Fuel system options for fuel injection

[John Slade]

Builders,

After three years of thinking about it, I'm still no closer to a decision on which way to go with my fuel system. Perhaps in this new unmoderated forum, I can get some useful ideas....

 

Current situation.

Cozy IV with plans fuel system, weatherhead valve in the seatback and 3/8 lines. No returns plumbed and no fuel level sensors fitted.

 

Problem

I'm installing a fuel injected Mazda 13B rotary which requires 40 + PSI fuel pressure and returns. I'd like two independant fuel supplies to the rail with a filter and pump on each. So, how do I handle switching the returns?

 

Suggested solutions

1. Andair duplex valve on the firewall with remote switching.

Pros - High quality part. Simple operation.

Cons - Price. Remote selector.

 

2. Draw always from tank A. facet tank transfer pump to regulate fuel levels, a la Tracy Crook.

Pros - Inexpensive

Cons - Pilot workload

 

3. A second weatherhead valve on the seatback for the returns.

Pros - Inexpensive

Cons - lots of pipes in the cabin.

 

4. A motorized soleniod on the firewall. Duplex valves are available, but with plastic fittings

Pros - inexpensive

cons - Plastic fittings

 

5. Go with a sump tank a la Velocity.

Pros - proven

Cons - proven to be a nuisance with irregular feeds. Header tank in the cockpit.

 

Every solution has its own set of advantages and disadvantages. I've listed some that I know of. There seems to be no RIGHT way to do this. I'd appreciate any words from the wise. Right now I'm leaning towards Andair.

[dust]

Suggested solutions

1. Andair duplex valve on the firewall with remote switching.

Pros - High quality part. Simple operation.

Cons - Price. Remote selector.

 

I wouldn't want a remote selector - denver effect

 

2. Draw always from tank A. facet tank transfer pump to regulate fuel levels, a la Tracy Crook.

Pros - Inexpensive

Cons - Pilot workload

 

Good way to run out of fuel on short final

 

3. A second weatherhead valve on the seatback for the returns.

Pros - Inexpensive

Cons - lots of pipes in the cabin.

 

Why do you need a valve on the returns??

 

4. A motorized soleniod on the firewall. Duplex valves are available, but with plastic fittings

Pros - inexpensive

cons - Plastic fittings

 

 

5. Go with a sump tank a la Velocity.

Pros - proven

Cons - proven to be a nuisance with irregular feeds. Header tank in the cockpit.

 

cons says it all

 

 

A good friend of mine was an aircraft accident investigator for 25 years, He rarely, I mean rarely investigated fires, but he often investigated fuel starvation from lack of fuel from tank being used!!! He thinks the fuel shutoff valve that the FAA requires kills many many people. He is also a 7,000 hour pilot, a&p,DAR etc etc

[John Slade]

Thanks for the input.

 

>I wouldn't want a remote selector - denver effect

Lets not get into a full John Denver discussion, but Andair's remote selector option is a real fine piece of work that you can REACH.

 

>Why do you need a valve on the returns??

to send the returned fuel to the tank I'm drawing from. I'm beginning to lean to the motorized soleniod on the firewall for this.

[dust]

some days I'm a little slow - so you just need to switch the return to the incoming tank, not to be able to turn it off?

 

Mike

[John Slade]

Correct

[dust]

Talked with my buddy today and he highly recommended a 1 lever valve system. Cesna has one that will change both the feeds and the returns at once. He also said to be sure to return the fuel to the top of the tank.

 

He has many stories and related this one, i'll not say the name of the airplane, but it was a major mfg, had planes falling out of the sky, and didn't know why and didn't believe it could be the fuel system. Well a factory rep took the plane from the factory and the engine stopped a certain amount away on his journey. He then went back to the factory and got another plane and took the same route and the same thing happened! This got the attention of the mfg and they began to look into it. No problems could be found with the planes or there systems. They then took a stripped down version , no interior panels, and clear plastic tubing for all fuel lines and returns and took off, well, at 9500 feet (the same altitude that the other planes had the problem, bubbles started forming in the return lines and stopped the flow back to the tanks and stopped the engine from getting fuel.

 

This required two changes - increasing the lines to 5/16 (another airplane mfg went to 3/8) and putting the tube to the top of the tank!

 

Mike

[John Slade]

Thanks for the input, Mike.

[John Slade]

Following various inputs and after agonizing on this for the longest time, I think I have a decision....

 

Based on the KISS principle, considering that I already have most of the parts, and given that hundreds of planes (including a turbo 13B powered Cozy) are flying with a similar setup, here's what I plan to do:

 

1. Use the Cozy plans fuel system currently installed including left / right off weather head valve in the seatback, 3/8 lines, facet pump and gascolator.

 

2. Output from the facet "boost" pump will go to a 3 gallon fiberglass header tank in the "hell hole".

 

3. The header tank will be sealed and will have no vent other than a valve to release the air when it's first filled.

 

4. Two finger strainers in the header will supply fuel to two EFI fuel pumps on the firewall. The main pump will draw from a point higher than the backup pump to allow an emergency reserve supply of about 1.5 gallons. I'll fit a low fuel sensor in the header.

 

5. The return will go to the top of the header.

 

If anyone see's a problem with the above, please let me know.

John Slade

[simmons3]

Will the boost pump be running all of the time pressurizing the sump tank? Will you have back-up boost tank? This system still has 50% of the cons of a sump system. Would it be easier to use the velocity sump system with 3/4" or 1" gravity drain lines with check valves in each line from the strake tank to the sump tank? I like the idea of venting through the strake vents. You can still do that with a gravity feed sump system. What are the feed problems associated with a sump tank?

[John Slade]

>Will the boost pump be running all of the time pressurizing the sump tank?

 

I expect to use the boost pump for take off and landing, but it shouldnt be needed in cruise. The facet pump allows fuel to flow through it when it's off and is plumbed in series with the mechanical pump for Lycomings.

The header is below the tanks and "should" get gravity feed. I'll measure the gallons per hour to see what max flow is going to be.

 

>Will you have back-up boost tank?

I assume you mean a back-up boost pump. No. I wasnt planning on it because of the gravity feed.

 

>This system still has 50% of the cons of a sump system.

Which cons are you referring to?

 

Would it be easier to use the velocity sump system with 3/4" or 1" gravity drain lines with check valves in each line from the strake tank to the sump tank?

Velocity has 3/4 or 1 inch lines? really?

 

>I like the idea of venting through the strake vents.

Venting of the main tanks will be as normal. The header tank shouldnt need a vent.

 

>What are the feed problems associated with a sump tank?

I know of one Velocity (Hi Jim) that has feed sporadic problems from both tanks.

[simmons3]

Will you have a mechanical pump in line with the boost pump? The mechanical pump may create some draw through the system. Do you think gravity will get the fuel from the strakes to the sump through the restrictions of the lines and valve in all attitudes?

 

<Which cons are you referring to?>

The cons you listed in your first thread.

 

<Velocity has 3/4 or 1 inch lines? really?>

I really have no idea. I was calling the system by the name you used. My plan is to use the larger lines to make sure the system stays full.

 

<Venting of the main tanks will be as normal. The header tank shouldnt need a vent>

I agree.

 

<I know of one Velocity (Hi Jim) that has feed sporadic problems from both tanks.>

I am hoping the larger lines eliminate this problem.

[John Slade]

>Will you have a mechanical pump in line with the boost pump?

No.

 

>The mechanical pump may create some draw through the system. Do you think gravity will get the fuel from the strakes to the sump through the restrictions of the lines and valve in all attitudes?

This is my main concern. I'd like to keep the fuel valve. Maybe I'll need to leave the boost pump on all the time.

 

<I know of one Velocity (Hi Jim) that has feed sporadic problems from both tanks.>

>I am hoping the larger lines eliminate this problem.

It should, provided you clean out all the crud from the main tanks and put screens on the feeds. That's what Jim is doing right now.

[Nat Puffer]

Dear John,

A remote valve on the firewall is a bad deal. It caused many accidents with the Varieze, and Burt outlawed it. But some builders did it on the Long EZ. The John Denver accident is most noteable.

A sump in the fuselage I think is a bad deal. An extra complication, more possible leaks, and vulnerable to a landing gear wipe out.

I always prefer the simplest solution to any problem. How about the plans fuel system (which you already have installed) and then a return to just one tank. This would be the minimum extra piping and the minimum pilot load. I would suggest you put the return to the right tank, because that is easiest to watch visually and the right seat will be empty some of the time. If you start out with full tanks, always draw from the right. When that is empty, switch to the left. This is the simplest piping job, the least number of tanks (no sump tank in the fuselage), and the least pilot workload, and therefore the safest.

Regards,

Nat

[John Slade]

Nat,

There's a LOT of fuel flowing through the return. Possibly as much as 30 gal/hr during idle and decent. Juggling tanks with this much return fuel flow would generate a fairly high pilot workload.

I think I can isolate the header tank from the cabin. There seems to be no "right" way to do this job.

Thanks for the input.

Regards,

John

[Dan Tomlinson]

Oone other option that should be minimal change with redundancy capability (but still the slight possibility of dumping fuel overboard in an emergency situation):

 

Assuming 6747 valve that you have, it appears that the OFF position flows from the top to bottom output, but not the sides. (Option: Add in series with this the FV-107702 shutoff valve at the seatback or similar shutoff only valve - always open. (emergency fuel off = closed)

 

Split the fuel feed line before the valve (at the sump?) on each tank to a double throw, double feed electric operated valve pair at the firewall, with the return line ONLY going through this "primary" feed and return fuel system. Now return goes to the feed tank, and switch tanks electrically for normal operation through the weatherhead "flow-thru" position. Adds 1 line in the cockpit.

 

Auto Valve Emergency fail: switch to L or R with weatherhead, return fuel goes wherever it was going.

 

It's easy to draw, hard to explain!

[John Slade]

I'm not sure I follow what you're describing, Dan, but it sounds like it involves running high pressure lines to the seatback which is something I'd prefer to avoid. Nat's comments raised the other issue I was trying to avoid, which was having a header tank in the cockpit. Also, another big negative, using the existing weatherhead valve and header tank means that I'll have no redundancy. OK, I think I have another plan...

 

Install AN867-3 aluminum flanges at the top and bottom of each tank. 3/8 pipe via AN elbows and 107702 manual shut-off valves to 230106 filters and EFI fuel pumps to a T at the rail. This gets me two redundant fuel systems. To get the return fuel back to the correct tank, I'm considering the ASCO 8300B410U 12VDC soleniod valve which would be switched depending on which pump is active.

 

Any thoughts? I see that a lot of people are viewing this thread, but I'm not getting much input.

John Slade

[Dan Tomlinson]

I'm sending a "napkin drawing" to your other address. It's a 16k jpeg.

 

There is NO high pressure to the seatback, just 1 additional low-pressure line. The solenoid valve is dual 3-way latching type - only powered during the switching. http://www.sbcontrols.com/products/3-way_latching.htm - I don't have a price (or similar type valve)

 

There is redundancy, as the solenoid flow only happens with the 6747 in the vertical position - thus the shutoff valve. The flanges etc are in top and bottom - return and feed. Filters I believe are not recommended on the "vacuum" side to prevent flow problems.

 

A small relay to indicate L/R tank and switch the fuel guage operated at the same time from the momentary switch which switches tanks.

 

Only drawback I see is still if it fails, you can switch the feed, but not the return using the 6747.

 

As my wife says: "a hand to hold, an opinion to reject"?

[Mevans]

Hello All;

 

I am putting a 13B in my Long EZ and I am at the similar planning stage for my injection. I was planning on running the rails out of a header tank in the hell hole or cold side of the firewall. Then run the return back into the right tank. I would always run out of the right tank when full, and switch tanks once the right is down sufficiently. Down side is my fuel system is not totally redundant as you were looking for. I will have two EFI pumps but they will draw from the same supply, (header tank.) I always felt that the main risk of fuel exh/starv. was no fuel or no pressure, so two pumps, but same supply (hopefully always with fuel.)

 

I am still early in planning, so if anyone thinks this is nuts, let me know without accusing me of throwing away the plans.

 

NO PISTONS!

 

Matt Evans

Clearwater, Florida

http://mevans28.tripod.com

[John Slade]

Hello Matt,

I've been occilating back and forth on this for years (literally).

One advantage of the header tank approach is that you can draw from different levels and thereby create a "reserve" capacity. Two disadvantages, that I see, are the potential for fuel leakage in the cockpit and the additional weight. Most header systems also receive the return. Plumbing it into the right tank removes the potential for air accumulation in the header, but introduces an increased pilot workload (which Tracy Crook is happily living with).

 

Will you tell me what you don't like about my (most recently) planned system?

See

Regards,

[Dan Tomlinson]

The only issues I see are:

 

Where do you physically locate the 107702 valves? Reachable for emergency shutoff only?

 

Lack of electricity results in no fuel - so backup power is required.

 

What is the state of the return valve without electrics - force right tank?

 

The benefit is you can use your setup as a low-volume transfer pump!

 

Regards,

[Mevans]

John;

 

I am still a neophyte when it comes to fuel systems, but here goes. I thought about having a header tank on the hot side, I saw someone with a similar aluminum type tank, that would keep it out of the cockpit area. I would like to avoid the sumps in each tank and I thought that was one way to handle that.

 

There is a workload in always feeding to one tank, but that does not seem more workload than remebering to switch the return valve when you switch tanks. Whichever workload fits with your thinking is fine, I do not think either system would matter, it would be whichever system you were comfortable with. I would rather have fewer valves and connections, than more valves and connections,(less to leak, break, fall apart, explode, etc. .) but again that is a matter of taste. Afer looking at your drawing it makes sense to me, with one caveat. In the scenario where you run one tank dry, your other tank is now feeding from a system with only one fuel pump. A highly unlikely scenario, and again, a minor element when all is considered. You do have two independant fuel supplies, but they are not necessarily duplicate in all situations.

 

 

Your system looks fine, it is different from mine, but I don't think any better or worse. Go for it.

 

Matt

[John Slade]

Thanks for the feedback, Guys...

 

>I thought about having a header tank on the hot side

OK, I guess, provided you're not returning to it. Just so long as it doesnt leak

 

>There is a workload in always feeding to one tank, but that does not seem more workload than remebering to switch the return valve when you switch tanks.

The return valve will be switched by the same switch as the pump. Nothing to remember, provided the switching actually takes place

 

>Where do you physically locate the 107702 valves? Reachable for emergency shutoff only?

Under the back armrests (Cozy IV). The valves are really for maintenance, but I think I'll be able to rig an emergency shut-off cable to each from the front seat.

 

>Lack of electricity results in no fuel - so backup power is required.

Also results in no ignition. I'll be installing a backup battery.

 

>What is the state of the return valve without electrics - force right tank?

I'm looking at JC Witney's Pollock valve. Will test failure modes before flight. Fail closed would be bad.

 

>The benefit is you can use your setup as a low-volume transfer pump!

Fairly high volume, actually - but I plan to wire it with the fuel pumps for reduced workload. A bypass switch could be added if I find I need this facility.

 

OK. Thanks again for the feedback gentlemen.

I'm ordering the parts....

 

But, if anyone reads this thread the day before I fly and has a problem with the planned system, please let me know....

[Jim Sower]

John,

I still think your earlier idea of:

"Always draw from tank 'A' and use a Facet pump to even things out" and returning excess fuel from rail to sump is still the best idea.

Pros:

Simple, inexpensive, minimal plumbing

Allows for simple, accurate fuel flow

NO plumbing in cabin - and only minimal plumbing in hell hole

NO selector valve - remote or otherwise

Simple, easy fuel management

Best way NOT to avoid running out of fuel in short final

Cons:

Sump

 

If you are drawing from the Right (easiest to check as Nat suggested)strake into a sump, and from the sump into your fuel pump/filter/etc. system and back to the sump from the rail, you have traded having a sump for three fuel lines in the cabin ~ about even I'd say.

Now, "T" in a line from the Left strake to the Right strake line near the tank and put your Facet pump in that line. Then, install your fuel flow transducer between that "T" and the sump.

 

You -could- put in all manner of trick stuff to warn you when the Supply (Right) strake was too high or low, but you'd need that just as badly with the good old "Left-Right-Off" system. But let's just pretend, for the sake of argument, that you are capable of monitoring your fuel, at least in the supply tank, without AI, just like a real grown-up.

 

As a flight progresses, you'll ALWAYS end up transferring ALL of the fuel out of the Left strake into the Right one. Unlike the design "Left-Right-Off" system, which requires you to run one tank "Pump Empty" as the ONLY way to avoid the possibility of running out of fuel with 3-5 gal in the "unselected" tank. This scavenging procedure is dicey on a good day, and absolutely HARROWING in a situation when you really think you'll actually need that last few gallons. I know. I've been there. I don't ever want to be there again.

 

In my view, the compelling reasons for this configuration are that it has the simplest fuel management system, the best way to get accurate fuel flow, and the safest in that it gives you the MOST assurances that late in the flight you'll have ALL the fuel on board in the "selected" tank

 

Just a theory .... Jim S.

[John Slade]

Jim,

Thanks for the thoughts.

You don't mention the biggest con against the (Single Return Transfer (SRT) System. Single point of failure. From the main tank outlet to the rail, everything is a single point of failure. Anything (water in tank, fittings, pipes, blockage, filter, pump, sump tank) goes wrong in this system and you're an instant Simon & Garfunkle fan.

 

You're big advantage is the ability to draw the last drop of fuel from the secondary tank. Pipers have been flying for years without major problems (at least for those people who are smart enough to use the 45min rule and not rely on that last drop). Actually, I (and piper) could add this feature easily with a T and a facet pump, but it introduces additional complexity.

 

There seems to be no "right" way to do this job, but on balance I still think the Duel Feed Switched Return (DFSR) system is the least complex, lowest pilot workload and most reliable.

John

[niceez]

John,

 

After sitting on my hands for two weeks..... I feel the safest way is two independant tanks with a mechanical valve. KISS seems to be screaming at me. Return feeds could also have a mechanical valve to decide which tank to use.

 

Whatever yo decide will most likely work.