Fuel Pump speed control

[Mr2Mike]

Pat_T said:

Yes.

Less fuel heating = colder, denser intake charge. (and reduced fuel evap for OEMs)

Not forcing fuel through a regulator under pressure will reduce the amount of heat being put into the fuel, but then again with a return system you have a lot of fuel to put that energy into, and quite a lot of area for it to be radiated from. Does this actually give a useful improvement in practice?

Pat_T said:

Less fuel pump current draw = lower alternator load on engine.

Similar to the power loss you get with headlamps on. A few amps current draw is pretty irrelevant in terms of power at the wheels.

Pat_T said:

Potentially less noise from fuel pump as others have stated.

For a DIY installation this would be important, noisy fuel pumps are very annoying. Most production cars manage to isolate the sound of fuel pumps pretty well however (my MR2 excluded, though the pump was right next to you).

[anonymous-user]

It's not really about upheat to the fuel from the fuel system itself (pump etc) more about upheat from engine heat, transmitted from hot components in the engine bay. (like pipe work, and crucially the fuel rail) This limits the total thermal loading imposed on the fuel tank, and massively reduces evaporative emissions from the tank volume.


These days, every Watt counts in terms of fuel economy, and being able to reduce fuel pressure when it is not required means even more energy savings.

[stevieturbo]

Mr2Mike said:

Similar to the power loss you get with headlamps on. A few amps current draw is pretty irrelevant in terms of power at the wheels.

Depends really.

First time I did a current check on my car I was pretty surprised. I wasnt bothered about engine power, more the load on the alternator and ability to maintain good strong voltage, as well as current and cable sizes needed.

The more current you need, the bigger all those items need to be.

With a pair of 044's running full time, the engine etc and in daylight, I'm sitting around 45A.
Add some lights, radio, indicators, fan off/on and very quickly the load was up close to 80A and there was still more I could turn on without getting excessive. I was shocked at how much it added up, and at the time I only had a 105A alternator.
So at times it would be pushed near its limit which didnt really leave a lot of excess to charge the battery itself.

So if you were pushing hard for a lightweight efficient race car, then less electrical demand can give area for weight reduction and improvements overall

[226bhp]

Mr2Mike said:

Pat_T said:

Yes.

Less fuel heating = colder, denser intake charge. (and reduced fuel evap for OEMs)

Not forcing fuel through a regulator under pressure will reduce the amount of heat being put into the fuel, but then again with a return system you have a lot of fuel to put that energy into, and quite a lot of area for it to be radiated from. Does this actually give a useful improvement in practice?

The fuel temp problem can be dealt with as the OEMs do it, by fitting a fuel cooler (or two).

Edited by 226bhp on Friday 5th December 22:42

[shoehorn]

Max_Torque said:

I more about upheat from engine heat, transmitted from hot components in the engine bay. (like pipe work, and crucially the fuel rail)

Why do we stick with a fuel rail that can suffer from heatsoak rather than say a distributor block,away from heat with separate lines similar to how an old Jetronic system was set up?

[anonymous-user]

226bhp said:

Mr2Mike said:

Pat_T said:

Yes.

Less fuel heating = colder, denser intake charge. (and reduced fuel evap for OEMs)

Not forcing fuel through a regulator under pressure will reduce the amount of heat being put into the fuel, but then again with a return system you have a lot of fuel to put that energy into, and quite a lot of area for it to be radiated from. Does this actually give a useful improvement in practice?

The fuel temp problem can be dealt with as the OEMs do it, by fitting a fuel cooler (or two).

Edited by 226bhp on Friday 5th December 22:42

Not really no. OEMs will definitely NOT be fitting any fuel coolers, from both a cost and a safety point of view. Far better to not heat the fuel up in the first place. And as a returnless system that does that is both cheaper, and reduces build time/complexity and increases reliability it's a no brainer!

[226bhp]

Max_Torque said:

226bhp said:

Mr2Mike said:

Pat_T said:

Yes.

Less fuel heating = colder, denser intake charge. (and reduced fuel evap for OEMs)

Not forcing fuel through a regulator under pressure will reduce the amount of heat being put into the fuel, but then again with a return system you have a lot of fuel to put that energy into, and quite a lot of area for it to be radiated from. Does this actually give a useful improvement in practice?

The fuel temp problem can be dealt with as the OEMs do it, by fitting a fuel cooler (or two).

Edited by 226bhp on Friday 5th December 22:42

Not really no. OEMs will definitely NOT be fitting any fuel coolers, from both a cost and a safety point of view. Far better to not heat the fuel up in the first place. And as a returnless system that does that is both cheaper, and reduces build time/complexity and increases reliability it's a no brainer!

I was under a Golf the other day which had one, I then went to Ebay and searched 'Golf fuel cooler' and found rather a lot.
Although on second thoughts they are diesel (Petrol ones are available from Demon Tweeks IIRC) which may make a difference and agreeably, probably not something you want hanging under a competition car!

Edited by 226bhp on Sunday 7th December 12:20

[anonymous-user]

On diesels you pretty much have to have one, as the high fuel pressure and fact that you are returning injector "overspill" back to the system means lots of heat! Of course, as the fuel has such low volatility, evaporative emissions are not an issue either.




back OT:


I've got a VDO fuel pump driver module, Ford Part number 3W 83-9D372-AB, VDO number A2C53005489, off a Range Rover. It has a 6 pin connector which pinsout as follows:

1) 12v output to fuel pump
2) 0v return (low side PWM) from pump
3) pump drive signal from eng ecu, 60 to 330Hz, drive with open collector low side drive, 0 to 50% duty is 0 to 100% pump duty
4) not used
5) 12V power feed, ignition switched
6) 0v chassis ground

The unit uses a MLX16201 "smart" relay controller internally (no CAN coms) and an infineon BTS202Z low side driver to run the pump. I've not beasted mine yet, but it looks good for something like 10A continuously, so should run a reasonable sized fuel pump.

[Pat_T]

Max_Torque said:

back OT:


I've got a VDO fuel pump driver module, Ford Part number 3W 83-9D372-AB, VDO number A2C53005489, off a Range Rover. It has a 6 pin connector which pinsout as follows:

1) 12v output to fuel pump
2) 0v return (low side PWM) from pump
3) pump drive signal from eng ecu, 60 to 330Hz, drive with open collector low side drive, 0 to 50% duty is 0 to 100% pump duty
4) not used
5) 12V power feed, ignition switched
6) 0v chassis ground

The unit uses a MLX16201 "smart" relay controller internally (no CAN coms) and an infineon BTS202Z low side driver to run the pump. I've not beasted mine yet, but it looks good for something like 10A continuously, so should run a reasonable sized fuel pump.

Fantastic, thanks Max_T! I'm buying one, I was just looking for a wiring diagram last night.

(.3kHz is enough for closed loop control? As I think before you mentioned around 1kHz might be better..?)

[anonymous-user]

Pat_T said:

Fantastic, thanks Max_T! I'm buying one, I was just looking for a wiring diagram last night.

(.3kHz is enough for closed loop control? As I think before you mentioned around 1kHz might be better..?)

300Hz is the frequency of the control signal output to the amplifer, not the control bandwidth. The pumps mechanical inertia and no doubt the input signal filter in the amplifier will in reality limit the rate of change of pump torque to probably something less than 50Hz. However, you want to run your control logic at a higher rate, ideally 10x as fast, to ensure sufficient control headroom and reduce phase lag.

[stevieturbo]

Max_Torque said:

On diesels you pretty much have to have one, as the high fuel pressure and fact that you are returning injector "overspill" back to the system means lots of heat! Of course, as the fuel has such low volatility, evaporative emissions are not an issue either.




back OT:


I've got a VDO fuel pump driver module, Ford Part number 3W 83-9D372-AB, VDO number A2C53005489, off a Range Rover. It has a 6 pin connector which pinsout as follows:

1) 12v output to fuel pump
2) 0v return (low side PWM) from pump
3) pump drive signal from eng ecu, 60 to 330Hz, drive with open collector low side drive, 0 to 50% duty is 0 to 100% pump duty
4) not used
5) 12V power feed, ignition switched
6) 0v chassis ground

The unit uses a MLX16201 "smart" relay controller internally (no CAN coms) and an infineon BTS202Z low side driver to run the pump. I've not beasted mine yet, but it looks good for something like 10A continuously, so should run a reasonable sized fuel pump.

At 10A, sounds like it might be at it's limit with an 044 ?

[anonymous-user]

stevieturbo said:

Max_Torque said:

On diesels you pretty much have to have one, as the high fuel pressure and fact that you are returning injector "overspill" back to the system means lots of heat! Of course, as the fuel has such low volatility, evaporative emissions are not an issue either.




back OT:


I've got a VDO fuel pump driver module, Ford Part number 3W 83-9D372-AB, VDO number A2C53005489, off a Range Rover. It has a 6 pin connector which pinsout as follows:

1) 12v output to fuel pump
2) 0v return (low side PWM) from pump
3) pump drive signal from eng ecu, 60 to 330Hz, drive with open collector low side drive, 0 to 50% duty is 0 to 100% pump duty
4) not used
5) 12V power feed, ignition switched
6) 0v chassis ground

The unit uses a MLX16201 "smart" relay controller internally (no CAN coms) and an infineon BTS202Z low side driver to run the pump. I've not beasted mine yet, but it looks good for something like 10A continuously, so should run a reasonable sized fuel pump.

At 10A, sounds like it might be at it's limit with an 044 ?

You'd have to carry out a proper abuse test to see where the thermal limit of the driver were. However, unlike a return system, the pump is only driven at full power when the engine is at maximum fuel consumption, which is obviously a much lower duty cycle in effect.

(I can't seem to find a datasheet online for the BTS202Z low side PWM driver chip in the VDO module. That would give some numbers for RDSon and thermal impedance allowing one to estimate the maximum capability)

[stevieturbo]

I see a few of the Range Rover ones on egay, various prices.

Any source for the connector and pins ?

Almost looks like a Bosch LSU lambda sensor shaped connector ?

[Pat_T]

www.volkspage.net/technik/ssp/ssp/SSP_334_d1.pdf
VAG love producing these sort of documents. This one is an overview of their FSI fuel injection systems. Looks like they use a PWM controlled low pressure pump in the manner we are discussing, with a camshaft driven mechanical high pressure pump to feed the direct injection at 30-100 bar.

Both systems have their own pressure sender units (the low pressure of which might be suitable for me 0.5-6.5 bar range)

The low pressure side is maintained at 0.5-5.0 bar, except during cranking where it is raised to 6.5 bar.

They also let on that the ECU PWM output signal is at 20Hz to the fuel pump control module which then outputs at 20kHz to the low pressure pump! I am wondering if this was strictly neccessary as the higher the PWM frequency the higher the switching losses? Maybe it's of NVH benefit by moving the frequency outside our range of hearing?

[anonymous-user]

Pat_T said:

They also let on that the ECU PWM output signal is at 20Hz to the fuel pump control module which then outputs at 20kHz to the low pressure pump! I am wondering if this was strictly neccessary as the higher the PWM frequency the higher the switching losses? Maybe it's of NVH benefit by moving the frequency outside our range of hearing?

Because the system does not require the lift (low pressure) pump to rapidly change it's flow rate, tu can be controlled by a low (20Hz) frequency signal, effectively acting as a low pass filter. They use 20Khz to drive the pump for the following reasons:

1) Outside audible frequencies
2) Minimises torque ripple on the pump motor, and helps reduce commutator / brush losses (as it's many times quicker than the basic commutation frequency, which depends on pump speed obviously)
3) Reduces the size and cost of the EMC filter components necessary in the pump driver (big capacitors / inductors are expensive!) For a typical 10A RMS N channel mosfet, the iintrinsic Gate Source capacitance is pretty small, and at only 10A the resistive losses during non saturated semi-conduction are relatively small.

[anonymous-user]

stevieturbo said:

I see a few of the Range Rover ones on egay, various prices.

Any source for the connector and pins ?

Almost looks like a Bosch LSU lambda sensor shaped connector ?

Think i got mine for about £16 iirc, was a while ago mind.

Not sure what the connector is, certainly could be the same as LSU4.2s which use AMP SSC (Sealed Sensor Connector) series interconnects. Looks very similar (rounded ends internally, 1 square end externalls for keying, 6 pins, retaining clip on top side)

[Pat_T]

Max_Torque said:

Think i got mine for about £16 iirc, was a while ago mind.

Not sure what the connector is, certainly could be the same as LSU4.2s which use AMP SSC (Sealed Sensor Connector) series interconnects. Looks very similar (rounded ends internally, 1 square end externalls for keying, 6 pins, retaining clip on top side)

Just got one of these for £15 I'll let you know when it turns up.

Also now got my fuel pump, level sender unit, and a job lot of -6AN fittings and hose on the way, so looking forward to making some progress...

[demlotcrew]

Hi Gents,

How did everyone get on with this VDO driver?

I just sourced one on the bay based on the advice in this topic and its use is destined be for a DC EWP (yes I know what you all think of these pumps) as my ECU wont output a high enough continuous current (1.0A at 70ºC) so will this VDO should do what I need.

Looking to hear how others got on and wired in the unit?

[Pat_T]

I'm not there yet, but will be back on this thread once I start the wiring in earnest. My fuel system is plumbed in now, just need to make a start on the main engine loom!

[Pat_T]

Max_Torque said:

I've got a VDO fuel pump driver module, Ford Part number 3W 83-9D372-AB, VDO number A2C53005489, off a Range Rover. It has a 6 pin connector which pinsout as follows:

1) 12v output to fuel pump
2) 0v return (low side PWM) from pump
3) pump drive signal from eng ecu, 60 to 330Hz, drive with open collector low side drive, 0 to 50% duty is 0 to 100% pump duty
4) not used
5) 12V power feed, ignition switched
6) 0v chassis ground

The unit uses a MLX16201 "smart" relay controller internally (no CAN coms) and an infineon BTS202Z low side driver to run the pump. I've not beasted mine yet, but it looks good for something like 10A continuously, so should run a reasonable sized fuel pump.

Max_T, (hoping you're still reading this)

I'm now doing the schematic drawings for my engine and data looms. Am I right in thinking that using the Jag VDO driver module, I no longer need a fuel pump relay and I can run the PWM signal from ECU direct to driver module, then 12V output direct to fuel pump?

cheers,
Patrick