Header tank design and location

[AZBearhawk272]

A lot has been written and intimated on this thread and I will throw in a few nickels worth.
Just so happens I design fuel systems and components as my lively hood.
So here are a few points of note:

The system that Bob designed is simple and adequate for carb or fuel injected engines.
It works very well when implemented properly.
I have mocked it up in clear tubing to check flow and line refill, bubble formation and flow stagnation. I did this for two reasons: one to
investigate why A Bearhawk engine went quiet
and would not re establish fuel flow.
Two to make sure that in my installation,
IO-470 that fuel flow would be adequate and
un interupted.

Most general aviation aircraft that have a header tank have one for 1 reason.
to meet certification requirement that after engine fuel starvation of running a tank dry
that the engine can restart and produce power
within 10 seconds after switching to an available source.
The Bearhawk system as designed by Bob will go from a dry tank to fuel at the carb in 3
seconds!



Most production FI systems utilize an engine driven pump that is positive displacement pump some fixed displacement / some variable displacement. The pumps are sized to provide Approx 1/3 more fuel than needed
to support combustion. Fuel delivery is proportional to engine rpm. Un used fuel is returned to the “ system “

In most cases the header tank provides a means to capture the excess fuel bypassed and immediately make that fuel available for the combustion process.

I am not utilizing a header. Fuel will return to the left tank. Fuel selector will be R, Both, L, Off.
The primary flight mode is Both. Returned fuel
will level as the fuel seeks cross tank equalization. The fullest tank will have the highest head pressure and will drain the fastest.

Aircraft fuel pumps are not designed to
“ Suck “ and operation in that mode will destroy a pump quickly. Pump inlets should always be flooded by free stream supply.

A header tank is not a requirement but a means to meet a requirement.
Mock up and test your system with clear Tyron
tube to know before you build.

Kevin D # 272

[Seth Kroenke]

Kevin D,

Are you able to take a call from me regarding your information on this post? It appears you have the information that I am after as I complete my system. I'd love to ask you a few questions. As a side note, I wrote a post last night to the general group regarding my 3 remaining issues. Your advice is incredibly appreciated.

[svyolo]

Kevin D;
Awesome info. Thanks. But I still see a problem that I described before.

Whether you return to both tanks, in "Both" or only return to the left tank all the time, I still see draining the high tank very quickly in a continuous, or mostly continuous turn. Assuming a left turn, fuel valve in Both, fuel returned to Both, or in you system, to the left, I see the right tank emptying fairly quickly when low on fuel.

Assuming a 35 gph pump, and 7 gallons in each tank, my high (right) tank is dry in 12 minutes, and the left wing lines are seeing only fumes.

Am I missing something?

I will be flying from the left seat. Given a choice, I will be making left turns in the above scenario for better visibility. If I was going to only return fuel to one tank, I think I would make it the right tank vs the left.

[Ed.Meyer]

In a coordinated turn one tank is is NOT higher than the other as felt by the g forces in the airplane which is what the fuel will react to.

[Mark Goldberg]

Bob's system has worked on both carb and injected engines. I am not sure why any modification is needed. I do not see it.

John - if you are in coordinated flight the tanks will draw down equally - even in a turn. At least it would seem that way to me. Mark

[Ed.Meyer]

With Bob's design and the fuel valve set to both at least one tank port will be supplying fuel in any positive g configuration. Nose up, nose down, slip left or right. Using rear port only, nose down can leave both dry as Jim pointed out.

[AZBearhawk272]

As noted, in a positive G coordinated turn the fuel distribution in the wing tanks should be even and level across both tanks.
You tube has several nice Bob Hoover videos that illustrate this point well.

A valid concern is a prolonged slip to landing
in a low fuel condition. Many production planes are placarded against this operating condition. The “saving grace” in this situation is the fuel consumption is down in the 1 to 2 gph range. So 1 quart of fuel should provide
15 min of slipping.

I postulate that if fuel is low beyond reasonable landing reserves, and a half hour
slip to landing is required the issue is not in the fuel system.

As part of Phase 1 testing,fuel unporting tests
and slips to landing, dry tank switchover, should all be conducted at altitude over a suitable landing site.

looking forward to that myself. : )

[svyolo]

Well, I believe I had a disclaimer, "am I missing something"? In any coordinated turn (which is the goal), the lift vector (and load) is vertical with respect to the aircraft vertical axis. Most of us only try something different only in a crosswind landing. So my worrying about fuel being transferred to the low tank in a turn is.............somewhere between stupid, misinformed, or in my case, I think I will go back to stupid. I am an engineer by education, and I "missed something". I started this thread, and I learned a lot from it. Thanks to everybody. I think I will shut up now.

[Mark Goldberg]

Do not be too hard on yourself as these things being discussed leads to better information for everyone. I have learned from postings on this forum. And when Kevin D chimes in (Bearhawk272), it is always good info. Mark

[whee]

Well, I believe I had a disclaimer, "am I missing something"? In any coordinated turn (which is the goal), the lift vector (and load) is vertical with respect to the aircraft vertical axis. Most of us only try something different only in a crosswind landing. So my worrying about fuel being transferred to the low tank in a turn is.............somewhere between stupid, misinformed, or in my case, I think I will go back to stupid. I am an engineer by education, and I "missed something". I started this thread, and I learned a lot from it. Thanks to everybody. I think I will shut up now.

Don’t do that! Discussions like this are how we learn. There are no stupid questions and there is nothing wrong with being wrong or not knowing something. This is a great discussion, thanks for starting it!

[PaulSA]

I must admit that I have never quite understood the "need" for a return line with a FI system. I am currently running an RV-10 with an injected IO-540 with EDP and electric pumps. The electric pump has an idling circuit which deals with any excess fuel. No problems whatsoever...... For my Bearhawk, I have installed the Andair pump which has an internal bypass.

With a low-wing aircraft which generally switches tanks every 30 minutles or so, I see no issue with fitting a return - you just end up feeding off one tank longer that the other. But with a high-wing system designed to run on "Both", it just seems that you are building in potential problems........

As Svyolo said - "Am I missing something?"

[Ed.Meyer]

I imagine there are several reasons why some FI systems need return lines and others don't. With the EFII system we are using one benefit is that there are zero vapor lock or flooded hot start issues. With the fuel contantly circulating when the system is powered up, there is cool fuel all the way to the injectors all the time. Even on a hot day, and hot engine restart, it starts immediately and runs smooth. I know the same cannot be said for some FI systems...

[Bdflies]

Hey Ed, it sure would be great if you'd do a write up about your experience with the EFII system. Installation issues? Good documentation? Good support? Teething pains initially? Any quirky operating requirements? A Beartracks article would be great, or a thread on the forum.
Systems like the EFII are cutting edge (for recip airplane engines) and it would be great to hear your thoughts about your experiences.

Bill

[Ed.Meyer]

I will work on a write up. It is somewhat different than the norm.

[svyolo]

I believe the purpose of the fuel return was always to eliminate vapor lock. Cars used to always use it. I believe most have gone away from a return in the last decade or so for environmental reasons. Most now have an in-tank pump, and no fuel return. The tank is run at a slight negative pressure to reduce evaporative emissions.

[JimParker256]

Excellent discussion, and I'm learning a lot. And it's also stimulated my thinking on this subject. So, even though I am 99% likely to go with a carbureted engine for the simplicity of it, I'm still trying to think through the injected design side...

An FI system that "recirculates" the excess fuel would seem to either need a header tank (which I don't like for various reasons) or it would need one (or more) return lines to the fuel tank(s) themselves. I've heard some "interesting" stories from a friend with a Bonanza that had a total of 7 tanks (main, aux, and tip in each wing, plus a "baggage" area tank), but had a fuel system that aways returned the fuel to the left main tank... Fuel management seemed nightmarish to me, since he had to keep coming back to the left main to ensure it didn't overflow the returned fuel... Based on that, if I were using such an FI system, I would probably use a full-duplex fuel selector – plumbed such that the return fuel goes back to the same tank that is selected. Yes, that would require two return lines (one to each wing tank), but the additional weight would almost certainly be lower than the weight of a header tank, gauge, etc.

In Zzz's case, if a 1/2 inch fuel line is what is required to supply the "gravity-only" fuel flow rate to prevent "sucking" fuel with the pump, then I would use 1/2 inch fuel lines connected to both the forward and aft fuel pickup bungs. And agreeing with Zzz's (or was it Whee's?) about using the "T" or "Y" fitting for the returning fuel, it would probably be prudent to add an additional bung to the upper side of the fuel tank for the returned fuel.

But in any case, I would NOT do away with the "dual" fuel pickup points, for the reasons discussed earlier.

[Ed.Meyer]

I agree with Jim's comment and would that I would NOT do away with the forward tank pickups. We plumbed the fuel lines as Bob designed all the way to the gascolator. The dual fuel pumps (redundant) and filters are after that. We put in a duplex fuel valve and return lines to both tanks. Of course had to add fittings to the tanks which we placed at about the center of a rib lightening hole for easy access and about the center fore and aft. I don't think it much matters whether high or low.

An interesting observation occurred yesterday and I had noticed i before but this thread came to mind this time. When I first powered up for start and the fuel pump came on, I could hear bubbles in the fuel tanks for a couple seconds as air was being purged out of the system.

I gave many hours of tought and some lost sleep over many of the concerns and ideas, including header tank, expressed here before deciding on the final design.

As to the concern regarding sucking air from an unported tank outlet, I worried about that as well. Apparently it does not suck hard enough to overwhelm gravity. Tested this before first flight by measuring fuel flow in hard nose up attitude with only 5 gallons in each tank. Was the same whether selector was on both or either individual tank. This is with 3/8 lines throughout.

As others have said, interesting thread...

[jim.mclaughlin924]

I am using Bob's design with 3/8 tubing for my FI engine except I am only using the forward bung on the left tank where the return goes. I have a lot of time in Bonanzas with the return to the left main only and am very comfortable with the operation. Just takeoff and land on the left main. If there is a problem, the left main always has some fuel in it. Low wing planes do not flow any fuel without a pump and do fine with 3/8 inch tube. I have an Titan engine and understand the return is less than a couple gallons an hour, similar to the Bonanza.

Newer RANS aircraft have a header tank under the baggage compartment and skip the forward tank drawoff.

[Ed.Meyer]

Not clear to me but sounds like you are using the left forward bung for the return. Interesting idea. I would think though that if you select the left tank only then you feeding from only one bung, the left rear. Wounldn't you then risk sucking air to that one feed with low fuel and nose low or left slip? I would think it would be better to select both for critical operations.

[jim.mclaughlin924]

I welded in an extra bung for the return line.near the top of the tank.

[svyolo]

I am starting to like Whee's idea of 1/2 inch lines as well. If nothing else, extra capacity, stored in the lines itself. Sort of a built in small header tank in each line.

Lots of great responses. I learned a lot.

I still haven't quite come to closure over forgetting how loading works in a coordinated turn. 30 years of flying for a living hasn't taught me much I guess. LOL

[whee]

I am starting to like Whee's idea of 1/2 inch lines as well. If nothing else, extra capacity, stored in the lines itself. Sort of a built in small header tank in each line.

I’d wait till empirical data suggests my solution is suitable. Hopefully within the month I’ll be performing actual flows tests with a complete system. Testing the fuel system will be part of my phase one testing.