Hi

I have a problem with the fuel pressure regulation on my HDI diesel engine. After many hours of testing and replacing parts, I've come to the conclusion that there is a problem inside the ECU. The car is a banger, so it's not worth getting a new ECU, so I am trying to work out something to compensate for the fault.

The pressure control regulator is fed on one side with a constant 14.3 volts, the output then goes to the ECU, which regulates the current flow according to demand from the accelorator. At low demand (and thus low current flow) there is no problem, but above about 1500mA it appears that the control circuitry inside the ECU just shorts momentarily (about 1/4 second) every few seconds, thus sinking maximum current through the pressure regulator causing the car to surge sporadicaly.

I first thought of using the original control circuit to power a transistor (a TIP120 I had lying around) which would reduce dramatically the current flowing through the ECU and maybe eliminating the problem (if it was due to overheating), but if not, I could adapt the circuit with a capacitor to eliminate the surges. I've now given up on that idea because of the problem of gain using a transistor, this would alter dramatically the current output to the pressure regulator with only a small demand in change from the ECU.

Next I thought of using a capacitor to smooth out the 'dips', but I would need a huge capacitor as at maximum demand there is over 2 amps flowing through the regulator. An inductor would work, but again, perhaps I would need a very large one to make any difference?

I could restrict the current flow by putting a relay or transistor in circuit, that above a certain current flow (or voltage to the regulator) it operated and disconnected the ECU and connected to earth via a resistor, but the problem with that is the resistor would have to be a fixed value, so depending on when it operated, I would still either get a surge, or drop off in power momentarily.

The figures I have measured are: Current flow: At idle - 750mA, driving steady uphill - 1400mA, foot to the floor - 2000mA. Voltage across regulator: At idle - 1.94V, driving steady uphill - 3.6V, foot to the floor - 5.5V. The Vcc is 14.3V, I've calculated the resistance of the regulator at 2.6 ohms.

Does anyone have any suggestions? (apart from scrap the car!)

 

I’d be reluctant to give up after all the investigative work you’ve done.

I offer this circuit (as the basis) that will allow you to input the 1.94 – 5.5V with an increased current output.

Given the maximum output current of the op-amp, Q1 will need to be a Darlington transistor.

With the engine running and the circuit drawing 2A, Q1 will be dissipating 17W - so will need a heat sink. Alternatively you could wire a 3Ω (12W) resistor in the transistor collector.

The input to the op-amp is the 1.94 – 5.5V (to the non-inverting input), with the inverting input connected to the emitter of Q1.

I invite others to improve on this circuit.

 

Thanks Hymie, that's given me something to work with, I was starting to lose hope. I will look into this method.

Thanks again for taking the time to help.

 

You could use your TIP120 transistor as Q1.

R1 could be 100Ω.

The op-amp could be almost any (741 should be OK).

Given Q1 power dissipation, I’d give serious thought to a resistor in the transistor collector – minimising the heat sink required.

Don’t forget that the tab of the TIP120 is connected to the collector.

 

Which engine is this: - By the sound of it (hdi), it is either a 1.4, 1.6 or 2.0l PSA engine fitted to any number of possible vehicles. The valve you are referring to is called an inlet metering metering valve, fitted to the injector pump. What make of pump is fitted on your car ? Bosch or Siemens (aka Continental) . What is the make of the ECU Bosch EDC 16 or Siemens SID?
I would like to see an oscilloscope trace of the signal to the metering valve v's rail pressure sensor signal , before doing any mods. The rail pressure sensor is a 0-5V potentiometer as is the EGR position sensor and they share the same 5V supply and EGR problems are far more likely to be an issue than an ECU.
If you gave me all the details I will try and help you. If this is a 1.6hdi first thing is unplug the EGR valve (located at the rear of the engine right hand side (facing the engine.) and repeat your tests and see if the readings you have obtained are the same.

 

Voltage across regulator

How is the valve wired up to the regulator? I would expect them to be in series, but that would imply that increasing the voltage across the regulator would leave less voltage (and hence less current) to drive the valve, which is inconsistent with the figures you quote.

 

Hi Cork_ie, the engine is a 2.0 HDI on a Citroen xsara 2000 model. I have tested the rail pressure sensor and it is giving the readings I would expect, showing the spikes in pressure as the pressure regulator is given full current intermittently.

Is the EGR valve control the one with the pneumatic hoses going to it? If so, I have unplugged that (plugging the hose as per internet instructions) and there was no change in the behaviour. I've also (besides many other things!) unplugged the mass air flow meter, no change.

 

Hi Alec_t, sorry, what valve?

 

Ok this sounds like the old old 2.0l 8 Valve PSA engine ( code will be RHZ or RHY) . The pump is a Bosch CP1 with pressure regulator at the back and no metering valve, which is a later innovation. The pressure regulator is the only method of pressure control. With no power going to the regulator you should have a pressure of ~200 Bar , the more power the higher the rail pressure. Forget about the EGR valve as yours is a vacuum operated EGR without feedback.
The pressure regulators could be quite troublesome when aged and may be the source of your problem, again I would suggest you get someone to do an oscilloscope trace of the PWM signal to the regulator V's rail pressure. Take your readings at the ECU pins as there is a distinct possibility that you may have a wire intermittently touching to earth and that will drive the regulator into full on.You may also have a defective rail pressure sensor.

 

Yes, it is a RHY engine. I'm pretty much sure that the problem lies in the ECU. Your logic is spot on, which is why when I finally narrowed the fault down to this regulator circuit, I monitored it with an analogue voltmeter (much easier to see large changes than my digital one and I don't own a scope) and saw that above a certain pedal position (except foot flat down, curiously) there was this spiking to earth sporadically. The intensity and speed of change, led me to the conclusion that it was a fault rather than a response to an input. Like you, I suspected an earth fault, so replaced both the live feed wire to the regulator, and the wire between regulator up to the plug on the ECU. No change. In a final bout of optimism, I fitted a brand new regulator, no change. This is why I am fairly certain the fault lies in the ECU, hence my original post and why I am asking for help on an electronics forum, not an automotive forum. Unless you can think of some other explanation of the symptoms?

 

Hi Alec_t, sorry, what valve?

Ignore my comment. Brain in wrong gear .

 

Are you sure that the fault is not a noisy throttle potentiometer ?

Les.

 

Hi, lesjones, I had the same thought, so I tested the outputs from that and there were no spikes that I could see, just a smooth change in voltage output. If the ECU is really just responding to a faulty input, it is rather drastic, because with my foot flat to the floor I am getting about 5.5V to the regulator and the car is accelerating healthily, but the spikes when they occur are sending over 10v to it, with the current nearly doubling from 2A to 3.5A.

 

Do you know if the ECU monitors the current through the regulator to get feedback for controlling the PWM drive of the regulator?

 

I'm sorry, I've no idea. Don't know what a PWM is, but I'm hoping that the current to the regulator isn't monitored, because that would stymi my idea of using it to drive an external circuit!

 

Don't know what a PWM is

Pulse Width Modulation. My next question was going to be "Do you know what the PWM frequency is?". I ask because, if the regulator is indeed PWM driven as Cork_ie says, then it could help to know the frequency in designing some way of modifying the PWM signal to restrict its duty cycle and hence limit the voltage across the regulator.

 

I don't know for certain, but I don't think that the setup is that advanced, I think it is just a straight DC currrent to the regulator driving a solenoid inside. I did at one time check for any AC on the circuit, wondering if I had a faulty alternator and it didn't register any AC voltage (I know pulsed DC is not the same as AC, but it would show something on the meter). I think if the solenoid were driven by PWM, it wouldn't be drawing so much current? Could be wrong. I'm (again!) assuming that inside the ECU, the regulator is connected to the collector of a high-power transistor, the emitter of which is earthed. The ECU would vary the base current to control the fuel pressure and thus the power to the engine, but this is just hypothesis. I had an idea that the transistor was overheating at high current output.

 

Have you opened up the ECU and looked inside? Can you map out the circuit that runs the regulator circuit? Can you post pictures inside the ECU?
Is this problem temperature related or there 100% of the time?

 

Have you opened up the ECU and looked inside? Can you map out the circuit that runs the regulator circuit? Can you post pictures inside the ECU?
Is this problem temperature related or there 100% of the time?

The problem starts about 30 seconds up the road, then all the time, whatever the weather. Sometimes it is worse than other times, but I haven't been able to pinpoint anything that ties in with that.

I haven't opened up the ECU, I'm a bit wary of doing that. It is hellish hard to get to, I'm not sure if it does come apart easily and am worried about breaking something, then the car is not driveable at all. As it's my only means of transport, not sure I want to risk it. Has anyone here ever taken an ECU apart to fix a fault?

 

I think it is just a straight DC currrent to the regulator driving a solenoid inside. M, it wouldn't be drawing so much current? Could be wrong. I'm (again!) assuming that inside the ECU, the regulator is connected to the collector of a high-power transistor, the emitter of which is earthed. The ECU would vary the base current to control the fuel pressure and thus the power to the engine, but this is just hypothesis. I had an idea that the transistor was overheating at high current output.

It is a PWM dignal controlled on the negative side by the ECU, As far as I remember the output driver is a mosfet or IGBT

because with my foot flat to the floor I am getting about 5.5V to the regulator and the car is accelerating healthily, but the spikes when they occur are sending over 10v to it,

Where is this coming from? The regulator is supplied with + Ign voltage via the main relay (12V-14V nom.) and grounded by the ECU
The DC resistance of the regulator coil is only about 1.5 Ohm. Fuel pressure is a function of engine load and speed. Can you try and graph rail pressure v's engine revs on a diagnostic computer,while driving the car? Again I repeat you really need to check this with a scope before going any further. Almost any half decent garage has them nowdays.
I do this kind of stuff all day every day and have worked on at least 500 of these engines in various cars and 4X4's and I have never seen the symptom you describe, caused by an intermittent short in the output driver of the ECU