What's new
Van's Air Force

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

Bad Fuel Pressure Sensor? Or bad main fuel pump?

Saville

Saville

Well Known Member
So I was flying around the other day and I flipped to the screen, on my Trutrak EFIS GP FM, which shows Fuel pressure (among other things) and I saw this:




well I immediately flipped on the Aux fuel pump - and saw this:



and I immediately landed.

At no time did the engine falter in the slightest. I was at 5500 feet and the engine had about 149 hours TTSNEW. On the way back I shut off the aux pump and watched the indicator....it hung at 3.8 for just a second and then slowly wound down until it hit 1.9 whereupon it disappeared and displays what you see above. Aux pump back on...immediately comes back on and counts up to 3.8.

In talking with the TruTrak people, this is what happens when there is no data on the line.



Following weekend I pulled the cowl to check all the connections to the Fuel pressure sensor and they all seemed very sound.

So today I went up - within gliding distance of an airport - and took some data. "LEFT" and "RIGHT" mean left tank and right tank. MECH is engine driven pump; AUx is Aux fuel pump on

Startup Reading - just after startup - 1000rpm:

Pump LEFT RIGHT GPH
MECH 2.4 2.3 2.1
AUX 5.1 5.1 3.5


I idled it and got: 2.0 1.9 .5gph

Taxied to the runup area. By the time I got there the readings for MECH disappeared:

Runup Reading - 1800 rpm:

Pump LEFT RIGHT GPH
MECH --- --- 4.4
AUX 4.5 4.5 7.6


5500 feet 23 squared:

Pump LEFT RIGHT GPH
MECH --- --- ??/ can't read my writing
AUX 4.8 4.8 16.0

I may have been too lean here at 23 squared.

Ok so I head back and as I'm letting down I'm playing with the throttle: Prop at 2300.

I bring the MP back slowly to 15" stopping every couple of inches. At 9 inches the readings return - 1.9psi

Back up to 15" - readings disappear.....

Back down one inch at a time - readings appear at 9".

Back up 1 inch at a time to 15...reading disappears at around 12"

I repeat that two more times..same results.

Next time I do it, the readings do not disappear! Odd.

So I concentrate on landing, and take a reading at the hangar just before shutdown:

Hangar Reading:

Pump LEFT RIGHT GPH
MECH 5.2 5.1 1.7


Ok so my first inclination is to say that the Fuel pressure sensor may be bad. It was bought from TruTrak in 2010, 0 - 30 psi

SO# 27614

Looks like a plain old every day sensor:



But it is almost 6 years old.

I can't think of any other things to check. As I say the engine runs fine.

Any ideas welcome.

thanks!
 
Fuel Pump

Good luck on it being the sender. My pump was bad and I had similar indications. I just couldn't believe that with a 0 fuel pump pressure reading that the fuel pump could be bad and engine still run fine. I installed a mechanical gauge for testing and sure enough under high power the pressure dropped way off. Replaced fuel pump and all is well. Pump had about 400 hours if I recall.
 
That VDO pressure sensor has a reostat-like winding inside and it gets worn out at the spot that gets used most often. Hopefully it's that and not the pump.
Talk to Trutrak and see if you can use a voltage based Kavlico or Honeywell sensor. Those seem to be much more reliable.

Lenny
 
Jrskygod said:
Good luck on it being the sender. My pump was bad and I had similar indications. I just couldn't believe that with a 0 fuel pump pressure reading that the fuel pump could be bad and engine still run fine. I installed a mechanical gauge for testing and sure enough under high power the pressure dropped way off. Replaced fuel pump and all is well. Pump had about 400 hours if I recall.
Click to expand...

How did you run your tests with the mechanical gauge? On a ground run?

Do you happen to recall at what MP the reading went to zero?

thanks!
 
Guys---back in my old days of turning wrenches, we always ran a mechanical gauge test to verify the reading before condeming anything. The problem as I see it, (and I'm guilty too) is that we've become captive by all the electronic gizmos, and have moved away from what has worked for decades.
So when the pressure readings on the EFIS start to fluxuate, the most natural thing is to condem the pump.
I my old world, the natural thing was to condem the electronics----because it was the 'new thing'. And quite many times it was as simple a bad connection.
I'll go out on a limb and say thats probably still the case.

It would almost be prudent to place an extra bulkhead fitting in the firewall
(capped off of course) to allow a mechanical gauge to be temporarily installed in the cabin to check the pressures. The fuel pressuse hose could be re-routed to this fitting, and the gauge connected. Test flight and see if the symptoms repeat. If so, then you've eliminated the electronic devices, and can zero in of the pump. IF the symptoms go away, then zero in on the electronics.

We seen guys the spent countless hours and hundreds of dollars or more buying and swapping parts, only to find a mis-crimped terminal or a loose ground. Above all----be safe!
Tom
 
I am in agreement with Tom Swearengen. Even though the sender is screwed in to what looks to be a perfect ground time and sealant may have compromised that ground. Take a bit of sandpaper and clean up a spot on the sensor. Now fasten a ground wire to the body of the sensor with a hose clamp. The other end of the wire goes to a good ground on the airframe.
This solved a similar issue that I had. A simple thing to check, then a mechanical gauge. The sensors are not that expensive, you will eventually have to replace it, mine lasted 400 hours and the replacement has 550 hours on it and I have a spare on the shelf....
 
My vote for most likely solution . . .

Lenny Iszak said:
That VDO pressure sensor has a reostat-like winding inside and it gets worn out at the spot that gets used most often. Hopefully it's that and not the pump.
Talk to Trutrak and see if you can use a voltage based Kavlico or Honeywell sensor. Those seem to be much more reliable.

Lenny
Click to expand...

But do a diagnosis first. If you have an air pressure regulator, ohm meter, and manual pressure gage, you could pressurize and test the sensor. Since the above failure mode is sensitive to vibration, you might not learn anything, but you never know. This is what I would do, but I'm nuts.

The Kavlico is much more reliable and durable (non mechanical), if it works with your system. After replacement you can tear apart the old one and look for the spot on the wiper for confirmation.
 
Saville I did as Tom S. Suggested. I used a known good mechanical gauge plumbed into the fuel system. I installed a bulkhead fitting in the firewall and connected the line to the electric sender to it. I made up a line for the cockpit side and temporarily installed the gauge. When flying the pressure dropped to almost zero under power. Flipped on boost pump and all was good. Apparently pump provided enough volume to supply engine at power as the engine always ran fine.
 
Try adding a ground wire

I agree with Tom. I had this same issue with an oil pressure sensor. I added a ground wire, just like Tom stated, and the problem went away.
 
Thanks everyone for all the comments. Really very helpful.

Ok so I think what I'll do is progressively more complex tests starting with:

1) Tom Martin's suggestions of an alternate Ground using a hose clamp

If that fixes it then the permanent solution would be to unscrew the sensor from the bock, clean the threads, and re-attach?



If that doesn't fix it then:

2) Tom Swearengen's (and others) of testing with a mechanical pressure gauge.


My plan would be to plumb in a T fitting then hook one output back to the fuel sensor. The other output through a couple feet of hose to the mechanical gauge.

I have a couple questions on that:

- Will all that plumbing still be properly pressurized so that I get accurate readings?

- Can anyone recommend a good mechanical gauge?

- would the gauge first have to be calibrated?



I suppose I could just forego the T fitting and hook up a mechanical gauge, temporarily, for the test. But I would feel better if I saw the gauge reading pressure but the EFIS showing the "---". That way I would know the mech fuel pump is fine and the issue is somewhere between (and including) the sensor on down to the EFIS.

Any comments/criticisms of the plan are welcome.
 
Actually---if you wanted to keep the existing setup, AND tee in a mech gauge---that would cross test BOTH systems at the same time, inflight. So---if you get a low press indication on the EFIS, and the Mech guage is normal----that would indicate to me that we have a electrical issue, either sender, connections, or maybe even a software issue, but thats really doubtful. BOTH if both the EFIS and the gauge read the same, then its an actual pressure failure.
Then you can start looking at pump, hoses, fittings, etc.

Seems like alot of work and testing, but I know of a builder that paid alot of money for parts, and needed to replace a crimp connector at the sender.
Tom
 
TS Flightlines said:
Actually---if you wanted to keep the existing setup, AND tee in a mech gauge---that would cross test BOTH systems at the same time, inflight. So---if you get a low press indication on the EFIS, and the Mech guage is normal----that would indicate to me that we have a electrical issue, either sender, connections, or maybe even a software issue, but thats really doubtful. BOTH if both the EFIS and the gauge read the same, then its an actual pressure failure.
Then you can start looking at pump, hoses, fittings, etc.

Seems like alot of work and testing, but I know of a builder that paid alot of money for parts, and needed to replace a crimp connector at the sender.
Tom
Click to expand...

I'd rather test a lot and pin down the exact issue before I start spending a lot of money on replacing parts.....and/or pulling the engine.

To TEE in a mech gauge I would need to:

1) identify the T fitting I need to get.

2) Identify the bulkhead passthrough I need to get.

3) Identify the pressure lines I need to get.

4) Pick the pressure gauge. I have an unused 2.25" hole in the instrument panel that I could use for that purpose

Lots of research to do.
 
just get it over with....

Or you could just replace the mechanical fuel pump. Those things have a finite life span and I suspect this will cure your low fuel pressure. I had the same issue and the new pump brought everything back to normal.

If in the off chance the new pump doesn't fix the problem, you will then have a spare pump on the shelf. :)
 
Simplify

Since you have observed the errant readings on the ground (at your 1800 RPM runup), I see no reason to run a mechanical gage in a semi-permanent install through the firewall.

You can tee into the sender with a mechanical gage, and then with the cowling off have an observer watch the mechanical gage during your start-up, idle, and run-up. If you have a significant difference between the mechanical gage and the EFIS, then you probably have enough information to resolve the problem.

If you are uncomfortable with having an observer in proximity to a rotating propeller, then you can run a short hose from the tee to the gage so that it is visible from the cockpit.
 
SMRacer said:
Since you have observed the errant readings on the ground (at your 1800 RPM runup), I see no reason to run a mechanical gage in a semi-permanent install through the firewall.

You can tee into the sender with a mechanical gage, and then with the cowling off have an observer watch the mechanical gage during your start-up, idle, and run-up. If you have a significant difference between the mechanical gage and the EFIS, then you probably have enough information to resolve the problem.

If you are uncomfortable with having an observer in proximity to a rotating propeller, then you can run a short hose from the tee to the gage so that it is visible from the cockpit.
Click to expand...

That is actually my plan. Go from the NPT plug on the top of the manifold block to a gauge which is visible by me from in the cockpit. Run the plane with the upper cowl off (carefully watching CHT all the time) and note the results.

I would be very uncomfy with an observer near the prop.
 
Unlikely that it's a grounding issue. You have two terminals on that sender, there's no extra ground needed, it's already grounded through one of the terminals. If it was a Stewart Warner type single terminal sender then yes, but looking at your picture that's not the case.
The sender is about $50. You need a spare anyway, if it hasn't quit yet, it will soon. Don't go poking holes in your firewall just yet. ;)
I like Bill's idea to check the sender with a compressor that way you can sweep through the whole pressure range easily. That's how I calibrated mine.

Lenny
 
Lenny Iszak said:
Unlikely that it's a grounding issue. You have two terminals on that sender, there's no extra ground needed, it's already grounded through one of the terminals. If it was a Stewart Warner type single terminal sender then yes, but looking at your picture that's not the case.
The sender is about $50. You need a spare anyway, if it hasn't quit yet, it will soon. Don't go poking holes in your firewall just yet. ;)
I like Bill's idea to check the sender with a compressor that way you can sweep through the whole pressure range easily. That's how I calibrated mine.

Lenny
Click to expand...

If the grounding terminal is bad somewhere along the line, the grounding test will reveal that.

I do agree about having a spare sender.
 
Lenny Iszak said:
Unlikely that it's a grounding issue. You have two terminals on that sender, there's no extra ground needed, it's already grounded through one of the terminals. If it was a Stewart Warner type single terminal sender then yes, but looking at your picture that's not the case.
The sender is about $50. You need a spare anyway, if it hasn't quit yet, it will soon. Don't go poking holes in your firewall just yet. ;)
I like Bill's idea to check the sender with a compressor that way you can sweep through the whole pressure range easily. That's how I calibrated mine.

Lenny
Click to expand...

Ya can even use a small portable tank.

Screen%2BShot%2B2016-05-12%2Bat%2B12.13.54%2BPM.png
 
I don't see how you can have a sensor issue, as it continues to deliver clear and consistent results in your tests. how can a sensor only give bad readings when you turn the boost pump off and provide correct readings when it is on?

Your pump is failing and it cannot deliver the volume it should or once did. Remember, a pump will only produce pressure when it's supply has exceeded demand. This is why you sometimes see lower pressure at higher fuel flows. Your pump is flowing enough to satisfy demand at full power, but not producing enough excess flow to produce a positive pressure (a bad situation, as your margin is gone). At lower power settings, supply is exceeding demand and you see pressure. In my opinion, your tests very clearly show a pump that is not producing enough flow.

Larry
 
Last edited:
lr172 said:
I don't see how you can have a sensor issue, as it continues to deliver clear and consistent results in your tests. how can a sensor only give bad readings when you turn the boost pump off and provide correct readings when it is on?



Larry
Click to expand...

Perhaps because the lower pressure moves the sensor off the worn part as others have suggested.

Also, your analysis forgets that after a while the reading came on and stayed on at all power settings. Now it's true that could mean the pump is intermittently bad. But it can also mean other things are intermittently bad.

Also remember that the readout is behaving as if there is no signal... NOT zero. It's not displaying zero pressure. It's displaying no reading.


I'm not saying it isn't the fuel pump. But a simple gauge test will say whether the problem is upstream from the sensor (hose or pump) or downstream ( sensor, wiring EFIS).
 
Last edited:
Saville said:
Perhaps because the lower pressure moves the sensor off the worn part as others have suggested.

Also, your analysis forgets that after a while the reading came on and stayed on at all power settings. Now it's true that could mean the pump is intermittently bad. But it can also mean other things are intermittently bad.

Also remember that the readout is behaving as if there is no signal... NOT zero. It's not displaying zero pressure. It's displaying no reading.


I'm not saying it isn't the fuel pump. But a simple gauge test will say whether the problem is upstream from the sensor (hose or pump) or downstream ( sensor, wiring EFIS).
Click to expand...

Sorry, must have mis-read your initial post. Have you ever seen your pressure reading display 0? Are you sure that --- is not synonymous with 0?
 
lr172 said:
Sorry, must have mis-read your initial post. Have you ever seen your pressure reading display 0? Are you sure that --- is not synonymous with 0?
Click to expand...


Firstly, remember that at startup, I have a reading - then sometime during the taxi to the run-up area, the reading goes to "---". Taxi MP is very very low. So if your theory was correct, I would expect the display to NOT be "---". But it is.

I have never seen it display 0.0...or 0.1 or 0.3. This is a very subtle but important distinction. I interpret this to mean either the input signal is grounded, or - possibly - it's open.

Here's why:

My EFIS GP FM is a TruTrak which is a couple of generations old. IT has a module called the Engine Data Module (EDM) which allows you to connect various signal lines to the EFIS display - such things as CHT's, EGT, Fuel Pressure, and RPM.

Important fact: There is ONLY ONE input to the EDM for RPM.

Important fact: The P-lead on the Left Magneto is directly connected to this input. Both P-leads go to the ignition key.

During run-up, when I check the left mag, I see 1800 on the display, as expected.

But when I check the right mag, the display stays at 1800 for about a half a second and then rapidly counts down to .....

you guessed it: "---"

As we know, when testing the right mag, the left P-lead is grounded.

Exact same behavior with the fuel pressure indicator...aux pump on and I get a reading; aux pump off and it hangs for a second then quickly ramps down.


Watch this video and you will see precisely what it does. Let me know if you cannot view the video.


The yellow band of the fuel pressure indicator is set from 0.0 psi to 1.0 psi.
The video begins with the aux pump on, then I turn it off - watch the needle and the numeric value. The needle disappears, and the numeric vale goes to "---".

Then I turn the aux pump back on - you see the needle re-appear and move up to 3.8 again.

I would HOPE that if ground means 0.0psi on the sensor that the EFIS would display 0.0.


 
Last edited:
I watched your video and also saw the fuel flow drop with the pressure. As it returns, the fuel flow goes well above normal, presumably because the carb bowl got low as flow was previously decreased below the engines requirements. This still points to an intermittently failing fuel pump. I suppose you could have some wide scale problem with your elms, affecting multiple sensors, but the odds don't support it.


I don't want to seem that I am arguing with, so will not post further.

Larry
 
Not looking like the sensors

1. I bet the EFIS shows --- when the signal gets outside the expected range (such as zero)
2. A second instrument (FF) backs up the pressure sensor anomaly.

Real enough for me to buy a fuel pump before next flight....
 
jdeas said:
2. A second instrument (FF) backs up the pressure sensor anomaly.
Click to expand...

In my opinion, the behavior of the fuel pressure readings are not anomalous. They are relatively classic for a failing fuel pump. Yes, some intermittancy thrown in, but pretty classic. The fuel flow behavior fully backups up the diagnosis. I also believe that "---" means 0 or out of range.

Larry
 
Test #1 Complete

With some very interesting results.

Quick reminder:

Last time out I was getting a fuel pressure reading at startup but by the time I got to the run up area it was "---"

So today I started up and taxied a very short distance to a place where I could run up.

At startup, I got an FP reading of 4.8psi.

Taxied over to the area - 4.8psi

After a little warmup I went to 1800rpm as if I was doing a runup

4.8psi.

So I watched and waited. As things warmed up I could detect a slight wavering of the FP. it would bobble between 4.7 and 4.8......

Things got warmer and the FP started to ever so slowly drop.

The warmer the engine got, the faster the FP went down. At around 300 CHT you could see it winding down though not as fast as the movie up above in an earlier post.

The warmer the engine got...the faster the FP dropped. It was approx around 328 where the FP got to 1.9......

then 1.8.............

then "---" in a matter of 2 seconds. And it stayed that way.

Pulled the power back to idle which was about 9" Hg:

stayed at "---"

MP didn't matter - heat did.

Ok so I taxied back to the hangar and shut down. A few moments after the engine shut down I flipped the EFIS back on and saw this:



Well whaddya know.......Note the oil pressure dial on the lower left - 0psi (as we would expect). The fuel pressure? "---"

The EFIS is perfectly capable of displaying 0 or 0.2 or 1.1. Therefore "---" DOES NOT mean zero. Also recall that in all experiments, the EFIS never displayed a FP of less than 1.8....it would drop to 1.8 or 1.9 and then "---"

So this test has shown 2 things:

1) Whatever the problem is - it's heat related.

Twice, now, I noted an FP reading at startup but not after the engine had time to warm up.

I now think the reason the FP display was giving me a reading as I was letting down, on my last flight, was because the engine was cooler than cruise.

I wish I had the presence of mind to flip onthe EFIS before I started to see of the FP read zero. I will do that at the next test.

2) "---" does not mean zero. And the fact that the FP display will never show less than 1.8 but always goes to "---" means something.
 
Saville said:
With some very interesting results.

Quick reminder:

Last time out I was getting a fuel pressure reading at startup but by the time I got to the run up area it was "---"

So today I started up and taxied a very short distance to a place where I could run up.

At startup, I got an FP reading of 4.8psi.

Taxied over to the area - 4.8psi

After a little warmup I went to 1800rpm as if I was doing a runup

4.8psi.

So I watched and waited. As things warmed up I could detect a slight wavering of the FP. it would bobble between 4.7 and 4.8......

Things got warmer and the FP started to ever so slowly drop.

The warmer the engine got, the faster the FP went down. At around 300 CHT you could see it winding down though not as fast as the movie up above in an earlier post.

The warmer the engine got...the faster the FP dropped. It was approx around 328 where the FP got to 1.9......

then 1.8.............

then "---" in a matter of 2 seconds. And it stayed that way.

Pulled the power back to idle which was about 9" Hg:

stayed at "---"

MP didn't matter - heat did.

Ok so I taxied back to the hangar and shut down. A few moments after the engine shut down I flipped the EFIS back on and saw this:



Well whaddya know.......Note the oil pressure dial on the lower left - 0psi (as we would expect). The fuel pressure? "---"

The EFIS is perfectly capable of displaying 0 or 0.2 or 1.1. Therefore "---" DOES NOT mean zero. Also recall that in all experiments, the EFIS never displayed a FP of less than 1.8....it would drop to 1.8 or 1.9 and then "---"

So this test has shown 2 things:

1) Whatever the problem is - it's heat related.

Twice, now, I noted an FP reading at startup but not after the engine had time to warm up.

I now think the reason the FP display was giving me a reading as I was letting down, on my last flight, was because the engine was cooler than cruise.

I wish I had the presence of mind to flip onthe EFIS before I started to see of the FP read zero. I will do that at the next test.

2) "---" does not mean zero. And the fact that the FP display will never show less than 1.8 but always goes to "---" means something.
Click to expand...

This test further supports a failed fuel pump. It's capacity is reduced with increased engine heat. Rubber becomes more flexible as temp increase and the slit/tear/hole allows more fuel through (i.e. less flow/pressure out of the pump) as it warms. I just can't see a likely scenario where a sensor consistently changes it's behavior based upon temp. The sensor isn't mounted on the engine and will heat much slower than the engine itself.

You really need to work with the probabilities here. Troubleshooting rarely produces 100% definitive actions. You have to follow the path that has the greatest probability and in your situation, I feel that path is clear. Either way, it's your engine. Let us know how you finally resolve this. Maybe we'll all learn something.

Larry
 
Last edited:
No reading means bad sensor. I had those same sensors in my RV and was advised by Dynon to change them as they do not last and are unreliable. My opinion, change the sensor not the fuel pump.
 
duh.......

I'm having a hard time figuring out all the dilly dallying around. Fuel delivery problems account for most engine outages. If this was my airplane (and it was a few years ago) I would (did) replace the fuel pump with no second thoughts (my plane also got a rebuilt carb). If the problem still existed, then the sender would be replaced. That would yield a 'new' fuel delivery system that should work fine for several more years without all the protracted hand wringing about how an EFIS works.

Fuel pumps wear out, consider it an item that needs periodic replacement. Replace it and the sender before they put you in a farmer's field somewhere in the middle of nowhere with a bent airplane.
 
Last edited:
It may come in handy to know how the Efis works for that exciting moment when the oil pressure sender fails and there's no backup to just switch it on, like with the fuel pump, and you're not exactly sure if you have 5 more seconds before a piston comes through the cowling or it's just a joke courtesy of VDO.

Btw, also consider that the Efis doesn't show instant pressure values but averages over a few seconds, that's why it comes down gradually.

Now i'm too really curious what this ends up being. :) So please do let us know.

Either way, no need to be annoyed by differing opinions. We're all learning.

Lenny
 
Sam Buchanan said:
I'm having a hard time figuring out all the dilly dallying around. Fuel delivery problems account for most engine outages. If this was my airplane (and it was a few years ago) I would (did) replace the fuel pump with no second thoughts (my plane also got a rebuilt carb). If the problem still existed, then the sender would be replaced. That would yield a 'new' fuel delivery system that should work fine for several more years without all the protracted hand wringing about how an EFIS works.

Fuel pumps wear out, consider it an item that needs periodic replacement. Replace it and the sender before they put you in a farmer's field somewhere in the middle of nowhere with a bent airplane.
Click to expand...

Sam, the sensor is bad as evidenced by the "blank" reading. With a good sensor installed one can determine if the fuel pump needs changing.
 
Steve Iacoviello said:
Sam, the sensor is bad as evidenced by the "blank" reading. With a good sensor installed one can determine if the fuel pump needs changing.
Click to expand...

Where did you come up with that? The OP has posted numerous tests with readings other than blank. You're dealing with intermittent pressure readings here that, in theory, could point to either the sensor or the pump.

Regardless of who you believe here, I think most will tell you that flying this plane without replacing the pump, just isn't smart given the risks.

Larry
 
Saville said:
After 152 hours???
Click to expand...

It is a sandwich of rubber slices. Someone posted a teardown of a pump. I encourage you to look at his pictures. LOTS of small rubber parts. failure can happen at any time. It is not just a wear issue. I just had a pump fail with only 400 hours on it.

Larry
 
I came up with that by looking at my own RV with a new Lycoming. Now at 450hrs. The fuel pressure moves around with temperature, and rpm but always well above the minimum of 0.5 psi required by the engine. What I do know is that sensor was junk. Much more stable readings when I accepted Dynons advise and replaced it with a Klavico.
 
lr172 said:
Where did you come up with that? The OP has posted numerous tests with readings other than blank. You're dealing with intermittent pressure readings here that, in theory, could point to either the sensor or the pump.

We now know there's a problem with the sensor/EFIS because that a reading is flat out wrong. It should display zero, not "---". It should continue to count down to zero below 1.8..I should see 1.7, 1.5 1.2 0.9 etc. and then zero. I do not see that.

That is not right.

It doesn't mean there's not a problem with the fuel pump. But I think today's tests show there's a sensor or wiring problem.

We also know that the readings come and go with heat. So the "intermittency" has to do with heat. What the heat is affecting has yet to be determined. I know what you THINK. But I'm going to determine that.



Regardless of who you believe here, I think most will tell you that flying this plane without replacing the pump, just isn't smart given the risks.

Who said anything about flying the plane before finding the exact problem and fixing it? I knew I could cause the problem to occur on the ground, as I showed this morning, mainly because of the failure during the taxi to the run up area.

The only difference between my position and yours (and others) is that I'd like to actually determine where the problem is instead of just starting to replace expensive parts and keep replacing parts until the problem goes away.

Perhaps you are thinking that my resistance to diving in right now and replacing the fuel pump is that I don't believe the problem is with the pump.

That's not it.

My way is to figure out what the problem *IS* and then fix it. And I never said it couldn't possibly be the pump.
.

Larry
Click to expand...

Ten more characters so that I can make the post
 
lr172 said:
It is a sandwich of rubber slices. Someone posted a teardown of a pump. I encourage you to look at his pictures. LOTS of small rubber parts. failure can happen at any time. It is not just a wear issue. I just had a pump fail with only 400 hours on it.

Larry
Click to expand...

Yes...anything can fail after 20 minutes use.
 
Steve Iacoviello said:
Sam, the sensor is bad as evidenced by the "blank" reading. With a good sensor installed one can determine if the fuel pump needs changing.
Click to expand...

Exactly. Also I can determine of the fuel pump is bad by hooking up a separate pressure gauge to the manifold block and read that at the same time I'm reading the EFIS.

Or another way to look at it is this:

With a good sensor/EFIS I can know that replacing the fuel pump fixed the problem.

With a bad sensor I'll still get a bad reading. Did replacing the fuel pump fix it?

People are assuming the EFIS/sensor is absolutely perfect and that the fuel pump is the spawn of the devil. I wonder why they assume the former in the face of what I've found.

If the fuel pump is bad (and I never said it wasn't), then I have two problems don't I? Because that reading on the EFIS is all wrong even if there were zero PSI in the line.

And so what if I replaced the fuel pump - I would still get an erroneous reading wouldn't I? So how do I know the new pump is working?

Since I can cause the fault to occur on the ground, I think a little dilly dallying and testing to pin down the precise cause is not completely unwarranted.

Yes I know a lot of you already "know" where the problem lies. And you all may very well be right. Do not take my resistance to diving in right now and changing the pump to mean I think the pump is fine. It could very well be the culprit - I've never said otherwise.
 
Ok time for a little update:

Last weekend, I was testing out the fuel pressure sender per Van's instructions when I noticed some fuel dripping from the left hand fuel tank air vent inlet.

(See The Case of the Dripping Fuel tank vent" in Ongoing Maintenance)

So here is my latest working theory:

When I start the engine the fuel pressure reads fine. By the time I get to the run-up area the EFIS shows "---".

I talked to TruTrak and they verified that this means either No Contact (i.e. broken wire) or out of range.

After noticing the dripping I opened the fuel cap for that tank and heard the suction sound - the tank wasn't getting any air from the vent. The exit end of the tank vent is above the level of the fuel in the tank. I blew air into the tank vent inlet and felt both air and gas coming out of the exit. Then I put my ear right on the wing over the point where the air vent enters the tank and blew some air. I could plainly hear bubbles.

So the interior connection to the fuel vent is loose - though not completely disconnected. I did see some fuel dripping from the vent once before but I thought that was because I filled the tank to the absolute brim prior to a cross country I was going to do the next day. I figured fuel sloshed in the tank when I pushed the plane back into the hangar and merely dripped out. Now I know differently.

My latest working theory is to why the fuel pressure slowly falls after starting is that the tank is not being vented. The open connection at the tank wall is always "underwater", so no air gets in. Over time, it becomes harder and harder for the Main Fuel Pump to pull gas from the tank. The pressure in the lines goes down over time. Turning on the Aux pump provides added force and so the pressure goes back up.

Luckily the tank didn't implode.

Why did the fuel pressure reading come back on the last flight as I was descending to the airport? My theory is that I was going pretty fast and the ram air forced some air into the tank and equalized the pressure: the connection isn't completely loose - I could still blow air into the tank if i blew hard enough. So my initial thought that this is heat related loses some credibility.

Anyway that's the working theory for now. As I have to fix the vent connection anyway, we will see if that was actually the problem.

So this weekend I drain the tank, remove the large access plate, and see if I can get a wrench onto the connection.
 
Saville said:
Ok time for a little update:
The open connection at the tank wall is always "underwater", so no air gets in. O
Click to expand...

Maybe I am not understanding (wouldn't be my first time), but, if you are dripping from the vent port, then there has to be an open path from that port to the tank...in which case, fuel being pumped out of the tank will be replaced by air through the vent, regardless if the vent is above or below the fuel level.

If the end of the line, in the tank, is below the fuel level, then, in the opposite scenario, positive pressure in the tank (e.g. a low pressure system moving through the area) will cause fuel to exit via the vent line, rather than air venting through that line.

In either case, the tank pressure differential vs Atm should be negligible and would not affect pump effectiveness.

I occasionally get a pressure drop from the mechanical pump after a long climb. I attribute that to either a failing pump or hot fuel close to the vapor point boiling at the pump inlet. Aux pump cures this condition when it occurs, but I will be inspecting and replacing the pump soon.
 
istrumit said:
Maybe I am not understanding (wouldn't be my first time), but, if you are dripping from the vent port, then there has to be an open path from that port to the tank...in which case, fuel being pumped out of the tank will be replaced by air through the vent, regardless if the vent is above or below the fuel level.

If the end of the line, in the tank, is below the fuel level, then, in the opposite scenario, positive pressure in the tank (e.g. a low pressure system moving through the area) will cause fuel to exit via the vent line, rather than air venting through that line.

In either case, the tank pressure differential vs Atm should be negligible and would not affect pump effectiveness.

I occasionally get a pressure drop from the mechanical pump after a long climb. I attribute that to either a failing pump or hot fuel close to the vapor point boiling at the pump inlet. Aux pump cures this condition when it occurs, but I will be inspecting and replacing the pump soon.
Click to expand...

Yes there is an open path from the port to the tank and that path is inside the tank where that vent line goes through the tank. The vent line is connected to a fitting on the inboard tank wall. The vent line comes in 2 parts...intake (under the fuselage) to the inboard tank wall exterior. And then another tube from the inboard tank wall fitting INSIDE the tank, outboard to the end of the tank.

What is loose is the connection on the INSIDE of the tank closest to the fuselage. The dihedral moves the gas to that wall. The outboard end of the vent line is near the gas cap and is NOT submerged.

Since the effective intake of the vent line is now "underwater" at all times, there can be no equalization of tank pressure.

As the fuel pump pressurizes the system, gas is drawn from the tank but the vent cannot replace the lost volume with air because the effective exit of the air vent is underwater.

If the air vent lines were connected solidly, then there would be a path from atmospheric air to the air pocket inside the tanks. But the connection is faulty so the atmospheric air cannot get bast the volume of gas into which it is submerged. Gas is incompressible.

This is a test being done on the ground.
 
Saville said:
But the connection is faulty so the atmospheric air cannot get bast the volume of gas into which it is submerged.
Click to expand...

This is incorrect. As that tank develops a negative pressure (i.e. vacuum) it will pull air from the external, higher pressure source even if that source is submerged in an imcompressable liquid. The liquid doesn't need to compress in order to let the air in.
 
Last edited:
Saville said:
Ok time for a little update:

Last weekend, I was testing out the fuel pressure sender per Van's instructions when I noticed some fuel dripping from the left hand fuel tank air vent inlet.

(See The Case of the Dripping Fuel tank vent" in Ongoing Maintenance)

So here is my latest working theory:

When I start the engine the fuel pressure reads fine. By the time I get to the run-up area the EFIS shows "---".

I talked to TruTrak and they verified that this means either No Contact (i.e. broken wire) or out of range.

After noticing the dripping I opened the fuel cap for that tank and heard the suction sound - the tank wasn't getting any air from the vent. The exit end of the tank vent is above the level of the fuel in the tank. I blew air into the tank vent inlet and felt both air and gas coming out of the exit. Then I put my ear right on the wing over the point where the air vent enters the tank and blew some air. I could plainly hear bubbles.

So the interior connection to the fuel vent is loose - though not completely disconnected. I did see some fuel dripping from the vent once before but I thought that was because I filled the tank to the absolute brim prior to a cross country I was going to do the next day. I figured fuel sloshed in the tank when I pushed the plane back into the hangar and merely dripped out. Now I know differently.

My latest working theory is to why the fuel pressure slowly falls after starting is that the tank is not being vented. The open connection at the tank wall is always "underwater", so no air gets in. Over time, it becomes harder and harder for the Main Fuel Pump to pull gas from the tank. The pressure in the lines goes down over time. Turning on the Aux pump provides added force and so the pressure goes back up.

Luckily the tank didn't implode.

Why did the fuel pressure reading come back on the last flight as I was descending to the airport? My theory is that I was going pretty fast and the ram air forced some air into the tank and equalized the pressure: the connection isn't completely loose - I could still blow air into the tank if i blew hard enough. So my initial thought that this is heat related loses some credibility.

Anyway that's the working theory for now. As I have to fix the vent connection anyway, we will see if that was actually the problem.

So this weekend I drain the tank, remove the large access plate, and see if I can get a wrench onto the connection.
Click to expand...

Perhaps a review is in order:
- It makes no difference if the vent line inside the tank is below the fuel level, other than if it is below the fuel level you will syphon vent gas out of the tank when the fuel expands (typically after filling and the sun hits the wing). As long as there is a differential pressure between the inside of the tank (when your fuel pump suck out fuel) you will draw in an equal, by volume, amount of air into the tank via the vent. If the vent is below the fuel level the air will just bubble up to the top of the tank.
- As you heard sucking sound when you removed the gas cap, I would first look to a clogged vent line as your root issue. While you may have been able to blow air into the line that may not mean you do not have a clog.
- I would think a partially blocked vent line would explain why your fuel pressure indication came back on decent - at the lower power you were drawing less fuel so the tank vacuum was not a great.
- I partially clogged line would also explain your "If I blow hard enough into the vent line" issue. If the gas cap is off it takes little to blow air into the tank via the vent line, unless the vent line is clogged.

Carl
RV-10 with 3/8" vent lines that have screens across them to keep out mud dabbers, and a check valve inside the wing so the vent can suck in air if the outside gets clogged with ice.
 
Carl Froehlich said:
Perhaps a review is in order:
- It makes no difference if the vent line inside the tank is below the fuel level, other than if it is below the fuel level you will syphon vent gas out of the tank when the fuel expands (typically after filling and the sun hits the wing). As long as there is a differential pressure between the inside of the tank (when your fuel pump suck out fuel) you will draw in an equal, by volume, amount of air into the tank via the vent. If the vent is below the fuel level the air will just bubble up to the top of the tank.

I don't see how of the vent line is disconnected. If there's no clear bath from the end of the vent line to ambient, how does the pressure get equalized?.

- As you heard sucking sound when you removed the gas cap, I would first look to a clogged vent line as your root issue. While you may have been able to blow air into the line that may not mean you do not have a clog.

I reported that I blew air through the line. Both air and fuel got through. Because BOTH air and fuel got through, I suspect a bad connection. Though is some ways that may behave like a clog. Also don't forget the bubbles..

- I would think a partially blocked vent line would explain why your fuel pressure indication came back on decent - at the lower power you were drawing less fuel so the tank vacuum was not a great.

I tried several different power settings at altitude including idle and the pressure indication did not return. That explanation doesn't hold.

- I partially clogged line would also explain your "If I blow hard enough into the vent line" issue. If the gas cap is off it takes little to blow air into the tank via the vent line, unless the vent line is clogged.

You are forgetting (or may not have read the post) that I could plainly hear bubbles coming out of the vent line at the inboard tank wall. This would not happen if there was only a clog.

Carl
RV-10 with 3/8" vent lines that have screens across them to keep out mud dabbers, and a check valve inside the wing so the vent can suck in air if the outside gets clogged with ice.

I wonder why Van went to the trouble of designing such a complication venting system that requires plumbing through the tank and then the fuselage.
Click to expand...


Tough to describe this situation in words so if you will pardon my crude drawings.

The system as it is supposed to be (as I understand it):



I've drawn the tank in grey and tilted it up exaggerating the dihedral of the wing. The left side of the drawing is against the fuselage - the right side is towards the wingtip.

The red lines represent the vent tubing and the green line the fuel line. The blue line in the upper corner of the picture is the level of the fuel in the tank - everything to the left in the tank is fuel. Note that the vent at the upper right end is in the air pocket.

When the system is at equilibrium, fuel pump off, the air pocket is at ambient pressure. This is because there's a direct and clear path to the ambient air via the red vent tube.

Now I turn on the pump, fuel comes out of the tank. The air pocket expands and lowers the pressure in the air pocket. Because the pressure is lower than ambient, the air flows from outside and brings the pressure in the pocket up to ambient.

Fuel pressure stays constant.

This is what I think the situation is in my tank:



The line is open at C - inside the tank.

There is no longer a clear path from the ambient air to the air pocket at the other end of the tank.

At equilibrium, say after I open the fuel cap, the air pocket is at ambient pressure.

Now turn on the fuel pump.

Gas is pumped out of the tank.

The air pocket grows larger and the pressure of the air in that pocket is reduced.

I do not see how ambient air is going to be sucked from the outside into that air pocket, because the gasoline is incompressible, and there is no longer a clear path from ambient to the pocket.

The air pressure in the pocket stays low until I open the gas cap whereupon I hear the sucking sound - the pressure is NOW back at ambient. There was a lower than ambient pressure in the air pocket.

And I know it's not connected because I could hear bubbles coming out of it when I put my ear on the wing skin above "C".

Maybe I do not understand the system, please explain how air from the outside can get into that air pocket if the vent is disconnected. If so I'd appreciate being educated.

Thanks
 
Last edited:
lr172 said:
This is incorrect. As that tank develops a negative pressure (i.e. vacuum) it will pull air from the external, higher pressure source even if that source is submerged in an imcompressable liquid. The liquid doesn't need to compress in order to let the air in.
Click to expand...

I think you are wrong - or at best do not understand my description:

At no point did I say the liquid needs to compress in order to let air in. Air comes in a working system because the fuel level goes down - the air pocket in the tank grows - the pressure of that air becomes lower than ambient, and so ambient air is sucked in (when testing on the tarmac) to equalize the pressure and keep the gas flowing. When flyign you have ram effect but these are all ground tests.

The incompressibility of the gas is what PREVENTS air from coming into the tank to replace fuel volume when the system is busted.

Difficult thing to discuss with words - I put some crude pictures in my response to Carl below.
 
Backup fuel pressure guage

I have a Trutrak EFIS and the same sensor manifold as you. I just added a second VDO sensor (using a 45 degree NPT fitting) to my manifold and a 1 1/4" UMA fuel pressure guage as a backup for my EFIS fuel pressure. I like my Trutrak but have backed up all its essential functions in case it goes out on a cross country since they don't make them anymore. John
 
You must log in or register to reply here.
Back
Top