'FSI Fuel Rail Pressure' - Desperate
- Thread starter Jeremy
- Start date
- Status
No, Media screenshots .lbl. - Andy
Thank you for trying the download, not working is unexpected and something that needs sorting out. I have full VCDS which leaves me ignorant of the VCDS Lite file structure and somewhat unable to assist. Mac has registered Lite, best if I leave it to Mac to sort out.
Are you quite certain that you changed the file type? (If any help I find it easier to to do this with right click and change using Properties).
Are you certain you placed the lbl file in the VCDS Labels subfolder?
Grrrr computers.
Andy
StevieG
Member
Yes I did both of those things. But will double check
PlasticMac
Member
I'll copy the lbl from my working Lite, change the suffix, and PM it to you in a while.
Mac.
PlasticMac
Member
StevieG
Member
We took Jeremy's FSI for a drive along some twisty and undulating Cotswold lanes last night, logging measuring blocks 000 and 140 (using the .lbl file kindly supplied by Mac - no VCDS crashes with this version).
Initially the target and actual fuel pressures tracked quite well, but after a few minutes it was clear that the actual fuel pressure reading diverged downwards when we encountered a hill or anything that put load on the engine.
You can see this in the graph here:
The grey line is the target pressure, the yellow the actual reported fuel pressure.
I then took the wheel. During the initial period, there were glimpses of the sweetness and punchiness of the FSI engine briefly on acceleration before the reported actual fuel pressure fell away and the the engine bogged down. Then the car became very difficult to drive; you can see that the reported fuel pressure dropped and stayed low (3.8 bar) for the last two thirds of the circuit. In fact, 20m from the end, we came to a halt, the engine struggled to idle and any attempt to set off resulted in a stall. After 10 minutes sat in the middle of the road, we resorted to pushing it to a safe place. Here is a log of the second circuit with target pressure in blue, and reported actual fuel pressure in orange.
After 30 minutes switched off, the vehicle was able to be re-started and driven back up the hill.
The reported constant fault is 17428 Fuel Pressure Regulation Control Range Exceeded P1020, but the test drive also threw up intermittent faults (16684 Random/Multiple Cylinder Misfire Detected P0300 and 16686 Cylinder 2 Misfire Detected P0302.
I'm attached the 2 .CSV files that are the basis for the above graphs (I've changed the file type to .txt to comply with the forum upload rules)
As always, any insights or observations welcome!
Stephen
Initially the target and actual fuel pressures tracked quite well, but after a few minutes it was clear that the actual fuel pressure reading diverged downwards when we encountered a hill or anything that put load on the engine.
You can see this in the graph here:
The grey line is the target pressure, the yellow the actual reported fuel pressure.
I then took the wheel. During the initial period, there were glimpses of the sweetness and punchiness of the FSI engine briefly on acceleration before the reported actual fuel pressure fell away and the the engine bogged down. Then the car became very difficult to drive; you can see that the reported fuel pressure dropped and stayed low (3.8 bar) for the last two thirds of the circuit. In fact, 20m from the end, we came to a halt, the engine struggled to idle and any attempt to set off resulted in a stall. After 10 minutes sat in the middle of the road, we resorted to pushing it to a safe place. Here is a log of the second circuit with target pressure in blue, and reported actual fuel pressure in orange.
After 30 minutes switched off, the vehicle was able to be re-started and driven back up the hill.
The reported constant fault is 17428 Fuel Pressure Regulation Control Range Exceeded P1020, but the test drive also threw up intermittent faults (16684 Random/Multiple Cylinder Misfire Detected P0300 and 16686 Cylinder 2 Misfire Detected P0302.
I'm attached the 2 .CSV files that are the basis for the above graphs (I've changed the file type to .txt to comply with the forum upload rules)
As always, any insights or observations welcome!
Stephen
birtyA2
A2OC Donor
Trying to be logical, but coming from a position of ignorance, I would say the misfires might be attributable to the fact that the fuel pressure is low which would suggest the mixture will be off. To my eyes this is fuel pressure related, the ecu is asking for a certain pressure which is often never achieved, hence the 17428, but may well cause the misfires. So logically it points to the pump, or I suppose a leak but I’m sure that would be very obvious.
J
audifan
A2OC Donor
The HPFP is unable to maintain pressure when hot or under load. Would say there is a problem with internal sealing. Suggest a new HPFP be fitted A.S.A.P.
This very informative thread explores the internals of the HPFP so you can see the number of seals that could age harden or be damaged by the wrong fuel as well as other areas that could be the issue...
This very informative thread explores the internals of the HPFP so you can see the number of seals that could age harden or be damaged by the wrong fuel as well as other areas that could be the issue...
Audi A2 1.6 FSI High Pressure Fuel pump
Some pictures for all you 1.6 FSI owners that have high pressure fuel pumps. Mine started to fail due to worn seals. Normally a very expensive fix but i managed to sort it for the cost of rubber o rings. I hope this helps people understand how to fix it their selves in the future and save...
www.a2oc.net
PlasticMac
Member
Bit late to respond, busy doing an FSI ECU swap, (SKCs and VCDS), for @drwellsie
The HPFP has to be the most likely failure.
If anyone has one to spare, then maybe the failed one could be investigated at leisure, to see what's failed, and if a refurb is a practical idea.
Great example of the use of OBD Diagnostics to fault find.
Mac.
The HPFP has to be the most likely failure.
If anyone has one to spare, then maybe the failed one could be investigated at leisure, to see what's failed, and if a refurb is a practical idea.
Great example of the use of OBD Diagnostics to fault find.
Mac.
PlasticMac
Member
@StevieG Those graphs make interpretation of the data so easy, and, virtually at a glance.
Could you describe how you went from the VCDS Log file to those wonderfully explanatory graphics please?
No rush, as you've already gone "above and beyond" in your support for @Jeremy and, the A2OC spirt.
I'll tidy up the Lite versions of the FSI/BAD lbl file, and make it available for download in a few days.
Mac.
Could you describe how you went from the VCDS Log file to those wonderfully explanatory graphics please?
No rush, as you've already gone "above and beyond" in your support for @Jeremy and, the A2OC spirt.
I'll tidy up the Lite versions of the FSI/BAD lbl file, and make it available for download in a few days.
Mac.
PlasticMac
Member
After more thought about this, I don’t think the HPFP is the only suspect. A faulty Fuel Pressure Regulator, (FPRV), is also likely to give these symptoms.
The HPFP generates a high pressure at quite low revs, and the FPRV bleeds off that high pressure to reduce it to the pressure set by the ECU. This is done via a solenoid valve, driven by the ECU, with a varying pulse width voltage, depending on how much pressure bleed off is required. A partial failure of the FPRV valve would increase bleed off, lowering the pressure. The FPRV also has a safety pressure relief valve that opens if excess pressure is present. This too could fail. So, if an HPFP is not available, look for an FPRV.
Good luck.
Mac.
Good Evening All,
In two minds about posting this, it answers no questions but as I spent some time writing it decided to post (sorry).
I too have been thinking about the Regulator Valve, a bit more than a quick think with the aim of building hive knowledge.
Thank you Stephen(@StevieG) for attaching the raw logs of Jeremy's poorly FSI, the data is there just needs perusing. I have so far I have only looked at the first log when Jeremy was driving.
First I tidied up the log presentation, hiding columns, changing text ...... much happier for the elapsed time to be in minutes and seconds - I struggle with 1765 seconds for example! Most of the time spent relearning very basic Excel and like any of these highly sophisticated pieces of software only scratching the surface really. Time to stop rambling.
First, what is surprising is the Regulator appears to use only roughly a quarter of the range described by 'Quantity Valve Closing Angle' - I checked with my test drive log and the same, albeit up to third with mine, difference must be down to engines, drivers, test route...... but still only a small part
Secondly when the rail pressure is near balanced (Target and Actual are close) the sweet spot for closing angle is about 27%, you see many closing angles in the twenty something percent. I assume this corresponds with even driving whatever that is.
For a near balanced rail I expected tweaking the closing angle down if Actual is just a bit below Target, more shut I assume, to nudge the Actual up to Target and vica versa if Actual is a little above Target. But it does not seem to work like that??? True only trivial tweaking of closing angle but I cannot detect rhyme or reason, probably unseen load factors and bear in mind the lines in the log are only fractions of a second apart, the valve will take a small finite amount of time to adjust, action will not show instantly. Too complicated for me!
Both these points are illustrated by the first few seconds of the log, see for yourself
On 8 seconds the first event and I think understand it! The ECU closes the Regulator and jacks up the Target pressure sky high. Why? Look at the revs column, the ECU is panicking. Low rail pressure on 48, not surprising with the injectors sucking out all the needed fuel. Next line, revs fall , Regulator opened and set at 22, notice how the Actual pressure has shot up to 92 from the Regulator closure. Over the next few lines the status quo is reinstated.
No point really forensically examining the the next 600 lines but worth a look where Jeremy's car starts to misbehave. Things start to go wrong at 1 minute. After revs rising the Target has gone sky high, high 90's but the Actual creeps into 3 figures, something that looks wrong. I wonder if this is the cause of error code '...; control range exceeded'. Revs start to fall with high Target BUT Actual starts to fall, the ECU gets desperate shutting the Regulator but it makes no difference, the Actual loses all high pressure and bottoms out at low pressure pump 3.8 bar.
Doom carries on relentless
The rest is periods of normal but several 30+ seconds of continued Actual 3.8 bar with Regulator closed.
The regulator appears to be working normally (unless faulty and lying), after all it appears to take values in the 'standard' range.
Andy
In two minds about posting this, it answers no questions but as I spent some time writing it decided to post (sorry).
I too have been thinking about the Regulator Valve, a bit more than a quick think with the aim of building hive knowledge.
Thank you Stephen(@StevieG) for attaching the raw logs of Jeremy's poorly FSI, the data is there just needs perusing. I have so far I have only looked at the first log when Jeremy was driving.
First I tidied up the log presentation, hiding columns, changing text ...... much happier for the elapsed time to be in minutes and seconds - I struggle with 1765 seconds for example! Most of the time spent relearning very basic Excel and like any of these highly sophisticated pieces of software only scratching the surface really. Time to stop rambling.
First, what is surprising is the Regulator appears to use only roughly a quarter of the range described by 'Quantity Valve Closing Angle' - I checked with my test drive log and the same, albeit up to third with mine, difference must be down to engines, drivers, test route...... but still only a small part
Secondly when the rail pressure is near balanced (Target and Actual are close) the sweet spot for closing angle is about 27%, you see many closing angles in the twenty something percent. I assume this corresponds with even driving whatever that is.
For a near balanced rail I expected tweaking the closing angle down if Actual is just a bit below Target, more shut I assume, to nudge the Actual up to Target and vica versa if Actual is a little above Target. But it does not seem to work like that??? True only trivial tweaking of closing angle but I cannot detect rhyme or reason, probably unseen load factors and bear in mind the lines in the log are only fractions of a second apart, the valve will take a small finite amount of time to adjust, action will not show instantly. Too complicated for me!
Both these points are illustrated by the first few seconds of the log, see for yourself
On 8 seconds the first event and I think understand it! The ECU closes the Regulator and jacks up the Target pressure sky high. Why? Look at the revs column, the ECU is panicking. Low rail pressure on 48, not surprising with the injectors sucking out all the needed fuel. Next line, revs fall , Regulator opened and set at 22, notice how the Actual pressure has shot up to 92 from the Regulator closure. Over the next few lines the status quo is reinstated.
No point really forensically examining the the next 600 lines but worth a look where Jeremy's car starts to misbehave. Things start to go wrong at 1 minute. After revs rising the Target has gone sky high, high 90's but the Actual creeps into 3 figures, something that looks wrong. I wonder if this is the cause of error code '...; control range exceeded'. Revs start to fall with high Target BUT Actual starts to fall, the ECU gets desperate shutting the Regulator but it makes no difference, the Actual loses all high pressure and bottoms out at low pressure pump 3.8 bar.
Doom carries on relentless
The rest is periods of normal but several 30+ seconds of continued Actual 3.8 bar with Regulator closed.
The regulator appears to be working normally (unless faulty and lying), after all it appears to take values in the 'standard' range.
Andy
PlasticMac
Member
I think the Closing Angle data shows that the ECU is correctly driving the PRV in order to, (try to), deliver the set pressure.
The Actual Pressure shows that it doesn't always succeed. The symptom we see is sometimes the Actual Pressure is much lower than the Set Pressure. This could be the HPFP failing to generate the required pressure, or the PRV not closing enough, allowing pressure to bleed away.
From the logs, I'm not sure it's possible to say which it is.
The behaviour of the fuel pressure is erratic.
It's fine, (100+Bar), one moment, and then lost completely the next.
This doesn't feel, to me, like the failure of a high speed, three piston pump like the FSI's HPFP.
Mac.
The Actual Pressure shows that it doesn't always succeed. The symptom we see is sometimes the Actual Pressure is much lower than the Set Pressure. This could be the HPFP failing to generate the required pressure, or the PRV not closing enough, allowing pressure to bleed away.
From the logs, I'm not sure it's possible to say which it is.
The behaviour of the fuel pressure is erratic.
It's fine, (100+Bar), one moment, and then lost completely the next.
This doesn't feel, to me, like the failure of a high speed, three piston pump like the FSI's HPFP.
Mac.
PlasticMac
Member
PlasticMac
Member
I've been looking hard at this fault condition, bit sad, but gives me something to think about.
My conclusion is that it's the HPFP that's failed in the @Jeremy FSI.
Here's why:
The PRV, (Pressure Regulating Valve), is spring loaded to the closed, no bleed, position, where the full output of the HPFP, (High Pressure Fuel Pump), is applied to the Fuel Rail.
So the reference to "Closing Angle" is a bit odd, "Opening Value" would make more sense to me. The valve is controlled by a pulse width modulation, so the figures in the "Closing Angle" data block actually represents the pulse width, and, guessing here, an angle of 0 = no pulse, valve closed by spring, no pressure bleed, maximum rail pressure.
Looking at the data in the spreadsheet format from @Andrew what's very odd is that the Actual Fuel Rail Pressure never drops below 3.8 Bar, the output from the electric pump, in the tank.
If, as I previously thought, the low fuel pressure was being caused by the PRV not closing properly, and bleeding the rail pressure off, (via the return line), then the Actual Pressure would be zero, or very close to it, not 3.8 Bar.
So, it looks like the HPFP gradually loses pressure, and then delivers absolutely no pressure at all, as if it physically stops. After a while it starts again, then stops again.
This points very firmly to the HPFP as the culprit, but not O ring seals on the pistons, (as there are three separate pump assemblies, unlikely to all fail, and restart at the same time), but the shaft, cam, or lifting ring, (see pic below).
Although the HPFP is hard to come by, it is much easier to change than the PRV, which requires upper manifold removal.
Mac.
My conclusion is that it's the HPFP that's failed in the @Jeremy FSI.
Here's why:
The PRV, (Pressure Regulating Valve), is spring loaded to the closed, no bleed, position, where the full output of the HPFP, (High Pressure Fuel Pump), is applied to the Fuel Rail.
So the reference to "Closing Angle" is a bit odd, "Opening Value" would make more sense to me. The valve is controlled by a pulse width modulation, so the figures in the "Closing Angle" data block actually represents the pulse width, and, guessing here, an angle of 0 = no pulse, valve closed by spring, no pressure bleed, maximum rail pressure.
Looking at the data in the spreadsheet format from @Andrew what's very odd is that the Actual Fuel Rail Pressure never drops below 3.8 Bar, the output from the electric pump, in the tank.
If, as I previously thought, the low fuel pressure was being caused by the PRV not closing properly, and bleeding the rail pressure off, (via the return line), then the Actual Pressure would be zero, or very close to it, not 3.8 Bar.
So, it looks like the HPFP gradually loses pressure, and then delivers absolutely no pressure at all, as if it physically stops. After a while it starts again, then stops again.
This points very firmly to the HPFP as the culprit, but not O ring seals on the pistons, (as there are three separate pump assemblies, unlikely to all fail, and restart at the same time), but the shaft, cam, or lifting ring, (see pic below).
Although the HPFP is hard to come by, it is much easier to change than the PRV, which requires upper manifold removal.
Mac.
Last edited:
StevieG
Member
Thank you for the analysis and effort that has gone into this, Mac. In the absence of an available HPFP, we had been considering (reluctantly) looking to fit new O rings to the existing pump. However, this analysis does seem to suggest that would be futile. Jeremy is going to have to take some difficult decisions soon as it's his only vehicle and he lives out in the sticks...
PlasticMac
Member
If possible, taking the pump off the engine, would allow the internals to be the examined, and, just possibly, fixed.
Here's a how to:
Post in thread 'audi a2 1.6 fsi (bad) remove / refit instructions needed for EGR PIPE replacement diy' https://www.a2oc.net/community/inde...or-egr-pipe-replacement-diy.18799/post-142794
Mac.
Gentlemen, good evening!
I have just returned from my MOT which my little Fizzie failed. Predictably, it has a number of age related issues. I have had a good think and I have now decided, especially in light of the mysterious engine issue, it really is becoming a burden I no longer wish to carry. So, I will offer this car to any interested parties, breakers, fans or anyone who wishes to take on a project. I hope it goes to a good home, but if not it will be scrapped.
Images to follow.
Thanks to each and everyone of you for reaching out. I really appreciate the input. Such a shame.
Jeremy
I have just returned from my MOT which my little Fizzie failed. Predictably, it has a number of age related issues. I have had a good think and I have now decided, especially in light of the mysterious engine issue, it really is becoming a burden I no longer wish to carry. So, I will offer this car to any interested parties, breakers, fans or anyone who wishes to take on a project. I hope it goes to a good home, but if not it will be scrapped.
Images to follow.
Thanks to each and everyone of you for reaching out. I really appreciate the input. Such a shame.
Jeremy
- Status