What's new
Van's Air Force

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

Fuel Pressure decay with fluctuations and engine stumbles

WingnutWick

WingnutWick

Well Known Member
Hello all,

Took the 10 (IO-540 TITAN) to OSH this last week from CA and ended up having to leave her in Minneapolis due to fuel pressure issues that were concerning to me accompanied with some momentary pull backs of power.

Prior to OSH I noticed that the fuel pressures on a climb out began decaying slowly around 7 mins into the climb dropping into the yellow and if left into the red. Electric pump brings it back to nominal. Along with this I've noticed continuous fluctuations of a few PSI. I thought the fluctuations might be a sender issue so I replaced that prior to the trip which I thought fixed it on my local test flight but turned out it hadn't.

SYMPTOMS: With full power climb out the fuel pressures fluctuate +/- a few psi and then begin to decay to the yellow and red if left un boosted. Boost pump will get to normal numbers but the fluctuations remain even with boost pump on. At cruise, LOP the pressures will remain normal levels with engine driven pump but the fluctuations remain. If ROP, the pressures will decay, even at altitude. Fluctuations occur on deck even when at idle. VIDEO HERE or HERE

The last leg into OSH I noticed a little stumble/pullback on takeoff even with boost (electric) pump on (with fluctuations). Taking out of OSH multiple momentary decreases of power on climb out with boost pump on and larger fluctuations which tripped my comfort level and I ended up landing in KSTP and commercial air back. I have to return soon to fix it but wanted to get your inputs prior to doing so to see what thoughts are out there.

MY BEST GUESS: That the engine driven pump is failing with possibly a hole in the diaphragm which is causing it to be unable to keep up with extended demand and also possibly drawing air into the lines causing the fluctuations and power issues even with the electric pump on. Does this sound about right?

Standing by your valued inputs!

Videos of symptoms here:
https://drive.google.com/file/d/1orD02MdHaxMRLbIAUI87t_Ri7vMzYJ9T/view?usp=share_link
 
Last edited:
WingnutWick said:
MY BEST GUESS: That the engine driven pump is failing with possibly a hole in the diaphragm .... Does this sound about right?

Standing by your valued inputs!
Click to expand...

My first guess as well. Failing diaphragm starts the trouble and eventually creates a tear or hole that bleeds off pressure from the boost pump also. HOWEVER. I would NOT rule out intermittent blockage (the debris can move around, making it intermittent) or leakage somewhere downstream from the boost pump as a possible culprit, as it could give similar symptoms. I probably wouldn't add that IF the boost pump brought it all back consistently. That added symptom greatly broadens the possibilities. Less likely, but possible, is a leak upstream that is sucking air. Check the servo strainer as well. Once clogged, it will bypass and allow trash into the injectors. Did you examine the overflow port on the mech pump? Most failed diaphragms will expell the fuel from there and leave blue staining.

Larry
 
Last edited:
Boost pump restores the higher pressures but the fluctuations still occur which made me thing air coming in from the mech pump which I would think would still show fluctuations since the air is already there, but I am not smart enough on the mechanics of it all to know if that is a valid assumption.
 
WingnutWick said:
Boost pump restores the higher pressures but the fluctuations still occur which made me thing air coming in from the mech pump which I would think would still show fluctuations since the air is already there, but I am not smart enough on the mechanics of it all to know if that is a valid assumption.
Click to expand...

When the boost pump is on, any leakage points downstream from it will expell fuel from the leak and not suck air in through it. That phenomenon only occurs on the suction side of a pump or within the pump itself. WIth the boost pump on, the mech pump is in bypass mode and is not sucking anything. The boost pump is stronger than the mech pump and always wins. Boost pump should clear out all air in a few seconds at high power, so if the problem persists it is not residual air from a failing mech pump. Vapor lock is still a possibility with the boost pump on, if the heat input is great enough. SO check for things that have changed the could add more heat to the fuel system. While fuel upstream of servo is 25 PSI, it is MUCH less downstream.
 
Last edited:
Nice review, and timely thread as I noticed fuel pressure fluctuations on extended climb the other day (IO-320) No engine stumbles, however. This would be my third failed engine-driven pump in 420 hours (1 Lycoming and 2 Tempests). Following.
 
lr172 said:
When the boost pump is on, any leakage points downstream from it will expell fuel from the leak and not suck air in through it. That phenomenon only occurs on the suction side of a pump. WIth the boost pump on, the mech pump is in bypass mode and is not sucking anything. The boost pump is stronger than the mech pump and always wins. Boost pump should clear out all air in a few seconds at high power. Vapor lock is still a possibility with the boost pump on, if the heat input is great enough. SO check for things that have changed the could add more heat to the fuel system.
Click to expand...

The flight out east was very hot, as was OSH. That's the the only change. The decaying PSI's on extended climbs and rich does sound like a failing pump though right? I may be seeing two issues here. Plane did sit for a while before I purchased it.

As for Vapor lock, this is my first fuel injected motor so I am learning about that. Any things to prevent this / look for and fix?
 
WingnutWick said:
The flight out east was very hot, as was OSH. That's the the only change. The decaying PSI's on extended climbs and rich does sound like a failing pump though right? I may be seeing two issues here. Plane did sit for a while before I purchased it.

As for Vapor lock, this is my first fuel injected motor so I am learning about that. Any things to prevent this / look for and fix?
Click to expand...

While operating on the mech pump alone, Decaying pressure during climb would point to leaks on the suction side, failing mech pump or vapor production downstream. However, in the vast majority of cases, turning on the boost pump resolves or mostly resolves all of the above. The fact that you are still running rough with the boost pump on tells me there is a broader set of potential issues. I would say there is a decent change your mech pump is going south, but not convinced that a replacement will fully address the issue. Ths is not a clear cut case where we can point to the mech pump as the culprit.
 
Last edited:
WingnutWick said:
As for Vapor lock, this is my first fuel injected motor so I am learning about that. Any things to prevent this / look for and fix?
Click to expand...

Vapor lock on an FI lyc running 100LL with the boost pump on is pretty rare no matter how hot the ambients. I would start with the other potentials that I mentioned. Leaks upstream from the boost pump can produce vapor and cause the issues you describe, but that is not really vapor lock per se.
 
Last edited:
Both tanks same symptoms?
Fuel smells?

Thoughts - vent line issue?
Sucking air in the tank to boost pump lines?
Trash on the in-tank pickup?
 
WingnutWick said:
H At cruise, LOP the pressures will remain normal levels with engine driven pump but the fluctuations remain. If ROP, the pressures will decay, even at altitude. Fluctuations occur on deck even when at idle.
Click to expand...

This is to be expected with these symptoms. LOP is flowing less fuel, so you see less fluctuation (pump is not struggling to keep up). ROP is flowing more fuel (i.e. FP can no longer keep up with the flow requirements) and the fluctuations increase as it struggles to keep up.
 
Last edited:
IMHO, you need a knowledgeable person to go through your entire system from tank to injector looking for something that is not right.
 
lr172 said:
Vapor lock on an FI lyc running 100LL with the boost pump on is pretty rare. I would start with the other potentials that I mentioned. Leaks upstream from the boost pump can produce vapor and cause the issues you describe, but that is not really vapor lock per se.
Click to expand...

Thanks, so what I gather my next steps are going to be:

1. Replace mech boost pump (based off of decaying pressures in climb)
2. Run boost pump on ground with throttle and mix up and cowling off to check for leaks in the system.
3. Overall inspection of how the fuel system is set up to see any glaring issues that could contribute to vapor lock?

What else?
 
lr172 said:
This is to be expected with these symptoms. LOP is flowing less fuel, so you see less fluctuation (pump is not struggling to keep up). ROP is flowing more fuel (i.e. FP can no longer keep up with the flow requirements) and the fluctuations increase.
Click to expand...

Fluctuations happen at all mixture settings and rpm's now for the most part, the PSI falloff is caused by LOP vs ROP though which makes me think a new pump is in order.
 
WingnutWick said:
Fluctuations happen at all mixture settings and rpm's now for the most part, the PSI falloff is caused by LOP vs ROP though which makes me think a new pump is in order.
Click to expand...

Yes, but the fluctuations are minor and not causing major issues, though indicative of an issue. It is the the 15 PSI drop when ROP is the thing that will bite you. A little odd that I observed the PSI drop to 9 WITHOUT a drop in fuel flow. From what I have learned, the bendix starts to drop flow once the input pressure drops below about 12 - 14 PSI and that didn't occur in your video. Maybe a call to Don will be informative on that one.
 
Just to clarify, that when on the mech pump only you get 23-26 fluctuations consistently and goes really low only when ROP. With the boost pump on, you get 23-26 fluctuations, but doesn't drop when ROP. If that is the case, then my very first move would be a new mech pump followed by a test. Very possible that the pump failure is also causing the minor fluctuations in addition to the abnormally low flow rate that are causing the drop as servo demand increases.

Sorry, should have looked at the video before all my other posts. The minor fluctuations point to something, but is not an immediate threat. The drop is. I often get sporadic fluctuations in my 10 during climb but never drops below 20. that said my boost pump makes it go away. I am convinced it is heat related.
 
Last edited:
Did this occur at altitude only? I'd say it's heat. Flashing fuel. I've seen it occasionally with a hot engine compartment after a fuel stop and at higher altitudes/climbing. I need to install a cooling shroud on my main pump.
 
Last edited:
lr172 said:
Sorry, should have looked at the video before all my other posts. The minor fluctuations point to something, but is not an immediate threat. The drop is. I often get sporadic fluctuations in my 10 during climb but never drops below 20. that said my boost pump makes it go away. I am convinced it is heat related.
Click to expand...

How would you go about confirming or excluding the integrity and functionality of the engine-driven pump? Is there a specific testing protocol, or is this a diagnosis of exclusion and we just load the parts cannon?
 
Check the finger filter at the servo inlet, and the main filter (typically at the electric boost pump).

This could also be a 'flapper' in a hose. Similar to this SB https://www.vansaircraft.com/service-information-and-revisions/sb-02-12-1/
while it is unlikely this old SB applies... the possibility of improperly constructed hoses never goes away. As it appears you aren't the builder, we don't know who made the hoses, or when.

Lower flow rates... (from LOP), less flow demand, the flap doesn't block the flow as badly.
 
Wing it - I’d suggest checking the pipe flaring at the fuel tap as it is higher than the tanks (therefore fuel will not leak out ) as a bad flare can result in the engine pump sucking in air. Turning on the electric pump can improve pressure however the ‘problem’ will not go away without rectification - I have seen the same scenario on another 10. HTH.
 
MacCool said:
How would you go about confirming or excluding the integrity and functionality of the engine-driven pump? Is there a specific testing protocol, or is this a diagnosis of exclusion and we just load the parts cannon?
Click to expand...

IMHO, the video already confirms it. Run the engine at 17 GPH and FP is 25 PSI. Push the mixture to 21 GPH and FP drops to 10. Pull it back to 17 and it goes back up to 25. Turn the boost pump on and it goes to 25 at all flows. All of this means that the mech pump can't keep up with flows above 19-20 GPH, yet the system at large can handle 21 GPH while maintaining 25 PSi. Ignore the minor oscillations, as they are a head fake. Yes, a real issue, but not directly related to the delivery capacity problem, which is highly likely to be a failure of some sort inside the mech pump. I speculate that the oscillations are also related to what has failed in the pump, but that is a guess and why I said replace the pump and see if the oscillations go away. If not, more troubleshooting is necessary, but it seems obvious to me that the pump is toast and therefore no parts cannon here. His engine requires at least 25 GPH and his pump has proven to be unable to meet that flow requirement at a minimum pressure level set by the servo maker. Not replacing that pump would be in the not so smart category, so let's just use that as the first diagnostic test.

The pump is a design with an inverse relationship between pressure and flow. As flow increases, pressure drops. Due to the spring design regulating a max pressure (pump could produce much higher pressures at low flows without the regulation) and the pump operating well below it's design capacity, we typically don't see that pressure drop, as the pressure fall off starts above our normal range of use (the pump is rated for around 45 GPH). In this case, that total capacity has dropped due to some problem and the OP is seeing that curve at a much lower pressure now.

Larry
 
Last edited:
BillL said:
While it might well be a failing mech pump, since you bought the bird, I would recommend a fuel system air pressure test. The same as it should be done when building. This will pressurize the mech fuel pump as well and if it is the leak then air should come out the vent tube.

Tiny air leaks on the suction side will become worse with lower ambient pressures and higher fuel temperatures. I have never had a builder locally do this test and not find a leak. Suction leaks degrade the pumping effectiveness of the mechanical pump.

Send me a PM if you want to know more about the test. It requires a good pressure gauge, some adapters to get pressure into the fuel line at the fuse bulkhead, plugs for the engine end, valves to capture the volume of air and an adjustable air pressure regulator. I bet the parts will be a lot less $$ than a new pump, and you can always sell to a builder in your area. Maybe some have the setup already.
Click to expand...

This a great test and recommend that everyone does it. I did and it creates great comfort to know there are no leaks. However if you pressurize the system and air does not come out of the FP vent, it only means that the diaphragm is not ruptured. There are other failure modes in the mech FP besides a ruptured diaphragm, such as debris that keeps a check valve from closing or simply a failed check valve spring.

Not sure how you can get a positive diagnosis on a mech FP without the engine spinning at a decent RPM or some test rig that will move the lever up and down while measuring flow and pressure. That is a tall order for the OP to perform on a ramp 100's of miles from home, likely with no access to a compressor or possibly even power. An FP is only $300 and probably the best first step given the symptoms and circumstances. Air leaks bad enough to create large flow limitations don't often happen overnight, though certainly possible, and the OP implied all was good one day and really bad the next, implying something significant has changed or broke.

Larry
 
Last edited:
OK, I'll heap on more of the same indictment of the mechanical fuel pump: I've had a very similar experience that culminated in the engine nearly quitting on takeoff. Luckily the Garmin EIS immediately marked low fuel pressure in red and the boost pump made it all ok in a second or two. Continued the flight a short bit, tried with the boost off again and quickly had to put it back on as the pressure again dropped with the mechanical pump.

At first I couldn't believe it as the general rule of thumb had been that the mechanical pump lasts the rebuild time of the engine, and I only had 400 hours since new on the engine and Tempest pump. (Incidentally, that was a replacement pump after the Lycoming new pump (new I/O-390!) leaked oil within the first 50 hours!) Well, lasting 2000 Hrs, not so anymore! Put on a new pump and that fixed it. There seems to be a downward reliability trend in a number of areas, such as magnetos as well as fuel pumps for both materials and workmanship issues.
 
Last edited:
MacCool said:
How would you go about confirming or excluding the integrity and functionality of the engine-driven pump? Is there a specific testing protocol, or is this a diagnosis of exclusion and we just load the parts cannon?
Click to expand...

Here’s how I *knew the eng driven pump was going bad on my VW Bus..I could see thru the transparent fuel filter that it was pumping air. Not sure how you would replicate that on a Lycoming, although my RV3 had transparent glass fuel filters at each wing root. Food for thought. As a diagnostic tool only.

Larry makes a pretty compelling case and sounds quite knowledgeable on the subject but I hate troubleshooting by parts replacement and a Lyc fuel pump is the *last component I want to pull and replace. They are a real PITA and easy to strip the accessory case or not capture the drive pushrod properly.

Good luck Wingnut, you sure are building your A&P base of knowledge!
 
N49ex said:
There seems to be a downward reliability trend in a number of areas, such as magnetos as well as fuel pumps for both materials and workmanship issues.
Click to expand...

No joke. As the large companies swallow up the mom and pops this gets even worse. I have all but quit Quality Aircraft Accessories and Hartzell Engine Technology brands. They have people off the street in their assembly lines hacking on stuff. This lack of skilled labor is legal under the guise of being a Repair Station, regardless of quality assurance.
 
Charles----just for giggles-----run the boost pump throughout the climb. I "assume" this is a mechanical injection with an AirFlow Performance boost pump. When on, the pump will provide 28-29 PSI of pressure and should keep your running fuel pressure constant.
IF this is what you have, and with the pump ON and during climb the same symptoms occure, then I'd be starting at the tanks and move forward and inspect every hose/tube/connection for the possibilty of a loose nut, cracked flare, etc. Check the selector valve for possible internal bleeding, although unlikely.

Alot of these symptoms are difficult to recreate on the ground, so a new test session will be needed.

Tom
 
N49ex said:
OK, I'll heap on more of the same indictment of the mechanical fuel pump: I've had a very similar experience that culminated in the engine nearly quitting on takeoff. Luckily the Garmin EIS immediately marked low fuel pressure in red and the boost pump made it all ok in a second or two. Continued the flight a short bit, tried with the boost off again and quickly had to put it back on as the pressure again dropped with the mechanical pump.

At first I couldn't believe it as the general rule of thumb had been that the mechanical pump lasts the rebuild time of the engine, and I only had 400 hours since new on the engine and Tempest pump. (Incidentally, that was a replacement pump after the Lycoming new pump (new I/O-390!) leaked oil within the first 50 hours!) Well, lasting 2000 Hrs, not so anymore! Put on a new pump and that fixed it. There seems to be a downward reliability trend in a number of areas, such as magnetos as well as fuel pumps for both materials and workmanship issues.
Click to expand...

I'm currently on my third mechanical fuel pump in 400-or-so tach hours since the plane first flew in 2011 (one Lycoming and two Tempest pumps). All were new pumps...no rebuilds.

As to your takeoff experience, I'm a little surprised. FUEL PUMP ON has always been part of my pre-takeoff check list for any airplane so-equipped, and I leave it on until I've hit the altitude where I know I can make the return to the airport. It's part of every GUMP check I do. Ever since I had one fail, I find my eyes involuntarily glued to the fuel pressure gauge for several seconds after turning it off.
 
This video was in another VAF thread on the same topic. I found it very informative about pump operation and potential problems. https://youtu.be/fxT4RAEssnw

I do find it interesting that the Tempest published min pressure for the pump is 25 psi, and there is discussion of 20 psi being ok. I’m fighting a similar problem on my IO360, and get nervous when pressures are down to 20 psi, as the pump is not performing to spec.
 
Last edited:
Fuel Filter Issue??

One other thought that I'm surprised no one has mentioned yet:

Suggest you check the fuel filter for plugging contamination. In the Van's standard design RV-10 fuel system, there is a single in-line filter placed between the fuel selector valve and the electric pump inlet. This is located in the tunnel, so can be easy or hard to reach depending on the interior finish (easy of plain interior with easy ability to remove the top cover of the tunnel in order to reach the filter, but hard if there is a central console covering the tunnel cover). This filter should be disassembled and cleaned at very yearly condition inspection.

See the fuel system drawings attached.

Good luck,
 

Attachments

  • RV-10 Fuel System.pdf
    2.4 MB · Views: 50
Hello all,

Thank you for all the valuable information here. Learned a lot and very much appreciate it. I ended up heading to Minneapolis with a new Tempest pump in hand and after a couple hours working on the ramp in the heat and some swear words got the new pump in successfully. Run-up showed no fluctuations and test flight held steady at 27 PSI!!! Flew it back to CA yesterday with everything operating 4.0. So it appears that the mech pump was sick after all.

To note a couple comments, the pressure drops on climb out had been occurring for a little while and getting worse. The fluctuations started on this trip however and I think the heat and continuous use sent the pump downhill. I also do use the electric pump until a few thousand feet then secure it. Was getting stumbles even with it on.

Thank you for all the other info, tests, and things to inspect. I will make sure I run through all of these and change anything as needed to make sure the fuel system is all rigged properly and no leaks or blockages.

Thanks again for all your expert info!

PS: Special thanks to David Howe for his spiffy tool he created to hold the plunger up. Made this job much easier!
 
You must log in or register to reply here.
Back
Top