Fuel servo issues

[larrynimmo]

I bought my plane 6/2017...and there are no records ever of having the fuel servo rebuilt...and it functioned flawlessly.   Then in 9/2018 I installed a Lycomming “rebuilt” engine which came with a freshly rebuilt servo.  In April 2019 sent the servo back to get it repaired because the throttle became too stiff to operate....now after a 5 minute stop, I tried to restart the engine when the throttle wouldn’t move....and I ran the battery down enough to get a “hot shot” (first time ever). Eventually I got it started and flew it back to my home base and when I went to shut down I pulled out the mixture and it stayed running at 500 rpm...wouldn’t shut down...I shut the fuel selector to off and eventually no fuel to keep it running...tomorrow the servo is going to go somewhere!

[Yetti]

Check to make sure some fod did not go down the wrong hole.   One person had some rubber go down theirs.    Does not make sense that it is servo proper.   What kind of air filter?  When changed last?

[PT20J]

Don’t know if it’s related to your issues, but it appears that the coil spring that keeps the idle mixture adjustment screw from turning is missing.

Skip

Edited November 25, 2020 by PT20J
Added diagram

[carusoam]

Sure does sound unusual...

Lets invite the doctor to the conversation @M20Doc (fuel Servo question M20J)

Best regards,

-a-

[larrynimmo]

21 minutes ago, Yetti said:

Check to make sure some fod did not go down the wrong hole.   One person had some rubber go down theirs.    Does not make sense that it is servo proper.   What kind of air filter?  When changed last?

6 engine hours ago I installed a new K&N pleated filter...don’t see that as an issue... clean with no loose items

[larrynimmo]

21 minutes ago, PT20J said:

Don’t know if it’s related to your issues, but it appears that the coil spring that keeps the idle mixture adjustment screw from turning is missing.

Skip

Yes older servos did have that spring...purpose of that spring was to keep the mixture from coming out of adjustment

[Guest]

First thing, does the mixture arm hit the ICO pin?  If the controls travel fully, then sending the servo for a bench check seems proper.

Clarence

[PT20J]

7 minutes ago, larrynimmo said:

Yes older servos did have that spring...purpose of that spring was to keep the mixture from coming out of adjustment

That’s the spring’s purpose, but it should be on all servos. It’s on my Avstar servo that came with my Lycoming rebuilt IO-360-A3B6 in Sept 2018.

[Yetti]

Check all cables and linkages and make sure they have not slipped.  I vote cable slipping.

[RobertGary1]

Don’t dismiss fuel servo issues too easily. One of my engine failures was a servo. If it’s telling you something is wrong make sure you fix it good. 
 

-Robert 

[EricJ]

Replace your cable ends with spherical bearings while you're at it.

+1 on checking the rigging of the mixture and throttle cables before you go too far.  They should both be against the stops at the fuel servo when pushed full forward, with about 1/8" of "cushion" (remaining travel) in the cockpit controls that isn't used.

Also, pull the finger filter and check it for debris or sludge or whatever.   Mine had a bunch of rusty crap in it when I first bought my airplane.

K&N filters are famous for fouling MAF sensor wires in cars, and I sometimes wonder whether the oil from them and the Bracket filters doesn't goober up the servos a little bit sometimes.   When mine failed it had evidence of having been full of water at some point, and had a lot of sandy grit in it.

[larrynimmo]

On 11/24/2020 at 8:58 PM, PT20J said:

Don’t know if it’s related to your issues, but it appears that the coil spring that keeps the idle mixture adjustment screw from turning is missing.

Skip

When I sent it back I looked at it, and the spring was hidden in the back

[larrynimmo]

When I called AVstar direct, they had no rebuilt servos available....and they said if I want a new one, it is an 8 week wait.

their suggestion was to send it back and they would repair it (much less expensive) should be complete next week.

[larrynimmo]

Next follow-up...

I had sent my servo to AVstar in Jupiter Florida....the mechanic called me and stated to me that it had to be rebuilt at the tune of $1,300 including shipping.  He told me that there was evidence that fuel pressure had exceeded 100 PSI.  He also said there is no feature of a pressure relief device on the fuel servo.

to me, what is happening is that cold fuel is drawn from the tanks, and when the engine is stopped by pulling out the mixture a “column of pure liquid” fuel is there.   The sensible heat  of the shut down engine will likely heat this liquid at least 50 degrees F.  That should cause of volumetric increase of about 3%...since you can’t compress a liquid it creates a hydraulic pressure increase...where would this excessive pressure be bled to?   Should the mechanical fuel pump have a pressure release bleeding the pressure back to the tanks?  The electric pump is centrifugal, so it will bleed backwards, but the mechanical fuel pump has check valves in and out...as most diaphram pumps have...

Today I received the Servo back, with my AI we reinstalled the beast....and so far it works perfect.   I ran the plane for one loop around my home field and when I shut down pressure was at 25 psi...within 5 minutes it climbed to 30 psi before I bled it down into the injectors

Edited December 4, 2020 by larrynimmo
Add detsils

[carusoam]

Gravity.... On!

 

Tremendous observation Larry!

Very interesting...

Some Mooneys didn’t get a fuelp gauge...

The needle creeping into the red zone is starting to really have more meaning... 

Best regards,

-a-

[PT20J]

The mixture control on a RSA fuel servo (Avstar is a carbon copy) comes right after the finger screen and before the diaphragms (or any other internal parts). I have no doubt that there might be some pressure rise in the fuel lines with the hot engine shut down and the mixture in ICO, but I’m skeptical that it could exceed 100 psi and even if it did, the servo shouldn’t see it. If this were the cause, there would be a lot of aircraft having servo issues.

The engine driven pump should not be capable of creating that much pressure. Perhaps the boost pump might be able to if its pressure regulator failed.

Skip

[EricJ]

There's no pressure regulator in the engine-driven pump, the servo is the regulator.   If something went to 100 psi, it's because something in the servo failed, and I'd still be skeptical about that.

 

[larrynimmo]

Yet another update....after I installed the AVStar rebuilt servo I watched my fuel pressure after shutdown for 5 to 10 minutes and it did go above 30 psi but it didn’t peg...then one day, it did peg...and I gave it an extra couple of minutes, then I pushed in the mixture with no relief...very quickly I removed my upper cowling and used my preplanned procedure...first thing I did (even though I doubted it would be the cause) I undid my inlet fuel hose to the divider...no pressure of fuel...quickly hooked it back up, verified my instruments again...mixture wouldn’t bleed down.  Then undid the fuel hose at the sender unit and the pressure disapated instantaneously with some fuel spill.   Hooked it back up and gage read 4 psi, and that bled down instantly when I pushed in the mixture again.

the fact that the pressure dropped instantly tells me that there was a solid column of liquid fuel in the un-metered section of the fuel servo.

not wanting to spend another $1,300 getting my servo rebuilt again, my AI recommended that I adopt bleeding pressure down at shut down.  So what I did do is I open the mixture slightly, enough to bleed down the pressure and I leave it there.

According to mike Busch, there is no back pressure relief from the mechanical fuel pump ...it has spring loaded check valves which will keep it from itself pressuring fuel above about 25psi...but when you shut down the engine...cool fuel from the tanks can pick up heat...fuel in servo can pick up heat...why I have this issue I don’t know..but at least I have found a way to live with it.   And my “hot” start is perfect without issue.

Edited January 6, 2021 by larrynimmo
Spelling

[carusoam]

Larry,

Can you write a few lines of what your shut down procedure looks like now?

Type slowly... the problem could be at my end... 

Best regards,

-a-

[larrynimmo]

After arriving at my hanger...I shut down my electronics...I then run the engine up to 1700 rpm as lean as I can for about 30 seconds...then I go down to about 800 rpm idle for about 10 seconds then I pull mixture out and engine dies...then I let it sit and watch the pressure build from 25 psi to 30 psi...then I open the mixture only enough to bleed down the pressure and I leave it cracked so that pressure can’t build

[PT20J]

I’m confused Here you say that you cannot bleed off the pressure with the mixture control

4 hours ago, larrynimmo said:

Yet another update....after I installed the AVStar rebuilt servo I watched my fuel pressure after shutdown for 5 to 10 minutes and it did go above 30 psi but it didn’t peg...then one day, it did peg...and I gave it an extra couple of minutes, then I pushed in the mixture with no relief...very quickly I removed my upper cowling and used my preplanned procedure...first thing I did (even though I doubted it would be the cause) I undid my inlet fuel hose to the divider...no pressure of fuel...quickly hooked it back up, verified my instruments again...mixture wouldn’t bleed down.  Then undid the fuel hose at the sender unit and the pressure disapated instantaneously with some fuel spill.   Hooked it back up and gage read 4 psi, and that bled down instantly when I pushed in the mixture again.

But later you say that you can bleed off the pressure with the mixture control.

27 minutes ago, larrynimmo said:

After arriving at my hanger...I shut down my electronics...I then run the engine up to 1700 rpm as lean as I can for about 30 seconds...then I go down to about 800 rpm idle for about 10 seconds then I pull mixture out and engine dies...then I let it sit and watch the pressure build from 25 psi to 30 psi...then I open the mixture only enough to bleed down the pressure and I leave it cracked so that pressure can’t build

The mixture control in the idle cutoff position is a valve that cuts off fuel flow (and pressure) to the servo internals. The fuel pressure takeoff is after the finger screen and before the mixture valve. So with the mixture control in idle cutoff, it’s as if you removed the inlet fuel line and the fuel pressure line and connected them together. The servo is out of the picture and you are just measuring the pressure in the lines.

With the engine stopped and the mixture rich and the throttle opened somewhat, fuel should flow through the servo and out the nozzles - that’s how we prime. So, opening the mixture valve should relieve any pressure in the system.
 

Skip

[larrynimmo]

Skip,

if I let the pressure build beyond the “pegged” point,when I push in the mixture the pressure stays pegged...it does nothing...it is important to know that when I used a wrench to bleed off pressure at the gage transducer, the fact that it instantly relieved the pressure tells me that it was liquid bound (you cannot compress a liquid)

my point is that somehow liquid gasoline is being trapped in the servo, and with added heat created by relatively cold gasoline picking up heat causes a hydrostatic pressure increase and that nothing is releasing this pressure.

Looking at the drawings of a servo, I believe it is being trapped in the unmetered section of the drawing, and the pressure on the diaphram is causing the plunger to close off solid and not allowing it to bleed.

my servo was new (or factory rebuild) 2.25 years ago with the Lycomming engine.  1.75 years ago, I had to send the servo back under warranty as the throttle became way too stiff to reliably operate....then I started having issues when pressure would build to the point that pressure wouldn’t relieve and 1.5 months ago I sent it back and paid $1,300 to have it rebuilt.  The rebuilder mechanic from AVStar called me and told me that there was evidence that servo experienced pressure over 100psi.  And he said there is no design feature that allows for pressure relief from the servo...he suggested that I call Mooney or Lycomming...

the one common issue here is all of my issues are with the same exact fuel servo (repaired once, complete rebuild the second time). If it fails again I will try a different fuel servo.

[carusoam]

See if @M20Doc is stopping by to see this data update... (fuel pressure after shutdown, M20J)

-a-

[Guest]

It sounds like the mixture plates are sealing so well at shut off, that the fuel pressure is held long enough that thermal radiation further raises the pressure.  Larry’s method of dropping the pressure should be good for the servo.

Clarence

[PT20J]

Some of this still doesn't make sense to me (which is not to say Larry's observations are incorrect!)

First, how can it be that pushing in the mixture control relieves the fuel pressure when it is at redline or below, but not when it is pegged? Pushing in the mixture control admits fuel to the servo and should reduce the pressure and it shouldn't matter how high the pressure is. It seems to me that the most likely explanation is that the gauge is sticking when it is driven to the upper stop. I have the same gauges in my airplane and one of the fuel gauges likes to stick at the lower stop.

Another observation is that the pressure is zero at shutdown with the mixture in ICO but then slowly rises after a few minutes due to residual heat in the engine compartment. What would explain this? With the mixture in ICO, fuel cannot flow into the servo and the engine driven fuel pump valving prevents fuel from flowing back out the pump. That leaves the fuel line between the fuel pump and the servo full of fuel and this line is connected directly (through the inlet finger screen) to the fuel line that feeds the fuel pressure transducer. This latter line is about 18" long and likely has trapped air in it. There is no need to purge it for normal operation and I doubt anyone does. Even if you purge the line, it would be difficult to completely purge the transducer. It is likely this air that is expanding and raising the pressure.

I don't see how any of this can damage the servo. AvStar claimed that they had "evidence that fuel pressure had exceeded 100 psi." 100 psi is a LOT. I doubt that 100 psi would be achievable in the engine-heating-the-fuel-line scenario. But even if it could, the servo is isolated from the fuel line when the mixture is in ICO (except for some small leakage through the valve). The concern that pressure can build up in the unmetered chamber is not warranted. With no fuel flow, there will be no pressure drop across the main jet; any fuel pressure from the unmetered side will bleed through the main jet to the metered side equalizing pressure across the diaphragm.

I have a Lycoming rebuilt IO-360-A3B6 with an AvStar servo. It exhibits the same behavior as Larry's. I've been leaving the mixture in ICO. I'll let you all know if it blows up .

Skip