Has anybody got a drawing or description of a fuel tank with inbuilt swirl pot suitable for EFI.

One that actually works would be nice.

Better than a drawing or description of a swirl pot...
Rescued attachment Swirl pot.jpg

For my Fuel injected Nissan I'm going to reuse the plastic swirl pot out of the original donor tank ,only thing is that i have to make sure that I dont fry it while welding the tank up as it can't be fitted in side otherwise ]

best wishes
Iain

How are you going to fasten the pot to the inside of the tank?

Have you given any thought to an external swirl pot burton power do them for £70

I've read that the VW Golf EFI pump has a integral swirl pot.

Only added problem with external swirl pot is the need to have a low presure pump to feed it.

I have always been a bit confused over 'swirl pot theory' as to me, a lot of people's ideas on what constitutes a swirl pot is too complex, or doesnt work.


I cant see how an internal swirl pot would work......

I see the mechanism as follows. Low fuel in tank is sloshing around. You need a way to trap it. So, you have a small container in the tank, fixed to the lower surface of the inside of the tank, like marks pikkie. So, you need holes in the wall of the swirl pot to allow fuel to fill it. These will be the same holes that will allow it to drain right back out again....plus you have the efi pump emptying it, so

fuel in = thro the side holes.

fuel out = thro side holes AND from the pump.

This looks to me like a losing struggle.

If you route the efi return back into the swirl pot, that takes one thing out of the equation, but you still have an equal ability for fuel to exit the side wall holes at the same speed it went in!


why do you need a pump for filling a swirl pot? Just put it lower than the tank and gravity feed it thro a large dia pipe.


My theory was that fuel needs to be supplied to the engine when the tank level is 'sloshing'. Ideally, you need to maintain a reserve of fuel such that for an average slosh, the fuel will hold up until the next slosh, at whatever the flow rate is. Scientifially, I had no clue of what these factors were, so I used the American technique - IE make it as big as you can. In my case its not a huge swirl pot, just a 180mm length of 50mm dia tube, IIRC its about 1/4 - 1/3 litre - cant be bothered to work it out.

This tube is welded to an exit hole on the tank bottom - the hole is approx 20mm - to easily let fuel in. 'slosh' wont allow the fuel back out as easily as a surface mounted pot, as gravity and the 'walls' at the 25mm entryb hole will tned to turn the fuel back down into the pot.

so, as long as the average slosh keeps fuel in the tube, it says there, so that only the pump draws it out. so....


fuel in = thro the 20mm top hole.

fuel out = from the pump, and sloshing back out of the top of the 20mm hole (small due to gravity).



The only mod I would make to my scheme would be to allow the efi return to dump right back into the 20mm hole of the swirl pot. It doesnt - it just dumps in the region of the swirl pot via a tube to the base of the tank.


atb

steve
Rescued attachment swirl.jpg

The swirl pot in my tank takes the form of a cylinder thats been split longintudinally and the edges overlapped for 45deg, the gap between the edges is 15mm. This is sealed to the bottom of the tank inside.
The fuel pump sits inside the pot and the return line is fitted inside the overlap, so as to direct the return flow into the pot.
In this way the return flow creates a swirl in the pot and also acts as a non-return valve for the inlet to the pot (which is through gap where the overlapped section is), as the flow encourages fuel to enter the pot and stops it leaving.

I tested it with water using a washer pump to give the flow and it easily kept the pot full whilst there was practically nothing in the rest of the tank.

James.

is the term 'swirl pot' just a bit of miss-naming?

why do you need a 'swirl' as such?

atb

steve


link to fuel pumps with integral swirl

http://www.vdo.com/siemens/en/prod_fuel_pumps.asp


[Edited on 30/7/04 by stephen_gusterson]

The main problem, and I may be nit picking, with the second pic is having to set the fuel tank up high. Eight pounds a gallon starts to add up the higher it sits. I would problably try experimenting with a small sump and trap profiles to reduce sloshing, along with feeding the return line directly back into it ala the pic. Cheers!

Is this fuel swirling round the inside of the tank or is it sloshing round the inside of the tank or does it only swirl once it gets into the pot? I'm getting confused by the different terminology. It's a device to prevent the pump from running dry isn't it?. Set the fuel gauge to empty when it's on quarter and don't let it get lower than quarter so you always have half a tankful. Simple.

Steve, I agree with what you say, but my pot is welded to the baffle and the only route between the two sides of the tank is through the pot, so whichever side the petrol sloshes it has to pass through the pot, so I should have a constant supply

I suspect it is called a swirl pot because it's other function is to remove bubbles in the fuel feed to the injection pump that would then end up in high presure fuel feed to the injectors. Just like a tap running at high presure into a bucket you would get too many bubbles produced with the return pipe directed at the swirl pot inlet.

Found this out when I could hear air running through my pump with the car at rest and with half a tank of fuel.

[Edited on 31/7/04 by zetec]

Why not make two narrow vertical tanks each side of the boot connected with a T at the bottom?
The chance that both will rapture in a tailender is small.
Should cive you more bootspace too.
Just an idea