Finding the Leak...(Found the Leak)

[bjdecker]

For the hive mind.

My friend purchased an almost-completed RV-14A a couple of years ago and finished it up and now has 80 hours on it.

During the first flight, and every subsequent flight, the G3X + Aithre 3.0 indicated high CO levels (50+). The exhaust system was pressurized with a vacuum cleaner outlet and soapy water sprayed everywhere. As expected, all of the exhaust ball/slip joints were leaking, but nowhere else.

A bit of HVAC "200MPH" tape was applied to the gaps between the aft edge of the top and bottom cowl aft edge and fuselage, and the CO levels decreased to ~50ppm in climb and 10ppm in cruise.

We've since inspected the firewall, and all penetration points, and found no air gaps or light leaks. We did change the cowling-to-fuselage attachment scheme from hinges to a continuous underlapping flange and screw affair (read: no gaps).

We're still chasing non-0 CO levels (~30 on climb out, 6ppm in cruise) . The remaining points that need some sealing are the bottom left & right corners of the cowl/fuselage.

We're going to take a handheld CO detector with us on the next flight to double check the Aithre.

Anyone else have suggestions?

 

[rvbuilder2002]

Has a leak check of the entire firewall area been done using a bright light in a dark hangar?

 

[wcalvert]

Has a leak check of the entire firewall area been done using a bright light in a dark hangar?

Vents open/closed? Heat on/off? Canopy seal condition, including the leading edge? ROP? LOP? Calibrated CO meter?

 

[Kyle Boatright]

On the -10, there is a gap in the fuselage skins just below the leading edge of the wings. Does the -14 have the same gap? If so, make sure it is sealed.

 

[bjdecker]

On the -10, there is a gap in the fuselage skins just below the leading edge of the wings. Does the -14 have the same gap? If so, make sure it is sealed.

The RV-14 wing-to-fuselage interface...isn't...

There is a wrap around strip of .020" or .032" skin with some stiffeners but no seal along the inside edge -- this is per plans...and different than the preceding RV series...It's gappy, and I am going to seal that up next --- just to see...

 

[Kyle Boatright]

The RV-14 wing-to-fuselage interface...isn't...

There is a wrap around strip of .020" or .032" skin with some stiffeners but no seal along the inside edge -- this is per plans...and different than the preceding RV series...It's gappy, and I am going to seal that up next --- just to see...

I'm not talking about the wing/fuse interface. This gap (on the -10) is at the bottom corner of the fuselage in front of the leading edge of the wing. Probably by an inch or so.

 

[noahhl]

I've used a cheap ultrasonic leak detector (they're available on Amazon for 40 or 50 bucks) to find firewall (and other) leaks when a bright light isn't enough. Stick the transmitter in the cabin, close the canopy, and then go sniffing w/ the detector.

 

[rv8ch]

Obviously you know know this, but any CO in cruise is not good. If you want to know why, here's a video I found very helpful:

I have read here on VAF of others that have had CO come in from the back of the fuselage due to some low pressure areas there. Not sure if this is a problem on the 14 or not, but you might check to see if opening the cabin air vents, which should increase cabin pressure, has a positive effect on the CO reading.

Also, are your exhaust pipes the standard length? If they are "too short" they might allow exhaust fumes to flow in unusual ways that get into the cabin.

Another area is check the junction between the flaps and the fuselage.

 

[bjdecker]

Update 2/26/2024 -- Leak Found.

We carefully inspected the fuselage along its entire length, specifically following the discoloration from the exhaust stream. The spaces between the flanges along the bottom of the F-01412A&B bulkheads where they attach to the F-01479 are open; the exhaust stream makes a 180° turn into the area between the rudder and fuse at this location (evidenced by combustion residue on the VS-410PP rudder bottom hinge bracket). The lower-than-ambient air pressure in the cabin "sucks in" the exhaust gas and into the cabin.

Also, CO levels were verified with portable CO detector. The response time of the Aithre 3.0 was a bit slower than the portable unit, but within a 1ppm or so when stabilized.

Taping over the gaps in the F-01412A&B flange resolved the CO issue. Will slather some neutral cure RTV or Proseal in the gaps as a permanent fix.

 

[Mike S]

Brian, first congrats on finding the leak.

And, secondly thanks for reporting back on the fix

 

[blaplante]

On the 6/7/9 the corrugated baggage aft panel (visible behind me in my pic) doesn't seal off the empennage, because of the corrugations. Filling those gaps with foam helped with my drafts. Not sure if the 14 has the same design. I'm also guessing your canopy isn't sealing as well as it could - hence why the cockpit is low pressure and sucking air in via the tail.

 

[bjdecker]

On the 6/7/9 the corrugated baggage aft panel (visible behind me in my pic) doesn't seal off the empennage, because of the corrugations. Filling those gaps with foam helped with my drafts. Not sure if the 14 has the same design. I'm also guessing your canopy isn't sealing as well as it could - hence why the cockpit is low pressure and sucking air in via the tail.

Bernoulli's principle predicts why the cabin interior is a lower pressure than the free-stream/ambient; no changing that. However, I do think that some additional work could be done along the front and sides of the canopy-to-airframe interface -- we'll add that to the list now that we've addressed "The Big Leak"

 

[BobTurner]

On the 6/7/9 the corrugated baggage aft panel (visible behind me in my pic) doesn't seal off the empennage, because of the corrugations. Filling those gaps with foam helped with my drafts. Not sure if the 14 has the same design. I'm also guessing your canopy isn't sealing as well as it could - hence why the cockpit is low pressure and sucking air in via the tail.

Yep, the -14 and the -10 have the same design, and once I stuffed some insulation under the corrugations my rear seat passengers stopped complaining about the cold draft on the back of their neck.

 

[MekMouse]

Thanks for bringing this up and showing the fix. I will definitely address this shortcoming when replacing those LCPs on my -10

 

[bjdecker]

This particular -14A is equipped with the YIO-390-A and 4-2-1 exhaust setup. I'm wondering if this exhaust configuration, plus the distance from the exhaust exit to the skin (instead of down-turned pipes) contributes to the incursion of CO.

Also, of note - I looked at a friends RV-8A and noticed similar exhaust markings at the bottom of the Vertical Stabilizer/Aft-most bulkhead at the "mouse-hole." I mentioned it to him, and he doesn't have a CO detector installed...makes me wonder...

Anyone with an EXP119 equipped -14A care to weigh in with their CO readings? Anyone else seeing CO readings that make them cringe a bit?

"...It's a gas gas gas..."

 

[bjdecker]

I'm not talking about the wing/fuse interface. This gap (on the -10) is at the bottom corner of the fuselage in front of the leading edge of the wing. Probably by an inch or so.

I don't have a -10 to look at, but I don't see what you're describing on the -14.

 

[rvbuilder2002]

Yep, the -14 and the -10 have the same design, and once I stuffed some insulation under the corrugations my rear seat passengers stopped complaining about the cold draft on the back of their neck.

Or… maybe the issue could have been resolved by focusing on the actual root cause.
In this post from another thread I explain
the issue that I think is the cause for a lot of unwanted leakage into RV cockpits.

Heat in the cockpit - or lacking heat in the cockpit!

I posted about a trip we made where my wife and I froze but didnt want to get off topic on the weekend thread. We were so cold we had to land to warm up. My carbon monoxide detector said it was 21 degrees inside. We warmed up at the airport, found a hole in the clouds and were in sunshine the...

vansairforce.net

 

[PCHunt]

The takeaway of this thread and Dr. Blue's presentation for me, is to get and use a low-level CO Monitor in the cockpit!

I have one, and I have a problem with CO in the cockpit. As soon as I lean the mixture, the CO goes WAY down. Will be getting exhaust extenders, and sealing the tail-most bulkhead!!

 

[LR60]

I would like to seal up this area to possibly eliminate the exhaust smell I get when transitioning to low power during descent ( I get the smell but not a CO alert), my concern is if water will be able to drain out tail cone?

After flying in moderate rain I have noticed water draining out this area while sitting in hanger the next day, I believe the water comes in around the empennage fairing.

 

[bjdecker]

I would like to seal up this area to possibly eliminate the exhaust smell I get when transitioning to low power during descent ( I get the smell but not a CO alert), my concern is if water will be able to drain out tail cone?

After flying in moderate rain I have noticed water draining out this area while sitting in hanger the next day, I believe the water comes in around the empennage fairing.

I think a small drain hole (<3/16") in the lowest point in the fuse would be fine and not contribute to CO in the cabin.

The issue I described originally was caused by a large number of voids in the bulkhead-to-skin interface in an area of low pressure (the "cove" in front of the rudder). A hole drilled in the fuselage at 90° to the slip stream/boundary layer would not "suck in" too badly --- it might generate a whistle (think coke bottle).

 

[LR60]

Ok I just attacked this area today (flashlight showing gaps) with proseal.

Is that all you had to seal up to stop the CO intrusion?

Thanks

 

[bjdecker]

Ok I just attacked this area today (flashlight showing gaps) with proseal.

Is that all you had to seal up to stop the CO intrusion?

Thanks

This was the source of the CO incursion on our -14A. YMMV of course…

 

[descro]

Has a leak check of the entire firewall area been done using a bright light in a dark hangar?

Hi. I've seen your posts and recognize experience talking. If I may, I would like to ask you a firewall question. Building an RV14a and adding a doubler to the firewall (to support remote Air Wolf Oil Filter).

Threaded fasteners vs Cherry Max vs Monel - have you wisdom to share about which method is best at the firewall? (ideally, I will capture the firewall angle (photo) and make gussets for the corners of the doubler.

Thank you,

Des

 

[bjdecker]

Hi. I've seen your posts and recognize experience talking. If I may, I would like to ask you a firewall question. Building an RV14a and adding a doubler to the firewall (to support remote Air Wolf Oil Filter).

Threaded fasteners vs Cherry Max vs Monel - have you wisdom to share about which method is best at the firewall? (ideally, I will capture the firewall angle (photo) and make gussets for the corners of the doubler.

Thank you,

Des

Dan H can weigh in on this, but from a fire resistance perspective you should use Stainless/Steel, Monel over anything aluminum.

 

[rvbuilder2002]

Hi. I've seen your posts and recognize experience talking. If I may, I would like to ask you a firewall question. Building an RV14a and adding a doubler to the firewall (to support remote Air Wolf Oil Filter).

Threaded fasteners vs Cherry Max vs Monel - have you wisdom to share about which method is best at the firewall? (ideally, I will capture the firewall angle (photo) and make gussets for the corners of the doubler.

Thank you,

Des

If the firewall is built per the plans up to this point, it has all of the stiffeners and bracketry attached using aluminum rivets so they could be used for the doubler and gussets as well, but any of the hardware choices you mentioned could also be used.

 

[mturnerb]

Hi. I've seen your posts and recognize experience talking. If I may, I would like to ask you a firewall question. Building an RV14a and adding a doubler to the firewall (to support remote Air Wolf Oil Filter).

Threaded fasteners vs Cherry Max vs Monel - have you wisdom to share about which method is best at the firewall? (ideally, I will capture the firewall angle (photo) and make gussets for the corners of the doubler.

Thank you,

Des

I used AN bolts/3M fire seal: