Strange Fuel Flow and Pressure Oscillations

[romair]

Hello
I am trying to figure out some very strange oscillations in my fuel flow and pressure. The aircraft is a 1989, M20J, original engine, 1850 hours on it, A3B6D.
Fuel flow problem: As I lean from rich of peak to lean of peak, I notice 2 very peculiar behaviors that I cannot explain, especially as I get close to peak EGT. First, the fuel flow gets stuck at a certain value, despite me leaning. I can tell that the engine is getting leaned because the EGTs go up, yet the fuel flow does not budge. After a while, the fuel flow starts oscillating significantly, by up to 0.5 gph, without me touching the mixture knob at all. It oscillates so much, that I cannot tell when I reach peak EGT on each cylinder, so I cannot do the GAMI lean test. The fuel flow also oscillates even without actively leaning, but it is more pronounced near peak EGT and lean of peak.
Around 00:29 you can start seeing the problem, as I am leaning and get to around 10.2 gph. EGTs rising, but nothing on the FF.
00:30:39 finally the drop in fuel flow on the monitor
For the next 30 seconds the fuel flow varies by up to 0.5 gph, with me only leaning (never going back rich)
This behavior is very reproducible, and as I said those oscillations are so high that I cannot lean by 0.1 gph in order to get the GAMI lean test. Fuel transducer has been replaced and the avionics shop checked the connections - no change.

Fuel pressure during the same period - the fuel pressure also oscillates between a low of 24.1 and a high of 27.1. This is also quite strange. New fuel pump and overhauled injection system. The fuel pressure oscillations do not seem to correlate with fuel flow oscillations. Also, I tested this by using the boost pump which gave me a steady pressure of 27, yet I continued to have the fuel flow oscillations.

Any thoughts? Could this still be a bad connection somewhere? Could I be drawing in air from somewhere in the fuel system (fuel selector valve, gascolator) that would make it look like this?
Thanks
Stefan

 

https://www.savvyanalysis.com/flight/643936/59db91fe-753e-439c-895d-2b8a97fee127

[takair]

Really odd. I wonder if you have multiple things going on. You don't mention the type of monitor you are using. I'm no expert in this, but have a couple of thoughts. The fuel injector servo is measuring airflow and making minor adjustments to fuel flow. If you alter the speed of leaning (slow down or speed up too much) you may be interacting with this and seeing some anomalies. The EGT probes typically react at a different rate than the fuel flow and there may be some slight catch up. In other words, when the fuel flow goes flat, the true EGT may be near peak already, but not yet registered. That said, it seems your problem is the noise in the fuel flow to enable a constant rate of leaning. Try leaning at a constant rate of EGT, say 1 degree per second and see if the flat spot in fuel flow changes. It does appear that the EGT bumps around a little with the pressure. I also notice that EGT three does some strange moves at around :36. Not sure if this is a clue. Partial plugged injector? That said, I don't know if it would cause the rapid variation. It might be handy to see if someone else can post similar data for you to compare.

[Bob - S50]

Hello

I am trying to figure out some very strange oscillations in my fuel flow and pressure. The aircraft is a 1989, M20J, original engine, 1850 hours on it, A3B6D.

Fuel flow problem: As I lean from rich of peak to lean of peak, I notice 2 very peculiar behaviors that I cannot explain, especially as I get close to peak EGT. First, the fuel flow gets stuck at a certain value, despite me leaning. I can tell that the engine is getting leaned because the EGTs go up, yet the fuel flow does not budge. After a while, the fuel flow starts oscillating significantly, by up to 0.5 gph, without me touching the mixture knob at all. It oscillates so much, that I cannot tell when I reach peak EGT on each cylinder, so I cannot do the GAMI lean test. The fuel flow also oscillates even without actively leaning, but it is more pronounced near peak EGT and lean of peak.

Around 00:29 you can start seeing the problem, as I am leaning and get to around 10.2 gph. EGTs rising, but nothing on the FF.

00:30:39 finally the drop in fuel flow on the monitor

For the next 30 seconds the fuel flow varies by up to 0.5 gph, with me only leaning (never going back rich)

This behavior is very reproducible, and as I said those oscillations are so high that I cannot lean by 0.1 gph in order to get the GAMI lean test. Fuel transducer has been replaced and the avionics shop checked the connections - no change.

Fuel pressure during the same period - the fuel pressure also oscillates between a low of 24.1 and a high of 27.1. This is also quite strange. New fuel pump and overhauled injection system. The fuel pressure oscillations do not seem to correlate with fuel flow oscillations. Also, I tested this by using the boost pump which gave me a steady pressure of 27, yet I continued to have the fuel flow oscillations.

Any thoughts? Could this still be a bad connection somewhere? Could I be drawing in air from somewhere in the fuel system (fuel selector valve, gascolator) that would make it look like this?

Thanks

Stefan

 

https://www.savvyanalysis.com/flight/643936/59db91fe-753e-439c-895d-2b8a97fee127

I'm not a mechanic.

 

First, if you are able to operate LOP you don't really need to do a GAMI test because you probably don't need GAMI injectors and you can save the money.

 

Second, it could be an indicator problem or an actual problem.

 

Since you seem to have both pressure and flow fluctuations, assuming you are using two different transducers for the measurements, I would assume it is an actual problem.  If it was one transducer I might lean more toward an indication problem.

 

Since you've changed the fuel pump that most likely rules that out.

 

You said using the boost pump you get a steady fuel pressure.  If I remember right the transducer sits between the boost pump and the fuel pump.  The boost pump would most likely pressurize the line to the transducer and thus give you the steady indication.  That being the case, I would look for a leak prior to the fuel pump.

 

Do you have any fuel leaks between the transducer and the injectors?  I would think an air leak prior to the fuel pump (line, selector valve, gascolator) could cause the fluctuations.  A leak after the fuel pump should also result in a fuel leak.  If there is not leak between the fuel pump and the injectors I would turn on the boost pump (on the ground engine off) and look for leaks prior to the fuel pump.  If none are found I would look for leaks between the tank and the boost pump.

 

I would also think an obstruction could cause the problem.  You might try checking the filters on the gascolator and the fuel servo.  Could also be something inside the servo causing it to meter unevenly.

 

As a pilot, that's all I can think of.

 

Bob

[carusoam]

Based on Bob's points above...

One thing that can be checked regarding the health of fuel pumps...

They have drains to the outside for various modes of failure. Both seals or diaphragm type of leaks.

Check the drains for any fuel drips or build up of dark blue deposits exiting the drains. There shouldn't be any..?

There is a thread about cleaning out the FF paddle wheel. This may help when the FF indication seems to be an issue.

Best regards,

-a-

[romair]

My system is as following: fuel selector valve, gascolator, electric fuel pump, engine fuel pump, fuel flow transducer, bendix rsa system with the fuel pressure being a T connection at the RSA system., then into the firewall to the engine monitor. The rsa system was overhauled recently hoping it would change the problem. The fuel pump has about 100 hours on it, also was changed hoping it would solve the problem. I have the MAP 50 engine monitor. I had the avionics shop change the fuel transducer in case that's what was causing the fluctuations...no luck. The fluctuations occur on both tanks, which makes me believe it would be either the fuel selector valve, gascolator, electric pump or any of the lines as a problem. However, no sign of fuel leaking whatsoever. It is really strange...no blue staining anywhere. The fuel overboard lines look ok, not clogged. The engine does not like it lean of peak too much...it will run, but pretty rough and last couple of times it stumbled.

Is there a way to check what would be leaking without taking every thing apart? I was thinking to bypass the fuel selector valve with a hose, take it up see if any changes. I could do the same with the gascolator, as well as the electric fuel pump. Not the safest way to do things, but I think the risk is small if it's a one time deal and I stay around the airport.

Thoughts?

Stefan

[carusoam]

Another known issue for fuel transducers is the straight length of tube into and out of the sensor. The recommendation is something like 5 or 10X the diameter. Gentle curves are OK. Close proximity 90d elbows or Ts are not...

Definitely check with your mechanic for best ways to try possible answers.

It is not recommended to re-plumb and test anything on your own. That is one of the certified planes vs home built issues.

I am not a mechanic...

Best regards,

-a-

[takair]

I think if you had a leak significant enough to draw air, you would see fuel leaking when you shot down. The idea about the 90 degree fittings at the transducer is a good one. Has the avionics shop looked at the shielded wires and bonding? Has the aircraft always done this?

[Bob - S50]

I'm assuming your plane has not always done this.  That being the case, did you change anything just before the problems started?

 

New plugs?

Magnetor work?

New alternator?

New xxx?

Adjusted xxx?

 

In the safety world, when there is an accident, the first thing we looked at was the last maintenance performed.

 

Bob

[romair]

No 90d fittings at the transducer. The fuel pressure fluctuations have been going on for a while, I initially thought it was a fuel pump problem. I just added the MVP50 so now I can see the fuel flow oscillations, so not sure how long it has been going on for. No new significant work that would explain it. It's really strange to be close to peal egt, continue leaning with raising egts yet no change in fuel flow, only to all of the sudden drop, then continued oscillations. I just purged the line to the fuel pressure transducer and will see if that changes the pressure oscillations, but it should not affect the fuel flow.

[carusoam]

If you are leaning smoothly/consistently, the FF will be consistently slowing...

Consider cleaning the sensor...

.1 GPH is a number that can be set. If you want 9.5, Set it to 9.5.... Do not accept 9.5 +/- .5

The sensor is typically disconnected sprayed with solvent. The solvent is collected and inspected to see what kind of dirt it may have on it.

It's a device you want to have working consistently.

How old is the sensor? Might it be worn out?

Calling the manufacturer may help.

Best regards,

-a-

[romair]

Brand new sensor. This happened since install. Called EI, they sent me another sensor, still the same problem. Oscillations occur even when I don't touch the mixture. Connections have been checked as well...hence my concern this is real, and wondering what it could be. However, no evidence of a significant fuel leak.

[Bob - S50]

Have you checked the mixture control connections for looseness?

 

I was wondering if engine vibrations could be causing the mixture input to the servo to move causing a fuel flow fluctuation.

 

Bob

[romair]

Yeah....did that too. It's tight. Looks like I'm either drawing air in (although no fuel evidence of a leak) in which case possible culprits would be fuel selector valve, gascolator or a line, vs a connection issue with the monitor. Not sure what's the best way to investigate if I have a leak...

[Bob - S50]

Dang!  You've tried just about everything!

 

Another idea came to mind.  I know you said you changed the engine driven fuel pump.  Is it possible there is something wrong with the drive mechanism for that pump?  I'm not sure how the engine turns the pump, but I assume it is gear or cam driven.  Is it possible that the engine gears/cam are worn or damaged allowing slop and inconsistent pressure?

 

You said the FF hangup occurs right around peak EGT.  Is that true regardless of power setting (55%, 65%, 75%) or does it occur around the same FF (i.e. 9.5 GPH) regardless of power setting?

 

Are there any sharp bends in the fuel line between the boost pump and the engine pump that could be causing cavitation?  Or is it possible the flexible fuel line has a partial collapse inside that is causing a restriction that in turn causes some cavitation?

 

Might be time to call Mike Busch and Savvy Aviation.

 

Good luck,

 

Bob

[sailcail]

We had a similar issue with pressure and fuel flow fluctuations.  Above 9k the fluctuations would start, about half a gallon an hour and 5-8 psi.  The EGT's would also fluctuate slightly.  If I leaned more than 20 ROP the engine would occasionally miss.  Turning the boost pump on would stop the fluctuations.

 

The engine driven pump was replaced (twice), boost pump replaced, fuel hoses replaced, gascolator, fuel selector and servo overhauled.  Two shops performed pressure and vacuum tests on the fuel system and found nothing.

 

Finally we replaced the rebuilt fuel servo with a different unit.  The shop inspected removed servo and found it was hanging up in a particular spot.  Evidently this is where it operated at higher altitudes.  Anyway, after the second servo was installed the problem was solved.

 

Hope this helps, good luck,

 

Simon

[romair]

Would turning on the boost pump stop the fluctuations In fuel flow as we'll or just in fuel pressure? This is becoming one expensive problem that I'm chasing

[Lionudakis]

It might be a long shot, but as you've gone through quite a bit already, I have found leaks (in a fuel system) using vacuum as opposed to pressure ; air would go in via vacuum, but no fuel out. (I believe this was in a cessna) Your pressure vs. flow sounds like air (although not fluctuating at the same time, but still not impossible as the reference points are usually far apart) . 

 

 

Last year I was alarmed to see my fuel pressure sitting at zero a few minutes after takeoff, and my newly installed G3 showing something like 38 GPH   with an O-360, My initial panic was that the fuel jettison was turned on, but mine was not so equipped. The boost pump cleared it up within a minute or so, I could not find and leaks and I could not reproduce on ground, the next flight same thing but only a few minutes after takeoff. The fuel flow indication returned to normal numbers long before the pressure returned to normal after turning on boost pump, The suspect was the engine driven pump was either sucking air, or in itself pumping air along with fuel. Air bubbles in a flow transmitter will alter the readings quite a bit. Changed the pump out and pressure increased, and problem gone for good. 

[romair]

Lionudakis- that's what I was thinking may be going on. Not sure how to test for that. The fuel pump was changed because of the fuel pressure fluctuations on the old pump...I think it is unlikely to get a new pump and show the exact same behavior. Replacing the servo may end up the last step, but I want to make sure I look into everything before I take such a step. I just overhauled the system and it was not cheap. The airplane has the original 1989 engine on it with 1850 hours. The fuel selector valve and gascolator are also original and have not been overhauled. I plan on going beyond tbo if possible, but in the same time not sink in a lot of money on an engine that is near overhaul.

[KSMooniac]

Are the hoses still original?  I would look there next... and perhaps change all of them out and just plan to re-use them at overhaul time, whenever that is down the road.  (I'm all in favor of flying beyond TBO if everything looks good)

[romair]

New hoses last year firewall forward. Original lines from the tanks to the firewall. No evidence of fuel leaking tho, including no smell.

[KSMooniac]

I've heard on issues with hoses where there is a slice of rubber from putting the fittings on incorrectly that periodically flap into the flow. Not sure how you can inspect for that, though.

Sent from my VS985 4G using Tapatalk

[sailcail]

Would turning on the boost pump stop the fluctuations In fuel flow as we'll or just in fuel pressure? This is becoming one expensive problem that I'm chasing

 

Yes, turning the boost pump on would stop the pressure and fuel flow fluctuations.  It would also allow me to lean the engine without any missing.  My belief is that the sticky servo would some how get in sync with the pulses of the engine driven fuel pump.  Turning on the boost pump would provide a steady pressure that the servo could handle better, just a theory.  It would be nice to borrow a servo and see if that fixed the problem.  Unfortunately, is quite a bit of work to swap it out.

[romair]

Yeah, not easy to find a servo to borrow. Interestingly enough, turning on the boost pump gets rid of the fuel pressure oscillations (I basically get 27 psi constantly), but does not get rid of the fuel flow oscillations

[KSMooniac]

Have you talked with EI tech support yet?  Do you have slow- or fast-response EGT probes?  If they are the fast-response style (much preferred) then I would expect to see EGT fluctuations to follow fuel flow fluctuations.  I don't see that, so it leads me to believe you have an instrumentation error, likely in the connections.  I would ask the shop specifically how they "checked the connections" because a visual check is likely not sufficient.  There might be a loose crimp in the pins at the connector, or an incorrect junction in the harness for the FF and FP connections.  EI might be able to shed some light on the situation before you engage the shop again, or spend money switching parts.

[romair]

I did talk to EI, multiple times. It got to the point that they gave in and sent me another fuel transducer. I'll call them again specifically about the connections and will also check them with my mechanic. I have the slow response probes, so all bets are off regarding how fast fuel flow correlates with EGTs. Again, it's very strange to be leaning the engine, watch the EGTs go up, yet nothing changing with the fuel flow for a full minute, then followed by oscillations, with values that are at times higher than the value where it was initially stuck at.