What is this mystery fuel return line for?

[kenpilot]

I've been troubleshooting a few fuel issues/questions and I realized I don't really understand the fuel system as a whole. Tracing the lines left me more confused.

I consulted the RV-6A build manual I received when purchasing this aircraft (I'm the 4th owner), but I only see an extremely brief description of fuel hoses, boost pump, primer, etc. and a diagram of "typical fuel / oil hose routing". These resources don't actually describe the system as a whole, such as which component comes after which component in the path.

Anyway, here's the behavior that's puzzling me:

- With the engine OFF...
Flow: If I turn on the boost pump the fuel flow reads 2.0 GPH. Obviously there is no fuel flowing into the carburetor, so this seems to be just 2.0 GPH of fuel flowing around in a circle. This is what initially caused me to start tracing lines, wondering where the fuel was flowing to.
Pressure: However, what really concerns me here is that the fuel *pressure* remains at near zero (0.1-0.3 psi) when the boost pump is on. If the mystery line is some sort of relief valve, I'd still expect full boost pressure.

- When the engine is ON...
Flow: I see 4-5 GPH as expected (that is normal for taxi, right?). But if the pump is capable of circulating fuel here, then I'd expect to always always see a +2 GPH increase with the boost pump on, but I've never noticed that.
Pressure: Typically 2-3 psi, but that would be due to the engine-driven pump taking over the role.

- In flight, I see behavior that might be considered odd.
Flow: Occasionally in cruise, with no other factors changing (no leaning adjustments, no temperature changes), the GPH will gradually shift +/- 0.5 GPH.
Pressure: Fuel pressure is also constantly changing, with no apparent pattern, anywhere from 2.0 to 5.5 psi. Turning the boost pump on _seems to_ provide a pressure boost, maybe, hard to tell as it's not a definitive "boost" as I'd expect.

Could anyone help me answer a few questions...
1. What is the "mystery line" and the devices on each end for?
2. With this fuel system design, and based on the behaviors I described, is the boost pump actually providing any protection in case of engine pump failure?
3. What increase in pressure "should" I be seeing when the boost pump is on?
4. How do you all typically test this stuff?
5. Is there any official recommendations of fuel system design I just haven't found yet?
6. Is it just me, or is 2.0 GPH really really low for a boost pump. Shouldn't it be at least 8.0 GPH, to provide full cruise power for the engine?

 

[rvbuilder2002]

Without knowing what the mystery question mark devices are it is not possible to answer most of your questions.
What I can tell you is that your system is not typical of a carbureted engine installation on aircraft . Experimental or certificated aircraft.

System pressure with boost pump on and engine not running is usually somewhere between 4-5 psi.

With the engine not running fuel flow should technically be zero but there is usually a small amount indicated because of fuel pulsation in the system caused by the pump. The actual amount varies from airplane to airplane depending on where the pump and flow transducer are located in the system.

 

[kenpilot]

Hmm. That's what I was afraid of. To my knowledge, and the owner before me, this airplane has always had a standard Vans design. I see no build notes or logbook entries stating anything otherwise. I might not have dug into this, except for the concerning fuel pressure issue.

When I traced the lines, those devices looked small and rectangular, made of metal, without markings. I'll see if I can get some photos of those devices, and a mirror on the backside.

 

[BH1166]

Many times it is the measuring that is incorrect, nothing more. What EIS is in use, model? Nothing weeping out of engine driven pump ? What are you troubleshooting… issue?

 

[FinnFlyer]

Are they simply small blocks with three ports? Some kind of attempt of an air purge line? With tiny holes to the mystery line that may be possible.

However I don't understand how that would purge air. I would expect the purge line to go back to the tanks. But then I don't understand how the vent on the fuel pump works.

Finn

 

[lr172]

That appears to be a poor attempt at a fuel recirculation line. In my plane it is a fuel return line. It tee's off just before the servo, just like yours. It then goes to a pilot controlled valve and then to the left tank. A key difference on mine is that it sits upstream of the red cube. The fuel gets hot in some cases and can boil and create vapor lock or other issues. This is not much of an issue with 100LL in most cases, but is more so with mogas. Mine has a restrictor in the line that limits return volume to about 8-10 GPH, to avoid a variety of potential problems. I mostly just use it after a quick turn fuel stop to flush the very hot fuel back to the tank (it is being replaced with cooler fuel) before startup and leave it open until just before TO. This is the scenario that most have rough running due to boiling fuel. This is less of an issue with carb users, as the vapor created from boiling gets vented once hitting the carb bowl. However, if the fuel in the bowl starts to boil that is an even bigger problem; Good news is that is pretty rare, but can happen. This is why you don't go straight from startup to TO after a quick fuel stop on a hot day if you have a carb. Spend a bit of time running the engine to get that hot fuel out of the bowl and for the engine roughness to settle down.

The design here is poor, as it creates constant recirculation, doesn't go back to the tank where it needs to to get cool fuel, and can't be shut off. It mostly explains your symptoms. It likely has a restrictor in the line. I don't believe it is doing much to help and would either remove it or change it to a more traditional fuel return arrangement. This is making your fuel flow data almost useless.

Larry

 

[kenpilot]

Many times it is the measuring that is incorrect, nothing more. What EIS is in use, model? Nothing weeping out of engine driven pump ? What are you troubleshooting… issue?

Yes, small blocks with three ports. I'll do a more thorough check for markings next time I'm at the hangar. But if there aren't any, I wouldn't know what they are without removing and testing.

It's an EIS 4000 from Grand Rapids (GRT). Data is displayed on the GRT Sport SX. Yes, the engine-driven pump has a leak, but it appears to be oil wetness, not fuel; there's no blue stain. I've considered that to be a separate issue for a future date.

Primarily, I'm troubleshooting the fuel pressure alerts. In flight, I often see the pressure go down as low as 0.3 for a few minutes. That's close enough to zero that normal fluctuations cause alerts no matter where I set the alert point. I finally had to disable the alert. Looking at historical data, I see one flight where the pressure went to 0.0 right before starting takeoff. Now that I am in the system working on fuel leaks and other things, it's a good time to figure out why the pressure is so low in the first place.

Secondarily, I'm trying to figure out what "correct" looks like. I've visited other builders to see their fuel systems. But is there an official "Vans design" anywhere?

 

[lr172]

Are they simply small blocks with three ports? Some kind of attempt of an air purge line? With tiny holes to the mystery line that may be possible.

However I don't understand how that would purge air. I would expect the purge line to go back to the tanks. But then I don't understand how the vent on the fuel pump works.

Finn

The purpose of a purge valve and line is NOT to purge air, but to purge hot fuel and replace it with cooler fuel. There is no vent in a fuel pump. That small fitting poking out of it is NOT a vent as you are thiking. It is an exit path from the non fuel area of the pump. If a diaphragm breaks, it will spill fuel into that area. If that area wasn't vented, the pump would stop pumping fuel once that cavity filled with fuel. With that vent in place, a pump with a failed diaphragm can still pump a reasonable amount of fuel even though it is leaking. This is a limp home, safety feature.

 

[kenpilot]

The purpose of a purge valve and line is NOT to purge air, but to purge hot fuel and replace it with cooler fuel. There is no vent in a fuel pump. That small fitting poking out of it is NOT a vent as you are thiking. It is an exit path from the non fuel area of the pump. If a diaphragm breaks, it spill fuel into that area. If that area wasn't vented, the pump would stop pumping fuel. With that vent in place, a pump with a failed diaphragm can still pump a reasonable amount of fuel even though it is leaking. This is a limp home feature.

The thing I'm calling a vent is a clear vinyl tube that runs to the lower cowl edge, right alongside the engine breather tube. Thank you for explaining what it's really for!

 

[lr172]

Y

Primarily, I'm troubleshooting the fuel pressure alerts. In flight, I've had the pressure go down as low as 0.3. That's close enough to zero that normal fluctuations cause alerts no matter where I set the alert point. I finally had to disable the alert. Now that I am in the system working on fuel leaks and other things, it's a good time to figure out why the pressure is so low in the first place.

Yes, this recirc circuit is going to wreak havoc on your fuel pressure, in addition to the fuel flow readings. Again, I recommend getting rid of it, as mentioned above.

 

[kenpilot]

The purpose of a purge valve and line is NOT to purge air, but to purge hot fuel and replace it with cooler fuel. There is no vent in a fuel pump. That small fitting poking out of it is NOT a vent as you are thiking. It is an exit path from the non fuel area of the pump. If a diaphragm breaks, it will spill fuel into that area. If that area wasn't vented, the pump would stop pumping fuel once that cavity filled with fuel. With that vent in place, a pump with a failed diaphragm can still pump a reasonable amount of fuel even though it is leaking. This is a limp home, safety feature.

The thing I am calling a vent is a clear vinyl tube that runs down to the lower cowl alongside the engine breather line. Thanks for your explanation. Perhaps that also explains why the tube is so clean and has no blue staining. So long as my pump has no internal failure, nothing comes out of the tube.

 

[rocketbob]

Early model Bonanzas have a fuel vapor recirculation line. Pressure carb models return to the left tank. Later constant flow injected models go thru the selector back to the selected tank.

In this case your issue is where the recirculation line is plumbed to and if it were me I would check the restrictor size and plumb it ahead of the red cube. Would also return it to a tank rather than the selector, which might be a considerable project.

I have flown a number of carbed RV's that exhibited vapor lock on 100LL on very hot days and consider a vapor return line a good feature when properly done.

 

[FinnFlyer]

I would get rid of (or cap the purge line port) of the block closest to the fuel selector valve. Then add a purge valve and tee its output into a tank vent line as close as possible to the tank. Or better, if you are going to work on your tank or tanks, add a fitting in the root rib for the purge line return.

Or just remove or cap the two blocks. I suspect they may be made by the builder from alum bar. Curious to know the restrictor hole size.

I'm also wondering why the builder installed it in the first place. Fuel lines or booster pump mounted too close to heat sources? Perhaps heat shields would be a better solution.

Finn

 

[N427EF]

One thing at a time

I agree with the above poster.
If it was me, I would remove that line and cap the ports. Nothing else for the moment. This will simplify the installation and present you with a couple of options. Leave it be without return line or embark on a plumbing venture and install a properly designed return line.
Run the engine and see what your indicators tell you.
My guess is that your fuel flow and pressure indications will look "normal".
Later you can decide if you want to pursue the installation of a functional return line with expected benefits as described by Ir172.

As someone who has sold an airplane with a functional return line to one tank only, I recall solving the confusion for the buyer by having him remove the return line and cap the ports.
The return line was beneficial for the use of mogas but the buyer was only using avgas and hence had no real need for a return line.

In my RV-10, I use a constant flow return line as well but in this airplane I can select the return tank as well as shut it off.

 

[Stevea]

6/6A fuel system schematic

I've been troubleshooting a few fuel issues/questions and I realized I don't really understand the fuel system as a whole. Tracing the lines left me more confused.

I consulted the RV-6A build manual I received when purchasing this aircraft (I'm the 4th owner), but I only see an extremely brief description of fuel hoses, boost pump, primer, etc. and a diagram of "typical fuel / oil hose routing". These resources don't actually describe the system as a whole, such as which component comes after which component in the path.

You're right, there is not a fuel schematic in the original 6/6A build manual. Here is one I put together for my POH, based on the fuel system drawings (fuel tanks, in cabin fuel plumbing, and firewall forward plumbing). It is the standard Van's carburetor fuel system design with two additions:

• Electric solenoid primer system (Van's option)

• Fuel flow transducer, located between the engine driven fuel pump and the carburetor (Location based on JPI instructions for the transducer. The same location is recommended for the "red cube" transducer on a carb'd engine)

View attachment fuel system schematic.pdf

As others have pointed out, your system has some "non standard" parts in it, which may not be doing you anything good.

Edit: OK, just noticed my schematic doesn't show the mechanical fuel pump overboard drain.......but it really is there on my airplane. Time to edit my schematic.

 

[rvbuilder2002]

All good advice which reenforces what I wrote initially… that until you know what the question mark parts are, it is anyones guess.
Based on your fuel pressure value, my guess would be that there is no type of restrictor installed which is not good. If that is the case. There may as well not even be an electric backup pump because it can’t do much if anything at all.

 

[BH1166]

Yes, small blocks with three ports. I'll do a more thorough check for markings next time I'm at the hangar. But if there aren't any, I wouldn't know what they are without removing and testing.

It's an EIS 4000 from Grand Rapids (GRT). Data is displayed on the GRT Sport SX. Yes, the engine-driven pump has a leak, but it appears to be oil wetness, not fuel; there's no blue stain. I've considered that to be a separate issue for a future date.

Primarily, I'm troubleshooting the fuel pressure alerts. In flight, I often see the pressure go down as low as 0.3 for a few minutes. That's close enough to zero that normal fluctuations cause alerts no matter where I set the alert point. I finally had to disable the alert. Looking at historical data, I see one flight where the pressure went to 0.0 right before starting takeoff. Now that I am in the system working on fuel leaks and other things, it's a good time to figure out why the pressure is so low in the first place.

Secondarily, I'm trying to figure out what "correct" looks like. I've visited other builders to see their fuel systems. But is there an official "Vans design" anywhere?

Good I have same EIS4000 & Sport SX now 10.1 Sport. See if you can validate the color bands on the resistor that powers fuel pressure. Mine was wonky for years ( bought my 6A). Finally had enough, looked as I recommend here, wrong resistor. Changed ( pain in the arse) problem solved.

 

[kenpilot]

Went to the hangar this morning...

See photos. As I had remembered, there is no P/N on these mystery devices. Some marks that might be branding; not sure. They look like generic brass tee fittings you might find at Home Depot, or they are just fabricated in the shop as ___ mentioned.














I'm pretty sure there is no restrictor feature. The one on the carb has the mystery line on a side fitting; the one in the cabin has it on the center fitting. Neither has any special markings that would indicate those are restricted lines, and there's no dedicated restrictor adapter either.

I'll probably cap the cabin end of the line, to start with, and verify no negative effects. I like the idea of having a circulation line for vapor lock, but I'll save that for a future project, when I can get a proper pilot-operated valve for it.

One final question, if you don't mind: When I hear "vapor lock" I think hard to start the engine after it's been sitting in the sun, ground-level non-safety issues. But is there any "safety of flight" concern with vapor lock? If I postpone installing that circulation line to the fuel tank, am I taking on some flight risk I'm not aware of.

 

[kenpilot]

Good I have same EIS4000 & Sport SX now 10.1 Sport. See if you can validate the color bands on the resistor that powers fuel pressure. Mine was wonky for years ( bought my 6A). Finally had enough, looked as I recommend here, wrong resistor. Changed ( pain in the arse) problem solved.

Thanks for the tip, Butch. I'll consider this when I tackle the larger wiring project. At the moment, I have no idea where the resistors are, as the wiring is all bundled and unmarked.

 

[rocketbob]

Your fuel system needs a serious re-do.

 

[N427EF]

Your fuel system needs a serious re-do.

I agree.

They look like generic brass tee fittings you might find at Home Depot,

Nothing wrong with those except they are heavy but cheap and those brass fittings are not the problem in your system.
Van sells a similar fitting block made of aluminum.

In the meantime, your plan to cap one end of it is a good first step in verifying what you are dealing with. Make sure you cap both the hose and the brass fitting. For a quick check if you are impatient you could just pinch that "mystery" hose with a large clamp and do a run up before removing fittings.
It is evident from the pictures as well as the age of the plane (what is the age and hours on it?) that the fuel system installation is due for an overhaul/redo with proper hardware.

I like the idea of having a circulation line for vapor lock, but I'll save that for a future project,

Good plan!
Let us know what you find and stay safe.

 

[kenpilot]

It is evident from the pictures as well as the age of the plane (what is the age and hours on it?) that the fuel system installation is due for an overhaul/redo with proper hardware.

21 years since its first flight. ~600 hours.

Yes it probably needs a redo. I'd love to replace everything with flex lines. But I think all of that is more of a project than I'm able to take on right now, considering I'm losing my hangar at the end of the year due to factors beyond my control. With all the other issues discovered during this condition inspection, adding that would probably take me down all winter.

My friends tell me by the time I'm done with this condition inspection, I'll be ready to build my own RV. lol

 

[lr172]

I'm pretty sure there is no restrictor feature..

You can't really know that from an outside look. I used a std steel JIC fitting (same as AN fitting) and used silver solder to fill the internal passage, then used a small drill bit to get the orifice size I wanted to reach the target return fuel flow I wanted.

Larry

 

[kenpilot]

used silver solder to fill the internal passage, then used a small drill bit to get the orifice size I wanted to reach the target return fuel flow I wanted.

Oh! I'll probably still cap off this line. But now you have me curious. I should disconnect and put into a bucket and measure flow first.

 

[OKAV8r]

I've seen two RV-6As with similar setups. Both had shut off valves in the return line adjacent to the main spar carrythrough. One had a restriction fitting, and one did not. Both owners were not the original builders, and one did not know the line was there, or what it would do.
It would be interesting to hear back what you find as far as a line restrictor goes. Fortunately, a carb doesn't need much line pressure to operate.

 

[daddyman]

What I tried to say, yet more eloquently

That appears to be a poor attempt at a fuel recirculation line. In my plane it is a fuel return line. It tee's off just before the servo, just like yours. It then goes to a pilot controlled valve and then to the left tank. A key difference on mine is that it sits upstream of the red cube. The fuel gets hot in some cases and can boil and create vapor lock or other issues. This is not much of an issue with 100LL in most cases, but is more so with mogas. Mine has a restrictor in the line that limits return volume to about 8-10 GPH, to avoid a variety of potential problems. I mostly just use it after a quick turn fuel stop to flush the very hot fuel back to the tank (it is being replaced with cooler fuel) before startup and leave it open until just before TO. This is the scenario that most have rough running due to boiling fuel. This is less of an issue with carb users, as the vapor created from boiling gets vented once hitting the carb bowl. However, if the fuel in the bowl starts to boil that is an even bigger problem; Good news is that is pretty rare, but can happen. This is why you don't go straight from startup to TO after a quick fuel stop on a hot day if you have a carb. Spend a bit of time running the engine to get that hot fuel out of the bowl and for the engine roughness to settle down.

The design here is poor, as it creates constant recirculation, doesn't go back to the tank where it needs to to get cool fuel, and can't be shut off. It mostly explains your symptoms. It likely has a restrictor in the line. I don't believe it is doing much to help and would either remove it or change it to a more traditional fuel return arrangement. This is making your fuel flow data almost useless.

Larry

Kenneth,
This is what I tried to explain in my hanger at EZF.
I assume the builder wanted to put in a fuel inj system at some point in the future.
Daddyman

 

[kenpilot]

Posting an update to this fiasco...

Removal of the return line
Yes, I will be removing the fuel return line. But because I don't know what I don't know, I'm going to go the safe route first: cut the line and install a shutoff valve that will be in the OFF position in-flight. I'll fly for a dozen hours, and if I never need the line, I'll remove it.

I spoke with the local EAA chapter, who pointed out that the Facet 40108 boost pump I have (which is recommended by Vans in the original design) is marked by the manufacturer and Aircraft Spruce as "not intended for aircraft use". Supposedly there was a case where it failed in such a way as to block all fuel, resulting in fuel starvation and an off-airport landing. So before I take the return line out completely, I plan to install a short bypass line (with a check valve) around the Facet pump, to ensure it is not the "single point of failure" of the system.

In the meantime, this return line (with a shut-off valve installed) will be the bypass for this purpose.

Waiting on parts from ACS to get this done next week.

Hypothesis about the puzzling flow and pressure readings
Yesterday I had a hypothesis that what I was actually witnessing was pump cavitation. I kept asking myself why am I seeing zero pressure with the boost pump on, when I definitely do see pressure in-flight. If the return line is allowing 100% of the boost pump's flow back through on the ground, it would do the same in-flight.

But I put the main line into a bucket, capped the return line, and measured a full 15+ GPH going into the bucket. Reassembled everything and still saw 2-3 GPH and zero pressure. Made no sense.

I finally realized... all my ground tests here were AFTER I had done my carburetor inspection items for this CI (drained the bowl, moved the throttle lever back and forth to check smoothness and security, etc.). I introduced a bunch of air to the lines! Gravity would have drained the fuel from the lines going back to the highest points. By distance, probably 1/2 of the "circle" was air, while the other 1/2 was fuel.

In this condition, my boost pump would have been cavitating in all my tests, even that last test after using the bucket. Without the engine running, there was nothing to purge that air. This is why I saw only 2.0 GPH and 0.0 PSI, when the boost pump is clearly rated for 32.0 GPH and 6.0 PSI.

Confirmation of hypothesis
After some careful steps to ensure the system was reassembled with all lines purged of air, I now see 2.9 GPH (the flow back through the return line) and 2.8 PSI (now that the pump isn't cavitating, it is trying to push the fuel faster than the return line will allow).

That confirms the the boost pump really is providing the safety net in case of engine-driven pump failure.

Wow, this whole situation scared me there for a minute.

Two final tests
During this process, I had an opportunity to measure the output of my boost pump. While it's rated for 32.0 GPH, I believe my fuel lines and fittings are a slight bottleneck. I confirmed (using a bucket and timer) 18.7 GPH steady stream without air bubbles. And with everything capped for absolute zero flow, I got the pressure up to 6.0 PSI.

This also confirms the resistor on the fuel pressure probe is good.

EIS calibration
While fuel was pouring into the bucket at 18.7 GPH (measured with a timer), my EFIS said only 12.0 GPH was flowing. I have a hunch the engine monitor has a calibration factor to account for that ~3 GPH flowing back through the return line. I'll have to recalibrate it now.

Lessons learned?
If I ever hear a boost pump as loud as mine was, I'm not going to assume it's normal. These facet pumps should make a mid-volume chugging sound, but not the knocking-on-metal harshness that makes you want to turn it off as soon as possible.

 

[Dan 57]

Thanks for the feedback Ken.
Interesting case u had, and one for sure never stops learning.

The builder of my -6.9i designed the fuel system for Mogas use from the onset. A fuel return line sends some fuel back to the respective tank thru a dual fuel selector (Andair).
I use 99% Mogas or Autogas for my flying, and the only cases of vapor lock are after shutdown on a hot day, and I then don’t take the time for the fuel to circulate.

As u can see the system comprises the return line from the carb with a one-way check valve, the dual selector with a fuel return line to the selected tank, and dual electrical fuel pumps in parallel.