Fuel Leak

After my last flight, there was blue staining on both sides at the wing root and down the gear legs. Not a huge amount, but obviously needs fixing. Had the root panels off - no signs of leak. So seats out and tunnel panels off. Sure enough, the leak seems to originate in the tunnel and flowed out across to the wing root.

I should mention now that my fuel system is modified a little. UK regs require a gascolator but with the wing fuel drains being at the low point, it's a bit redundant. A mod has been approved to remove the inline filter and put the gascolator in the tunnel to act as a fuel filter only. Using the Andair version means easy cleaning by screwing off the bowl. I also use the Andair fuel selector. All piping is standard flared aluminium fittings.

Witness stains in the tunnel indicate that the leak is from the mid-section. This is good news as if it was forward by the pump or red-cube it would be a real PITA. I know that red-cubes have been a potential source of leaks but this does not appear to be the case.

Now here's the problem - I can't find the leak. This is bad news as a tunnel leak is potentially very serious. I initially just tried the fuel valve left/right to see if anything leaked. Then I tried again with the booster pump on. Nothing..... From what I can see, the leak has to be from the selector, gascolator or maybe the inlet to the fuel pump.

Could it be that the tunnel flexing in flight is causing a joint to weep? In which case it is going to be extremely difficult to track down. I think my next step is to take her out for a high power engine run and see if I can spot any drips. Obviously, though, I did that during the initial ground tests and all was well. Other than that, I could try tightening the joints but I don't want to end up over doing it. I could take them apart, check the flares and threads, clean and re-assemble but access is a bit tight with the panel and controls in place and I don't want to make things worse.

Any advice appreciated......

I initially had fuel leaking in the wing root around the fuel level sending units. I had used the neoprene seals provided with the sender. I changed them out to cork with a little Permatex gasket sealer and the issue is gone now for 500 hours.

Look closely around the bolts that hold the sending unit to the tank.

How about your tank vents, are they protruding enough into the air stream or could vented fuel be running back into the wing root cover?

I can't imagine there would be enough flex in the lines to produce leaking in the joints. I would inspect each joint for cracking in the flare if possible.

The first thing I would is to check all connections and see if any are loose. If not I would start disconecting each connection looking for bad or cracked flares. Access is a pain. On my ten I built in an access panel on the co pilot side but I also had a center console.

If you have a drain valve on the gascolator, I would suspect this first. Considering the locations, I don't see how a drain valve would be of any use.

Next would be the gascolator 'O' ring. As a cut ring is hard to see, just go ahead and replace it.

After the above two suspects, I'd guess you have a fuel line fitting problem. Assuming you already looked for a loose fitting, tightening of already tight fittings will not help. Under no circumstances should you attempt to "seal" the leak with RTV or other such stuff. The flex issue you describe is something to design around.

This leaves the fix as biting the bullet and remake the aluminum fuel lines and fittings, giving close attention to properly making the flared ends. If you are not using the right tool (like this; http://www.aircraftspruce.com/catalog/topages/flaringtools2.php?clickkey=4577 ) you will not be successful. Pay attention to how much you are "pulling the fitting together" when you tighten the flare nuts. This is a tight area and you can easily deform the fit and create a lead. For the flexing issue, avoid straight runs between hard mounted objects.

Carl

Leak is definitely in the tunnel, not the roots. No drain plug on the gascolator.

Guess my job tomorrow is to start disconnecting pipes and inspecting - but how can a leak be intermittent? I pressurize to 25psi with the electric pump and nada.....

I read about this somewhere. Wrap suspects areas with aluminum foil. Leaks will show up underneath. I havent had opportunity to actually try this method yet.

jchang10 said:

I read about this somewhere. Wrap suspects areas with aluminum foil. Leaks will show up underneath. I havent had opportunity to actually try this method yet.

Click to expand...

Even better is wrapping with white Kleenex. Even a very small amount of blue dye will be very obvious

Thanks everyone. Off to the hangar today. I'll try the tissue tip before I tear anything else apart!

Or I could just leave it. Doesn't seem to be leaking now - I mean, what's the worst that could happen?

paul330 said:

Thanks everyone. Off to the hangar today. I'll try the tissue tip before I tear anything else apart!

Or I could just leave it. Doesn't seem to be leaking now - I mean, what's the worst that could happen?

Click to expand...

I think we already lost one RV due to a fuel leak in the tunnel area, I would seriously recommend you find it and fix it before you fly.

Paul,
One additional thing to check. I assume you are using the Andair selector...
The input ports for the selector are held in place with 4 #4 screws. These screws should be peened in place after you have selected the orientation required for your particular installation. If these screws were not peened and have loosened up you may be leaking fuel around the connection to the selector body around the O-ring on the outlet port.
With the amount of fuel you have observed, you may not yet see any staining on the metal as it may be "washing" it clean.
Just a thought.....
Bill

paul330 said:

Leak is definitely in the tunnel, not the roots. No drain plug on the gascolator.

Guess my job tomorrow is to start disconnecting pipes and inspecting - but how can a leak be intermittent? I pressurize to 25psi with the electric pump and nada.....

Click to expand...

You can only pressurize with the boost pump from the pump to the firewall. If no leaks there, your leak is in the unpressurized portion from the tank to the boost pump. If you have a means of applying pressure to the fuel vent tube with a NON-electric pump up to no more than 1/2 psi (to not damage your tanks) you might then be able to see where the leak is.

Would UV dye be helpful in locating this leak? I found it to be very useful in locating a very small leak on my accessory case. Once I found the right combination of dye and UV light, it was obvious where the leak was.

Carl Froehlich said:

After the above two suspects, I'd guess you have a fuel line fitting problem. Assuming you already looked for a loose fitting, tightening of already tight fittings will not help. Under no circumstances should you attempt to "seal" the leak with RTV or other such stuff.

Carl

Click to expand...

I use SealLube on all npt fittings, carefully appied to ensure that there is none on the first couple of threads. It has helped me fix very pesky tiny weeping of fuel. Numerous pros have told me this is an acceptable practise. Certainly not rtv!

Thanks everyone for the tips. Seriously, I was joking when I said "let it go" - I was fully aware of the lost RV to the tunnel fire

Anyway.......

I have pressurized the system to the firewall. I have slightly pressurized the tanks. I can find absolutely no evidence of a leak anywhere So, I am beginning to think I am chasing ghosts. There were relatively large stains around the gear legs and belly and only minor ones in the cross bay and tunnel. I am beginning to think that there was some sort of spillage at the roots which migrated inwards rather than the other way round.

On my last flight, I went through a mountain pass and we had a couple of pretty large jolts in turbulence. I'm feeling that there was probably fuel that came out of the vents - I had about a 3/4 fuel load at the time. Now, if the vents are leaking, then it needs fixing but it's not nearly as serious or dangerous as a tunnel leak. Even with full tanks, though, there is no sign of a vent leak. I am therefore thinking this was a one-off spill caused by a couple of negative-G bumps.

I'm going to clean off the stains, button everything up and try a local flight to see if there is any recurrence.

Thanks again for the help. I'll let you guys know how it goes. Due weather and Christmas shopping, Wednesday looks favourite......

I wrap all my fuel joints in the tunnel and the wing roots with a small piece of white paper toweling held in place by a piece of tape. I inspect these whenever I have access and replace. If a single drop of dyed fuel finds its way past a joint, I will know where it came from.

9GT said:

I wrap all my fuel joints in the tunnel and the wing roots with a small piece of white paper toweling held in place by a piece of tape. I inspect these whenever I have access and replace. If a single drop of dyed fuel finds its way past a joint, I will know where it came from.

Click to expand...

Good tip - I'll do it next time I have the tunnel panels off. However.......

Just done a 50 minute flight. Steep turns, stalls, half a dozen circuits. Not a sign of any fuel leak anywhere. Go figure......

OK, two flights (one 1/3 tanks, one full) - nothing. Third flight, fuel stains around the roots and legs and smell of fuel.

The last flight, I again had some moderate turbulence. Could fuel leaking through the filler caps migrate to the roots and get in that way?

Fuel leaking through the filler caps will be going straight back in flight, doubt it would even make it to the wing root when parked.

You haven't by chance filled the tanks tight full in cold temps and move to warm hangar? The fuel will expand and leaks may become apparent.

It it possible you have a leak on the rear baffle or the root rib of the fuel tank... Both leaks could travel in front of the main spar to the root and into the floor of the cockpit.

You might have to pull the tank, you might actually try (with precaution) using fuel then in the tank to detect the leak.

Have you disconnected and physically look at the flare of each fitting? Over tightening a flared fitting will cut (smash) the flare right off inside the nut and it will leak

If the stains appear only when tanks are full, then you almost certainly have a leaking vent. It could be the bulkhead fitting isn't sealed up, or the flare fitting is defective. Wrap you tissue around the fitting where the vent exits the tank, and see what you have.

Have you looked at your fuel level senders? I had a leak from my left tank -- after one flight I had a strong fuel smell in the cabin. A quick inspection showed blue stains on the left bottom wing root fairing. Took of the top fairing and there were blue stains forward and aft of the spar in the wing root area but nothing outboard indicating a baffle leak nor inboard under the pilots seat. There was also nothing discernible from any of the fittings or line in the root area. Ended up wiping each part with a paper towel and bingo it was weeping from the sender. Took it off, applied some fuel appropriate permatex gasket maker to the sender and that fixed the problem.

I've looked in all the areas mentioned. I have just removed the neoprene sender seals and replaced with tank sealant. I can see no evidence of leaks anywhere - just stains down the legs and some very minor staining in by the spar and in the tunnel which appears to have migrated inwards - there simply isn't enough to account for the leg stains if it comes from the tunnel.

What is baffling me is that it doesn't always occur - even with full tanks. The only common thread I can find at the moment is turbulence but even then I can't find where it is coming from. It's only happened twice in 20-odd hours flying........

As suggested, I think it is time to start wrapping white tissue around the place. I won't be flying for a while so I'll get back to you with results in the new year.

Thanks again for the input.

Paul,
One additional item. Don't know if you built your tanks, but some of the q build tanks did not have the "B" nut INSIDE the fuel tank tightened. This allowed av gas to leak around the b nut and flare fitting from in the tank and exit the fuel vent line. The fuel level would have to be above the vent AN fitting, or sloshing around it.
Bill


http://www.vansairforce.com/community/showthread.php?t=74258&highlight=fuel+vent

Could fuel coming out of the vents make its way into the root area back up through the grommet in the bottom root fairing? Common sense says it will simply flow back in the airstream but flow patterns do strange things......

Paul, the vent fittings that Bill and I are referring to are in the wing root. There is a bulkhead fitting at the inboard end of the tank and a flare fitting attached to that to carry excess fuel out the bottom of the wing root. If either of these are leaking, it might produce the symptoms you are describing (especially the flare fitting).

OK, see where you are coming from. The root pass-through is not leaking because that would be visible all the time the tanks were more than about half full. But if, in turbulence, fuel got in the vent line and there was a leak around the nut, then this would end up in the root area - this would not be obvious on the ground.

I have to say that I don't think it's the case - the flares have been checked and the joint looks good. However, for elimination purposes, I'll tissue up the joint and see.

Not flying now till after the holidays so I'll get back with further results.