Edumacate me on gravity feed swirl pot design plz

A long term project is building EFI for the Jensen Healey. I hates fuel starvation in turns, so my plans were to buy or build a swirl pot but I don't want to run a lift pump and then a pressure pump. There's plenty of room under the back of the car for a swirl pot. I'm thinking ~1/2 gallon, probably made from round aluminum tubing etc. Anyone have practical experience to share?

How much lower than the bottom of the fuel tank can you get the bottom of the swirl pot?

Probably 6-8" at the bottom, if I use 5" diameter aluminum tube mounted horizontally. It might could go lower but I'd have to put it further forward meaning the feed lines (I was thinking one from each end of the stock tank) will have to go over the rear axle and that would leave a high point in those lines at about 2" or so below the stock tank.

I also was considering running the return line back to the swirl pot instead of the main tank. Good/bad idea?

I'd just bite the bullet and use two pumps. The payoff is flexibility to locate the pot, and elimination of the need for it to run at injector pressures.

Or, ISTR a Bosch combo surgetank/pump available form VW/Audi cars?

Carter

Mount the tube vertically.

I thought return lines were supposed to go back to the swirl pot. Anyway, I use a similar setup on Dr.Linda's Sportster:

Gas tank gravity feeds a large EFI metal fuel filter which acts as a swirl pot, sort of, which feeds the fuel pump, to the actual fuel filter, to the rail, to regulator, back to a T in the line between the gas tank and the filter (swirl pot). It works well. The large metal filter holds about a half cup or so, enough to make sure the system stays fed if cornering or accelerating with a low tank. She's never had a problem, where as the factory (HD) EFI systems on Sportsters in particular (not sure about the others) report problems when cornering or accelerating with low fuel levels. You can see the filter and fuel lines near the teardrop S&S air cleaner.

My thought was to run two -6 or (more likely for flow purposes) -8 AN lines from the bottom of the tank, 1 at each end, to the top of the swirl pot (possibly into a T fitting), have the pot below the main tank's level. There would need to be a fuel filter in each of these lines to keep big pieces of trash out of the swirl pot, or bring them together into a T, then through a filter, then to a single fitting at the top of the swirl pot.

The fuel pump would feed from a -6 fitting in the center bottom of the swirl pot through a main filter to the injectors and then have a return line to dump the returned fuel directly into the pot. That way the swirl pot is not under system pressure, the only part of the system under pressure would be from the pump to the regulator. If the return overwhelms the suction side (not real likely), all it would do is return up through the feed lines to the tank. I'd rather not run a low pressure pump to the swirl pot because it means a second pump running all the time.

I guess what I'm wondering is would gravity feed be enough to keep the swirl pot full under acceleration and cornering? Hate to run out of fuel halfway through a turn like our LeMons Civic does when it gets below about 3/8 tank.

I also thought about an in tank pump, but with the stock baffling when it gets to around 1/4 tank the 'rattler' low pressure pump goes batE36 M3 in a long curve (think cloverleaf). I'm sure that could be overcome but it probably means I'd be building a tank from scratch.

I think that would work. You're basically making a sump, not a swirl pot so much. My Esprit's right tank has about half of a 1 lb coffee can sized sump welded to the bottom of the tank. The pump/pickup sits down in there.

The pot/sump does not have to stay full under acceleration/cornering. That's kinda the point. Say your doing something that has all the fuel pushed somewhere where it won't feed the sump. The pump then continues to supply the motor with fuel until the fuel supply from the tank comes back online. As long as your sump is big enough to supply enough fuel until the tank starts feeding again, you're OK.

The pressure you can get to drive the fuel through the lines is equal to the density in pounds per cubic inch (0.026 lb/ci for gasoline) times the height in inches from the top of the fuel in the gas tank to the point you're measuring, which is going to be at the filter if you want to see if you can drive it through a real filter. Let's say you have six inches; that's 0.16 psi (rounded to two significant figures). Looks like if you use a really big filter like an SX 41002, that's enough to drive a good a good 200 gph through it. At a 2" drop, you'd get around 100 gph.

I'd probably use large fuel lines and try to get the surge tank as close to right under the fuel tank as you can. If it's more than a 45 degree angle of a run, you may find acceleration holding you back more than cornering. A bike is often a better application for gravity feed.

I thought it might work and couldn't think of a reason it wouldn't. I like bouncing ideas off the brain trust here because someone will usually say 'Hey Skippy, did you think about...' I guess it could be called a remote sump, so to speak.

I was working on a car recently that was getting fuel starvation in long left handers with about 3-4 gallons left in the tank. We rerouted the return line so it dumped right by the pickup into the little baby in-tank sump. That solved the problem - even with 0.25 gallons, we had uninterrupted fuel supply.

Keith wrote: I was working on a car recently that was getting fuel starvation in long left handers with about 3-4 gallons left in the tank. We rerouted the return line so it dumped right by the pickup into the little baby in-tank sump. That solved the problem - even with 0.25 gallons, we had uninterrupted fuel supply.

was this a miata?

my 99 does this on high g left hand turns when it's below ~1/4 tank... (I think that's what it is anyway!)

Of course it was a Miata The stock fuel return shouldn't give any trouble, the car in question had been "improved".

Keith wrote: Of course it was a Miata The stock fuel return shouldn't give any trouble, the car in question had been "improved".

ah! well it looks like I have some more troubleshooting to do then.

Do they make an Accusump that handles gasoline?

Maybe that's what this is:

http://www.ebay.com/itm/New-Bosch-Fuel-Pressure-Accumulator-0438170004-Mercedes-/290558765456

Q: When air is seperated from fuel, where does the air go?
Atmosphere? Top of fuel tank?

Carter

Autolex wrote:

Keith wrote: Of course it was a Miata The stock fuel return shouldn't give any trouble, the car in question had been "improved".

ah! well it looks like I have some more troubleshooting to do then.

You might still be able to help it a bit, it's worth having a look inside.

erohslc wrote: Do they make an Accusump that handles gasoline? Maybe that's what this is: http://www.ebay.com/itm/New-Bosch-Fuel-Pressure-Accumulator-0438170004-Mercedes-/290558765456 Q: When air is seperated from fuel, where does the air go? Atmosphere? Top of fuel tank? Carter

That's an accumulator for a CIS system, I saw them mostly on V8's. Rough way it works: when the plate in the CIS moves a bunch, there can be a momentary drop in fuel pressure at the injectors, that's a hesitation or cough. That doohickey stores fuel at rail pressure and releases it when the pressure drops low enough. The push on fitting is for a vacuum line, high vacuum pulls a diaphragm back. when the vacuum drops, a spring behind the diaphragm pushes fuel out of the exit port and into the fuel delivery system. So yes it's very much like a small scale Accusump. It works for small volumes and short times (2-3 seconds) but I don't think it would work for extended periods, say 10-15 seconds.

Air should go to the top of the tank. From there it's vented to the atmosphere, usually through a vapor separator that allows fuel vapor to be returned to the tank.

Keith wrote: I was working on a car recently that was getting fuel starvation in long left handers with about 3-4 gallons left in the tank. We rerouted the return line so it dumped right by the pickup into the little baby in-tank sump. That solved the problem - even with 0.25 gallons, we had uninterrupted fuel supply.

The Jensen tank is a rectangular metal box with two vertical baffles. There's nothing as sophisticated as a sump area. Hell, the way Jensen cut corners I'm surprised they didn't just give the owner a 10 gallon bucket with a hose stuck in it.

I'll bet you could easily put a baffle in there. Something as simple as a cylinder around the pickup with some flaps or even just holes so that the gas takes a bit of time to flow out. Route the return back into it.