MK7 2.2tdci FWD 2010 won't start

[Cavesurfer]

The 100ml + leak-off on injector #4 will definitely be the cause of failing the learns. Injector coding is relatively easy, it's just the inputting of the 16 digit "correction factor" codes into Forscan or IDS. Then rerun the pump, pilot/injector learns and you should be golden.

Lord Knobrot

So, I have made some progress after a long delay in getting hold of injectors. Now I have 4 x reconditioned (by denso) injectors fitted, new leak off lines and new fuel filter fitted. I flushed out the fuel rail and the high pressure pipes as well before it all went back together.

Coding the injectors was a little more challenging as one of the reconditioned injectors did not have it's correction values stamped on very clearly, so it was a bit of guesswork. For these injectors reconditioned by Denso, the correction values are stamped into what appears to be a brass ring wrapped around the body of the injector...not printed onto the plastic body of the wiring loom connector as it would be on a new injector. The alignment of the letters on just of of the injectors was slightly awry such that all of the letters were low enough on the ring as to make reading the difference between the letter 'F' and the letter 'E' problematic.

With the aid of a magnifying glass I read the code as; F000 2C31 1E3D 3EF0

I'm confident that the first 'F' in the first set of four digits is correct as all of the injectors in the set started with this sequence 'F000'. The other three occurrences of 'E' or 'F' in the final two sequences of four digits may not be correct however.

Initially when I tried to enter the correction values for all four injectors the Forscan software produced an error code and would not let me save the values and it suggested I contact the developers. Assuming that one of the 'E' or 'F' values may not have been correct, I tried all 8 combinations of these three values and the FOrscan software gave the same effort for every permutation. The advice on the Forscan website associated with this error was to download the latest version of Forscan which I did and was then able to run this procedure again with the codes being accepted.

Anyway, job done, I thought and I primed the fuel system and tried to start the engine. A little reluctantly it started and ticked over nicely. No fuel leaks anywhere so I let it warm up with the intention of running the pilot learn which had previously failed.

On connecting to Forscan I get the following:

Connection to Forscan 14-9-2021.JPG

There appear to be two problems here;

One problem is the P167B - Fuel Injector Learning Not Done

Is this the Pilot learn that you say I need to do?

The other problem that worries me is this;

Additional Fault Symptom:

- General Checksum Failure

Does this indicate that I still have the injector correction values wrong for that one injector with the dodgy printing?


In my ignorance I assumed that all I needed to do was re-run the Pilot learn procedure so I tried this and it failed, with an error code indicting that the fuel rail pressure was either too low or too high...shame it can't be a bit more specific. Out of interest I ran the following using the oscilloscope tool in Forscan:

FRP_DES & FRP - 14-9-2021.JPG

This was run at tickover with the engine well up to temperature and seems to show that the fuel rail pressure is in excess of the desired fuel rail pressure. However, is the Pilot learn failing because it is too high though?

You mentioned that I need to re-run the pump learn as well. Is that a pre-requisite to being able to run the pilot/injector learn successfully?

Also...and excuse my ignorance here...is the pilot/injector learn the same procedure?

Lastly, since the pilot learn failed I now have this very unspecific message when I turn on the ignition;

Engine Malfunction 14-09-2021.jpg

The message remains when the engine in running but does not seem to have any impact on anything. I took the van for a short runaround after it was all up to temperature and ran sweet, pulled fine with no problems.

Still not very happy starting though, but is that a function of the pilot learn not having completed sucessfully?

Thanks again for your time on this. It is very much appreciated. I'm learning a lot but still have a lot to learn clearly.

Regards

[knobby1]

Reconditioned injectors "should" come with "new" correction factors as they will be different to the original numbers. Were they supplied with new numbers or did you use the old ones..??

Lord Knobrot

[Cavesurfer]

Reconditioned injectors "should" come with "new" correction factors as they will be different to the original numbers. Were they supplied with new numbers or did you use the old ones..??

Lord Knobrot

Yes, the reconditioned injectors were supplied with a new set of correction values, but there is this potential issue with the correction values for just one of the injectors being difficult to read. The old values had been removed by the Denso by painting over them on the electrical connector, and the new values were stamped or engraved into this brass ring set into the body of the the injectors.

I should have taken a picture before fitting them to the van ideally, but didn't think of it at the time.

[knobby1]

Reconditioned injectors "should" come with "new" correction factors as they will be different to the original numbers. Were they supplied with new numbers or did you use the old ones..??

Lord Knobrot

Yes, the reconditioned injectors were supplied with a new set of correction values, but there is this potential issue with the correction values for just one of the injectors being difficult to read. The old values had been removed by the Denso by painting over them on the electrical connector, and the new values were stamped or engraved into this brass ring set into the body of the the injectors.

I should have taken a picture before fitting them to the van ideally, but didn't think of it at the time.

Strange they would "overwrite" the numbers on the injectors, normally with "recon" injectors they supply a separate tag or sheet of paper with the new correction factors on them. Just take a pic with your phone and "zoom in"...

Desired fuel rail pressure at idle is usually ~230bar. After the pump, pilot and injector learns are "completed successfully"...the desired and actual FRP should be reasonably close to each other.

Lord Knobrot

[Cavesurfer]

Just take a pic with your phone and "zoom in"...

Desired fuel rail pressure at idle is usually ~230bar. After the pump, pilot and injector learns are "completed successfully"...the desired and actual FRP should be reasonably close to each other.

Lord Knobrot

Okay, so I'll try and get a photograph of the brass ring tomorrow and then hopefully you'll see what I mean about the difficulty with reading the correction values for this one injector. The other three injectors were fine as the alignment of the letters was better. It occurred to me I should get in touch with TPUK or Denso and ask them for clarification or a replacement.

So do you recommend that i run the pump learn again with the new injectors fitted?

The top value for the desired fuel pressure is around 239 bar, but the sensor is picking up 300 to 330bar which is quite a bit more. Would this be down to the pump learn needing to be run again do you think?

If so, this would explain why the pilot learn failed with the message the pressure too low/too high.

[knobby1]

Just take a pic with your phone and "zoom in"...

Desired fuel rail pressure at idle is usually ~230bar. After the pump, pilot and injector learns are "completed successfully"...the desired and actual FRP should be reasonably close to each other.

Lord Knobrot

Okay, so I'll try and get a photograph of the brass ring tomorrow and then hopefully you'll see what I mean about the difficulty with reading the correction values for this one injector. The other three injectors were fine as the alignment of the letters was better. It occurred to me I should get in touch with TPUK or Denso and ask them for clarification or a replacement.

So do you recommend that i run the pump learn again with the new injectors fitted?

The top value for the desired fuel pressure is around 239 bar, but the sensor is picking up 300 to 330bar which is quite a bit more. Would this be down to the pump learn needing to be run again do you think?

If so, this would explain why the pilot learn failed with the message the pressure too low/too high.

Yes...best to rerun the pump learn as it may have changed with the installation of the new injectors. Some people don't advise doing the pump learn unless the pump itself has been replaced...but the pumps, injectors and valves etc.. do wear over time and IMHO will need "recalibration" to keep all the calibration and correction numbers in the ball-park.

Lord Knobrot

[Cavesurfer]

Yes...best to rerun the pump learn as it may have changed with the installation of the new injectors. Some people don't advise doing the pump learn unless the pump itself has been replaced...but the pumps, injectors and valves etc.. do wear over time and IMHO will need "recalibration" to keep all the calibration and correction numbers in the ball-park.

Lord Knobrot

Okay, so success!....sort of...

Pump Learn and Pilot/Injector learns completed successfully and I now have no DTC codes reported, pump pressure is right where it needs to be and the engine is running nicely.

However, it's still a beggar to start. Prior to all this it would fire up as soon as I turned the key, but now it's cranking for maybe 6 to 8 seconds before it slowly comes to life, and then sometimes it doesn't start at all.

I found this though which I think maybe a clue...

Pump feed.PNG

I'm assuming this is the fuel feed from the filter to the pump. When the engine was running there was a steady stream of bubbles heading from the filter down this pipe to the pump.

I've read elsewhere that this can be a problem with the filter housing or the return pipes from the injectors or the fuel rail. I replaced the return pipe that connects the injectors and the fuel rail, but i see that this in turn connects to the filter housing via another pipe.

Do you have any suggestions about which if these I try and eliminate first? I'm guessing the pipe is cheaper than the filter housing so maybe start with that?

Thanks

[Cavesurfer]

Yes...best to rerun the pump learn as it may have changed with the installation of the new injectors. Some people don't advise doing the pump learn unless the pump itself has been replaced...but the pumps, injectors and valves etc.. do wear over time and IMHO will need "recalibration" to keep all the calibration and correction numbers in the ball-park.

Lord Knobrot

I also spotted that this was disconnected...

Fuel Filter connector.jpg

I know this comes from the fuel filter housing, but I don't know what it connects to.

Any ideas?

Thanks

[knobby1]

I also spotted that this was disconnected...

Fuel Filter connector.jpg

I know this comes from the fuel filter housing, but I don't know what it connects to.

Any ideas?

Thanks

It's the "water in fuel" sensor wire.....goes on the bottom of the filter, (usually a yellow fitting).....somtimes the "WIF" fitting gets forgotten and thrown out with the old filter..... and the fitting which comes with the new filter stays there, nowhere to plug the wire to though.

Lord Knobrot

[knobby1]

Okay, so success!....sort of...

Pump Learn and Pilot/Injector learns completed successfully and I now have no DTC codes reported, pump pressure is right where it needs to be and the engine is running nicely.

However, it's still a beggar to start. Prior to all this it would fire up as soon as I turned the key, but now it's cranking for maybe 6 to 8 seconds before it slowly comes to life, and then sometimes it doesn't start at all.

I found this though which I think maybe a clue...

Pump feed.PNG

I'm assuming this is the fuel feed from the filter to the pump. When the engine was running there was a steady stream of bubbles heading from the filter down this pipe to the pump.

I've read elsewhere that this can be a problem with the filter housing or the return pipes from the injectors or the fuel rail. I replaced the return pipe that connects the injectors and the fuel rail, but i see that this in turn connects to the filter housing via another pipe.

Do you have any suggestions about which if these I try and eliminate first? I'm guessing the pipe is cheaper than the filter housing so maybe start with that?

Thanks

Filter housing itself can crack and let air in, low pressure/vacuum fuel lines on filter head, to/from tank and to/from pump have O-rings in them which can perish and let air in. This can cause hard starting especially after sitting for a while or overnight as it allows fuel to slowly siphon back to tank.....then takes a while to draw the fuel back up to the pump.

Lord Knobrot

[Cavesurfer]

The 100ml + leak-off on injector #4 will definitely be the cause of failing the learns. Injector coding is relatively easy, it's just the inputting of the 16 digit "correction factor" codes into Forscan or IDS. Then rerun the pump, pilot/injector learns and you should be golden.

Lord Knobrot

It's been a long time since my last set of questions as I've actually had to go and do some work and take a sabbatical from my residence under the bonnet of my Mk7.

So the latest state of play is that I've replaced just about everything that I think could lead to the starting problems. This now includes injectors, PRV, SCV, EGR, leak off pipes, main leak off pipe back to filter, Main supply pipe to pump, Filter body and the filter itself.

Unfortunately the beggar still won't start...it turns over very rapidly like it's firing but just not quite enough to keep itself going. The only way I can get the thing to start is with just a tiny squirt of Easystart. Then it's good to go.

All the learns completed successfully, the pump pressure is exactly where it needs to be, there are no error codes and no warning lights. Tickover is super smooth and I swear the engine runs quieter now than before I replaced all these parts. Moreover, it now pulls like a whole fleet of locomotives, runs super smooth and is returning a fairly comfortable 38/39 miles on a gallon of juice. There's no sign of smoke, even when under heavy load, no blow by into the crank case, no leaks anywhere and it drives like a dream.

I wondered briefly whether the fast turning over when I'm trying to start it was down to no compression. So I disconnected the fuel supply and it's very clear that the compression is fine as it turns over like you would expect it to with no fuel going into the cylinders, i.e. a lot slower.

So, there's clearly some fuel getting in there when I try to start the thing, just not enough to kick the beast into life.

So any ideas where else should I be looking?

I know there are several temperature and pressure sensors on the intake side that can give inaccurate readings when they get oily. Is this the next place to look?

Cheers

[knobby1]

Still sounds like air ingress somewhere.....but you've done your best to sort that..

Another thought, have you had a look at the cam and crank sensors??...If the crank sensor has crud on it from the DMF disintergrating it can cause issues. Also the cam chain tensioner or a stretched chain can cause grief if it's on it's last legs....timing can be out slightly until oil pressure comes up and pulls the chain tight, then you get cam/crank sync and she then starts. Check for cam/crank sync during cranking in Forscan.

Lord Knobrot

[Kandilx]

The 100ml + leak-off on injector #4 will definitely be the cause of failing the learns. Injector coding is relatively easy, it's just the inputting of the 16 digit "correction factor" codes into Forscan or IDS. Then rerun the pump, pilot/injector learns and you should be golden.

Lord Knobrot

It's been a long time since my last set of questions as I've actually had to go and do some work and take a sabbatical from my residence under the bonnet of my Mk7.

So the latest state of play is that I've replaced just about everything that I think could lead to the starting problems. This now includes injectors, PRV, SCV, EGR, leak off pipes, main leak off pipe back to filter, Main supply pipe to pump, Filter body and the filter itself.

Unfortunately the beggar still won't start...it turns over very rapidly like it's firing but just not quite enough to keep itself going. The only way I can get the thing to start is with just a tiny squirt of Easystart. Then it's good to go.

All the learns completed successfully, the pump pressure is exactly where it needs to be, there are no error codes and no warning lights. Tickover is super smooth and I swear the engine runs quieter now than before I replaced all these parts. Moreover, it now pulls like a whole fleet of locomotives, runs super smooth and is returning a fairly comfortable 38/39 miles on a gallon of juice. There's no sign of smoke, even when under heavy load, no blow by into the crank case, no leaks anywhere and it drives like a dream.

I wondered briefly whether the fast turning over when I'm trying to start it was down to no compression. So I disconnected the fuel supply and it's very clear that the compression is fine as it turns over like you would expect it to with no fuel going into the cylinders, i.e. a lot slower.

So, there's clearly some fuel getting in there when I try to start the thing, just not enough to kick the beast into life.

So any ideas where else should I be looking?

I know there are several temperature and pressure sensors on the intake side that can give inaccurate readings when they get oily. Is this the next place to look?

Cheers

Hi i am new here and i have seen this post i have almost the same issues with my citroen jumper with 2.2puma engine the car takes a couple of tries before finally starts, did you find the problem with yours?

[Funkycarldog]

Hi I'm having the same problem with mine sometimes it starts first time but then it could 5 or 6 times did you manage to fix your issue I have no fault codes