Upgrading fuel filtration on an IDI VW

[TurboJ]

R.O.R-2.0, do you have a picture of your lift pump setup?

I'm considering this already... Even though I practically swore I wouldn't fit one. Could be a nice safety feature, BUT, I would be
scared of the lift pump ceasing to work for any reason and cutting the flow to the injection pump -> boom.

Not very likely of course, but a lift pump would be one extra part to have the ability to go wrong...

Please show me how you did yours!

[745 turbogreasel]

I have a generic electric lift pump before the filter too.  I only use it when I have an air leak, need to purge(closed loop return while running WVO), or fill the filter after a change.  the rest of the time it's not  powered, tandhe vane pump pulls through it no problem.

[R.O.R-2.0]

stock 15m unit in the stock spot.

Racor 2 micron ahead of it..

works amazing..

this was when it was still setup in my mk1..

im pimpin a mk2 Jetta now. it used to be gas powered.

right where the old gasser high pressure pump used to be, thats where the lift pump is now..

i had a spot to mount a pump, a break in the lines , and a power source even (because the old fuel pump used to be there)

so, the pump got mounted to the original spot for the high pressure pump, or the water separator, if your car had one..

i really like being able to see the water separator on the racor filter.. kinda useless tho, ive never had to do jack with the racor.. not even change the element yet..

[TurboJ]

Are you guys running your transfer pumps directly to the VE-pump's inlet (through the filters of course) ?

Won't that pressure alter the VE pump's internal pressure?

I'm thinking a separate catch tank for me if I go electric pump...

[theman53]

Yes it alters the pressure but it doesn't change the timing curve. Giles and BYM put pumps in front of the VE pump and pressured them to something stupid like 40psi with no timing advance change.

[745 turbogreasel]

I'm guessing if anything it helps compensate for the pisswater thin diesel they sell us.

[TurboJ]

Is there any particular reason to fit the transfer pump before the filter?

Filtering the fuel before the pump should reduce wear on the pump, no?

Anyway, there are two general types of a transfer pump; 1. feed pump and 2. suction lift pump.

...those are stupid names just invented by me right now, but the idea is that 1. needs to reside in the fuel tank or below it,
or at the very least at the same low level as the fuel tank's bottom.

2. Can be fitted in the engine compartment, and will suck fuel out of the tank even though fitted higher up than the tank.

Are there any drawbacks to using an in-bay pump, or the type 2. as explained? Would you need a one-way valve to stop this pump from running dry at start-ups?


*** I learned that some of the BMW 3 and 5 series 2.4 and 2.5 liter IDI diesels use a feed pump (transfer pump fitted inside the fuel tank) as stock.
Do you guys know of other OEM designs that combine a VE pump with an electric feed pump?

[vanbcguy]

Is there any particular reason to fit the transfer pump before the filter?

Filtering the fuel before the pump should reduce wear on the pump, no?

Anyway, there are two general types of a transfer pump; 1. feed pump and 2. suction lift pump.

...those are stupid names just invented by me right now, but the idea is that 1. needs to reside in the fuel tank or below it,
or at the very least at the same low level as the fuel tank's bottom.

2. Can be fitted in the engine compartment, and will suck fuel out of the tank even though fitted higher up than the tank.

Are there any drawbacks to using an in-bay pump, or the type 2. as explained? Would you need a one-way valve to stop this pump from running dry at start-ups?

A 'pusher' pump can overcome a lot more resistance (ie multiple filters, very very low micron high restriction filters) than a 'suction' pump.  That would be the biggest reason - it is a lot easier to push fluid through a filter than it is to suck it through.  You are correct though - if you have the pump before the filter it is going to get all the unfiltered nastiness of the fuel fed through it.  That said, a suction pump that is working against a restrictive filter can cavitate, which will wear it out quite quickly too.

*** I learned that some of the BMW 3 and 5 series 2.4 and 2.5 liter IDI diesels use a feed pump (transfer pump fitted inside the fuel tank) as stock.
Do you guys know of other OEM designs that combine a VE pump with an electric feed pump?

The Cummins engines do.  I believe many non-VW applications with VE pumps have lift pumps.  VW gets away without one because the tank is not that far away from the injection pump (compared to a truck) and the injection pump is not really that much higher up than the tank.

[92EcoDiesel Jetta]

No matter the type, pumps works better if it does not have to "lift", i.e. better to mount the pump lower than the tank so it is gravity fed than to mount it in the engine compartment where it has to fight gravity.

Whatever "lift" pump you choose, I'd suggest installing gauges before the IP to make sure the additional pump is actually helping and not presenting a restriction. I think by going to 1/2" fuel lines may be enough by itself w/o the additional "lift" pump. Just my gut feeling, no experience. Instrumentation don't lie.

[TurboJ]

Thank you, kind people!

This is an interesting subject for sure.

I have done some research and I found a good solution ( I think ) to overcome most of the possible problems.


As far as I know, the problems to overcome in designing a high performance fuel system for the VW IDI engine are:

1. Flow restrictions that cause either air leaks or the VE pump to cavitate
2. Lack of adequate extra fuel flow for the VE pump's cooling and lubrication
3. Possible water contamination
4. Leaking return lines
5. Uneven flow or pressure to the VE pump that affects timing / internal pressure / throttle response
6. Too high fuel temperature resulting in pump wear and reduced power
7. If a transfer pump is used, providing enough flow and not making this extra pump a restriction
8. If a transfer pump is used, protecting it by adequate filtering of the largest particles without causing unnecessary restriction.



SO. My solution looks like this:

 


I drew this up using a gasoline EFI catch tank design as a template.


Tell me what you think!

The big idea in this design is to allow plenty of cool fuel to the VE pump at all times, with minimal restriction to the VE pump's intake.
The catch tank in this design is a non-pressure one to remove as much stress from the transfer pump as possible.
The transfer pump is protected by a rough filter and also by a water separator.

The only questions that come to my mind right now are:

1. Will there be reverse flow in the fuel tank's return line?
2. Will the transfer pump be over loaded by three different filters?

...At least the VE pump should meet no resistance nor pressure when sucking in fresh fuel.

[R.O.R-2.0]

Thank you, kind people!

This is an interesting subject for sure.

I have done some research and I found a good solution ( I think ) to overcome most of the possible problems.


As far as I know, the problems to overcome in designing a high performance fuel system for the VW IDI engine are:

1. Flow restrictions that cause either air leaks or the VE pump to cavitate
2. Lack of adequate extra fuel flow for the VE pump's cooling and lubrication
3. Possible water contamination
4. Leaking return lines
5. Uneven flow or pressure to the VE pump that affects timing / internal pressure / throttle response
6. Too high fuel temperature resulting in pump wear and reduced power
7. If a transfer pump is used, providing enough flow and not making this extra pump a restriction
8. If a transfer pump is used, protecting it by adequate filtering of the largest particles without causing unnecessary restriction.



SO. My solution looks like this:

 


I drew this up using a gasoline EFI catch tank design as a template.


Tell me what you think!

The big idea in this design is to allow plenty of cool fuel to the VE pump at all times, with minimal restriction to the VE pump's intake.
The catch tank in this design is a non-pressure one to remove as much stress from the transfer pump as possible.
The transfer pump is protected by a rough filter and also by a water separator.

The only questions that come to my mind right now are:

1. Will there be reverse flow in the fuel tank's return line?
2. Will the transfer pump be over loaded by three different filters?

...At least the VE pump should meet no resistance nor pressure when sucking in fresh fuel.

i think half of that stuff is completely un necessary..

the only reason the EFI cars had a reservoir, was so the pump was always 100% submerged in fuel..

there is no need for a fuel catch can to feed your injection pump..

BYM and Giles fed a bosch VE pump with 100psi *(MAYBE 150psi even)* and nothing was different about the timing of the pump..

only thing you have to worry about, running such high pressures, is that you may push seals out of your pump...

[TurboJ]

That was not the question, but thank you for your comment.

The question was if the design in the picture would work the way I meant it to. And this I need answered.
So, if I use a catch tank, do you see something in the design that would not work or what should be done differently?

Comments regarding that are welcome!

[theman53]

I would definately eliminate the catch tank. I can see no purpose to it other than heat the new fuel coming in. If you do keep it then add another filter before the IP. If the IP or injectors do have contaminates then they will go right back into the IP again. Also there will be nothing stopping the fuel from the pump pushing it into the catch tank and right back into the main fuel tank.

If cool filtered fuel is your main concern I think you should attack it this way. Run a separate pump into a fuel cooler rad, that all its job is to cool the fuel and return it to the tank. Then run your filter, water separator, inline pump, final filter to the IP. Send the return back to the tank. Two separate fuel loops, but you will be able to run the fuel on a temp control and keep it exactly where you want it. The cooling loop would be able to transfer more turnaround and should be more effective.

[745 turbogreasel]

My car has a sock on the  end of the fuel pickup that seems to be fine for what the lift pump can swallow.

[R.O.R-2.0]

AGAIN, for those of you who missed it in the first EPISODE:

you can feed the injection pump with both VACUUM and PRESSURE..

pressure does not change the timing advance curve..