2,5 tdi AKE running rough when warm.

[ametlib]

Have a kind of update.
Today I hooked up my laptop again, engine temp 91 degrees.
Measuring blocks 13 & 14. Readings right bank -2,3 , - 2,3 and -1,8.
Then I rew it, back to idle the readings settled at around -1.
5mins. of idle later the readings where still around -1 .
Then I stopped the engine, started again and the readings settled around -0,3.
5 mins later right bank was stll around -0,3 !? Will test out some more things and write an update.

[ametlib]

Ok, went outside again ( freezing cold here in Norway)
Engine temp. 85 degrees. This time the right bank readings where actually an positive experience.
+1.3 , +0,8 and +0,8.
Took her out for a short spin. Back again the readings where still positive, 0,3 , 0,3 , and 0,5.( engine temp 91 degrees)

Then a flip at the pedal, and suddenly the sun came up and the birds was singing, readings 0.08 , 0 , 0 , 0.08 , and 0.
Stayed like that for a couple of minutes until I turned of the engine.

This time I took several visits to basic setting, gr. 4. and fully retarded timing readings was +0,5 every time ( exactly where I put it)
So I guess this is not an belt/ tentioner issue.
If this is the way things are supposed to be, I would'nt spend much time looking at measuring blocks 13 & 14.
Can someone shed some light upon this ?

[ametlib]

No one wants to take a shot at this ?
I just can't see any part in the Injection pump ( maybe except the brain at the top?? ) doeing this.
If I find the time this weekend, I'll take a look at the cam belt and especially the hydraulic tentioner and
return with an update. Will appreciate any ideas ( good , bad, and anything between ) because I have actully never
seen this fault before. ( belt, rollers and tentioner is new, oem)

[ametlib]

Starting to get the feeling of making this thread just by myself .
Guess that's ok.
Checked the tentioner, all ok. Just to be sure I hooked up an old diesel timing strobe.
Regarding to G28/ flywheel position, the camshafts are running steady. ( Checked when the right bank readings were around -2,5 and around 0 )
Today I hooked up my vcds and check measuring blocks 13 & 14 during warm up.
And ( this is getting interesting) right bank readings where around -5 !
Engine sounded just like that too. Guess -5 means almost total starvation of the right bank.
In my book this leaves me with just two alternatives, ECU or quantity adjuster in the IP.
As I just hate to replace stuff that ain't broken, I could need some help here. Addams?

[PaulF2.5TDi]

It seems strange that the readings fluctuate so much. Is there any possibility of leakage from the return lines?

I found my IQ changed very slightly for the better when I sorted a tiny leak from the copper washer on my wifes A4 Avant. The car also lost some of it's harsh 'bark'. I didn't check 13 and 14 though.

As far as I can work out, the cylinder balance readings are worked out from the speed of rotation of the crank, using No 3 as reference. The fuel is adjusted +or- to speed up or slow down the engine for the relevant cylinders power stroke to give a smoother delivery. This could mean if you have an issue with number 3 the others will look odd - likewise if the sensor detecting the rotational speed is off.
The other issue could be crap in the injector(s) if you've had the pipes off a few times, it's easy to get some small foreign debris in there. This could also look intermittent depending on how the FOD sits in the nozzle and changes flow.

Bearing in mind how it works, -5 may suggest the other bank is faulty as the ECU is trying to slow it down by -5 to match..........

[ametlib]

It seems strange that the readings fluctuate so much. Is there any possibility of leakage from the return lines?

I found my IQ changed very slightly for the better when I sorted a tiny leak from the copper washer on my wifes A4 Avant. The car also lost some of it's harsh 'bark'. I didn't check 13 and 14 though.

As far as I can work out, the cylinder balance readings are worked out from the speed of rotation of the crank, using No 3 as reference. The fuel is adjusted +or- to speed up or slow down the engine for the relevant cylinders power stroke to give a smoother delivery. This could mean if you have an issue with number 3 the others will look odd - likewise if the sensor detecting the rotational speed is off.
The other issue could be crap in the injector(s) if you've had the pipes off a few times, it's easy to get some small foreign debris in there. This could also look intermittent depending on how the FOD sits in the nozzle and changes flow.

Bearing in mind how it works, -5 may suggest the other bank is faulty as the ECU is trying to slow it down by -5 to match..........

Thank's for answering!
I took off both return hoses at the pump, blinded the pumpconnections and just put the two rubber hoses in two bottles. Drowe around like that all day but nothing changed.
I don't think there is an issue with the needleliftinjector or any contamination in the injectors eighter, I have replaced all 6 injectors and nothing changed.
The flywheel on theese engines have 6 steelpins ( axially attached) to trigger the engine speed sensor .I have always believed that the ecu measured the time elapced between the pulses, then at the NEXT REVOLUTION adjusted fuel
to the respective injector.
Agree with you, it seems like the left bank " is faulty" and the ECU is trying to slow things down by starving the right bank. In that case the ECU balances out a faulty injector by adjust fuel to the next fireing cylinder.
Interesting..
What I really would have liked to know is witch ECU is doeing this "job" The main ECU or the small one at top of the pump ?

[PaulF2.5TDi]

Did you get any further with this one?

I would guess the main ECU holds the parameters for the pump ECU to do the work to as it is the main ECU that you reprogram for tuning. This would indicate these individual fuelling adjustments are done by the pump ECU to keep in line with the main set parameter - unfortunately I don't know for sure and there is little information on how these engines work at this level. It might be worth looking on a tuning/mapping forum as they may know the in depth workings better.

I would think that one bank off/out would not be the main ECU at fault as I would expect any main ECU fault to be across all cylinders. I would expect to see a fault on the return circuit, or the pump delivery system (mechanical or electrical) assuming your engine is mechanically sound.

[ametlib]

Did you get any further with this one?

I would guess the main ECU holds the parameters for the pump ECU to do the work to as it is the main ECU that you reprogram for tuning. This would indicate these individual fuelling adjustments are done by the pump ECU to keep in line with the main set parameter - unfortunately I don't know for sure and there is little information on how these engines work at this level. It might be worth looking on a tuning/mapping forum as they may know the in depth workings better.

I would think that one bank off/out would not be the main ECU at fault as I would expect any main ECU fault to be across all cylinders. I would expect to see a fault on the return circuit, or the pump delivery system (mechanical or electrical) assuming your engine is mechanically sound.

Exactly mine thoughts too.
I will just leave the main ECU for now, as I have actually not yet seen one faulty, and as you say, it would probably not make this fault.
Map sensor and G28 woudn't eighter make this fault ( Guess G28 could, but should have thrown a faultcode)
The rest is replaced/ checked out. Leaves me with just the I.P. (quantity adjuster ?)
Am keeping an eye at ebay for a Injection pump.Will post an update after I have replaced it.
Could take some time though, maybe not before the spring because its freezing cold in my garage theese days ( seems like
the global warming missed Norway, AGAIN.) Thank's for answering!

[Crasher]

I just want to clarify that you have checked the pumps dynamic timing, the Commencement Of Injection setting using the Basic Settings 004 screen, to make sure it is as close to 2 degrees ATDC as possible? This is not the pinned and plated static timing of the cams/crank/pump and you did do the static timing using a crank pin, head plates and pump pin didn't you?

[ametlib]

I just want to clarify that you have checked the pumps dynamic timing, the Commencement Of Injection setting using the Basic Settings 004 screen, to make sure it is as close to 2 degrees ATDC as possible? This is not the pinned and plated static timing of the cams/crank/pump and you did do the static timing using a crank pin, head plates and pump pin didn't you?

Yes everything is done " by the book " Have changed the dynamic timing from -2 to +0,5 degrees as this seems to take some of the "bark" out of the old lady.
Cam timing done with camplates, crank- and pump pins, checked twice after cam-replacement, is ok.