The car idles fine, no problems, is fine under strong acceleration, will show no problems on a flat road at a constant speed, but as soon as there is an incline and I move the pedal down to maintain speed the motor keeps surging. It is as though it is trying to provide maximum throttle every couple of seconds or so. It keeps doing this until I either put my foot flat to the floor or the road levels off again, I lift the throttle and it carries on as normal. There are no fault codes.

 

Here's a guess as to a possible cause of the problem.
Since the regulator coil is inductive, I would expect it to have in parallel a suppressor or reverse-biased diode to reduce the voltage spike which occurs each time the PWM switches the coil current off. When there is a demand for higher pressure, the PWM duty cycle is increased and hence the coil current increases. This in turn increases the voltage spike. If the supressor/diode is failing, there could be a throttle setting at which the spike causes breakdown of the switching transistor in the ECU, resulting in the anomalous behaviour you experience.
You could test this theory by connecting a suitable reverse-biased power diode across the regulator coil and seeing if the regulator behaviour changes. No guarantee of a cure though .

 

The car idles fine, no problems, is fine under strong acceleration, will show no problems on a flat road at a constant speed, but as soon as there is an incline and I move the pedal down to maintain speed the motor keeps surging. It is as though it is trying to provide maximum throttle every couple of seconds or so. It keeps doing this until I either put my foot flat to the floor or the road levels off again, I lift the throttle and it carries on as normal. There are no fault codes.

When you face an incline and press the accelerator down you are moving into high fuel demand, i.e. you should get increased fuel pressure and max out on the pulse width of the main injection pulse(ignore the pilots)
If as you are suggesting the injection pressure is going way too high then you should have an exceptional cloud of black smoke out the rear.
If you don't get a large puff of black smoke then respectfully you are absolutely barking up the wrong tree. If this is the case you may not be building sufficient fuel pressure at peak demand and I would start looking at low pressure fuel supply (Filter?)
If you are getting black smoke, that indicates either excess fuel or insufficient air supply. The first suspect at this point is the EGR valve , which can be easily tested by plugging the vacuum hose going to it temporarily with a 5mm bolt. If smoke still persists then you need to check that the engine can breath i.e. that the intake isn't badly sooted up , limiting the max air supply. When you are then satisfied that all those parameters are Ok you need to again start looking at your fuel pressure readings.
I do this kind of work day in day out, for a living and have worked on this engine for almost 20 years since it first appeared and am very doubtful that this is an issue with the pressure actuator driver in the ECU

 

Here's a guess as to a possible cause of the problem.
Since the regulator coil is inductive, I would expect it to have in parallel a suppressor or reverse-biased diode to reduce the voltage spike which occurs each time the PWM switches the coil current off. When there is a demand for higher pressure, the PWM duty cycle is increased and hence the coil current increases. This in turn increases the voltage spike. If the supressor/diode is failing, there could be a throttle setting at which the spike causes breakdown of the switching transistor in the ECU, resulting in the anomalous behaviour you experience.
You could test this theory by connecting a suitable reverse-biased power diode across the regulator coil and seeing if the regulator behaviour changes. No guarantee of a cure though .

Hi Alec_t, I tried your idea of putting a diode parallel to the regulator (flywheel diode?) but there was no difference unfortunately. Thanks for the suggestion though.

 

When you face an incline and press the accelerator down you are moving into high fuel demand, i.e. you should get increased fuel pressure and max out on the pulse width of the main injection pulse(ignore the pilots)
If as you are suggesting the injection pressure is going way too high then you should have an exceptional cloud of black smoke out the rear.
If you don't get a large puff of black smoke then respectfully you are absolutely barking up the wrong tree. If this is the case you may not be building sufficient fuel pressure at peak demand and I would start looking at low pressure fuel supply (Filter?)
If you are getting black smoke, that indicates either excess fuel or insufficient air supply. The first suspect at this point is the EGR valve , which can be easily tested by plugging the vacuum hose going to it temporarily with a 5mm bolt. If smoke still persists then you need to check that the engine can breath i.e. that the intake isn't badly sooted up , limiting the max air supply. When you are then satisfied that all those parameters are Ok you need to again start looking at your fuel pressure readings.
I do this kind of work day in day out, for a living and have worked on this engine for almost 20 years since it first appeared and am very doubtful that this is an issue with the pressure actuator driver in the ECU

There is no puff of smoke appearing. I have also done the EGR valve test previously.

Unfortunatly I can only find a sensor for the high pressure side of the fuel supply, do you know of one on the low pressure side that I can take a reading from? I couldn't think of a way of measuring the low pressure side, without spending a lot of time and money making a test rig with some kind of spare hose and a pressure sensor.

What you are saying makes sense to me, but why I had mostly dismissed that idea, is because of the fact that the symptoms disappear when I put my foot to the floor. Perhaps the car is not accelerating at it's full capacity (I can't remember how well it accelerated before this problem, it has been so long!), but if the high pressure pump is being starved of fuel, wouldn't the engine stutter and/or bring up the fuel pressure fault code? As it is, it accelerates reasonaby well, given the age, enough to overtake in safety.

Edit: I forgot to mention, perhaps more pertinent, the symptoms can appear going as slowly as 30 mph, but I can drive at 80 mph for hours with no problem. If the pump was being starved of fuel at 30 mph, how could it have enough fuel supply to do 80 mph? I wish it was fuel starvation, it would be much easier to fix, but I can't fathom how it is possible.

Can you think of an explanation that covers this anomoly?

What would be your next step, if you were trying to isolate the fault? Many thanks for your input.

 

I am not familiar with Citroen engines so I am just going to add a couple of thoughts to the mix.
1) Does this car use a lock-up torque converter in the transmission? I am not even sure it is automatic. Possibility it is jumping in and out under load?
2) Is the ignition system 100%? I have seen where ignition coils will break down under light load and nowhere else through the power range.
I may be off track but just food for thought.

 

I tried your idea of putting a diode parallel to the regulator (flywheel diode?) but there was no difference unfortunately.

In that case any existing protection components are probably working well. That's one thing we can cross off the list . Cork_ie is clearly the expert here, so hopefully he'll come up with the solution.

 

I am not familiar with Citroen engines so I am just going to add a couple of thoughts to the mix.
1) Does this car use a lock-up torque converter in the transmission? I am not even sure it is automatic. Possibility it is jumping in and out under load?
2) Is the ignition system 100%? I have seen where ignition coils will break down under light load and nowhere else through the power range.
I may be off track but just food for thought.

The car is manual transmission and diesel, so no ignition coil.

 

Today I ruled out the accelerator pedal position sensor. I had an old one lying around, that I had changed because it was throwing up a fault light. I re-installed it and had exactly the same symptoms, so have put the other one back on. Oh well!

 

I am not familiar with Citroen engines so I am just going to add a couple of thoughts to the mix.
1) Does this car use a lock-up torque converter in the transmission? I am not even sure it is automatic. Possibility it is jumping in and out under load?
2) Is the ignition system 100%? I have seen where ignition coils will break down under light load and nowhere else through the power range.
I may be off track but just food for thought.

It's a diesel!!!

 

This looks very like some kind of fuel starvation to me. Are you 100% sure that the low pressure pump is OK and the fuel filter is clean?
You need around 3 Bar (45 psi) coming to the high pressure pump.
At what RPM does the surging start at ? & approx what RPM does it disappear at, when you accelerate again?
Another possibility is the MAF (mass airflow meter) Can you just try driving the car with the MAF disconnected and see if there is any difference in the RPM range that the fault occurs at. You will most likely notice a serious drop in overall power but this is just to confirm whether it is influencing the surge or not.

 

This looks very like some kind of fuel starvation to me. Are you 100% sure that the low pressure pump is OK and the fuel filter is clean?
You need around 3 Bar (45 psi) coming to the high pressure pump.
At what RPM does the surging start at ? & approx what RPM does it disappear at, when you accelerate again?
Another possibility is the MAF (mass airflow meter) Can you just try driving the car with the MAF disconnected and see if there is any difference in the RPM range that the fault occurs at. You will most likely notice a serious drop in overall power but this is just to confirm whether it is influencing the surge or not.

The garage changed the fuel filter when I first had the problem, but it was nearly new anyway. They said they checked the low pressure pump and it was OK. Have now done a test, will post the results separately.

The surging doesn't seem to be engine rev specific, more throttle position, but it starts around 1800 rpm and up to about 2300, I don't generally go above that except when accelerating. It will disappear when I press the throttle fully no matter what the revs.

I have previously disconnected the MAF, no difference. It is a new one I replaced last year.

 

I have just done a flow and pressure test on the low pressure side. It was easier than I thought, I was previously trying to connect to the hose, but when I looked again today, it is much easier to connect to the outlet pipe of the filter, this also tests the flow through the filter.

I must add, the voltage to the pump when I tested was not up to the level it would be with the car running, also the battery is a bit tired as well. So getting about 12.5v instead of the 14.3v with the alternator running. So I don't know if that would change the results significantly?

I tested it a few times, with little variance. 2.6 Bar pressure and a flow rate of 330 ml in 4.6 secs or 4.3 L/min. I couldn't test for longer than the 4.6 seconds as it turns off after that. I can't test the pressure when running either with the equipment I have, as I have to disconnect the hose from the filter.

2.6 bars to me sounds just about on the edge of acceptable, and 4.3 L/min sounds like plenty to run the engine? Thoughts?

 

Well I went 0/2 on my last try. I am with Cork_ie on the fuel starvation. Any chance you can run this with an oscilloscope on the fuel pump during a test run? Again, not familiar with this engine. Assuming electronic injection, can you scope the injectors on a run?

 

Well I went 0/2 on my last try. I am with Cork_ie on the fuel starvation. Any chance you can run this with an oscilloscope on the fuel pump during a test run? Again, not familiar with this engine. Assuming electronic injection, can you scope the injectors on a run?

Are you referring to the low pressure fuel pump? If so, that is definitely fed with straight 12v D.C. so not much point in checking it with a scope. I have checked it with a meter on a run, and it is getting a good voltage (13.7v I think it was) with no change.

 

Hey Dumpster. Sorry I wasn't looking for voltage reading. Watching current fluctuations.

You may be caught between a rock and a hard place with a little more sophisticated equipment. Scope will tell you a lot more detail than a meter will and you can catch glitches a lot easier.

 

Hey Dumpster. Sorry I wasn't looking for voltage reading. Watching current fluctuations.

You may be caught between a rock and a hard place with a little more sophisticated equipment. Scope will tell you a lot more detail than a meter will and you can catch glitches a lot easier.

Sorry, I misunderstood your reason for a scope test, my bad. I'm not sure if I can manage that, but I have thought about a way of testing for a pressure drop in the low pressure side of the fuel supply. I'm not rich and I am reluctant to spend good money on test equipment that I would probably never use again, so it is kind of like one of those game shows where you have to accomplish a task with only a limited amount of equipment! Still, it stimulates the old grey cells coming up with ways round these problems!

I don't have a pressure transducer, but I might have hanging around, or may be able to get my hands on a pressure switch from a well pump (lots of people use them out in the sticks around here believe it or not), if I can then somehow 'T' it into the supply line, I can keep change the setting on it to see how low the pressure drops on a run to activate the switch. The switch is a simple matter to monitor with a continuity meter inside the car. Will see what I can find this weekend. Eat your heart out Mcgiver! (have I spelled that right? Never watched it).

 

Update: I managed to 'T' off the low pressure inlet to the high pressure pump. I connected a pressure gauge to it, with enough tube to tuck the guage under the windscreen wiper, but had to leave the bonnet on the safety catch.

With the engine running there was a constant 3 bars of pressure to the pump, it didn't vary even a fraction on my test run, no matter what the engine load or speed, but the fault was still evident.

I think I can finally rule out a fuel supply problem, so back to my compensating circuit - I will start on it this week.

 

Can you use your scope setup to monitor your injectors or are they mechanical? Again, I know nothing about Citroen but I do know diesel engines.