Header tank design and location

[svyolo]

I have read (by googling) the threads that popped up about header tanks on FI Bearhawks on this forum. I think I want to use one as well for the SDS EFI that I want to use. I have no experience with designing a header tank.

Most tanks should be filled from the top, and "feed" fuel or fluid from the bottom. An aircraft header tank will also be vented to both wing tanks, with the vent lines coming from the top of the header tank, vented to the top of the main tanks.

Is shape important? If the system is designed correctly, the header tank should remain full until 1 or both main tanks are empty. Would there be any problem "filling" the header tank from the bottom? That would make a few things easier. The returned fuel should be able to force the rest of the header tank up, and they should still vent upward. At least that is my thinking.

Has anybody else put much thought into this, as well as what shape, and where to locate the tank. Tall/thin, mounted up by the firewall? Low, flat, under the floor or under the seats (front or rear). 3 gallons minimum si recommended by the EFI manufacturers to prevent excess fuel heating in the header tank.

I won't let this delay the build. I am buying a rebuilt carburetor from Bob, and it will run on that first. If I think the EFI will delay the first flight, it will also fly on the carburetor first. I sort of want to do that anyway.

Thanks for any input.

[Mark Goldberg]

What some FI systems require is a return line to a tank. You should not need a header tank from what I have heard. On my RV8 injection system I just put in a loop that brought fuel back upstream of the boost pump. So I suggest looking at the different FI systems and the requirements for the system you want to use. Mark

[Chris In Milwaukee]

Gerhard Rieger posted something recently about a brake system setup, but it prominently featured the header tank he put together for his Continental IO setup. Very similar to the way Maule does it. Take a look at this one and see if it will accomplish what you’re after. Perhaps Gerhard will answer some additional questions.

https://bearhawkforums.com/forum/bearhawk-general-discussion-news/37603-brake-system

[Ed.Meyer]

We have the EFII system on our Patrol with no header tank. We had to add a fitting to the fuel tanks for return lines and installed a duplex fuel valve that switches the return as well as the feed. Valve has left, right, both, and off settings. It was recommended not to use a header tank for the return to avoid fuel warming issues. Further, if the fuel is returned to a header that is vented up to the wing tanks you have no control over where fuel returns to. So far, our system works great. One more point is that switching from carb to EFI is not simple task in that carb needs no return nor fuel pumps nor fine filters that EFI requires.

[whee]

I put a lot of thought into it and did the same and Ed. I studied many Cessna Illustrated Parts Catalogs, maintenance instructions, etc when making my decision. I do like Gerhard’s header tank under the floor and that is what I would have done had I decided to use a header tank.

Actual shape isn’t important but it is important to keep the supply and feed ports separated so that warm fuel and vapor isn’t recirculated back to the engine. For this reason I would feed from the bottom and supply from the top. On the side near the top is actually where I’d place my supply/return ports.

[svyolo]

I guess my reason for using a header tank is that is seems most high wing aircraft with fuel injection seem to use one if there are return lines, i.e. Continental, and now Rotax. I can think of ways to get into trouble using "both" on takeoff and landing with low fuel without a header tank. It seems like using a header tank in this set of circumstances is "convention". Cessna (Conti motors), Maule, and the newer fuel injected Rotax's all seem to use a header tank.

Without a header tank, you need to run a full size fuel line from the duplex valve to each main tank. With a header tank, you STILL need to run two fuel lines to the top of both main tanks as vents.

I am using the stock fuel feed to the fuel valve layout no matter what I end up with for as far as a header tank or not. It appears to work as designed, especially when run in "both". The valve with be in the stock location. SDS recommends a minimum of 3 gallon header tank (if using one) to avoid the problem of excess fuel heating of the tank.

I still might start with a Duplex valve and no tank to start with. The feed side of the fuel system will be identical no matter what. There is plenty of room for the pumps and return lines to be added in the beginning, or later.

I will contact Mr. Rieger

[svyolo]

Schu;
I was leaning toward putting the header tank under/right behind the pilot seat, or under the rear seat for the reasons you mentioned, including crashworthiness and ease of inspection. Header tanks are not normally pressurized, although the ones that they seem to use on the FI Rotax's are shaped like a pressure tank, so maybe they are pressurized. the BH already has 4 gravity fed fuel lines running through the cockpit. What additional risk is another small tank?

With or without a tank, you need to run two lines up the the mains for either vent or full flow return. If using a tank, it is best if the tank is in a convenient location to run the two lines, with no low spots, up to the mains.

For some reason high wing airplanes use header tanks a lot, and I think most if not all injected Conti motors use them. I am not 100% sure why, but I think the engineers that chose to do that know something about fuel systems that I haven't thought of yet. All that being said, the fuel feed side of Bob's fuel system appears to function perfectly, and I will utilize that layout.

I was even thinking about putting in a very small header tank in just the feed side of either the front or rear fuel lines.

As for ultimately, why EFI? Other than personal preference, which is big. I almost talked myself out of it, and using MFI. But I finally was able to justify it to myself. I have spent an adult lifetime leaving home for 6 months to 2 years, and coming home. It didn't take long for me to learn to disconnect the battery before I left. Old tires used to "flat spot" and be ruined when you came back (this was fixed). When I came back to a carburated vehicle, I was never sure if it would even start, let alone how it would run. A few years later I was running modern EFI vehicles. 6 months or 2 years, it doesn't matter. As long as the battery can crank the engine over, the engine starts, and runs perfectly, every time.

My airplane is going to sit unused 6-8 months a year, unless I take on a partner in the plane. I want to know that when I come back, connect the battery, and turn the key, that it will start and run perfect.

[svyolo]

If there was more room up in the boot cowl, I would probably go with that location. Some applications use a tall, thin, flat tank. That still might work as well, situated above the rudder pedals. I think from a crashworthiness perspective, I would rather have it somewhere between beneath me and slightly behind me.

If I haven't sorted it out within 3 months, I will fly it first with a carb. That is the last major system I haven't sorted out in my mind.

[Bcone1381]

Svyolo, May I ask a question out of curiosity? Ed Meyer in post #4 has is flying an EFII installation without a header tank. It functions as it is designed and as you hope to have yours function. What aspects of a header tank installation is appealing, or what concerns do you have about a headerless fuel system design? I am curious, have no intentions to stir up controversy, nor am I trying to persuade you for or against. Just trying to understand. (I use to lean towards a header tank install in this type of system, but Ed's report has me doubting myself now.)

[whee]

I was even thinking about putting in a very small header tank in just the feed side of either the front or rear fuel lines.

I considered something similar. I was going to replace the AN T fittings at the forward door posts with a 1ish gallon tank. The was primarily to prevent un-porting a tank when low on fuel and maneuvering. I decided I didn't need this or a header tank in my airplane.

Maybe worth noting:

The Cessna 206 doesn't have a "both" position on the fuel selector valve. There are two header tanks under the floor and the fuel selector valve is a duplex valve that returns the fuel to the appropriate header tank.

The Cessna 185 has a single header tank has a simple On/Off fuel valve that in placed after the header tank. The fuel tanks all feed the header tank with no valve in between.

The Cessna 337 does not have a header tank or a 'both' position on the fuel valve. Fuel from the front engine is always returned to the left main tank and the fuel from the rear engine is always returned to the right main tank.

I plan to primarily use mogas in my airplane and after a conversation with Peterson Aviation I determined that a fuel injection system with a vapor return line was necessary on my airplane. Peterson spent a lot of time and money trying to develop a mogas STC for fuel injected Lycomings (no vapor return) and eventually gave up. He couldn't keep the fuel from vaporizing in the fuel lines. I know many RV guys successfully run their fuel injected Lycomings on mogas but after talking with Peterson that was one experiment I wasn't willing to test.

[Zzz]

I'm building my 4-place as a FlyEFII machine. I have an Andair duplex valve, and am returning fuel to the tanks. In this kind of system, the header tank is really only good for the unporting issue. I am using only the rear pickups on the tanks, with fuel returning through a newly added bung near the top of the tank, more forward to keep it further from the pickup.

One benefit of returning fuel to the tanks is that the extra travel gets you good cooling. A header tank can accomplish cooling as well, but needs a minimum volume of like 5 gallons. I don't want a 5 gallon tank in my fuselage. I figure if I'm totally dependent on the fuel pumps anyway, might as well just go with return lines. The header tank does nothing for you if you lose both high pressure pumps.

I got over my need for a BOTH selector after flying old Pipers and a Bonanza.

[Mark Goldberg]

Why would you NOT use both the forward and aft tank outlets? Mark

[JimParker256]

I have the same question as Mark G... My concern about not using the forward wing pickup is that in a descent, with the nose-low, the aft pickup may be surrounded only by air because the fuel is sloshed forward in the tank, while the forward pickup point should be awash in fuel. By feeding from both fwd and aft ports, the nose-up / nose-down attitude of the airplane becomes a non-issue.

I attended a "fuel systems design" seminar at OSH a couple of years ago. The instructor (whose name escapes me) was an engineer who designed the fuel systems for several military aircraft, as well as consulting on many homebuilt designs. When I received my Patrol plans, I was pleased to see that Bob's design was EXACTLY identical to the "optimal design for high-wing homebuilts" that the instructor provided us. Right down to the gascolator being the lowest point in the fuel system... And he highlighted the importance of using two fuel pickups per tank (forward and aft)

By the way, his personal pet peeve (he's a DAR as well) was those small in-line fuel filters that: A) have no automatic "bypass" provision in case the filter becomes clogged), and B) have non-transparent bodies (so you cannot even see if fuel is flowing or if it is totally clogged). He simply won't sign off an airplane with those installed, because of the high danger of fuel starvation in the event of a clogged filter. He doesn't much care for the non-bypassing "transparent" filters, either, but reluctantly approves them, after extracting a promise from the builder that they will include "replace fuel filter" in their 50-hour preventive maintenance schedule, along with oil and filter change...

[Zzz]

Why would you NOT use both the forward and aft tank outlets? Mark

It's not as if it didn't cross my mind also as the ideal solution. I can't remember where I discussed this, but I did at length, about the potential to suck air in the unported forward pickup. Would the rear pickup and line keep everything primed? When the two lines are Y'd or T'd together, it gets a little more head-scratchin'... This isn't just atmosphere or even an engine-driven fuel pump, it's high pressure, target being 35 psi. Suckage of air could happen fast and a lot of it.

I figured the most likely scenario, and the most deadly, for unporting would be in a steep climb, so the rear pickup is the primary.

I'm all ears if someone can figure out how to incorporate the front pickup too without the complication I mentioned.

[JimParker256]

If you explained the issue, Zzz, I must have missed it. You mentioned you were using a "bung" near the top to return the fuel, and only using the rear pickup to deliver fuel. It would seem to me that you could use both front and rear bungs (at the bottom of the tanks supplied with the QB kit, and shown in the plans for scratch-builders) for your fuel pickup, and add one near the top to return the fuel. If absolutely necessary, I suppose one COULD even use a "Y" or "T" fitting and use either the front or rear pickup point as the "return" point as well.

To reiterate my concern, there are TWO situations you will encounter on many flights: The first is the high-angle climb-out, where the fuel will shift aft, and (especially in low-fuel situations) the front pickup point may be uncovered for the duration of the high-angle climb. The second is a descent from altitude, where the nose is lowered, and the fuel will shift forward, and (again, especially in low-fuel situations) the aft pickup point may be uncovered for the duration of that descent. The steeper the descent, the greater the likelihood.

[whee]

Jim, The issue is the amount of fuel being drawn from the tanks by the fuel pump. It is entirely possible that the fuel pump will actually be sucking the fuel out of the tank because head pressure will not deliver fuel fast enough to keep up with the pump. If the pump is pulling fuel from the tank and one of the pickups un-ports then the pump will suck air and not fuel. You’ll be left with whatever fuel gravity can supply through a single pickup which may not be enough.

Also, simply having the return line T’d into one of the supply lines at the tank isn’t sufficient. The returned warm fuel and vapor needs to be cooled and vented in a tank.

Z, we discussed this at length a while back. You also talked to a buddy of yours in AK that suffered a partial power loss because he un-ported one of the pickups on his EFI cub.

The Continental IO360 fuel system flows 30-35gph pretty much constantly, it varies a little with rpm, but the max that is supposed to make it to the cylinders is around 20gph. The remainder is returned. I wanted to use both ports in the tank to supply fuel so I did some testing and decided that I would be comfortable with using both ports if I used 1/2” fuel lines at the rear port. After installing the lines we performed a preliminary flow test. As we hoped the 1/2” line alone will meet the 125% flow requirement. In my system, which is basically what Bob specified except the 1/2” rear fuel lines, gravity should provide more fuel at the fuel pump than required. The pumps should never have to suck fuel from the tanks making un-porting a nonissue.

[svyolo]

Svyolo, May I ask a question out of curiosity? Ed Meyer in post #4 has is flying an EFII installation without a header tank. It functions as it is designed and as you hope to have yours function. What aspects of a header tank installation is appealing, or what concerns do you have about a headerless fuel system design? I am curious, have no intentions to stir up controversy, nor am I trying to persuade you for or against. Just trying to understand. (I use to lean towards a header tank install in this type of system, but Ed's report has me doubting myself now.)

I originally was wanting a header tank because it seemed like that was the convention in high wing planes with FI, and I didn't know any better. I still feel that way. After lots of thought, my biggest concern with not having one is

1. Low on fuel, landing pattern, or looking for a place to land.
2. Turning in one direction for 15 or 20 minutes.

Because it is a side by side airplane, I will mostly turn left when given a choice for visibility reasons. While the engine may only be burning 7-11 gallons an hour, the FI is using a constant displacement pump, (25 or 35 gph I can't recall which at the moment). So I am feeding out of the high (right) wing at a rate of 35 gph, how long until that tank is empty, assuming I had 7 gallons in it to start?

If I am running the fuel valve in "right", and all the fuel is returned to the right tank, no different than a carb or MFI. If I am running in "both" as recommended, I simply don't know.Assuming that the fuel follows the path of least resistance, I could be feeding 35 gph out of the right tank, and returning 28 gph to the left tank, as it is several feet lower when in a 15-30 AOB turn to the left. I could possibly run the right tank dry in as little as 10 or 12 minutes when running in both. At the moment, this is a concern, but I don't know if that will happen for sure. I could fly for 500 hours with no problem. Then, at 501 hours, I am having a hard time deciding where to land, or am practicing touch and goes, and ......................

Which gets me back to "why a header tank". Some engineers with a lot more experience than me designing high wing airplanes thought it was a good idea. Mostly I am not 100% either way, and am leaning toward following "convention".

[Mark Goldberg]

It would seem to me that if you are in a coordinated turn the fuel would be flowing equally from both sides. Mark