Fuel Pump Voltage: Affect on Flowrate
Turbo Mini Forums

Pump: Walbro 255
Fuel lines: 3/8 feed and return
Injector used for test: Rover MJY100460
Duty cycle: 50%
Rail Pressure: 55psi gauge indicated
Voltage measured at pump contacts
Test carried out over 2 minutes to increase resolution of results.


Test 1:
Pump Start voltage: 11.9v
Pump End Voltage: 11.8v
Average Voltage for test duration: 11.85v
Injector voltage: 11.85v
Flow in two minutes: 483cc = 241.5cc/min

Test 2:
Start voltage: 10.45v
End Voltage: 10.20v
Average Voltage for test duration: 10.33v
Injector voltage: 11.9v
Flow in two minutes: 470cc = 235cc/min

Test 3:
Start voltage: 13.45v
End Voltage: 13.45v (connected to running car)
Average Voltage for test duration: 13.45v
Injector Voltage: 13.45v
Flow in two minutes: 500cc = 250cc/min


Test 4: (to re-confirm nothing had changed)
Start voltage: 11.8v
End Voltage: 11.7v
Average Voltage for test duration: 11.75v
Injector Voltage: 11.75v
Flow in two minutes: 481cc = 240.5cc/min


Summary:
Well - it changes, no doubt about that. In this particular installation, you have to ask if the ballast resistor setup is worth implementing. It must be remembered that the test was conducted with just one injector. In a turbocharged mini, you will likely be running two injectors in parralell at idle, so the flow difference will likely become more apparent.
Certainly, with a different pump - an OE one for example - it is likely a greater change would be experienced; the walbro is basically a race-pump with a large flow-rate at high rail pressures.
Finally - Audiably, the 'pitch' of the walbro changed with applied voltage, but the actual audiable volume was little different. It is a NOISY pump!
How well i will cope with this in my road-going mini we will have to wait and see...

Edited by TurboDave16V on 10th May, 2006.

interesting results, I'm shocked that it made any difference if fuel rail pressure was was the same.

What voltage did the standard turbo pump run at when it was on in the metro with the ballest resistor setup?

Team www.sheepspeed.com Racing

I assume that you were firing the injectors for a fixed length of time at a fixed frequency.

Could the variation of the results be due to the speed of the injector opening slowing at lower voltages?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'm guessing the injector was always operated at the same voltage, and it was only the voltage of of the pump which changed.

Team www.sheepspeed.com Racing

BMW use PWM to alter supply qualtity as there is no return line from the injector rail anymore.

Alex

AlexF

As do many OEM's.

You need a pump with an internal pressure relief aswell however.



As regards the injector voltage, I've ammended the first post to include that. Basically it's to simulate driving the pump through a resistor - hence the injectors are still at 'normal' battery voltage.

I have an MPi tank here Peter.
I'll re-do the test sometime in the next few weeks with that and see. I should find some spare time inbetween making my manifolds and assembling my engine and trans.

Hi guys,

I thought I’d chip in with a few comments here that may or may not be useful to some of you.

Firstly if a constant rail pressure was maintained for these tests (in this case 55psi) then in no way should decreasing the pump voltage to that level specified affect the fuel delivery volume. Was the rail pressure maintained absolutely throughout all three tests? A tiny drop in rail pressure (as a result of the lower pump voltage decreasing available volume) would account for the small flowrate difference.. that I could believe. The variances seen between a couple of those tests will undoubtedly be from the decreasing voltage applied to the injector and the associated affect on injector opening time as suggested by some of you. Like a couple of you have said, I find it a little odd to think the results are purely from pump voltage whilst at a constant rail pressure.

Dave, is this a new pump in perfect condition?

The reason I ask is the Walbro 255 (I’m guessing either a GSS341 or 342 in this case?) is a phenomenal pump, if a little sensitive to voltage (well documented elsewhere). I do find it hard to believe that firing only one 440cc/min injector in this test saps enough flow to work the pump enough to see reduced flowrates at the lower voltage……Just my 2p.

This pump is used in many 2JZGTE applications at around 12v for over 600 flywheel bhp. To put that into perspective that’s typically 6x 650cc injectors at a differential of approx 45psi. I’m running one of these pumps in my car and that pushes out just over 420bhp – not even close to the limit of the pump. The general bhp guide for 2JZ applications is passing 380-400bhp you may want to think about swapping out the OEM for a Walbro 341/342.

I just didn’t want people to assume from Dave’s test that this would be a good pump for a turbocharged A series. At 12v the flowrate would be so high the main result would be overheated fuel – nasty in any installation as this raises a whole host of issues….

Good to see people getting stuck into these kind of tests though – generates healthy discussions.

Cheers

Dan

Edited by Daniel on 10th May, 2006.

Dan,

It is indeed a new, genuine Walbro 255. 342 model i think (i'll check).

I can believe the results, but feel (in this example) that it's not down to the flow of the pump being challenged. I feel it's the micro-changes in pressure as the injector opens and closes, that the pump has to keep up with.
That's why i want to test the Rover MPi tank pump aswell. The difference in this will be more down to a shortfall in overall flow if anything.

I'll also be running a fuel pressure damper in my fuel rail. I've a feeling that these are key, especially with huge batch-fired injectors - in reducing this deviation.

Reason I chose the walbro was because i'm going to be running higher rail pressures, and huge injectors - and need a pump that i KNOW will be more than upto the task. Thing that must be remembered, is that the pump needs to flow what can come out of your injectors at the time they're open... Basically - a lesser pump would happilly run a set of (say) 300cc injectors than a set of 450 cc units, as the flow out of the injectors at the time the injector is 'open' is less on the smaller injectors. The smaller ones just rely on being open longer to get the fuel in. With the A-series, we have a different set of rules; where we're limited by duty cycle, NOT available pulsewidth.
I can cope with excess flow. I'm not prepared to take a chance on 'just enough' however. Basically - the Walbro was chosen based on a selection criteria, not just because I found one cheaply....


As regards the temperature; well the tank was around 1/4 full last night. I did a 'hand' test on the tank before the 12v test. I then ran the pump at 70psi for 30, maybe 40 minutes to drain the battery down to 10.5 volts. I repeated the 'hand test' and I could identify no difference at all... Sure - a thermocouple would have been more usefull though!!!
I WAS expecting the tank to be a lot warmer however.

Naturally - with less fuel, it will heat up more. But with a half, or full tank - it'd be less...

If it was looking like a problem, it is pretty easy to make a fuel cooler from a peltier heater. I'll probably do this at some point in the future...

I'm liking the sound of these motors of yours - how about mentioning them in 'show us yours'? They sound worthy of our attention!

Edited by TurboDave16V on 10th May, 2006.

on the subject of peltier elements, I had to get some for work, got them from greenweld.

http://www.greenweld.co.uk/cgi-bin/ss00000...R=-1&TB=A&SHOP=

going o/t but i was toying with useing peltiers for cooling inelt charge whilst waiting to stage at the strip... but a simple alcohol or co2 spray is much simpler. anyways, sorry, back on topic.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

In addition to what has been discussed, could there be an effect due to the location and type of fuel pressure regulator?

Also, you don't mention what pulsewidth you're using in the measurements. I assume you want lower pump voltage at low engine load which means very small pulse widths and very low duty cycle. So if the effect you're seeing is due to injector opeing then it will become larger as the pulse width goes down. If it's something else, it could be affected by duty cycle. So redoing the test with different pulse widths and duty cycles could reveal something about the cause.

Jean

http://www.jbperf.com/

Good point Jean!

Probably over the weekend, when i'm re-testing the cossie injectors - I'll repeat the test with different pulse-widths.

If anyone is interested, the unit I'm using for my injector test rig is from http://www.dysys-controls.com/ by the way.

It's a smart little unit that has three settings:


1: Simulated Engine speed (500 to 9900 rpm)
2: Injector duration (0 to 9.9ms)
3: Duty % (calculated from the above)

Hence I can wind it up/down/whatever within the range to get a more representative figure.

If i have time - I'll repeat with the MPi tank aswell.

Edited by TurboDave16V on 11th May, 2006.

jsut a wee somehting you may wana think about mate. a normal car would have a fuel rail with a load of fuel in it and a deent abmount of rubber hose honiing it all together. i know a liquid is theoretically incompressible, but say in such a system there is enough liquid and rubber to damp the "weirdness" the injectors are causeing the pump . may be worth repeating your experiments with, say, an extra meter or two of fuel hose. also mayb try an oe regulator (assumign your useing a large aftermarket one, which by its physical size will ahve more mechanical inertia and thus less ability to quickly react to changes in fuel presssure)


makes for very interesting reading.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

A fuel cooler off a diesel with Common Rail should be more than enough for our purposes...

The look either like a bog brush (pipe with wires for greater area) or a very small oil cooler (which is my personal plan as I have a few spare - like 9!!).

Alex

AlexF

Well they are only oil coolers....

best pic I can do is on here http://www.batinc.net/coolers.htm

The one I plan to use has tig'd ends for 8mm pipes.

Alternativly a good coil of copper fuel line curled up and in the air flow will help.

To be honest I don't think for what we are doing we will need to worry to much about the temp of the fuel.

Alex

Edited by AlexF2003 on 12th May, 2006.

AlexF

Ive seen newish ( say 2001 era ??? ) BMW 320d's with a nice little "fuel" cooler, with 8mm barbs on it. No idea what year, but it sits near the headlight. I seen one ages ago.

Also seen another cooler on a VW Polo tdi I think it was, mounted under the drivers floor.

FWD Toyota Corolla twin cams of old, have a nice little oil cooler up front, with 10mm barbs ( I'm using one as a PAS cooler )

Or you could use any sort of cooler really intended for oil etc.

The Walbro as mentioned is a superb pump, and more than capable of supplying 500bhp+ at circa 60psi +

That said, at higher pressures flow does start to fall off badly.

Lots of good info here
http://www.stealth316.com/2-fuelpumpguide.htm

Most Bosch pumps will handle higher pressures than the Walbro, and still maintain flow.
The 044 would be serious voerkill, but a lot of guys within the Subaru community are using a Bosch "909" pump. Cheaper, push on fitting for the inlet, and it maintains excellent flow, at high pressures...

Although for Minis, I dont think any of the above pumps would be pushed too hard at an rate ( perhaps with the exception of a methanol fuelled car )

editted to subscribe ( how do you automatically subscribe to threads ?? )

Edited by stevieturbo on 12th May, 2006.

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Good link Stevie,

Saw it ages ago, but didn't bookmark it and never found it again - till now!

I'm guessing if peeps are testing pumps at over 13.5 volts, you must be able to purchase step-up transformers for such a task?

well you can mess with the alternators regulating circuitry to give out more power. been done alot by people who run "SPL" car's to get ever last dB out there sound system.

"steeping up" DC is pretty tricky.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

http://www.jbperf.com/

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

Edited by Andymini on 26th May, 2006.