What's new
Van's Air Force

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

Engine Stumble Switching Tanks

eastonale

Active Member
Patron
On my RV-10 I had a recent engine stumble that I'm trying to wrap my head around. After fueling we took off on the first flight of the day from northern CA eastbound and climbing though 17,000' I switched from left to right tank. As soon as I moved the handle the engine quit so I promptly rotated the fuel selector back and she came back to life. We decided to make a precautionary landing nearby to check things out and upon landing ran up the engine on both tanks and couldn't duplicate the problem, of course.

I'm running EFII System32 which has a recirculating fuel line that maintains constant fuel pressure to each injector and a return line to the same tank. There is no engine driven fuel pump but rather a primary and secondary electric pump, one of which is operating at all times. When this event happened the system operated properly switching from pump 1 to pump 2 when it detected the fuel pressure drop. We checked the tank vents, refueled, ran her up thoroughly and decided to continue our trip. No further hiccups.

So, the question is, who's got a theory as to why this happened? Did an air bubble form just prior to the fuel selector in the unused tank feed that hit the system when I switched tanks? Perhaps if I hadn't have switched back to the left tank so quickly after the engine stumble it would have cleared. By the way, I don't have any hot tunnel issues so I don't think heat would have been a factor. I do have a bit higher than average vibration issues and am having the prop balanced soon. Could a small amount of water become frozen in the right fuel line? This was about 25 minutes into the flight and the OAT was 20F at altitude. The day prior I had flown 2.5 hours from my home base to CA and used both tanks doing it. I also make a point to make sure I run up the engine from both tanks before departing so this wasn't the first time it had run off the right tank that day. I had checked a fuel sample from the tank drains after fueling that morning and didn't find water. I've since learned, by the way, that the Andair fuel selector is designed so that there is virtually no "dead spot" between positions and have verified this on the ground.

Any ideas? Other than the PTSD I now have every time I switch tanks, I haven't even had a hint of a problem since.
 
To begin with, I don't particularly like the EFII System 32, but that's not what caused your problem here. The same problem sometimes manifests in the SDS system as well.

When you're on the ground with ambient temp (or sun-soaked higher temp) fuel sitting in the tanks, there is more than enough pressure to keep everything in liquid form. When you take off and climb relatively rapidly, the pressure outside drops (so does the temperature) and you get closer to the vapor pressure limit where the fuel is more happy in a vapor state than a liquid state. The vapor pressure of the fuel is a function of temperature - so once you settle at altitude for a while and fuel in the tanks chills to the cooler temperature, life is good - but sometimes you can climb fast enough and high enough that you approach the vapor pressure limits before the fuel can cool off.

The fuel pumps used by both EFII System 32 and SDS do not like vapor - they are very good at pumping liquid, and very cranky about pumping vapor. If there is no realistic head pressure that the discharge is pushing against, then they can still pick up their prime and begin pumping - but both systems usually operate with a fuel loop in the 40psig range, and the pumps simply cannot pull up a vapor-laden fuel supply inlet against that much back pressure - it can sometimes take a few nerve-racking seconds.

What will happen initially is that you switch tanks (or pumps - same effect) and there will be an initial vapor bubble in the line from the warm fuel which has now reacted to the half-pressure outside atmosphere at 17,000' by partially flashing to vapor, and the pump will choke on this vapor and not be able to push it forward. The engine continues to burn fuel, but the pumps can't replace it, and the fuel pressure FWF will fall, rapidly approaching single digits, the engine stumbles because the injectors need about 40 psig, but now the pump can finally work the vapor out against the lower head pressure and it will pick up its prime and start moving fuel again. Depending on specifics, this process can be completely un-noticeable or 10 seconds, or anywhere in between. Definitely worse if you are running mogas or autogas with a higher vapor pressure.

After the first tank switch, the fuel in the tanks has cooled to ambient and the vapor pressure falls, and is usually not a problem.

One way to help this is by bringing online the second pump - watch the fuel pressure go up by a couple pounds, indicating both pumps are now pushing fuel, and then switch tanks. It's more likely that the total low-pressure episode is reduced in time this way that you don't even notice it without looking at the fuel pressure.
 
Last edited:
Fuel Pressure readings

What where the fuel pressure readings from your EFIS system before and after the stumble?
 
To begin with, I don't particularly like the EFII System 32, but that's not what caused your problem here. The same problem sometimes manifests in the SDS system as well.

When you're on the ground with ambient temp (or sun-soaked higher temp) fuel sitting in the tanks, there is more than enough pressure to keep everything in liquid form. When you take off and climb relatively rapidly, the pressure outside drops (so does the temperature) and you get closer to the vapor pressure limit where the fuel is more happy in a vapor state than a liquid state. The vapor pressure of the fuel is a function of temperature - so once you settle at altitude for a while and fuel in the tanks chills to the cooler temperature, life is good - but sometimes you can climb fast enough and high enough that you approach the vapor pressure limits before the fuel can cool off.

The fuel pumps used by both EFII System 32 and SDS do not like vapor - they are very good at pumping liquid, and very cranky about pumping vapor. If there is no realistic head pressure that the discharge is pushing against, then they can still pick up their prime and begin pumping - but both systems usually operate with a fuel loop in the 40psig range, and the pumps simply cannot pull up a vapor-laden fuel supply inlet against that much back pressure - it can sometimes take a few nerve-racking seconds.

What will happen initially is that you switch tanks (or pumps - same effect) and there will be an initial vapor bubble in the line from the warm fuel which has now reacted to the half-pressure outside atmosphere at 17,000' by partially flashing to vapor, and the pump will choke on this vapor and not be able to push it forward. The engine continues to burn fuel, but the pumps can't replace it, and the fuel pressure FWF will fall, rapidly approaching single digits, the engine stumbles because the injectors need about 40 psig, but now the pump can finally work the vapor out against the lower head pressure and it will pick up its prime and start moving fuel again. Depending on specifics, this process can be completely un-noticeable or 10 seconds, or anywhere in between. Definitely worse if you are running mogas or autogas with a higher vapor pressure.

After the first tank switch, the fuel in the tanks has cooled to ambient and the vapor pressure falls, and is usually not a problem.

One way to help this is by bringing online the second pump - watch the fuel pressure go up by a couple pounds, indicating both pumps are now pushing fuel, and then switch tanks. It's more likely that the total low-pressure episode is reduced in time this way that you don't even notice it without looking at the fuel pressure.

Thank you Greg for your advice. With the exception of your first comment, which is unnecessary, I think it makes a lot of sense. I appreciate your explanation.
 
Thank you Greg for your advice. With the exception of your first comment, which is unnecessary, I think it makes a lot of sense. I appreciate your explanation.

My opinion is based purely on the technical merits of the two systems, and is strictly that - my opinion.
 
The recirculating action of the efi (note lower case, brand agnostic) is magic on hot starts. The hot fuel firewall forward is quickly flushed out and replaced with the cooler stuff from the tanks - generally before you are even ready to hit the start button. That said, it might be a good practice to run on both tanks to run fuel through each side before you even start? Would take an extra 10 seconds.

As for the vaporization action once "up and away", I did add a Facet "clicker" cube pump to the outlet of one tank just in case the main pumps wouldnt draw vapor. So far it has not been needed, even with auto gas.
 
Doesn't the fuel valve have a close-before-open design for the supply and bypass? That would mean for the closure of the supply (and return) there would be no reservoir of fuel like is contained in the spring loaded fuel pump.

You should be able to calculate the loss of fuel by the time from decline to restoration of the pressure times the fuel flow. Does Ross know about this and recommend an accumulator?

Whose fuel valve? There are multiple vendors on the market. The Andair, *I think* has an overlap where there is no pressure loss - not 100% certain on that although I can say with 100% certainty that I have not experienced a fuel pressure drop using the Andair when switching tanks, in over 1000 hours of operation.

What Ross knows (or not), you'll have to ask Ross about. He does not recommend an accumulator.
 
Point of reference only;

The six-way Andair I have (and I assume all of them) appear to be a manifold type. Byproduct = the intermediate position allows flow from both inlets and a slight bit of capacitance in the valve body.

A ported ball valve, two coupled valves here, applied to a modern full flow FI systems would be more susceptible to moments of insufficient pump NPSHA and would behave in the way the OP described. Altitude obviously makes this worse. Very possible it was the perfect storm of conditions that day. Will be interesting to hear what type of fuel selector is involved. I'd also check for any evidence of fuel leaks in the associated lines.

Keep us informed.
 
I have the Andair valve and it does, in fact, have the capability of feeding from both tanks while in the "middle" position between left and right. I believe this to be required by law. I've verified it's operation while watching fuel pressure on the ground as fuel is recirculating at 35psi even when the engine isn't running.

The way the system works is that when the ignition system is powered up, it senses low fuel pressure, because it was off, and automatically switches to and engages pump 2. Fuel pressure is checked. The preflight procedure to then reset the pump system and that reengages pump 1. Pressure is checked again. So yes, both tanks are recirculated prior to start. I also want to make sure that both tanks have the ability to run the engine prior to takeoff so before doing the runup, I switch the fuel selector to the opposite tank.

I'm assuming the vapor theory is valid so I believe I'm going to adjust my fuel tank switching schedule in flight. Previously I had just switched every 10 gallons burned from alternating tanks and that's why this event happened about 30 minutes in at 17,000'. I'm now going to make it a habit to do a routine tank switch climbing though 10,000' if heading to high altitudes before a vapor bubble has a chance to develop. At least that's my idea at the moment.
 
Cool. Thx.

Do verify you're using 100LL.

With the a manifolded xfer valve, 100LL, good vent, clean tank pick-up screens, clean filter, etc. my suspicion moves toward a leaky fitting allowing air to creep in under the right conditions or water in the fuel. This would be a fun one to instrument and troubleshoot on the ground.

Keep us informed
 
I have the Andair valve and it does, in fact, have the capability of feeding from both tanks while in the "middle" position between left and right. I believe this to be required by law. I've verified it's operation while watching fuel pressure on the ground as fuel is recirculating at 35psi even when the engine isn't running.

The way the system works is that when the ignition system is powered up, it senses low fuel pressure, because it was off, and automatically switches to and engages pump 2. Fuel pressure is checked. The preflight procedure to then reset the pump system and that reengages pump 1. Pressure is checked again. So yes, both tanks are recirculated prior to start. I also want to make sure that both tanks have the ability to run the engine prior to takeoff so before doing the runup, I switch the fuel selector to the opposite tank.

I'm assuming the vapor theory is valid so I believe I'm going to adjust my fuel tank switching schedule in flight. Previously I had just switched every 10 gallons burned from alternating tanks and that's why this event happened about 30 minutes in at 17,000'. I'm now going to make it a habit to do a routine tank switch climbing though 10,000' if heading to high altitudes before a vapor bubble has a chance to develop. At least that's my idea at the moment.

Keep us in the loop Mark. Curious to hear if it happens again.
 
Probably a long shot, but each valve inlet has retaining screws that must be installed before use. If they are missing or loose, it is possible to suck air. Don’t ask me how I know!
 
Last edited:
Probably a long shot, but each valve inlet has retaining screws that must be installed before use. If they are missing or loose, it is possible to suck air. Don’t ask me how I know!

Robert, can you elaborate this one? Can’t say that I remember this.
 
Robert, can you elaborate this one? Can’t say that I remember this.

Hi Mark. The Andair valve I have has two 90 deg inlet fittings attached to the main valve body with small screws. The screws are not installed at the factory (at least on my fittings which are 90 deg to allow being repositioned to best suit fuel line routing from the wing tanks). They leave it to the builder to accomplish during installation. There are adequate warnings included in the instructions but …….?
BCBB6E67-60A0-42DE-BD4A-D66FCCB6E98B.jpg
 

Attachments

  • 7D39168E-797B-4D4A-B613-752099ABB88A.jpg
    7D39168E-797B-4D4A-B613-752099ABB88A.jpg
    105.3 KB · Views: 95
Hi Mark. The Andair valve I have has two 90 deg inlet fittings attached to the main valve body with small screws. The screws are not installed at the factory (at least on my fittings which are 90 deg to allow being repositioned to best suit fuel line routing from the wing tanks). They leave it to the builder to accomplish during installation. There are adequate warnings included in the instructions but …….?
View attachment 25717

Thanks, and we’re all good there.

For those interested, we’ve got 150 hours on the airplane and this is the first time we’ve had even a hint of an episode like this. However, now that I think of it, this was the first time that I had done an initial climb from sea level to over 17,000’. I’ve had it that high before but I stepped up later in the flight.
 
Thanks, and we’re all good there.

For those interested, we’ve got 150 hours on the airplane and this is the first time we’ve had even a hint of an episode like this. However, now that I think of it, this was the first time that I had done an initial climb from sea level to over 17,000’. I’ve had it that high before but I stepped up later in the flight.

I'm running 93E10 instead of 100LL by preference, which is more prone to vaporizing. Occasionally during the summer months I'll see something similar to what you describe, but rarely and it's easy to clear the bubble within a few seconds. Yesterday I departed my homefield at about 3pm, mid-90's on the ground, and climbed direct to 17,000' without any sign of unhappiness from the fuel system.

https://flightaware.com/live/flight/N16GN/history/20220511/1955Z/73XS/KGGG
 
I have the Andair valve and it does, in fact, have the capability of feeding from both tanks while in the "middle" position between left and right. I believe this to be required by law. .

It is most definitely not required by law and can be dangerous to the unaware on low wing airplanes. In non gravity feed airplanes, if the selector is set to both, the engine will stop running once one tank runs dry, as the fuel pump can no longer create a suction to pull fuel from the tank with gas in it, as it keeps pulling air from the empty tank. I recommend placarding your plane to prevent people from using that position.

Larry
 
Last edited:
It is most definitely not required by law and can be dangerous to the unaware on low wing airplanes. In non gravity feed airplanes, if the selector is set to both, the engine will stop running once one tank runs dry, as the fuel pump can no longer create a suction to pull fuel from the tank with gas in it, as it keeps pulling air from the empty tank. I recommend placarding your plane to prevent people from using that position.

Larry

I didn’t mean to say there was actually a “both” position. There isn’t. The requirement is that there isn’t a dead spot between the left and right to prevent the stoppage of fuel flow while switching. Just trying to say that my engine stumble wasn’t likely caused by the fuel selector.
 
Back
Top