What's new
Van's Air Force

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

Fuel Flow Difference Between Tanks

iamtheari

Well Known Member
I finally got to test the fuel system on my plane tonight. I am curious about a difference I saw between the two sides of the system. I purchased my fuel lines (firewall to selector, selector to fuselage sides, and tanks to fuselage) from Aircraft Specialty. I used the kit stock Andair fuel valve, boost pump, and filter. The tanks were built to plans. Nothing comes to mind that would be different between the two sides.

However, the flow is definitely not the same. We started on the left, then did the right, and then did the left again. I disconnected the fuel line from the input to the fuel flow servo, selected a tank, and ran the boost pump. We got the pump and lines primed by flowing about 10 oz of gasoline out the line before we did the measurement. Then we did a 30-second timed run. The left side put out 26 oz in that time (24.3 gph). The right side put out 43 oz in the same time (40.3 gph). We tested the left side again and got the same results. The flow from the right side also appeared smoother than the left.

Other than an obstruction in the fuel lines somewhere, are there any likely causes to check for?
 
The first thing I would do is try that test again with the gas cap off. That will either confirm or eliminate a vent problem.

Not familiar with the 14 but does it have separate filters for each side? If so, that would be the second place I would check.

You might also peek down in the tank with a scope and make sure the pickup isn't partially blocked with construction debris i.e. a chunk of pro seal or a piece of tape.

Like previously mentioned, routing and # of bends or elbows will have something to do with it, but one side putting out 100% more than the other seems like a lot unless your plumbing is a real rats nest.

I know that FAR 23 doesn't apply here, but just in case you don't know, FAR 23 says that min fuel flow should be 125% of flow at max takeoff power.

Edit- had one more thought-

Did it have the same amount of fuel in each tank and was it more or less level? Might be a dumb question, But I've seen stranger things. Like I said, I'm not familiar with the 14, but If you only had a couple of gallons in the tank, I could see a scenario where you just discovered your unusable fuel load. Especially since you said the flow was kind of rough on the one side
 
Last edited:
You might have a leak somewhere in the left system, prior to the selector valve, allowing air to be sucked in. That might also explain the difference in the "smoothness" of the flow. Depending on how much fuel was in the tanks, the leak could be inside the tank (if a fitting is above the fuel level).
 
Andair?

There are some good suggestions above. The only other thing that comes to mind, from my armchair, is the Andair selector itself. I know they have a good reputation for quality, but perhaps there is an internal alignment issue or restriction of some sort? Diagnosis would require removal and blowing through to check.
 
Did you do an air pressure leak down test from the fuse bulkhead to the servo inlet? I did this to ensure all the fittings were properly sealed. I convinced two friends to do the same an each found a tiny leak. Just close off the system and test each side (switching the fuel selector) separately. When you test each opposing side, open that line so you will catch a leak in the selector.

Not only does this tell you if you have a leak, but with air pressure on it you can use snoop to find the specific leak. It may not be your fitting that leaks!!

Once you pass that test, do this. You should do the flow check with a restriction valve on the exit to reduce the flow to the FAA recommended amount - I forget if it is 1.25 or 1.5 max fuel and then see what pressure is generated. Also have a 1 ft section of clear tubing to see if there is a suction leak - on the pressure side of the valve.

You would be amazed how little air it takes to destroy suction and capitate the pump.

It is either a suction leak or restriction, these test will tell which. Then find the location.
 
Last edited:
A lot of good suggestions above. When I did this on the 10 last year, I got 60 GPH on both sides (AF"P pump/filter with Andair valve)

I will add the suggestion to possibly pull the filter as well, in case it caught a bunch of trash on the first run. Even the 43 number seems low to me.
Larry
 
Last edited:
The 14 kit has the fuel go tank to selector, pump, filter, then engine driven pump. So any pump or filter problem should affect both sides. The lengths and bends are the same on both sides, unless the Andair valve is internally a rat’s nest.

I haven’t built the vent lines yet so I just have an open fitting on each tank. I removed my painter’s tape for the test.

Didn’t find any gas leaking anywhere, but air leaking in is possible.

Thanks for the suggestions so far. Looks like I have some work cut out for me!
 
A lot of good suggestions above. When I did this on the 10 last year, I got 60 GPH on both sides (AF"P pump/filter with Andair valve)

I will add the suggestion to possibly pull the filter as well, in case it caught a bunch of trash on the first run. Even the 43 number seems low to me.
Larry

I calculated 26 gph as the 1.25 figure for my engine. So I am comfortable with 40 gph but not so much the 24.
 
The rigid tubes on the 14 from the selector valve to the fuselage, from fuselage to tanks, and valve to filter are all CNC bent 5052O aluminum. The discharge hose from the pump to the firewall is teflon. The common component to both sides is the valve, and forward. Possibly the valve wasnt fully in the detent position on the left side. Since the flow through the filter forward was acceptable on the right tank, it looks like a left issue.
Valve detent, vent maybe, tank strainer screen, possibly some debris from the tank got into a line.
I'd repeat the tests and verifiy it, then check the left side plumbing. Valve, left side tube, bulkhead fitting at the fuselage, the 'S' wing tube---make sure it wasnt kinked during install, then the tank strainer. I'd check the valve first to make sure when you select the left tank, the spool in the vale actually uncovers the ports. Make sure the selector handle is secure and doesnt rotate on its own, been known to happen.
Blow through all the lines--even the valve.

Tom
 
In addition to Tom's suggestion and if all those checked out and no air was leaking, maybe you would need to check the pickup line inside your left tank. This is obviously a bit more work considering your tank is installed but something is restricting the flow from the left and as others have said, the lines for left and right are pretty much the same.

As for the leak test, a Brake Bleeder And Vacuum Pump from HBF is an easy way to test any leaks in your system. With the Andair valve on closed, it will not leak air even the tiniest bit.
 
Borescope inspection of fuel pickup

Tight fit but I was able to get my Vividia borescope (minus blue sleeve) up through the drain valve fitting and get a good look at the pickup in my RV3.
 
I have seen a circular spiral drill shaving inside a fuel hose that caused a few partial power interruptions, then a total power loss with the M1 Cassutte crashing right in front of me...

I have more than once seen a flap of rubber inside of a hose formed by not getting the mandrel in properly when end fittings were installed.
This little rubber flap is no problem if flow pushes in to the outside, but it will act as a one way valve it pushed to the inside. If that flap is big enough it would totally shut off flow....
 
WELL, the RV14 has CNC bent rigid tubes from the valve to the tank strainer nipple. That nipple is just a AN816-6-6D with a screen. Easy to pull it from the tank once the fuel is drained. The only hose---at least in our configuration---is the pump discharge hose, and in this situation is down stream and common to both sides of the system.
Ari indicates the issue is just on the left side---so that tells me that its something only on that subsystem.

Tom
 
Right side is in line with typical flow numbers for the Andair in a -14, about 40 GPH. If you have the AFP pump in a -14 it's closer to 60 GPH for normal. Left side definitely has an obstruction or the valve not opening properly when set to left.
 
I tested it a bit more tonight. I put 4 gallons into the tank this time, to see if fuel level affects the results. And it does, a bit. The flow is consistently 30 gph when the tank has more than 2 gallons in it. But it’s still slightly aerated in appearance.

The selector valve is in the detent so, barring a defective selector valve, I have the tank correctly accepted.

I don’t see or smell any fuel coming out of the lines in the fuselage or between the fuselage and the tank.

In between tests, I take 10-20 seconds to empty my measuring bucket. When I start the next run, it takes a moment for the flow to get up to speed again. I can’t recall if it did that on the right side between test runs the other day.

I think an air leak is more likely than an obstruction, as the screens went in clean and the tanks were sealed since then. The fuselage lines were sealed from installation until the wings went on and the connecting fuel lines went on.

I dislike that the fuel line fittings inside the fuselage are so hard to get a wrench on. I do not foresee much fun in my immediate future thanks to those. :confused:
 
Close

I think you are very close to finding the issue. You have it narrowed to the left side for sure and the fact the appearance is aerated it indicates sucking air. I know what you mean about those fittings. If it is a defective valve that could be where the air is coming in.

Hope you find it soon!
 
I am going to re-suggest you do an air pressure test for the section between the fuse and servo. The fuel should be gently blown out. Then put a gage and valve on the inlet end, pressurize to 30 psi (use a regulator for safety) and leave it for a while. An air suction leak makes a huge difference and it is best to find the specific leak and fix it not to over tighten all the fittings. I have two friends that tested theirs as suggested and each found a leak on the suction side, one was in the factory filter housing.

You are clearly investing labor to solve this, so why not use a definitive test?
 
I am going to re-suggest you do an air pressure test for the section between the fuse and servo. The fuel should be gently blown out. Then put a gage and valve on the inlet end, pressurize to 30 psi (use a regulator for safety) and leave it for a while. An air suction leak makes a huge difference and it is best to find the specific leak and fix it not to over tighten all the fittings. I have two friends that tested theirs as suggested and each found a leak on the suction side, one was in the factory filter housing.

You are clearly investing labor to solve this, so why not use a definitive test?
I appreciate your suggestion. I've been contemplating the best way to set it up. Because of my experience chasing leaks between my test apparatus and fuel tank when I air-tested the tanks using the schraeder valve and balloon method, I don't want to base any decisions on whether or not the system holds air pressure.

I am leaning toward putting a small amount of air pressure on the fuel system and spraying soapy water on it, despite the irritation of having soapy water in my fuselage. I guess that's why there are drain holes in the fuselage. The soapy water test will show where any leaks are located without having to rule out leaky test apparatus.
 
I appreciate your suggestion. I've been contemplating the best way to set it up. Because of my experience chasing leaks between my test apparatus and fuel tank when I air-tested the tanks using the schraeder valve and balloon method, I don't want to base any decisions on whether or not the system holds air pressure.

I am leaning toward putting a small amount of air pressure on the fuel system and spraying soapy water on it, despite the irritation of having soapy water in my fuselage. I guess that's why there are drain holes in the fuselage. The soapy water test will show where any leaks are located without having to rule out leaky test apparatus.

Great idea, use a small artist's brush for applying it, and you won't have much mess. Soap bubble are about the most sensitive test you can get, provided all the fuel is blown out first.
 
I appreciate your suggestion. I've been contemplating the best way to set it up. Because of my experience chasing leaks between my test apparatus and fuel tank when I air-tested the tanks using the schraeder valve and balloon method, I don't want to base any decisions on whether or not the system holds air pressure.

I am leaning toward putting a small amount of air pressure on the fuel system and spraying soapy water on it, despite the irritation of having soapy water in my fuselage. I guess that's why there are drain holes in the fuselage. The soapy water test will show where any leaks are located without having to rule out leaky test apparatus.

Funny, I had the same "test apparatus" leak experience with my static rig too.

Yes, do a solid job with hardware for the leak test. I use a 30-50 psi gauge (~$8) with a tee, a 1/4" ball valve, and NPT- AN adapter for the connection. Then it is easy to soap test the external fittings to ensure they don't leak. Making the test rig is a time investment, but still should save much more time achieving a successful result. And it is reusable in the future after service. Otherwise it is a lottery.


I use the method Alex mentioned, a small brush (acid brush) , Dawn and DI water, I also have the store-bought stuff but the dawn works pretty well. Cleaning with distilled water as I go.

Note: pressurize from the tank end as the check valves in the servo will prevent back flow - the reason the fuel pressure sensors failed recently.
 
I finally got around to some testing. I hate when people leave a thread like this unresolved and I find it by googling, so even though it’s embarrassing I’ll go ahead and post the answer...

We put the air valve on the vent fitting on the tank and disconnected the fuel line from the fuel flow servo forward of the firewall. First we ran it open with some air to clean fuel out of the line. Then we put a balloon on the fuel line forward of the firewall and pressured it up. There was an audible whistle from inside the airplane. We followed the sound and came to the valve as the source.

But then my A&P friend who was in the plane looking for leaks felt the fitting between the left main fuel line and the selector valve rotate, meaning it was loose all along. Air was therefore able to leak in at that point when the boost pump was drawing fuel from the left tank, leading to my decreased and aerated flow.

My best guess is that, two years ago when I put the fuel lines and selector into the fuselage, I had an “out of sight, out of mind” moment or couldn’t get a wrench on it and thought I would come back to it later. Anyhow, the problem is solved. Fuel flow from the left tank is now >40gph and smooth.

Thanks for everyone’s help in this thread. I would have only been thinking about obstructions, rather than air leaking in, if I hadn’t posted. Of course, I wouldn’t have posted if I had tightened the fitting first. :)
 
Wouldnt be the first time that this has happened. The obvious thing to do first is check to make sure the connections are secure. We've heard of builder tearing apart the systems, including the tanks looking for a similar air leak, and finding a loose fitting at a valve.

If you are just mocking up the install, or doing a final and didnt tighten the fittings, TAG THEM, so you, or someone with good eyes can see the tags. Torque seal is great---IF the fittings were actually tightened before the torque seal was applied!

Seemingly insignificant things can create a major bad day-----be careful!

Tom
 
Back
Top