Fuel System Simplicity

[Chris Kouba]

Thinking about how to run the twin main tank system with a swirl pot for my car, I am trying to keep things as simple as possible. I have come up with my ideal solution but from my current level of exposure to this topic, there doesn't seem to be many people running in this manner.

I know there is an extensive amount of knowledge on this board. My question is will this work?

I have a stand alone EFI system (Edelbrock Proflow). My chassis (RCR) has a small tank built in to the tail end of the left sponson (3-5 gal est) as well as two full size tanks in the traditional sponson locations.

My idea is to pull fuel with a low pressure (i.e. carb) fuel pump from each sponson, through a one-way check valve into the swirl pot. Basically two main tanks with one way low pressure feeds to the swirl pot. The EFI pump would then pull fuel from and return it to the pot for running the motor.

This would theoretically give me the lower ambient temp fuel supply from the circulating HP loop as well as the ability to control which tank supplies the swirl pot (each main tank pump individually switched).

Pros: no pollack valve, fancy plumbing and associated headaches; much simpler
Cons: would the one way valve do its job and stop the supplying tank from pushing fuel to the opposite sponson when the swirl pot is full? Don't know.

I've bounced the idea off a few people and it seems like it'd work. I want to know what this target audience thinks... Thoughts, opinions, recommendations?

Thanks in advance!

 

[Ian Anderson]

Chris

No reason it would not work
You'd need to size the LP pumps correctly to ensure the HP pump cannot draw more fuel than the LP pumps can supply

Also you will end up running 3 pumps in total - 2 LP and one HP

One downside I can see is if you are sat in traffic idling you only have the swirl as a heatsink for the fuel - yes at 3 gals is should be sufficient but I would think this is the low point of your system

On the one way / non return valves - is you are worried put 2 in series - you will automatically have a backup if one fails

Ian

 

[Jonathan Strain]

There is one thing that I don't think you have addressed. How do you plan to keep the swirl pot from overfilling?

 

[Chris Kouba]

Jonathan-

The active LP pump will fill the swirl pot until it's full. At that point it will stay on to replenish the fuel consumed by the motor and will effectively apply a base pressure to the swirl pot of whatever its regulator is set to- analogous to a low pressure pump pushing fuel against the needle valve/float assembly on a carburated car.

I don't plan on venting the HP system. It would be a closed loop with the base pressure of the LP pump regulator and the rails at the HP system pressure.

With the HP pickup at the bottom of the swirl pot, if it's maintained at full capacity there shouldn't be any air introduced into the HP circuit.

 

[Ian Clark]

Hi Chris, Why not put a balance pipe across the sill tanks? Then it just one big tank, a lot less plumbing, one pump in the surge tank, one return to the surge tank, how simple is that...

 

[Ian Anderson]

There is one thing that I don't think you have addressed. How do you plan to keep the swirl pot from overfilling?

Chris

Jonathans is correct

Problem exists because at some time you will hit the brakes and the LP pump will suck on air

It will pump this into the swirl and in your system the swirl cannot bleed air off and eventually the swirl will be full of air

Ian

 

[Bill Musarra]

Ian(Clark),
I have already tried this. I used an AN-8 line with a one way valve that has no drag. It doesn't allow the crossover fast enough. Just sitting on the jack stands and gas in the righthand side tank with the overflow to the lefthand side, the primary tank on the right empties completely before enough comes back over. I could literally watch the fuel gauge drop. The logical thing to think is that the level in the overflow tank won't allow any flow until the level is higher than the primary tank.When I shut it down,and switch it back on later, the tank shows at least 3/4 full!! The resulting action is the pump would start sucking air. Not good. Imagine what would happen with the car moving and sloshing with turns etc. This also tells me that if the left tank is full as well, the result would cause overfilling of that tank. Again, not good. For now I have the return piped into the primary tank(essentially running on one tank) til I test plan B. Using a bigger line might work, but with a moving car I don't think that is the answer.

My plan B is to use a low pressure pump in the left side and have it controlled with a relay and/or timer controlled relay. The timer is triggered when the low level fuel light comes on. Most of the timers I have seen run for a max of 100 seconds. This leads to the problem of how you keep the pump from cycling on when the left side tank is empty, or runs empty before the timer turns it off. There is a way to do it with relays, but I haven't had a chance to work it out yet. Still chasing down an overheating alternator. I am trying to keep from having two sets of fuel pumps, switches and valves. The other pieces are in place and its just a matter of working it out. An in tank pump cost virtually nothing.

Chris,
You might think of a valve closing mechanism on the return line much like a float valve in a carburetor that would stop flow to the tank that became full first causing the return to go to the other tank. Only down side I can see is you would have to have this on both tanks.

Bill

 

[Mike Drew]

Ian(Clark),
I have already tried this. I used an AN-8 line with a one way valve that has no drag. It doesn't allow the crossover fast enough.

That's much too small. A handful of original GT40s were equipped with mondo crossover pipes. One of the two cars that was displayed 'in progress' at Elkhart Lake was so-equipped. They used a rubber fuel hose about 3 inches in diameter, running side-to-side right at the base (front) of the seats, connecting the two tanks. The hose was squashed into an oval shape in the middle, where it passed through the center of the monocoque.

This was done (I believe) only to speed up refueling. The Gulf cars only had a fuel filler on one side, so I would assume they had some sort of mondo interconnect too. The standard GT40s used a much smaller hose (probably a garden-hose-size hose) connecting the tanks at the rear. Do you have the image of the GT40 fuel system, from the original parts manual? Here it is:

If you click on it, it will enlarge; click on it again and it will enlarge further....

 

[Jim Cowden]

Chris

My system sounds similar to what you are considering and I dont have a problem with it.

My engine is EFI

#1 The main tanks supply pumps are running constantly and supply the surge tank.
I never bothered with L and R pumps as I have the tanks linked so I cant see the point.

#2 the HP pump exits the surge and immediately is T'ed of to the reg and the return goes back into the surge tank.
This stops the HP pump out running the LP pumps.
The reg is at the surge tank, this will drop fuel temps dramatically as I have no return line from the fuel rail (blocked off) to send hot fuel back to the surge tank.

#3 at the surge I bleed of fuel back to the main tanks so the surge does not build up pressure (this will change your mixtures if it gets any pressure on an EFI system)

#4 I did put a cross over in 5/8 or 3/4 I cant remember.

You dont need 1 way valves if the 2 lift pumps are both running and the surge spills back to the main tank the fuel will take the easiest path.
I would not personally have a base pressure in the surge tank as you have mensioned.
Doing that can cause inconsistencies in fuel pressure and mixtures ect,just really on the pump to supply the eng.

All my pumps are intank to reduce plumbing.

So far no problems

Jim

 

[EGLITOM]

Hi Chris

Basically i have my system designed the very same way, except the swirl pot because i run a carbed engine.
Don´t know i f it will work, but i think so. In your case the only thing i would add is return line from the swirl pot. Do this with a switched 3 way valve which routes the return flow according to which of the sill tanks is activated.
POLLAK® ELECTRIC FUEL TANK SELECTOR VALVES - CarParts Quality Discount Parts and Accessories

my system

TOM

 

[Bill Musarra]

Mike,
I am familiar with the fuel system on the originals(have done some work on them). Your drawing is an original, but it doesn't show a crossover line. It shows the selector valve with its tubing . There is no return line from the webers or crossover in the drawing unless I am missing something. Must be in another drawing I am sure.
A garden hose is about the size of a -8, maybe a little larger, not much though. Most I see at the stores are 1/2 and 3/4. A -8 is 1/2. I am tempted to try a larger line except I don't like making the holes in the tank unless I know how well it will work. What really concerns me are a full set of tanks. I will have to install a rollover valve in the vent line in the front, which I have been meaning to do, and install one on the rear as well, which I haven't gotten around to doing yet. This would allow minimal overfil before forcing the fuel over to the main tank.

Jim,
I am having a hard time visualizing your setup. If your regulator is before the injectors, and your return line is part of that, it desn't seem that you are keeping pressure on your circuit to the injectors. For your engine to be running I am sure it is. Most every setup I have seen has the regulator after the injectors to keep that pressure on the injectors. I am sure it is something I am not "visualizing".

Bill

 

[Bill Musarra]

I believe I have an answer for the cross over line problem that I noted in an earlier post. My system was setup like this:
8An tank pickup, to filter to LP pump to swirl tank, to HP pump, to miro filter to transition to 6AN line to FI, to regulator, to swirl pot, to overflow line to secondary tank, to 8AN to primary tank.
The problem I was having was that when the two pumps were turned on, a full tank would literally go to empty before enough fuel would get back to the LP. When the pumps would be turned off, after a minute or two, the fuel gauge would register over half a tank of fuel.
I experimented with a secondary pump in the secondary tank, but figured out that triggering it would lead to problems that would eventually leave me with a worn out pump or stranded on the road.
So here is my solution.

The rear ends of the tanks will be walled off with a trap door to ensure adequate volume. As fuel flows into the secondary tank, the flow will cover the trap door closing it. When that happens the vent at the top will be sufficiently small enough that a pressure will be built up and force fuel over to the primary tank. The primary tank will upon return flow of the gas ensure that the fuel pickup will have sufficient volume as to not be starved and suck air. If the returning volume closes that trap door, the fuel will continue to circulate until enough fuel has been used to allow the trap door to open and replenish it. The warning lite should not come on(and stay on) until both tanks are low enough to trigger it.
The baffles in a DRB tank are vertical with a vertical slit in them to restrict the free flow of gas from one end to the other. Of course I didn't include with the drawing the needed roll over vent tubes at both ends of the tank to relieve any pressure build up from a hot tank or surroundings.
Does anyone see a flaw in this system?

Bill