want to call it but it is spelled out in the plans that way. Someone with little or no experience in this hobby is going to follow the plans...

...which is a very good point.

That inexperienced builder may wish to learn more. Like most issues, God is in the details.

At typical operating temperatures the snap bushings will be fine.

Snap bushings combined with a good sealant choice actually performed at a reasonable level in a fire resistance test, maintaining a gas tight seal for roughly 1-1/2 minutes when adjacent to, but not encompassed by the hot spot. That said, sealant choice is an integral part of bushing performance, and there were significant differences in sealant. The popular orange Permatex was not a good choice, as it appeared to be highly flammable, and would serve as a fire transfer mechanism, bringing open flame and smoke to the cabin side.

In the same hot test, a simple tubular pass through filled with intumescent sealant performed much better than the basic snap bushing and sealant. It never lost a gas tight seal, even when subjected to the full 2000F, and did not itself transfer flame to the cabin side. The test was ended when the cabin side tefzel wire insulation ignited via its red hot copper core.

The popular eyeball style pass throughs were not tested, but their performance is going to be predictable. Aluminum eyeballs should be expected to melt if encompassed in the hot spot, but maintain integrity if held below 1000F, more or less. Steel eyeballs would maintain integrity regardless, but may not remain gas tight when the wire insulation melts.

Read here:

https://vansairforce.net/community/showpost.php?p=549778&postcount=38

https://vansairforce.net/community/showpost.php?p=549798&postcount=40

https://vansairforce.net/community/showpost.php?p=549839&postcount=42
 
Dan, thanks for posting this-----I have tried to find these tests in the past, but they seemed to be lost in never-never land.

Moved to the Building Tips forum to make it easier to find in the future.
 
It's Dan's testing that caused me to think a lot harder about my FW penetrations, and make them better. In the end it cost very little additional time or money. Still far from perfect, and no idea if it will make any difference if I have a FWF fire, but I think they have to be better than a snap bushing with a bit of RTV.

Here's how I did it:

IMG_1642.jpg

IMG_1632.jpg
 
Wraps

It's Dan's testing that caused me to think a lot harder about my FW penetrations, and make them better. In the end it cost very little additional time or money. Still far from perfect, and no idea if it will make any difference if I have a FWF fire, but I think they have to be better than a snap bushing with a bit of RTV.

Here's how I did it: ...

What is the shielding on the exhaust pipe?
What is the wrap on the Intake pipe? Silicone?
I may need to protect things as well.
My penetrations are very similar. Steel eyeballs and fire sleeve pathways.
 
I have similar radiant heat shields to protect the fuel lines and the control lines from the radiant heat generated by the exhaust pipes.
 

Thanks Mickey, I forgot about tubulars with a firesleeve wrap. Should have tested it with the others, but I suspect they will perform well, in particular if used in firewall locations which are not likely to be subjected to full fire intensity. The basic concept is that fire follows airflow, so locations near the outlet are likely to be a lot warmer than a location at the top of the firewall, as you pictured.

FWIW, I am also using a wrapped tubular pass though for the one penetration which carries a bunch of small wires. It's far over on the right side, out of the likely high temperature region. Used a wrapped fitting there so I can easily add and remove wires for my experiments. All the others are filled with FireBarrier 2000.

BTW, pick up some ordinary shiny aluminum tape, as used in modern residential ductwork. A reflective wrap on those throttle and mixture cables will significantly slow radiant heat transfer.
 
Last edited:
On melting...

...Looking at the myriad of AN426AD rivets that attach the firewall to the substructure and imagining what happens when those melt; little jets of fire come in to the cabin from everywhere...

Those probably should be solid monel rivets if we're going to be serious about it.
 
Those probably should be solid monel rivets if we're going to be serious about it.

Different subject, easily addressed. No worries here.
.
 

Attachments

  • 5 min, Aluminum Angle Small.jpg
    5 min, Aluminum Angle Small.jpg
    275.6 KB · Views: 385
What is the shielding on the exhaust pipe?
What is the wrap on the Intake pipe? Silicone?
I may need to protect things as well.
My penetrations are very similar. Steel eyeballs and fire sleeve pathways.
The exhaust pipes have "heat shield" on them to try to reduce the radiant heat that goes to things near the exhaust, like cables and hoses. I used both store-bought heat shield from Van's and "home made".

http://www.rv8.ch/heat-shield-home-made/

I don't have any wrap on my intake tubes but they have rubber joints that connect the tube to the "sump". I don't think these are silicone but some kind of rubber. I've got more pictures here: http://www.rv8.ch/engine-photos-after-test-flights/

I've updated the clamps there to "constant pressure" clamps. Just added some pics: http://www.rv8.ch/sds-intake-tubes/

@DanH - thanks for the tip on the aluminum tape - will have a look at that. So far I think the heat shield on the exhaust pipes is working pretty well, but I have not pulled the cowl in a good 20 hours - need to do that soon.
 
Last edited:
firewall

One easy firewall penetration to deal with is any fluid lines, especially main fuel line. Bulkhead fitting with all stainless components. A bit hard to find but they are available.
Most engine fires will be exhaust and or fuel related. All you need is enough time to get the fuel shut off.

Do this with the fuel shutoff valve, not the mixture.
 
Steel

All my penetrations and fittings are steel or stainless steel. All lines custom made by TS Flightlines. Fire sleeved. Firewall titanium over Fiberfrax.
 
The discussion of fire and heat protection can get out of hand in a big hurry---especially when the subject of aluminum rivets enters the picture.
So I ask, what about the fiberglass cowling? Anyone flown an RV without a cowling, under the supposition that it would have burned off with an inflight fire?
I'll bet no is the answer. I know Dan hasnt tried it, and I doubt seriously anyone else has either.

You can do the best, safest preparation you can to mimimize the risk, but there is always the possibility.

Tom
 
... Anyone flown an RV without a cowling, under the supposition that it would have burned off with an inflight fire?...
Tom

It is pertinent that the cowling, the baffle seals and maybe an air box are flammable. If it is hot enough to burn through the firewall grommets, then surely it will be hot enough for these items to melt or burn. When they separate it may not be clean. I was taught to immediately turn off the fuel valve and close any vents and then try to extinguish an engine bay fire with a sideslip. Not sure if this applies for an RV?
 
The discussion of fire and heat protection can get out of hand in a big hurry---especially when the subject of aluminum rivets enters the picture.
So I ask, what about the fiberglass cowling? Anyone flown an RV without a cowling, under the supposition that it would have burned off with an inflight fire?
I'll bet no is the answer. I know Dan hasnt tried it, and I doubt seriously anyone else has either.

You can do the best, safest preparation you can to mimimize the risk, but there is always the possibility.

Tom

Yep, where do you draw the line?

It is all about risk mitigation; and everyone has their own comfort level...

If you want to be completely safe in an airplane, don't fly...
 
The discussion of fire and heat protection can get out of hand in a big hurry--

Yes, it can.

Above all, primum non nocere...first do no harm. It's far more important to show builders why they should stop doing really dumb things on and around firewalls, than it is to add any sort of extra fire resistance. Rubber, plastic, and polyester cabin side "firewall insulation" is a good example. It's unbelievable how bad some of them are in terms of flame and boiling black smoke. I once had to drag the fire test rig away from my shop before the burning sample melted the vinyl siding.

Second, it's not about making the airplane fireproof. It's about buying some finite amount of time. An extra 60 seconds to get the fuel source shut down is a big deal. Three minutes is huge.

Third, gas tight and no smoke is just as important as keeping flame at bay. Can't breath or can't see has pretty much the same result.

No reason to worry about melting cowls. As noted previously, fire follows airflow, meaning the critical area is the center and bottom of the firewall, as well as the aluminum belly skin just aft of the exit. There have been enough actual fires to know it's true.
 
Aerotronics found sealed metal-shell CPCs for my RV-10 panel. The shell is aluminum. I just discovered that TE (www.te.com) makes what appears to be steel-shelled CPCs. For the large gauge power wires (battery and starter), I used a marine transom pass-through. And for the LSE ignition coil wires, I found metal BNC connectors (with a nylon spacer to isolate them from the firewall). I'm not sure how the latter will hold up, but I expect it will be better than just passing the wires through a hole and depending one sealant. And, I get the benefit of being able to disconnect at the firewall. After having to deal with the wiring when I nearly nosed my RV-6A over, resulting in damage back to the firewall, and then again last year when I upgraded the panel, I really hate having wires threaded through the firewall.
 
Firewall

With my homemade pass thrus there is no concern about wire chafing on a sharp edge. The wires are wrapped in fire sleeve. No sharp edges on the pass thrus.
Appropriate size stainless tubing welded to a flat stainless plate with screws at the four corners. All stainless hardware for the fasteners.
Regarding fiberglass I have had two exhaust failures with piston twins that burned holes in fiberglass nose cowl but did not support combustion. As soon as the hole in fiberglass reached a certain size the hole stopped growing.
In the second event the temperature at takeoff was 120 F. I was about 20 miles from the nearest airport and not confident that I could make it on one engine so I kept the bad side running at very low power until I had the runway made.
People are misusing the word flammable. Cowl, baffles, baffle seals, airbox etc are not flammable by definition. They likely don't even meet the definition of combustible.
The real concern with normally aspirated engines in a combination of failed exhaust and failed fuel component. Shut the fuel off immediately. Seconds count.
 
What about the aluminum floor just aft of the firewall? If there really was a fire big enough to worry about breaching (poorly done) firewall penetrations, the blowtorch coming out the cowl outlet would surely melt through the floor first.

Some RV's have stainless-lined cowl outlets I think: -8's and -14's? But the rest don't.
 
What about the aluminum floor just aft of the firewall? If there really was a fire big enough to worry about breaching (poorly done) firewall penetrations, the blowtorch coming out the cowl outlet would surely melt through the floor first.

Some RV's have stainless-lined cowl outlets I think: -8's and -14's? But the rest don't.

The floor melt through has happened in more than one incident. That's a really bad thing. I don't have any protection against that in the -6, but my -10 has a stainless overlay with fiberfrax under it in that location.
 
I used 1/16" Fiberfrax and some .020 grade 2 titanium over that for a belly overlay since on my RV-3B, the floor is the belly skin. Here it is at a very early stage.

Ti Belly Overlay in Place S.jpg

Dave
 
Firewall

If you have the IF to watch a nightmare scenario, search for the video of Beech Duke fire in flight May 15, 2019.
Aircraft just out of shop with completely new panel and avionics. Out of annual, no ferry permit. Ferry pilot.
Fuel line was not properly torqued. In a few seconds the airplane was engulfed in flames and went straight in. Both fuel selectors still in on position. Very close to the runway at Loveland, CO
 
I was taught to immediately turn off the fuel valve

If the fire, smoke, heat (and smell??) mainly go out the cowl outlet there may be a critical short time before one would realise there is a fire to turn off the fuel valve. With this in mind I have installed a external temperature probe in the fuselage floor just aft of the cowl outlet to measure the cowl outlet air temp. The temp is displayed on the EFIS with an alarm set for over normal temps.

Fin. 9A
 
Last edited:
If the fire, smoke, heat (and smell??) mainly go out the cowl outlet there may be a critical short time before one would realise there is a fire to turn off the fuel valve. With this in mind I have installed a external temperature probe in the fuselage floor just aft of the cowl outlet to measure the cowl outlet air temp. The temp is displayed on the EFIS with an alarm set for over normal temps.

Fin. 9A
Interesting idea, Fin. If you are flying, what kind of probe did you use and what values do you see? I'm thinking something like a CHT probe might work, but would it pick up heat from the nearby exhaust pipes. Since many of us have engine monitors that will work for 6 cylinder engines, and only have 4 cylinders, using #6 EGT or CHT could work. Quick notice of engine fire could be very helpful.
 
Mickey,

The probe is an MGL air temp probe that connects to an aux channel on my MGL Xtreme EFIS/EMS. Can’t recall the typical cowl outlet temp but the alarm is set at 210 degrees F and over many years has never been tripped. The probe is not ideal for this application as it is aluminium and would be destroyed in a fire. However it would set off the alarm first and if it subsequently stops working that confirms excessive heat/fire.

An EGT k type probe would be better but I couldn’t set a seperate alarm temp for say, probe #5 on the MGL. The MGL Xtreme has two aux temp channels with alarms but they won’t take k type probes so my only option was the probe shown. Maybe some other EMS/EFIS have a separate k type probe channel with an alarm??

The only unlikely problem I can think of is if the probe simply decides to fail showing a false over temp when there is no fire - might end up doing an unnecessary forced landing :eek: I guess an option would be to also install an EGT probe in the same location on say, channel #5 so an over temp alarm could be confirmed. Perhaps I may be overthinking this as most aircraft just rely on one oil pressure sender!

Fin 9A.

141-E1-FCE-F243-4340-821-B-80-AE077-EE1-A0.jpg
 
Last edited:
Gentlemen: I CERTAINLY didnt mean to open up a can of whoop ***. I was mearly making a statement. Dan, I do agree with you 100%, finding things that work, and are safe, and passing that information on is critical. We are all safer because of forums like this, and those that came before us ( even though I'm not flying). WE can put that information in to the memory banks ( I think I have one of those) and apply what we've learned.

Tom
 
..
So I ask, what about the fiberglass cowling? Anyone flown an RV without a cowling, under the supposition that it would have burned off with an inflight fire?
...

Tom

I think that Tracy Crook did one flight in his RV-8 without cowling during cooling troubleshooting/solving. Probably not very fast. BTW, that led to the discovery that air outlets were as important as inlets, if not more.

Finn
 
Back
Top