Fuel Pressure Regulator

[spdingtkts]

With a fuel pressure regulator you can go two ways, return type or non-return type.

Without factoring in the ease of not have to plumb the extra line for the return style, which setup is better? Is one better for carbs vs fuel injection?

 

[EngJW]

EFI systems use the return type. They must maintain a constant pressure in the rail. Flow is either on or off, being controled by the opening of the injector. Without the return line, the fuel pressure could spike when the injector is closed.

Compare this to a carburetor where there is pretty much a steady flow of fuel, so a non-return regulator will work. With a non-return regulator, if don't have flow then the pressure will build up to as much as the pump will put out.

 

[richdubbya]

IMO it's best to use the return type with carburetion also. The return type will keep fuel temps more constant, and prevent fuel pressures to increase when the engine is shut off hot, which can cause flooding and raw fuel emissions. The fuel pump volume must be adequate to supply the engine and all the return fuel.

 

[spdingtkts]

Thanks for your input guys.

I have always thought that the return configuration was superior to a non return setup for the fuel temperature reason previously sited.

The reason I asked this is that in the past few years I have witnessed fuel injected Japanese sport bikes go from utilizing a return style fuel system to a non return style system in recent years.

I am trying to find out if there are any benefits of running the non return style that I am unaware of.

I think that the Japanese factories may have done this in order to reduce pump size in order to save weight, reduce electrical draw, and to reduce fuel heating from larger in the tank pumps.

 

[GregLocock]

Non return is the way of the future, because you are heating the fuel in the tank, which increases evaporative emissions. It is also lighter and uses less power.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[NeilDS]

Many of the automotive non return systems just use a return style regultor mounted in the fuel tank. This does reduce the the fuel heating and and ends up cheaper since only one line needs to be run up to the fuel rail. When they started swithching over to the non retrun systems the did find pressure pulation problems in some of the systems. They ended up putting dampers on the fuel rails to solve the problem.

Neil

 

[spdingtkts]

I have seen the dampers. If I remember correctly, Chrysler had a recall on the Caravan to install the dampers.

I would call the in the take configuration more of a bypass setup, similar to an OEM oil system than a return style system.

A friend of mine is plumbing up a fuel injected motorcycle in a non stock application and he wants to install an after market fuel pressure regulator so that he can adjust the pressure to what he wants. I always thought that a bypass configuration (at the rail) is/was the way to go. I wanted some other opinions on the subject.

 

[EngJW]

Does the non-return efi system use a pulse width modulated fuel pump instead of a regulator?

 

[spdingtkts]

It still has a regulator, it just has it in the tank.

As for the pulse width question, I do not know.

 

[NeilDS]

I know that pulse width modulated pump have been in development for auto fuel systems, but as of a few years ago all the non return system I worked on just used in tank regulator modules. These regulators were the same ones that were put on the rails for the return systems, just moved back to the tank.

 

[schmidtj86]

I think Ford may be using a voltage regulated pump in some applications.
There are many OEM reasons to go to a deadheaded system (like no more engine bay heat being put into the return line to the tank, and one less line), but few aftermarket reasons. It takes lots of engineering to combat the water hammer like pulsations, even the type of fuel line makes a difference. The pulsations and pressure waves can cause uneven distribution at best, and the harmonics can cause low pressure during the injector opening.
Sequential injection systems will not suffer much from these effects though, in which case a simple damper can suffice. Some OEMs have developed flexible fuel rails that slightly balloon with the pulsing. Most aftermarket systems are batch or bank, in which case it's best to avoid deadheading.
I'll never recommend to anyone to try to setup a deadheaded system on their own, and expect it to perform as well as a crossflow. Heck, just purging the thing takes luck and risks engine or injector damage. And if the system isn't 100% sealed at key off, it'll likely have to purge each time as heat can vapor lock a low pressure rail.

 

[EngJW]

NeilDS or anyone else-

Where the regulator has been moved to the tank, does it still function as a return type regulator? That is, does it bleed off excess flow and return it either to the tank or back to the inlet of the pump? Also, is the regulator a separate part or is it part of the pump?

 

[NeilDS]

For the systems I worked on at Siemens, they were just pressure relief type regulators that were put in an intank pump module. They were not part of the pump, still a seperate part. From what I remember they just returned back into the tank. Here is some info on the regulators.

http://www.synerject.com/regulators_fuel.html

Neil

 

[WGJ]

Ford are using the PWM fuel pump control system in the Lincoln Town car (I think) - I worked on the Jaguar X400 and X350 models which use the same basic control system.
It gets complex as it's another map that the engine control unit needs to refer to. The ECU puts out a variable frequency to another module which contains the power electronics and puts out a PWM signal to the fuel pump.
This has a lot of advantages for a (luxury) production car:
- Minimised pump noise
- Minimised fuel flow noise
- Extended pump life. The systems we worked on would support over 500bhp in the vehicle and you don't really want to be circulating all that fuel all the time
- Cooler fuel, as already mentioned elsewhere

 

[spdingtkts]

Barry Grant sells an electric control unit that drops voltage to the pump at lower RPM's for the same reasons you stated above.

 

[pulsewidth]

I've been playing around with solid state relays to vary pump speed, especially the large Weldon pumps. I've found these pumps need at least a 2000hz PWM drive. The relays I've used are fast-switching sinking types not
sourcing. It's amazing how much diode suppression the pumps need when operated at less than 100% duty. My goal is to reduce the return flow by PWM to a level that does not overflow a standard off-the-shelf Bosch regulator while maintaining a manifold-referenced fuel pressure in spite of fuel consumption. A three dimension map of duty cycle with Manifold pressure and RPM as table axis should let me tailor the fuel to the demand. The Ford system described by WGJ goes one better by closing the loop with a differential pressure sensor as feedback.

 

[spdingtkts]

The Barry Grant step down device is not very sophisticated.

I thought it tied into the ignition and had some type of "map" based on RPM's, but all it has is a WOT switch and a manual bypass switch.

 

[WGJ]

The trouble with step-down devices is sometimes associated with variability in the electric pump's performance.
The pump will have been designed to provide an optimum performance at 12 or 13.5 volts. If you take a batch of pumps and then look at how thay do in the range 7 to 9 volts, the variability in performance will be much greater than the 12 - 13.5 volt performance, making it more difficult to get a good 'fit' with your engine requirements.

On the relays to provide a sort of PWM - the inertia of the moving parts of a mechanical relay will work against you at some frequency. Contact wear will be a problem, and as already pointed out the back-EMF generated when the relay is switched off can be a real problem to deal with. Fast switching relays also cause problems in terms of pick-up of noise in control circuits on the engine.

May I suggest a look at

httP://www.4qdtec.com?

You'll find descritpions of electronic switches called "H-bridges". These circuits are used to PWM switch electrical loads like fuel pumps and cooling fans.

 

[pulsewidth]

I should clarify my comments about PWM and relays. I'm only using solid state (electronic) relays. Mechanical relays would probably give up the ghost in under 5 minutes!

 

[EngJW]

We have an engine in the shop that was set up two different ways:

1. We used a Bosch electic pump mounted externally and connected to a fuel rail. A pressure regulator is at the other end of the rail and has one return line to the tank and another line for map reference. The engine was calibrated this way and sent out for field test.

2. When we got the engine back, it had a Ford pump, mounted externally, but one end of it has a fuel line returning to the tank. The flow from the pump is through a regulator and into the rail where it dead ends. The regulator is referenced to map. There is also a small in-line device that might be a damper.

I am trying to find some information about system #2 in order to understand how it works. I found one article on the net saying that Ford uses a constant pulse width injector but regulates fuel flow by continuously varying the regulator pressure. If this is true, the calibrating should now be way off.

Could someone please englighten me on this?

Thanks