Transit fuel pump learn

[jm7812]

Hi,

Attempting to do a fuel pump learn on my transit and the first part “temporary learn” works perfect however when attempting the next part it keeps failing, I have tried this 4 times now.
This is the error:
‘Suspect an intermittent fault:
-Fuel rail pressure sensor
- short to VBATT
‘
Any ideas what I can do?

[l200lc]

im having the same problem. i keep getting fuel pressure too high.

[Gilles]

Same issue from my side, I posted detailed informations few weeks ago, it loks like nobody found the solution yet. I am thinking about a bug in Forscan. It probably detects something wrong but the error message is not reflecting the reality: the sensor and wiring are generally OK.

[boosted_diesel]

Might be a bad fuel rail pressure sensor or even a flow co trip valve on the hp pump. If either of those are bad it’ll fail.
You could monitor voltage with a meter and see if it does actually spike which would mean either a bad sensor or it’s unable to control the pressure properly.

[Gilles]

Thanks for the advice. This is also what I was expecting. The pump was is new (not a fake) so I suspected a pressure sensor failure or a wiring issue.
Here is a short description of my investigations:
- continuity <10 milliohms from sensor to ECU (all pins)
- insulation > 10 MegOhm from wire to wire , wire to ground, wire to Vbat,
Wiring is OK:
Signal quality measurement with an oscilloscope:
The signal is pretty similar to the one displayed by Forscan and looks OK:
- Dc voltage at the sensor output= 1.35 V at idle --> sounds OK
- AC noise +- 30mV peak is it to high ? I did not see any spikes above 1.65 V
Enenthough the noise level seems to me beeing acceptable I decided to reduce it by inserting a first order filter.
- The noise went down to +- 15 mV peak which is OK in this environment.
I started again the pump learn service. Unfortunately it failed again at 46% with always the same error message saying: ""suspect an intermittent fault: short circuit to VBATT.

After conduction of all these investigations and reading many posts on several forums reporting the same problem without finding the solution I am thinking about a bug in Forscan or, at least a mistake in the reason of the abortion. Because there is no glitch nor spike to be interpreted as a short to
supply voltage (5 V from the ECU).
Since the engine is running fine I give up for the moment and stay with a temporary status with the pump learn.
I am not satisfied with this situation but I have no idea on how to go further from my side. I would really appreciate if a Forscan expert could check in the code which condition(s) and /or voltage limit raises the short circuit to VBATT ...
I would like to thank everyone who had the patience to read to the end.

[boosted_diesel]

Thanks for the advice. This is also what I was expecting. The pump was is new (not a fake) so I suspected a pressure sensor failure or a wiring issue.
Here is a short description of my investigations:
- continuity <10 milliohms from sensor to ECU (all pins)
- insulation > 10 MegOhm from wire to wire , wire to ground, wire to Vbat,
Wiring is OK:
Signal quality measurement with an oscilloscope:
The signal is pretty similar to the one displayed by Forscan and looks OK:
- Dc voltage at the sensor output= 1.35 V at idle --> sounds OK
- AC noise +- 30mV peak is it to high ? I did not see any spikes above 1.65 V
Enenthough the noise level seems to me beeing acceptable I decided to reduce it by inserting a first order filter.
- The noise went down to +- 15 mV peak which is OK in this environment.
I started again the pump learn service. Unfortunately it failed again at 46% with always the same error message saying: ""suspect an intermittent fault: short circuit to VBATT.

After conduction of all these investigations and reading many posts on several forums reporting the same problem without finding the solution I am thinking about a bug in Forscan or, at least a mistake in the reason of the abortion. Because there is no glitch nor spike to be interpreted as a short to
supply voltage (5 V from the ECU).
Since the engine is running fine I give up for the moment and stay with a temporary status with the pump learn.
I am not satisfied with this situation but I have no idea on how to go further from my side. I would really appreciate if a Forscan expert could check in the code which condition(s) and /or voltage limit raises the short circuit to VBATT ...
I would like to thank everyone who had the patience to read to the end.

Well it certainly seems you’ve done your due diligence.
If you aren’t seeing a spike in pressure or the signal anywhere when it does then it could be a software issue. It is strange it points it’s finger at a particular thing yet it can’t be seen happening.
Do you see any pressure spikes while normal driving? Something like a kinked return pipe could cause pressure issues.

[Gilles]

Thanks for your time, I realize that reading such long post may be a little bit boring...
Back to your reccomendations:
- I did not monitor the signal while running yet. I will do it through forscan oscilloscope soon.
- Regarding return pipes I did not notice any issue when I replaced the 4 injectors but I will check anyway.
The outcome will come soon.

[boosted_diesel]

Yeah keep us posted, will be interesting to see what it turns out to be.

[colin39]

Have you done a pilot learn on the inlet metering valve on the fuel pump

[Gilles]

Hello, I'm back as promised, unfortunately I'm still at the same point. The engine runs fine with the temporary status. I don't have any error codes but the learning procedure always stops at 46%. I guess that the inlet metering valve learn is part of the pump learn procedure. I think that I have covered most of the possible causes so as long as I don't know the exact conditions which arise the flag in Foscan software I stop the investigations. Running with temporary calibration parameters is not ideal but it works so I will stay with that but I am still looking for any news on dedicated forums.