Fuel Filter selection

[r2d2hp]

I can confirm what Stevie has just said.

My setup is very similar to number one - the only change being my feed back to the tank is from the top of the swirl pot.

On the way to the Retro Show I suffered from the HP Pump cutting out due to the fuel vaporising. The swirl pot was almost too hot to touch.

I have since changed the setup similar to that in number 2 but as yet have not tested to see if the problem is cured.

[Ian Anderson]

Another thing to remember is that the return to swirl and also return to tank has to be a pipe of big enough bore to allow for maximum flow rate of the pump or your system will end up working at a higher pressure.

Imagine yur car is at idle and the fuel pump going flat out regulator is allowing it to flow back but a restriction further back will increase the pressure in the fuel rail - probably ending up richer than you expect.

Ian

[stevieturbo]

Another thing to remember is that the return to swirl and also return to tank has to be a pipe of big enough bore to allow for maximum flow rate of the pump or your system will end up working at a higher pressure.

Imagine yur car is at idle and the fuel pump going flat out regulator is allowing it to flow back but a restriction further back will increase the pressure in the fuel rail - probably ending up richer than you expect.

Ian

The FPR will usually present the biggest restriction.

My feed line is 1/2", and the return 3/8"

I have no issues.

I think you would need a very small return line, bordering on silly, for it to be a problem. Even if there was a small restriction, the FPR would still be the main controlling factor.

[chodjinn]

Second schematic looks ok. I suppose a second filter pre-HP pump would be ok as well.

Any chance i could T the two fuel returns together (only have 1 return line in tank)?

Or would that cause problems as the return from the fuel rail will be under more presure than the return from the swirl pot?

[Ian Anderson]

Stevie

I agree re the pipes but I have seen some people try to yse the original 1/4 inch feed for the return and run into problems. (Converting a small engine car similar to MGB to V8)

Try to get 45 gal per hour from a red top through a 1/4 inch pipe!

I would consider it foolish not to return the fuel rail to the swirl pot - if you look at the reason for the swirl pot it is to ensure constant flow to the engine. Swirl pot is say 2 litres and EFI pump will run about 90GPH or about 400 litres. You have about 20 seconds running in the swirl!


Ian

[chodjinn]

Hmm, getting conflicting views here.

Stevie reckons the second schematic is better as less vaporisation, cooler fuel etc.

Ian reckons first schematic is better; why would it be foolish not to return the fuel from the rail to the swirl?

If using the first schematic, the size of the lift pump doesn't matter so much - the Facet I have would suffice as that's what the car was using before I started to fit EFI and there hasn't been any poweer hikes since.

Logistically I would prefer the first option, as I can use the existing outlet/return on my tank, and means I only have to buy a swirl pot + filters. If I used the other layout I'll need e.g. a holley and possible a new tank.

[r2d2hp]

I had the first option and I had problems so am going with the second. That way any heated fuel from manifold goes back to the tank and the swirl pot should never get hotter than the temp of the tank.

The pressure in the swirl should never be higher than the facet pump, I think that is was being presurised due to the return from the manifold as the return from the swirl to the tank is quite small.

Second option should eliminate all these problems in theory.

[stevieturbo]

Stevie

I agree re the pipes but I have seen some people try to yse the original 1/4 inch feed for the return and run into problems. (Converting a small engine car similar to MGB to V8)

Try to get 45 gal per hour from a red top through a 1/4 inch pipe!

I would consider it foolish not to return the fuel rail to the swirl pot - if you look at the reason for the swirl pot it is to ensure constant flow to the engine. Swirl pot is say 2 litres and EFI pump will run about 90GPH or about 400 litres. You have about 20 seconds running in the swirl!


Ian

If I recall correctly, I think I used a 1/4 return into the fuel tank of my Mini, which ran a Bosch efi pump. Its been so long since I looked, but I am fairly sure.

Again, its no big deal if there is a restriction in the return line. As long as the restriction is less than the FPR, then the system will still work properly.

Thats a big mutha of an efi pump at 400l/m !!!!!!! Even 250 l/m is pretty big, so I think your figs are a little bit off there.

And a reservoir of 20 seconds should be plenty. How often would there be fuel starvation in the main tank, actually lasting 20 seconds ??

Or would that cause problems as the return from the fuel rail will be under more presure than the return from the swirl pot?

No it isnt. All return lines will, or should be virtually zero pressure.
The only restriction posed to them, is the pipe back into the tank.
I see no problem tee-ing them like you describe.

[Ian Anderson]

Have a look at
http://www.nfauto.co.uk/
http://www.rallydesign.co.uk/catalog/default.php
http://www.demon-tweeks.co.uk/

Plus I am sure many others and then search for swirl pot.

all are tanks with 4 outlets

Top is return to tank
Next is return from fuel rail
Next is feed from tank
Bottom is feed to HP pump and fuel rail

Seems like you are trying to reinvent the wheel - why move away from a tried and tested sytem?



Swirl ports on above sited tend to be 1.5 litres
Walbro fuel injection pumps 255 litres per hour
1.5/255 = 0.00588 hours on 1.5l
0.00588 * 60 = 0.3529 minutes
0.3529 minutes = 21.176 seconds

Also Walbro is 255 litres per hour
Red top is 45 gal per hour = about 200 l per hour - it will never be able to get the swirl pot to stay full as you are drawing fuel faster than it can fill

Yes I have this length of time on the GT40 as the tanks are long along the length of the car. Less than 1/4 tank and normal street braking causes the Red top I use to hammer on air! It can take about this long for it to quieten and thus start pumping fuel again. This is on a road car and not a race version where the surging would be worse and also coupled to turning forces possibly moving fuel away from pick up!

Ok if you are worried about hear build up in the swirl from returned fuel increase the size of the lift / low pressure pump so it will move a lot more fuel than the engine will consume.

Also remember with all this fuel running you should run the pumps through safety switches that will kill the pumps in case of pipe rupture / engine stopping as this much fuel makes a BIG fire!

Ian

[r2d2hp]

Ian,

Your figures seem to show how long it takes to fill the swirl pot, not how long it takes to consume the fuel. Are you suggesting that underbraking you will consume the content of the swirl pot in 20 seconds.

I guess what your saying is - that if the HP was drawing on the swirl pot and feeding back to tank and the LP feeding the swirl and back to the tank that in that situation underbraking the HP could empty the swirl pot when LP was sucking air.

Have I understood correctly.

[Ian Anderson]

Yes that is correct

The calculation is what the HP pump will pump i.e draw from the swirl pot.

If the feed to the swirl pot is not working (pumping air by the low pressure pump) this is how much running time you have. After that the HP pump will pump air the engine die and if the HP pump left to run will also overheat and die.

So return the unused fuel from the fuel rail back to the swirl pot and (although the fuel may now be hotter than the tank fuel) the car will continue to run longer without the low pressure pump. Also less chance of killing the High pressure pump by running on air.

Ian

[katanaman]

Wouldn't it just be a whole lot easier to just weld in a swirl pot into the original tank? That's all I did with mine all it took was a morning which included going to scrap yard to get an old Sierra tank to cut the pot out of. Only one pump and less space used and is now the same as just about every other injected car on the market. Granted if you don't have the gear you will have to get someone to weld it back up for you but that's going to be a lot less than the cost of a swirl pot. There might be times when you cant do this for some reason but there cant be many.

[chodjinn]

Ian, I'm not trying to reinvent the wheel, dunno where you got that from. I'm just trying to sort my fuel system out in the best way, preferably using the bits I currently have rather than buying more.

I can get a Holley pump from the states for about £60, which isn't too bad I guess.

Will check out what the sitrep is with my tank, then I'll be able to finalise the fuel system as a whole.

thanks

[stevieturbo]

I can get a Holley pump from the states for about £60, which isn't too bad I guess.

Isnt that roughly the same Price Real steel sell them for ?? Carter/Federal Moghul low pressure pumps, circa 100gph.

IMO, internal collectors can be good, but no internal collector will match the abilities of a swirl tank. Unless of course your tank is tall enough to mount a small diamater swirl tank inside it lol

[Ian Anderson]

Chodjinn

Sorry no the comment was not pointed at you but to some others who thought it better to run the fuel rail return to the tank. This would mean buying a swirl pot then having it modified to block a pipe off. Or commission a build to your specifications both cost more than standard units!


Swirl pots are used for a reason and lots are as I have described 4 outlet / inlets

Yes a good sump in the tank and an in tank pump will also work very well and probably cost less than 2 pumps, swirl pot, pipes, connectors etc.

Als before you spend ££££ check with your class racing regs you may need to fit a fuel bladder / cell as these reduce fuel spills in a shunt. Again re engineering always costs - research first and do the job once.

To others I apologise for the rant earlier

Cheers
Ian