Returnless FI?

[Archer39J]

svyolo Great stuff John, thanks for sharing your setup!

That last part about putting the regulator in or on the surge tank I assume means a pressurized tank? If it's being returned after the pumps. Yeah I don't know if I'd be ok with that lol.

That's what I liked about the SDS fuel rail, it could even be possible to mount it on the cabin side of the firewall and just have the pressurized lines penetrate through to the injectors. None of the returned fuel would be heat soaked by the engine. But an engine side mount would be the place to start, then shielding, then cooling, but I don't think those would be necessary. I recall the pressure being 45-50 psi, so once the pump is on there's no danger of vapor lock on the engine side. The concern I've heard is with the pump inlets supposedly "sucking" hot fuel from the regulator return and vapor locking, but only in the context of low wings that are drawing fuel up from tanks. On the Bearhawk a quick calc says we get 1.13 psi at roughly minimum fuel, not accounting for head loss which can be not insignificant.

With the "stock" design returning straight to the tanks via duplex valve I would be worried about supply flow at 45 gph too. Keeping a closed loop eliminates this concern, similarly to your surge tank.

[svyolo]

Dave;
Returning fuel via a duplex valve at first seemed pretty complicated. But I soon realized that was not the primary problem. The problem was FEEDING the 45gph through 3/8" fuel lines, and the small duplex valves. I don't believe the lines will do it, and I thing the valve is a restriction as well. I read a few threads on BH's doing fuel flow tests, and none of them that I recall approached the required fuel flow. I think, reading the regs, we would need 125% of 45 gph, which is 57 gph. Continuously. Groco, if I remember right, has a marine duplex valve with much bigger ports, but the whole thing was very big and heavy.

I think you would need 1/2" fuel lines, and a large capacity valve. Returning to a small tank fixes this, as the BH fuel system only has to supply engine fuel.

Returning fuel directly to the small header tank does not pressurize the tank, as the tank has to be vented to the main tank(s). It doesn't matter if the pressure regulator is in the tank, or on the other side of the firewall. I guess the header tank is slightly pressurized by the 3 ft of head pressure from the mains.

If I do have excessive fuel heating I will move the fuel block behind the engine tins, but still at the high point of the pressurized fuel system. Having a return bleed from the high point of the pressurized section makes removing any air very simple, and continuous.

Right now I am only planning on a single 3 ft line from the pump module to the firewall fitting that is pressurized on the back side of the firewall. All other pressure fittings are ahead of the firewall. I would like to keep it that way.

The surge tank and the pumps sit under the right seat, inside the primary structure. If I had a tank with internal pumps, I am pretty sure I can get 2-3 gallons sitting there.

[zkelley2]

45GPH, what engine are you running? That's ~500HP.

[svyolo]

I am pretty sure you could get a return-less fuel system to work. I think it still involves a small header tank. For now, I am satisfied that what I have will work. My only concern, and it is minor, and fixable, is fuel heating.

[Ed.Meyer]

On our Patrol with a Bob O360, we used the EFII system. I spent many hours thinking about all this and went with return lines to the tanks via a duplex SPRL fuel valve. Had to add bungs to the tanks for the returns. According to Robert at EFII, 3/8 lines are big enough and after about 140 hours flying there have been zero problems. I normally fly feeding (and returning) with “both” selected. The fuel supply lines are as Bob designed through the fuel valve and gascolator then through an EFII supplied in-line filter to the pumps. If I was to build again, I would do the same thing. Just another data point...

[Ed.Meyer]

One additional thing: often when first powering up the system activating the fuel pumps, within seconds I hear bubbles in a fuel tank. I have no idea how air (or whatever gas) is getting in the system while sitting but I like knowing it purges right away.

[Archer39J]

Returning fuel directly to the small header tank does not pressurize the tank, as the tank has to be vented to the main tank(s). It doesn't matter if the pressure regulator is in the tank, or on the other side of the firewall. I guess the header tank is slightly pressurized by the 3 ft of head pressure from the mains.

Oh yeah I know yours isn't like that. Just the configuration you described with the surge tank teeing in after the pumps would result in 45-50 psi in the tank, which I would not want sitting under my seat.

If I do have excessive fuel heating I will move the fuel block behind the engine tins, but still at the high point of the pressurized fuel system. Having a return bleed from the high point of the pressurized section makes removing any air very simple, and continuous.

Right now I am only planning on a single 3 ft line from the pump module to the firewall fitting that is pressurized on the back side of the firewall. All other pressure fittings are ahead of the firewall. I would like to keep it that way.

Ah, you did the case mount fuel block then? My order has that now but I’m switching to one I can remotely mount. I’d figure the fleet history and the fact that you have wing tanks will prove your setup sufficient. I don’t know about trying to remove air from the pressurized system, that sounds complicated.

I too am trying to minimize pressurized lines in the cabin, as well as return lines or tanks in general. The less fuel I can surround myself with the better I figure.

On our Patrol with a Bob O360, we used the EFII system. I spent many hours thinking about all this and went with return lines to the tanks via a duplex SPRL fuel valve. Had to add bungs to the tanks for the returns. According to Robert at EFII, 3/8 lines are big enough and after about 140 hours flying there have been zero problems. I normally fly feeding (and returning) with “both” selected. The fuel supply lines are as Bob designed through the fuel valve and gascolator then through an EFII supplied in-line filter to the pumps. If I was to build again, I would do the same thing. Just another data point...

Nice, thanks for the pirep! Your setup is the same as I’d planned for my return. Even got the same Newton valve, maybe the last one from ACS actually as they don’t seem to make that version anymore.

I want to be careful not to dissuade anyone reading this from going the full return or header route. Between the two main suppliers there are thousands of RVs out there with either header tanks or full returns using 3/8 lines working just fine, and those are drawing fuel UP from a single tank. 45 GPH doesn’t cause them any problems and with our high wings this is even better, Ed’s install suggests this isn’t an issue for our planes. EDIT: Brooks brought up that a full return could over tax the stock supply capacity, so that's something to consider, as well as the flying examples...

I am pretty sure you could get a return-less fuel system to work. I think it still involves a small header tank. For now, I am satisfied that what I have will work. My only concern, and it is minor, and fixable, is fuel heating.

Yeah it’s looking like I may be able to. Issues are heat rejection and purging air from the loop.

Fuel heating can be largely avoided to begin with, and though I haven’t done an in-depth analysis yet I expect the head pressure from our tanks will obviate the need for further cooling. Heck if you really wanted to you could plumb in a little fuel cooler, heat exchangers of that type are relatively cheap, just looking at diesel examples.


The other issue is sucking air into the lines but as I mention this isn’t possible with the Bob fuel system without a concerted effort. Not to confuse folks, one commenter was asserting I made claims I didn’t, unporting the inboard tank pickups obviously happens in normal (uncoordinated) flight and probably when getting thrown about. But.. this should be a limited event, this would only be an issue at minimum fuel and, if you’re running BOTH this isn’t an issue at all. If these assumptions weren’t correct we’d be hearing about stuttering engines and this would affect anyone using Bob’s design, though I understand this kind of deductive reasoning is too much for some.

[Archer39J]

Talking with the designer it looks like I'm on the right track and have identified the salient issues. We agree running a tank dry should not be a SOP and running on BOTH eliminates this possibility.

[Archer39J]

For clarity, he didn't see anything wrong with this design or my assumptions. He did confirm this would be the first example of headerless loop return. He also mentioned that a return vent wouldn't be necessary from a header tank, so although it looks like I'm good to not use one it can be added later if need be.

[svyolo]

One of the "ultimate" tests for the fuel system, is when you are very low on fuel, and are in a climbing or descending turn, maybe only 1 fuel line feeding, does your engine still turn? With Bob's system, feeding the engines' requirements, it seems the answer is yes. I don't personally have the answer to the question when I have to feed the EFI pump. A small header tank answers it for me. BH's are not certified, and I doubt anybody has done that test with EFI pumps.

The auto surge tanks have two functions. Off road and racing guys like them for the same reason I do. They provide a small reserve of fuel to the feed side of the pump, in ununusual attitudes or G forces.

This is experimental aviation, this was the solution that made me comfortable. Maybe I will not have thought of something, and it doesn't work.

[svyolo]

Dave;
I found this about 3 link inside a link deep.

I would add that "driveability" is far less important than in a car or motorcycle. We tend to run engines at a constant speed most of the time.

https://www.onallcylinders.com/2017/...ar-the-engine/

[Archer39J]

Interesting article thanks. I figure on keeping the regulator where it's recommend and just having the pressurized lines penetrate the firewall.

[svyolo]

I think I went down the same logic train you did, 9-12 months prior. I got to the point that I felt I could do a return-less system and have it work. But it included a small, "bleed" line to purge air or fuel vapors that came out of suspension. If I need a small return, a full size one back to the header tank didn't seem any harder, which left where I have been for the last 9 months or so. Small header, full return to the header. Header is vented to the mains. Mostly happy at 2 liters.

In some ways it sounds more complicated than a duplex valve. In reality it is about equal. I did have to buy (70 bucks if I remember) a small tank, but that is cheaper than a duplex valve. I think the complexity is similar, the tank provides more of a reservouis, and it is actually cheaper. In addition, I need a MAP line back to the header tank. So I backed off, and decided on a simple full return system, with the return to the header tank. No MAP line.

I hope it works. It sounds good, in theory.

[Bcone1381]

I want to be careful not to dissuade anyone reading this from going the full return or header route. Between the two main suppliers there are thousands of RVs out there with either header tanks or full returns using 3/8 lines working just fine, and those are drawing fuel UP from a single tank. 45 GPH doesn’t cause them any problems and with our high wings this is even better, Ed’s install suggests this isn’t an issue for our planes.

This response deals with the quote above. It does not deal with a fuel system that has a header tank. I fear the quote has some assumptions made that could lead to poor fuel system installations if the above is applied. So I'll share my viewpoint. But it contradicts the quote above.

If a Bearhawk fuel system demands 45 gallons per hour, and it has 3/8" supply lines, then I wonder if the fuel system is still feeding via gravity. It may in fact be sucking fuel out of the tank rather than gravity feed. My gut tells it would be sucking. If the Fuel Flow Test wont drain 45 gallons per hour into a bucket from the tank to the pump, then it fair to assume its sucking fuel out of the tank under normal operation.

An RV system sucks fuel out of a tank. A Piper Cherokee sucks fuel out of a tank. (In these aircraft each fuel tank has only one fuel line supplying the fuel system, not two like a Bob's design) These low wing fuel systems don't not have a "BOTH" selection on the fuel sector. Rather its merely L - R - OFF. Thats because if there was a BOTH, then as soon as one tank runs dry the fuel pump will suck air from the empty tank instead of feed fuel from the full tank because its easier to suck air than it is to suck a fluid. So, the engine will fail with fuel in one tank.

So, we must conclude that an RV has a very different fuel system that does not apply to a Bearhawk.

If any aircraft is sucking fuel out of a tank, and one of the four ports gets uncovered, that will cause fuel starvation and engine failure.

For there record, I am not installing a FI system like what is being discussed in this thread. If I were running this kind of system I would want to know if its sucking for feeding fuel via gravity.

Realize if you keep your tanks full and avoid extreme attitudes, it may fly fine for years before a problem (one tanks fuel port comes uncovered) arises.

Lastly, I want to ride in an aircraft that can run one tank dry without having to worry too much. There are benefits to running a tank dry. Merely having to deal with a cough, with an immediate resumption of power when the selector is twisted to a tank with fuel is what my experience was once upon a time in a previous century.

Lastly #2 If I wanted to supply 45 gallons per hour thru 3/8 lines, I would have only one fuel line exiting each tank feeding a vented 5 gallon header tank installed aft of the cargo bay, that is gravity fed from the R and L tanks. Engine Return fuel would go to the header tank. A simple on-off fuel selector would make it a very simple safe install I believe. But I would not make this change without talking to Bob first. I think I just described a C-185 fuel system.

[svyolo]

Brooks. I couldn't agree more. At first I was going to return to the mains, but didn't like it. After reading a couple of BH fuel feed tests, I knew that the pumps would be sucking at least some of the time. Maybe not in straight and level flight, with the tanks full. It might work most of the time. Maybe 99.9%. Someday I might find the .1%. A header tank, any size, with return fuel going to the header tank, and the header tank vented to the main(s), allows to stock fuel system to feed engine demand only, instead of 45 gph continuously.

The fuel pump then sucks fuel from near the bottom of the header tank, with a little boost from gravity. The return line and vent from the header deals with any air, or fuel vapors that form after engine shutdown. At least that is what I hope...........

[Archer39J]

This response deals with the quote above. It does not deal with a fuel system that has a header tank. I fear the quote has some assumptions made that could lead to poor fuel system installations if the above is applied. So I'll share my viewpoint. But it contradicts the quote above.

If a Bearhawk fuel system demands 45 gallons per hour, and it has 3/8" supply lines, then I wonder if the fuel system is still feeding via gravity. It may in fact be sucking fuel out of the tank rather than gravity feed. My gut tells it would be sucking. If the Fuel Flow Test wont drain 45 gallons per hour into a bucket from the tank to the pump, then it fair to assume its sucking fuel out of the tank under normal operation.

An RV system sucks fuel out of a tank. A Piper Cherokee sucks fuel out of a tank. (In these aircraft each fuel tank has only one fuel line supplying the fuel system, not two like a Bob's design) These low wing fuel systems don't not have a "BOTH" selection on the fuel sector. Rather its merely L - R - OFF. Thats because if there was a BOTH, then as soon as one tank runs dry the fuel pump will suck air from the empty tank instead of feed fuel from the full tank because its easier to suck air than it is to suck a fluid. So, the engine will fail with fuel in one tank.

So, we must conclude that an RV has a very different fuel system that does not apply to a Bearhawk.

If any aircraft is sucking fuel out of a tank, and one of the four ports gets uncovered, that will cause fuel starvation and engine failure.

For there record, I am not installing a FI system like what is being discussed in this thread. If I were running this kind of system I would want to know if its sucking for feeding fuel via gravity.

Realize if you keep your tanks full and avoid extreme attitudes, it may fly fine for years before a problem (one tanks fuel port comes uncovered) arises.

Lastly, I want to ride in an aircraft that can run one tank dry without having to worry too much. There are benefits to running a tank dry. Merely having to deal with a cough, with an immediate resumption of power when the selector is twisted to a tank with fuel is what my experience was once upon a time in a previous century.

Lastly #2 If I wanted to supply 45 gallons per hour thru 3/8 lines, I would have only one fuel line exiting each tank feeding a vented 5 gallon header tank installed aft of the cargo bay, that is gravity fed from the R and L tanks. Engine Return fuel would go to the header tank. A simple on-off fuel selector would make it a very simple safe install I believe. But I would not make this change without talking to Bob first. I think I just described a C-185 fuel system.

This is a good point. I didn't go too far down that path since this isn't the setup I'm considering, though it is the setup recommended by the fuel system designer and flying on at least one Bearhawk. A return loop, header or not, avoids any of this.

Definitely sounds like a case where "just trust the designer and don't ask your own questions" might not be the way to go though, go figure...

More editing:

I too think it's possible to have a problem lurking for years, so testing at the extremes should be considered required.

Another subject of never ending debate, I do not consider intentionally running a tank dry to be an acceptable procedure. But I also understand the issues if you do with an EFI system like this. The design I propose mitigates the risk of unintentional dry running sufficiently for me.

[svyolo]

One more thing about return less. When cars starting doing it they were having some hot start issues. They ended up raising the pressure they run from 30-45 up to 58 psi, or higher. That kept fuel vapors from forming, and fixed it. SDS or EFII could probably run at the higher pressures, but the stock tuning tables would need to be adjusted.

[svyolo]

Changing my fuel system to return-less would involve nothing more than moving the fuel pressure regulator from the firewall on the downstream side of the fuel block, to the pump outlet. I remember finding a thread on VAF where someone asked one of the vendors if they could do that, and return fuel back to the inlet side of the pump. They admitted that it would work, but seemed a bit reluctant. In my case it would go back to the small header tank. If I wanted to run higher pressure, the regulator is adjustable, and I could adjust the mixture with the manual knob.

I am going to break my engine in on the ground. I might try it just for kicks.

[whee]

I know my opinions on FI BH fuel systems are not popular so I’m not going to share them here; I’ve shared them several times elsewhere. Just please be careful guys. We all know fuel system issues are the number one cause of EAB accidents, let’s try to not add any of our planes to that statistic.

[Archer39J]

I know my opinions on FI BH fuel systems are not popular so I’m not going to share them here; I’ve shared them several times elsewhere. Just please be careful guys. We all know fuel system issues are the number one cause of EAB accidents, let’s try to not add any of our planes to that statistic.

Absolutely, this is why I'm here talking this out.

[svyolo]

Me too. Asking the stock BH fuel system to feed 45 gph, in all flight conditions and fuel levels, Might be a bit much. If I was going to do that, I would ask Bob. I would not expect a definitive answer from him.

I wanted to use the BH fuel system, for its intended purpose. That is to feed the amount of fuel consumed by a 260 hp Lycoming.

[Archer39J]

Changing my fuel system to return-less would involve nothing more than moving the fuel pressure regulator from the firewall on the downstream side of the fuel block, to the pump outlet. I remember finding a thread on VAF where someone asked one of the vendors if they could do that, and return fuel back to the inlet side of the pump. They admitted that it would work, but seemed a bit reluctant. In my case it would go back to the small header tank. If I wanted to run higher pressure, the regulator is adjustable, and I could adjust the mixture with the manual knob.

I am going to break my engine in on the ground. I might try it just for kicks.

I'm not sure I understand but I know you have experience with FI and know what you're doing. That sounds interesting, I'd be interested to hear how it works.

A header tank, any size, with return fuel going to the header tank, and the header tank vented to the main(s), allows to stock fuel system to feed engine demand only, instead of 45 gph continuously.

The fuel pump then sucks fuel from near the bottom of the header tank, with a little boost from gravity. The return line and vent from the header deals with any air, or fuel vapors that form after engine shutdown. At least that is what I hope.........

I planned to return just before the gascolator, wanting to keep the advantage of its fuel volume, even if small. But the advantages of a vent line are compelling. If I can't easily add one to the gascolator I may make my own that incorporates it. Something to think about at least...

You're using 1/4" vent line right?

[svyolo]

Dave;
As designed most return systems return fuel AFTER the injectors. SDS does it after the fuel block on top of the engine, if that is where you mount the fuel block. They recommend not on the engine, but mounted on the firewall. EFII does it after the fuel rail, like older cars did it.

I can pretty easily remove it from the firewall, and put it right at the outlet of the pump. It is then returning fuel in the feed line, right back to the tank, which is a few inches away. That is one of ways return-less works. It still returns, but just from the pump outlet. Either in the tank, or sitting on top of the tank. I read in theory that the fuel pressure can be less stable, but I doubt in our application it would be much of a problem. We run steady rpms the vast majority of the flight. I would only need to make a jumper to replace the regulator to make it work. The other way return-less works is electronically controlled pump motors, with some sensors downstream. A few companies sell them aftermarket as well.

If I do try it, and I probably will, it will be on the ground, tied down, while I am running in my engine.

Why am I not planning on doing that permanently? It only saves me 4 ft of return line. Returning fuel is preferred for most reasons. But it should work, either way.

I will probably just use a 1/4" vent line from the header to the main.

The question is can you skip the whole header tank, and return fuel from the pump outlet, right back to the pump inlet? Maybe a small manifold to stabilize pressure fluctuations. I am not going to try that. But it might work as well. I prefer the header tank, and returning fuel too it 4 ft from the regulator.