Pls help me to understand this signla from the scope

[Chriss]

Hi!
I'm dealing with a car electronic repair.
The high pressure fuel regulator didn't regulate correctly.

I took my scope and checked the pwm signal which is coming from the engine ecu and drives the fuel regulator which is a simple solenoid.
So far as I know, the ecu is switching the negative side of the solenoid, but both wires from the solenoid is connected to the ecu and
not like the hot wire or plus wire to the battery plus.
But however. The signal; looks for me very strange for a good pwm signal.

When I done the measuring, I measured the signal with my scope respect to the battery ground.
On one wire I have this signal which looks like a pwm but with lot of spikes and some distortion,
I assume when the solenoid goes of (is de energized).

The other wire is constantly around battery voltage.

Can somebody help me and discuss about the signal from my scope?
I'm not sure what is wrong and in which direction should I turn in my repair process.

Thank you for any help, suggestion and your time.
My best regards.

Here are the pictures of my scoping results:
1. when the ignition is on and the engine is not running:


2. when the engine is running but fuel pressure is very high over 1000 bar on idling:

[wraper]

Inductive kickback from coil. https://www.underhoodservice.com/fuel-injector-circuit-waveforms/

[Chriss]

Inductive kickback from coil. https://www.underhoodservice.com/fuel-injector-circuit-waveforms/

Thanks for the quick reply.
Do you think this ecu drives the solenoid correctly? I mena, is this output really in good condition?

Here is how I see this:


Somehow this is strange for me. I never saw this kind of signals on the fuel pressure regulator solenoid before.
And I tried on another car to take samples and I don't have that spike there and that discharge time curve.
I have to say the comparing measurement was taken with the exact same scope settings.

[aqibi2000]

And I’m surprised you misinterpreted it this bad.

He clearly stated fuel pressure regulator not fuel injector

[Johnny10]

My Mistake!
Looked exactly like Fuel Injector waveform to me.

Not enough Coffee this morning?

Thanks for the observation.

[Chriss]

So guys, what you think?
Is the output of the ecu bad or I've have another situation?

[CaptDon]

I don't see anything on that waveform I don't like.
Is it bad, well, that depends??? What is wrong with
your fuel pressure??? Too low, too high? To low, maybe
pulse width is to short or not enough pulses in a given
time. Opposite if the pressure is high. Sometimes the
regulator gets crud in the modulated needle valve and
the pressure drops and car runs bad. We would probably
need to compare to a 'good' waveform to tell if this one
is bad.

[Chriss]

Thanks for the reply and your time.

The pressure is to high, around 1500 bar or even more on idling.

The regulator valve is ok and the rail pressure sensor is also ok and the ecu also recognise the pressure value from
the rail fuel pressure sensor correct.
How I know this?
Cos I attached a pwm controller to the fuel pressure regulator valve and set it up to deliver around 350 bar on idling.
So to say, I regulated the pressure by hand and not with the ecu.
On the wiring of the ecu I connected another fuel pressure regulator just to make the ecu happy, not to fire a dtc.
In this scenario the engine was running perfect on idle, the fuel pressure dropped to 350 bar what was feeling by the
sound how the engine was running.
The ecu also registered the 350 bar fuel pressure or any other fresurre what I set up manually.

So, I assume the input/fuel pressure monitoring stage on the ecu is working correct, but I cant figure out why the ecu didn't
regulate correctly the fuel pressure regulator when it is connected.

I dont have the same car to make some tests and compare the signal shape, but I saw many times signals from the fuel pressure regulator from other cars, and all of them was nice square wave pwm signal, without any spikes etc.

I think, the high pressure scenario could be because a prolonged period of OFF time, what we can also see on one of my pictures where I was draw the red curve on the signal.

That part of the signal, the red curve duration is around 2ms where the valve is maybe closed.
And based on this theory the pressure could be high in the rail.

If the signal would be much steeper then the OFF time would be shorter and the valve would be much more in an ON state,
and therefore the pressure would be also lower in the rail.

What you think about my theory?
Could that cos a high pressure scenario?

If yes, what could cause this kind of signal distortion?

Note:
I also investigated how high is the voltage of the pwm on other cars on the fuel pressure regulator.
They were around the battery voltage and not around 40-50v.

Thank you very much.

[Johnny10]

What kind of car are you working on?
Engine/Make/Year ?

[Chriss]

It is a Fiat Doblo 1.3 MultyJet 2007.
The engine code: 223A9

[mikerj]

I agree the second trace looks very suspicious, the recovery time after the PWM signal switches off is extremely long, almost as though there is a significant leakage current to ground on the FPR return.

When you used the manual PWM controller to set the pressure, do you see the expected waveform?  Is so it could point to a problem in the driver within the ECU, or possibly a problem in the FPR wiring.

[m3vuv]

It maybe a good idea to ask on digital kaos garage section,lots of clever guys on there with fuel injection,i assume its a marelli ecu and common rail injection?

[Chriss]

Yes. It is a Mareli and a CR...
I also put on several automotive forums my question but without any answers, just viewing...
Do I ask some stupid question here or mehanics guys not really dealing with measuring and tools and they not sure what I'm talking about?
I dont know...

[Chriss]

mikerj:

When you used the manual PWM controller to set the pressure, do you see the expected waveform?

No, I didn't measured when I used the manual method.
But I'm pretty sure the engine was running perfectly and the ecu also recognized very accurately the fuel pressure.
And I never saw before this kind of signal on other type of engines with similar fueling systems where are also the same
fuel pressure regulator are used...

[aqibi2000]

Can I suggest you add a 10k him pull down resistor between ground and the output of the fuel
Pressure regulator which you are monitoring

And then observe the slew of the fuel rail pressure

[Chriss]

If I understand you corret, I should put a 10k to pull down the regulator?
I'm not sure what you mean by "output"?
The ecu is switching the ground side but the positive side is always on during the ecu is on.

Do you try to discharge the coil until the ecu is not switch to ground?

[bdunham7]

In the first instance you have a short on time which means the energy stored in the coil is modest and the inductive kick is also modest.  In the second, you have much longer on time (remember--the ON time is the LOW (0 volts) section) which leads to much more energy stored and a much larger inductive kick.  Part of that kick is clamped by an overvoltage protection circuit (the top flattish section) and then after the energy subsided a bit the voltage goes down below the clamping voltage and dissipates normally (the curved portion you highligted).  This is not a transistion, the next transition occurs when the ECM turns the current back on near the end of the curve.

I have no idea if this is 'normal' for your car, that would be a matter of the exact circuitry and the characteristics of the regulator.  Does a higher duty cycle equal more fuel pressure, or is it the opposite?  Is there a fuel pressure sensor for feedback?  I would think the first step would be to see if the ECM is correctly reading the fuel pressure.  Which brings up a different question--how are you determining that the fuel pressure is high?  ECM data or physical fuel pressure gauge?  If one, can you get a reading the other way?

EDIT:  I just saw your other post.  If the ECU reads the pressure correctly according to your gauge, you'll have to ask it why it is setting the pressure where it is.  It clearly has the capability of changing the PWM, so there must be some control logic leading it to do what it is doing.

[Chriss]

I completely understand you and here are some answers:
More duty cycle - lower pressure.
I have no gauge that grade to measure 1000 bar but the ecu
is showing the pressure even it is really high corect.
I'm sure about that cos I put on the regulator a pwm controller and the engine was running fine and the ecu was reading the correct pressure and didnt set any more the "high pressure" meaning error code ( dtc ).
And the engine also didnt give that hard noise which is present when the pressure is to high.

There is a fuel pressure sensor which is also checked and changed for testing and it is ok.

I measured before many many cars and never saw this kind of fuel pressure regulator signal.
All they was a nice square wave pwm signal without any spike and curve...

[bdunham7]

I completely understand you and here are some answers:
More duty cycle - lower pressure.
I have no gauge that grade to measure 1000 bar but the ecu
is showing the pressure even it is really high corect.
I'm sure about that cos I put on the regulator a pwm controller and the engine was running fine and the ecu was reading the correct pressure and didnt set any more the "high pressure" meaning error code ( dtc ).
And the engine also didnt give that hard noise which is present when the pressure is to high.

There is a fuel pressure sensor which is also checked and changed for testing and it is ok.

I measured before many many cars and never saw this kind of fuel pressure regulator signal.
All they was a nice square wave pwm signal without any spike and curve...

Then it is likely that the fuel pressure regulator is bad, or for some reason cannot respond properly to the PWM signal.  I have no idea of the particulars of this specific system, so take my advice guardedly.  The ECM is going to a high duty cycle to try and lower the pressure, but it isn't able to do so.  One reason you may not have seen a signal like that is perhaps because in normal operation at a lower duty cycle it doesn't look like that.  You have another fuel pressure regulator handy?  Can you just measure the DC resistance of the coil?  If the FP regulator coil were partially shorted  (like halfway) it would draw more current, not work properly and might have a kick signature like that.

[Chriss]

I put new FPR and FPS.
Also checked the resistance of the FPR and they are 22 Ohm.

As I wrote, when I manually managed the pwm signal to the FPR i got a nice low pressure on the ecu. Which was as the ecu requested on idling 350 bar.

[bdunham7]

Well then!  I guess you've covered those items!   

What did your manual PWM signal look like on a scope?  How did you generate it?  What duty cycle got you the 350 bar? 

It's tempting to say that the ECU is the only thing left, but I never liked replacing ECUs unless I absolutely knew why, partly to avoid replacing a good part and partly to avoid wrecking a new part.

[bdunham7]

Two other questions.  You said the other side of the FPR coil comes from the ECU.  What does that voltage look like on a scope?  And when you did your manual PWM experiment, what did you do with that supply-side wire--leave it connected to the ECU, connect it to B+, ??

[Chriss]

I didnt checked how the signal was when I connected the pwm controller to the FPR.
I can do that tomorrow.

The two pins from the FPR was connected to the pwm controller. So the ecu was practically disconnected but I connected to the ecu another FPR just to make the ecu happy. So the ecu won't drop a dtc for a missing FPR.

Here is a correct signal how it should look like from ESITRONIC app.

[bdunham7]

I didnt checked how the signal was when I connected the pwm controller to the FPR.
I can do that tomorrow.

The two pins from the FPR was connected to the pwm controller. So the ecu was practically disconnected but I connected to the ecu another FPR just to make the ecu happy. So the ecu won't drop a dtc for a missing FPR.

Here is a correct signal how it should look like from ESITRONIC app.

OK.  Scope the other (power) wire running it with the ECU as well because even if it drops the voltage on the control side, if it can't source enough current to pull the solenoid in it won't operate properly.  I'm not sure why the inductive kick would be so high in that case.  And while you're at it, it might be instructive to scope both wires on the dummy FPR while you are using the manual control method.

Is the ESITRONIC example from your specific application?

[Chriss]

Ok.
I will do what you suggested and lost.

Btw:
Bosch ESI[tronic] is a diagnostic software products for the maintenance, diagnosis and repair of vehicles. The software includes basic vehicle data, important diagnostic functions, maintenance documentation and diagrams.