What's new
Van's Air Force

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

Fuel problems

Hotscam

Well Known Member
The fuel pressure running the mechanical pump varies between 22 and 26 psi in a 1 second interval. Is this normal or a bad relief valve ? Looks pretty nervous.
When switching the E pump on the reading is very stable 28 psi.
I purchased the 10 a year ago and it was like this all the time.

Another problem is occasional short pressure drops to zero only when flying 5000 ft and higher. Pressure return immediately but the Garmin lady starts shouting.
This happens 2-3 times an hours and makes me nervous.
The relation to altitude makes me suspect the sensor.

Any thoughts?
 
If it was a sensor , wouldn't the pressure fluctuate when the e pump was being used as well.
There are some post on here talking about fuel pressure fluctuation on climb that maybe some interesting read. If I knew how to link them I would find them for you.
 
Might this be bubbles in the line from the suction side out to the fuel system? Going to zero occasionally seems less like a snubber fluctuation and more like a pump loosing it's prime momentarily.

I had a small leak in the tunnel past the aux pump but before the engine pump. I would get a slight fuel smell whenever I had the boost pump on for more than a minute or two. I didn't notice any pressure fluctuations with the boost pump off, but also never looked closely. I'm sure there was at least some minimum air draw when the line was under suction.

An air leak between the tank and the aux pump that draws in under suction might be a possibility? Does this happen on both tanks? Tank full or near empty?
 
Fluctuation

Might this be bubbles in the line from the suction side out to the fuel system? Going to zero occasionally seems less like a snubber fluctuation and more like a pump loosing it's prime momentarily.

I had a small leak in the tunnel past the aux pump but before the engine pump. I would get a slight fuel smell whenever I had the boost pump on for more than a minute or two. I didn't notice any pressure fluctuations with the boost pump off, but also never looked closely. I'm sure there was at least some minimum air draw when the line was under suction.

An air leak between the tank and the aux pump that draws in under suction might be a possibility? Does this happen on both tanks? Tank full or near empty?

I had an Arrow once that had this problem almost exactly. Turned out finally to be the fuel pump itself leaking air on the suction side but never leaked fuel, only air sucked in while pumping. It was nearly impossible to find. We were quite certain it was air, but with no leaks, how to find it?

I used some temporary fittings and clear plastic hose at various points in the system to run at low RPM while observing the fuel in the lines and then moving the line to another location until we found that there was no air in the line to the pump but air in the line after the pump. Replaced the pump ($1200) and problem resolved.

Be sure to have someone in the cockpit and someone with a fire extinguisher at the ready...clear plastic hose is definitely not the thing to be casual about with fuel in them under pressure. We only replaced one line connection at a time with clear plastic until we found the problem.
Ed
 
Fluctuation

Might this be bubbles in the line from the suction side out to the fuel system? Going to zero occasionally seems less like a snubber fluctuation and more like a pump loosing it's prime momentarily.

I had a small leak in the tunnel past the aux pump but before the engine pump. I would get a slight fuel smell whenever I had the boost pump on for more than a minute or two. I didn't notice any pressure fluctuations with the boost pump off, but also never looked closely. I'm sure there was at least some minimum air draw when the line was under suction.

An air leak between the tank and the aux pump that draws in under suction might be a possibility? Does this happen on both tanks? Tank full or near empty?

I had an Arrow once that had this problem almost exactly. Turned out finally to be the fuel boost pump itself leaking air on the suction side but never leaked fuel, only air sucked in while pumping. It was nearly impossible to find. We were quite certain it was air, but with no leaks, how to find it? The problem would only show up when running the engine without the boost pump and go away with running the boost pump but the engine pump and boost pump are in series.

I used some temporary fittings and clear plastic hose at various points in the system to run at low RPM while observing the fuel in the lines and then moving the line to another location until we found that there was no air in the line to the boost pump but air in the line after the pump. Replaced the boost pump ($1200) and problem resolved.

Be sure to have someone in the cockpit and someone with a fire extinguisher at the ready...clear plastic hose is definitely not the thing to be casual about with fuel in them under pressure. We only replaced one line connection at a time with clear plastic until we found the problem.
Ed
 
The sensor is a UMA N1EU70A

This problem has already taken many hours but no luck.

It happens on both tanks and as I also suspected air leaks I checked all piping and hoses with a vacuum pump. All fine.

As a last resort I purchased a new fuel pump but no change.

I assume the snubber may reduce the pulses on the indicator but not the dropouts. Where do I find info about these and where to buy.

I am tempted to replace the sensor as the relation to altitude could be a clue.
The varying pressure happens all the time independent of altitude but dropouts only happen at 5000 ft and higher

But I hate to buy more new expensive parts that do not solve the problem.

Thanks for thinking with me
 
When the FP dropped to 0, even for a second, did your engine stutter? Mine did at 10 PSI. New FP fixed that for a while. About 250 hours later it was fluctuating again bit always remained above 20 PSI. Changed out the pump again, Spruce warranted it, and its been fine ever since.

With a 10 year old pump, I’d have no problem taking a chance with a new pump. Might not fix the problem, but worth a try.

Sorry, just re read the last post and you bought a new pump already.

Sounds like air getting in the system, as said previously.
 
The fuel pressure running the mechanical pump varies between 22 and 26 psi in a 1 second interval. Is this normal or a bad relief valve ? Looks pretty nervous.
When switching the E pump on the reading is very stable 28 psi.
I purchased the 10 a year ago and it was like this all the time.

Another problem is occasional short pressure drops to zero only when flying 5000 ft and higher. Pressure return immediately but the Garmin lady starts shouting.
This happens 2-3 times an hours and makes me nervous.
The relation to altitude makes me suspect the sensor.

Any thoughts?

Firstly, when Scott asks you if you have a "snubber" on your fuel pressure sensor hose I assume he means a restrictor which is in effect just a narrow orifice usually installed in the fitting at the bottom of the fuel sensor hose to reduce the magnitude of pressure fluctuations. It fits inside a standard fitting so you will not know if you have one without unscrewing the bottom fitting and checking. But my guess is that it is probably installed but you don't know because you didn't build the aircraft.

At any rate I don't think that the existence (or not) of a restrictor in the fuel pressure hose has anything to do with your problem. Nor do I think that a faulty sender is your problem from what you have said.

In my experience a mechanical fuel pump pressure of 22 - 26 psi sounds pretty normal to me. When you ask whether these pressures might be caused by a "bad relief valve" what do you mean? A boost pump pressure of 28 psi also sounds normal.

Please advise if your engine misses when your pressure drops to zero. That is crucial information you have not provided.
 
Last edited:
When the restrictor is screwed inside the hose I have not seen it.
Will have to take it apart. The sense hose is directly screwed into a sensor manifold block with the sensor on top.

The pressure variation does not look normal to me moreover as it stops with both pumps.

When dropout of pressure occurs it takes a fraction of a second but drops all the way to zero and returns strauight away.

I don’t think the engine stutters but with ANR on and the Garmin making loud alarm noises it is difficult to say for sure. I guess you would feel the power drop as well which I don’t

On the pump assembly in the tunnel there is a presuure reliaf valve which I assume only works when the E pump is on, but what if it leaks to the return at lower pressures ?

The reproduceable relation of the drops to altitude puzzles me.
 
I can't say it's the issue, but it may be suction restriction between the tanks and the main pump. This would explain the altitude effect. Higher altitude means lower air pressure, and lower air pressure means easier for fuel to vaporize. Without the boost pump, the engine pump must pull fuel from the tanks, and the only thing "pushing" on the tank side is ambient air pressure and gravity. Too much line loss in the piping could mean that the fuel is borderline to vaporizing. Get a flash to vapor, momentarily starve the pump, flow rate drops, line loss drops, and pressure at the pump rises, no more vaporization. Flow increases, repeat the cycle.

If this is the issue, you might be able to note very subtle changes between a full tank and a near empty tank. With a full tank, you've got more pressure from the greater fuel depth. The onset of the problem should occur at a *slightly* higher density altitude than with low fuel level.
 
When the restrictor is screwed inside the hose I have not seen it.
Will have to take it apart. The sense hose is directly screwed into a sensor manifold block with the sensor on top.

The pressure variation does not look normal to me moreover as it stops with both pumps.

When dropout of pressure occurs it takes a fraction of a second but drops all the way to zero and returns strauight away.

I don’t think the engine stutters but with ANR on and the Garmin making loud alarm noises it is difficult to say for sure. I guess you would feel the power drop as well which I don’t

On the pump assembly in the tunnel there is a presuure reliaf valve which I assume only works when the E pump is on, but what if it leaks to the return at lower pressures ?

The reproduceable relation of the drops to altitude puzzles me.

Neither my 6 nor my 10 have restrictors in the FP sensor line. My readings are quite stable. I occasionally see some moving around between 22-25 (climbs, switching tanks, etc.), but never something that swings across that range once every second, back and forth. I have seen occasions where it will drop to around 14-15, but it takes quite a few seconds to get there and then it recovers over a few seconds or longer (likely due to bubbles,boiling, etc.) and it never occurs below 8K'. This only happens on the 6. The 10 never drops below 22, unless I get hot in a long climb and is pretty much expected.

If you are dropping to 0 and returning to 25 in the course of 1 second with otherwise no major movements or abnormal FP behavior, this IMO points to an instrumentation problem. Floating around between 22 & 25 is tougher to pinpoint a cause, but but the consistency of the frequency leads me to think instrumentation vs actual changes in pressure. I have NEVER seen a highly consistent FP oscillation like that. Vapor lock, boiling, etc. rarely appears and then disappears in 1 second.

Larry
 
Last edited:
The fuel pressure line snubber is, in my understanding, not to stabilize readings but to limit the volume of gasoline being fed to a fire if the line ruptures and the exhaust lights it off.

I'm chasing a fuel pressure drop issue similar to this which is undoubtedly vapor lock, and so far seems to have responded to removal of the original plumbing in my kit as-purchased including some pretty iffy 40 micron JEGS billet filters under the seats and some 90 degree flare fittings. So far it seems like the new teflon hoses and Holley hi-flow filters are a good fix (thanks, Aircraft Specialty) but won't know for sure until I burn off more avgas and replace with 93 mogas on subsequent flights.

My current pressures are 23-24 psi in cruise at 5k, rising to 26 with the electric boost pump. System is Bendix injection with no return lines.

My original issue was a fluctuating drop in pressure on the left tank only down to about 10psi with a return to normal readings for awhile after hitting the boost or switching to right tank and back to left again, repeating every 30 seconds or so. At one point I let it go to see what it would do, and it went to near zero, red EGT bars (lean!) and engine most definitely in idle-cutoff territory. Recovered immediately with boost pump.

Changing to 100LL reduced the severity of the problem on the left side, but did not relieve it. Re-plumbing seems to have made the difference. But as I said, still running 100LL/mogas blend now and need to work my way to pure mogas to be sure it's gone for good. Right side never did act up although plumbed identically to left.

This will sure get your attention when it happens. The feeling of having fuel unavailable without the boost pump running is not a comforting one in early in Phase 1 or any other time. Now I have a suddenly inop Red Cube and I need to see if I bumped some wiring while wrenching in the tunnel on fuel lines or if it's T. U. Fun times.
 
When the restrictor is screwed inside the hose I have not seen it. Will have to take it apart.

The restrictor is typically installed in the fitting where the fuel pressure hose connects to the pump output.

On the pump assembly in the tunnel there is a pressure relief valve which I assume only works when the E pump is on, but what if it leaks to the return at lower pressures ?

With the boost pump OFF, there is no positive pressure at the pressure relief.
 
Pressure works too.

Turned out finally to be the fuel pump itself leaking air on the suction side but never leaked fuel, only air sucked in while pumping. It was nearly impossible to find. We were quite certain it was air, but with no leaks, how to find it?

Ed

You method seems to be quite effective. I recommend pressure, drain and use air pressure between the servo/carb and fuse bulkhead fittings. I use a shutoff, 0-40 psi gauge, pressure regulator and add air at the bulkhead to 30 PSI and shut off the valve. A sealed system will hold that easily for 24 hrs. Even a tiny leak will cause a steady pressure drop. Then use snoop etc to find the air leak. I found my leak, and two friends found theirs. All in odd places.

Your technique, Ed, to narrow the field - is also effective in reducing the scope of "opportunity". i.e. fuel valve leaks.
 
The fuel pressure line snubber is, in my understanding, not to stabilize readings but to limit the volume of gasoline being fed to a fire if the line ruptures and the exhaust lights it off.

SNIP.

The restrictors do also serve to stabilize readings, owing to the analog to digital conversions going on to display pressure. After installing a new panel (G3X Touch) I had oscillating manifold pressure indications. The pressure indications varied cyclically from, say, 19 to 24, about once every couple seconds. By changing RPM by a bit either way, I could eliminate these oscillating indications.

I knew for a fact that the mean pressure was not oscillating at the observed frequency since there is no physical reason for it to do so. I believe it has to do with a beat frequency between the data sampling frequencies vs the cyclical (one cycle every 2 rotations) real variation of intake manifold pressure. (It would be easy to fix this in software or electrical hardware, but that would be up to Garmin.) I fixed it by decreasing the size of the restriction in the MAP pressure line. I do not know what the internal sample rate parameters are for the Garmin unit, nor do I know if there are analog damping algorithms built in before the A/D portion. I gave up a little bit of response time, but not noticeably.

I have to assume the same thing could be happening in the fuel pressure line.

To really know, one needs to record the analog signals coming from these sensors, something I've threatened to do but haven't.
 
The G3X system seems to be more sensitive to pressure pulsation and fluctuations than some of the other EFIS systems. Probably because of the way they analyze the sensor output.

I think Garmin specifically specifies in their install manual, that one be installed. When we upgrade our RV-10 demonstrator to the G3X panel, we initially had erratic fuel pressure readings because for the previous engine monitor we had only a restrictor fitting at the sensor hose connection at the output of the engine driven fuel pump.

Once the specified snubber was installed, the fuel pressure indications were stable and normal.

Note: Garmin's official name may not be snubber... I don't remember what they call it, but if you can't find it in the install manual I am sure that Justin at G3X Support could help you out.

[email protected]

EDIT: after seeing Alexs' post it makes me wonder if I am thinking of a MP indication problem, and not fuel pressure. It was almost 3 years ago...
At any rate, contact Justin, he will be able to help you confirm if your install is done properly.
 
Last edited:
A word of caution on MAP snubbers. I split the MAP line and use two - one which provides MAP pressure to the electronic ignitions (I happen to have LSE), and another which provides pressure to the G3X. The hole diameter and length of the restriction I put in-line to the G3X's sensor was quite small, and I didn't want the ignition to have a delay in retarding the timing upon rapid throttle up.
 
Restrictor and Snubber

I have an IO-360-A1B6 and G3X Touch (GEA 24) and observed continuous fluctuation in fuel pressure when the engine was running (+2/-2 PSI around 27PSI median). Engaging the electric boost pump would raise the reported pressure and smoothed out the variations.

Subsequently I installed an Omega Engineering PS-8E at the transducer end of the fuel pressure line (Van's VA-102, IIRC) which resolved the issue.

In the FWIW category, I didn't observe this problem with my Advanced Flight Systems AF-3400EM with VDO Pressure transducer in prior aircraft (Same engine)

Cheers!
 
FWIW, I installed Ross' 6 into 1 MAP manifold, with restrictors in each line at the primer port, and a Y-connector to split the MAP to the ignition and the EFIS. Added cost, weight and complexity but it's smooth :rolleyes:

I don't do rapid throttle-up - because crankshaft pendulum counterweights.
 
Back
Top