Surge Tank for EFI

[neuropel]

One of the significant issues with converting from carburetion to EFI is making sure that the high pressure EFI pump isn't starved for fuel. Some clever solutions to this problem have been detailed here previously, but they often require extensive tank modifications. So it seems that if one is unable or unwilling to have the fuel tank cut apart to install baffles or special pickups, a surge tank is required (or keep the tank 1/4 or more full all the time). I've been leaning more toward the surge tank approach. This issue has been briefly discussed in a number of threads, but here are a couple of solutions that I've come across that look promising - one is dirt cheap, the other looks like a nice aftermarket solution. Both incorporate the EFI pump inside the surge tank, which is an approach I really like.

antisurgetank

The second product I found at the website below:
FI on a Spitfire- The Fuel Delivery system

I was unable to find it at Summit, but it is for sale here:
HOLLEY - Fuel Pump With Accumulator Tank, P/N 12-923

Any comments on these various approaches? I found the first one a bit scary but innovative - the material used is polypropylene and it's claimed to be holding up well after 3+ years.

[opelnut_1]

This is a bit pricey but.. if necessary looks to be a solution, I was planning on mine to just run the return into the tank with a pipe that exited in the bottem of the tank its self.

no I will need to rethink this idea as I want to make sure there are no issues

[namba209]

Guys, here's some things to consider, and I'm not knocking either of what Todd linked to. First, I'm gonna presume that both units are sealed, so there will always be an air bubble at the top of each one, unless the air is bled off on initial installation and each time the tank goes empty. If you run a return line to the bottom of the fuel tank, the return fuel will have to overcome the gravity of fuel already in the tank, so the return pressure has to be greater than weight of fuel covering the line. If it does not, the fuel rail pressure will be greater than the rail pressure relief setting, and the engine will run richer due to increased pressure in the fuel rail. Another thing to consider is the placement of the "surge' tank. If it is at the rear of the car, it should not pose a problem, if it is lower than the fuel outlet, or reasonably close. if you mount it up front, then you'll need a pump in the rear to feed the "surge" tank, along with the wiring and fusing. To see what I used, check out Phase VI in the Articles and Projects forum. It may give you some more ideas. HTH.


After looking at both surge tanks again and reading the documentation, they are both similar to what I have in Willit?, they both have provisions for a return back to the fuel tank, which means they will have to be plumbed into the ruel rail return line, and the air bubble I mentioned earlier will not be a problem. The old "Open Foot, Insert Mouth" syndrome.

[neuropel]

Ron brings up several good points. Just to clarify, almost all applications using a surge tank require two fuel pumps - a small feeder pump for the surge tank is required in addition to the high pressure EFI pump.

Todd

[RallyBob]

If you are building a car with EFI that will pull some 'G's', then a surge tank (or more correctly, a swirl pot), or trap door system will be required to prevent fuel starvation.

If you have a fuel cell, then the trap door system is a no-brainer as they are commercially available and only require one pump, and they are literally a 'drop-in' for a fuel cell.

But if you intend to keep the stock fuel tank, then a swirl pot makes a lot of sense. You can feed the swirl pot direct from the fuel tank with a low pressure high volume carbureted pump. Then use an external EFI pump to draw from the lowest point of the swirl pot. The swirl pot idea works very well as it eliminates air pockets and bubbles in a similar manner to a dry sump tank. And you will always have a full 'column' of fuel available to the EFI pump as long as the gas tank doesn't run dry. The return fuel line is plumbed into the swirl pot as well, and when the swirl pot is 'full', any excess returned fuel is dumped back to the main tank.

I've built a number of these systems for Subarus that are raced. Even though those cars come stock with EFI, when raced on very wide competition tires (285 to 315 mm width!), the stock EFI pump starves in right hand turns, literally shutting the engine off. The swirl pot solves the problem completely.

Bob

[davegt27]

Todd I have the same links saved

I even called Holley but there fuel accumulator had been discontinued years ago

Right now I am in the process of making my own

I went to Alreco a local Aluminum supply house and bought a 5" diameter tube off there scrap pile

Bought a 5" hole saw

Cut the ends with a hole saw using some 1/4 aluminum plate I have laying around

I had both ends of the aluminum tube welded with the plates

Haven’t touched since I had the ends welded on ha ha

The plan is to have the top where a stock Toyota FP assembly will go right in

My collision repair class finally got there tig welder working so I hope to have 3 of them done soon

You can also buy them

http://www.more-japan.com/Fuel-Colle...p-1-c-544.html

Davegt27

[neuropel]

Here's an idea that was thrown out initially in the Megasquirt forums and involves modifying a fuel filter for farm equipment. This looks very "doable".

http://madscientistmatt.blogspot.com...-overkill.html

If you go through the blog, the final implementation is less complex than what he showed in the original article above. The biggest obstacle would be finding a suitable place to mount it.

Todd

[opelnut_1]

So for street application. is the general consensus that this would not be needed: is there some sort of race foam that could be added to the stock tank to prevent sloshing.

Also where should one put the return line if its not a good idea to plumb it back to the bottem of the tank. The top maybe?

[davegt27]

When I converted my street GT to 75 FI

I ran brake line from the rear of the car to the front

Then used the stock fuel line as the return line

On the passenger side of the car at the fuel tank I took a drill and drilled out the vent then hooked the return line there

Worked great for like 3 years
The only time I had problems is when I autocrossed the car

I had to have at least a 1/2 tank of gas in the car or it would fuel starve


Davegt27

[RallyBob]

Originally Posted by opelnut_1 So for street application. is the general consensus that this would not be needed: is there some sort of race foam that could be added to the stock tank to prevent sloshing.

Foam does very little to suppress sloshing. It's mostly a suppressant in the event of an explosion (slows down the flame travel supposedly). But yes, in a fuel tank not designed for EFI, even on the street you'll have problems if you like to corner reasonably hard.

Also where should one put the return line if its not a good idea to plumb it back to the bottem of the tank. The top maybe?

Top is a better idea. But it 'may' be noisy when the tank is low, it'll sound like draining water back there.

[RallyBob]

This got lost the other day when the site crashed, but here's what a proper swirl pot design looks like. http://www.nfauto.co.uk/fuel_swirl_pot.htm

Note the angled entry (3rd from top) from the feeder pump, which induces a swirl into the tank (hence the name). At each of the necked-down areas of the tank would generally be a piece of screen or mesh to break up any air pockets. Coupled with the 'swirl' effect, only 'solid' fuel is picked up by the high pressure EFI pump fitting at the bottom of the tank. The second from the top fitting is usually the return line from the fuel rail, and the very top fitting is the overflow, which returns back to the main tank. When fuel demand is low (idling or cruising aound town), more fuel is returned from the fuel rail than is consumed, and the swirl pot will be filled right to the top. Any excess simply overflows into the gas tank.

[namba209]

Originally Posted by RallyBob Foam does very little to suppress sloshing. It's mostly a suppressant in the event of an explosion (slows down the flame travel supposedly). But yes, in a fuel tank not designed for EFI, even on the street you'll have problems if you like to corner reasonably hard. Top is a better idea. But it 'may' be noisy when the tank is low, it'll sound like draining water back there.

Bob, there's no supposedly about it. At one of the bases I was stationed at, An AC-130, Spectre Gunship, had the wing fuel tank take a direct lightening strike. They took the panel off over the tank and pulled out the foam, and could trace the path of the lightening through the foam layers where it melted. The tank was full of JP-4 at the time.

On Willit? with the return fuel line plumbed in to the filler neck, I don't hear the fuel running in to the tank unless I pull the gas cap. Of course that could be because of the loud exhaust or the rotary pumps working.

[RallyBob]

Originally Posted by namba209 Bob, there's no supposedly about it. At one of the bases I was stationed at, An AC-130, Spectre Gunship, had the wing fuel tank take a direct lightening strike. They took the panel off over the tank and pulled out the foam, and could trace the path of the lightening through the foam layers where it melted. The tank was full of JP-4 at the time. On Willit? with the return fuel line plumbed in to the filler neck, I don't hear the fuel running in to the tank unless I pull the gas cap. Of course that could be because of the loud exhaust or the rotary pumps working.

Thanks for the insight Ron. I remember a TV show back in the 1980's and they blew up two cars with onboard dummys, one with foam in the gas tank, one without. The one with the foam just went 'poof' and burned up. The one without the foam went 'ka-boom!' then burned up. The dummys on board would have both eventually melted, but the car without the foam caused much more initial damage to the occupants thanks to the initial concussion. If the 'boom' doesn't get you, you have a greater chance of escaping the car before catching fire I guess!

BTW, welcome to the 5000 post club!

[namba209]

Originally Posted by RallyBob BTW, welcome to the 5000 post club!

Thanx Bob, I feel extremely lucky to be in association with you in this club, but I am subordinate to you in the knowledge base, for sure. Could be Margaret is right, I am vociferous, at times.

[davegt27]

when i did my GT i did hear the fuel going back to the tank


some more info on fuel surge tanks

http://toyotaperformance.com/surge_tank.htm

http://www.tweakit.net/shop/product_...roducts_id=123

the Bosch FP mounting bracket is cool also

http://www.tweakit.net/shop/product_...roducts_id=371

David

[Hiro]

I made a very easy swirl pot for my race KadettC:
I used a tall & narrow can (as tall as the fuel cell),
Drilled a couple of holes in the bottom of the can,
Inserted this can in the fuel cell & blocked it with FIA security foam,
Both return lane & pick up lane come directly to the bottom of the can,
No problem even in harsh G turns!
Hiro

[oldopelguy]

I built the one I'm going to use on the turbo Kadett project out of a very small air compressor tank, approx 4" round and 10" long. I went quite a bit different direction than most of those pre-built ones, though.

I made the top of mine incorporate the fuel sending unit from something that has all the fittings in it and holds an in-tank fuel pump in place inside my little swirl pot. The "vent" line goes back to my regular car tank as the fuel return, and the normal fuel supply and returns on the sending unit go to and from the EFI system. All I had to plumb in was the low pressure pump discharge line into the small tank, and it will go in the air fitting bung on the side of the tank.

I then should have two fuel system loops, one going out the normal fuel outlet on the normal fuel tank, through a filter and very low pressure fuel pump, into the swirl pot, then back out the swirl pot's "vent" back to the regular tank. As long as the small pump is running this loop should always be full and overflowing back to the regular tank, and all the lines should have very low fuel pressures.
The other loop starts in the fuel pump mounted inside the swirl pot through it to the only real high pressure line in the system, to the injectors and fuel pressure regulator and then back to the swirl pot. Since my swirl pot will be mounted in the front of the car, the only high pressure fuel line should be something like 3' long, the rest of the lines should be very low pressure. Sloshing in the main tank, which isn't as big a concern in the Kadett, should have very little effect on the fuel supply to the injectors.

BTW, a lot of the sprint car guys use one form of mechanical injection or another, and they need a return line for their system to work properly as well. Lots of them use a fuel filter with two inlets on it to route the return from the injection back to the suction side of the filter in front of the pump instead of back to the tank itself. Since the fuel pump should be sending more fuel to the front than the return flow, it would still be drawing the fuel out of the tank at the rate the engine uses it.
http://www.speedwaymotors.com/pl/751...th-Return.html

[neuropel]

Well, I think this may have become a non-issue for me. There's a place called Moyer Fuel Tank Renu less than an hour away. I spoke with them yesterday and they said more and more people are coming to them to convert tanks over for EFI. I think they might be able to set it up with baffles and an in-tank EFI pump. The tank will come back with a lifetime rust-through warranty - The typical cost for a tank that comes in in decent shape can get the renewal process and the return fittings installed for $200-$250 (rough estimate). Since I have a couple of fuel tanks laying around, this is the easy route.

One other question: I have a '75 Manta fuel tank and a spare '74 Manta tank. Does the '75 tank have appropriate internal baffling to take care of this problem? If so, I can take both tanks in and just have them make the '74 like the '75, though I notice that the sending unit is in a different location on the '75. I've never dealt at all with the '75 Opel EFI fuel tanks or fuel pumps, but I wouldnt' want to use the old EFI pump anyway.

Todd K.

[RallyBob]

Originally Posted by neuropel One other question: I have a '75 Manta fuel tank and a spare '74 Manta tank. Does the '75 tank have appropriate internal baffling to take care of this problem? If so, I can take both tanks in and just have them make the '74 like the '75, though I notice that the sending unit is in a different location on the '75. I've never dealt at all with the '75 Opel EFI fuel tanks or fuel pumps, but I wouldnt' want to use the old EFI pump anyway. Todd K.

Todd, just use a higher volume external pump (Bosch or whatever), and use the 1975 tank. Problem solved. It was designed for EFI from the outset.

[neuropel]

I was planning to do exactly that for one of my cars, but I have two cars that will eventually get EFI so I figure I can get both tanks done at the same time. I only have one '75 tank, but I believe they can install baffling in the '74 tank and add the fittings I'll need.

One of the cars is getting the built up 2.4L with the Extrudabody ITB stuff (further down the road). The second car is a stock '74 1.9 manta automatic that I'd like to get my first Megasquirt system up and running on first. That one will use the 75 EFI manifold and the Nissan throttle body. I think getting a '74 tank modified will be cheaper and easier than dealing with the surge tank.

Some additional questions for anyone that might know:

Will any of the pieces from a 2.4 EFI setup work with the 1.9 EFI plenum? I haven't tried it yet, but the 2.4 throttle body looks to have similar mounting. Also wondering about the fuel rail.

Todd K.

[namba209]

I'm gonna ask for some expertise here, just for my own piece of mind regarding the setup I have on Willit? I have a high pressure pump that has a pressure relief and bleed port on it that dumps excess pressure (above 70 PSI) and continually bleeds a small amount of fuel back to the tank to keep the pump cool. My surge tank/swirl pot has a stand pipe in it that also dumps back to the tank when the fuel level in the tank gets above it. Also the same return line for the tank overflow and pump bleed is plumbed in to the fuel rail pressure bleed. My question is: Will the fuel, continually recycling between the fuel rail and surge tank. as shown in some applications, affect performance by getting too hot or creating a back pressure in the fuel rail? That was the one reason all my fuel returns go back to the fuel tank. First to keep cooler fuel to the fuel rail and second to eliminate the possibility of back pressure at the fuel rail. Whatcha think?

[RallyBob]

Originally Posted by neuropel Will any of the pieces from a 2.4 EFI setup work with the 1.9 EFI plenum? I haven't tried it yet, but the 2.4 throttle body looks to have similar mounting. Also wondering about the fuel rail. Todd K.

I never bothered to check the 2.4 TB, but the 2.4 phenolic injector plates, the 2.4 injectors, and the 2.4 fuel rail all fit the 1.9 EFI intake manifold. The only thing questionable would be whether the 2.4 fuel rail interferes with the 2.4 TB. The 2.4 fuel rail is very close to the Nissan TB on my 1.9 intake.

Bob

[RallyBob]

BTW, there's quite a bit of general FAQ-type information on EFI installations available on S.D.S's (Simple Digital Systems) website. Most of it is aimed at do-it-yourselfers.

http://www.sdsefi.com/tech.html

And specifically, surge tanks.

[jordan]

Ideally where would one of these tanks be located?

[namba209]

Jordan, my personal opinion, and what I tried to do on Willit? was to mount the tank, fuel pump and filter, next to the battery, but with the AC, I couldn't, so mine is in front of the structure, behind the front grill work. I do believe there is ample room below and forward of the MC using the holes in the radiator support to route the lines through.