Technical P2919-64 Gearbox control unit initialization failed

[Sparks1061]

So have finally worked out the right model and gearbox, so now have the following error message when I start the engine
P2919-64 Gearbox control unit initialization failed

No messages before I start the engine and I am able to run the various actuators with no error message.
Not sure how it can be a problem with the control unit when it clearly works with the engine not running and I can hear the actuators working,

Anyone have any ideas?

 

[Sparks1061]

And so back to this subject again, today finally managed to plug my OBD reader into a working Ducato with a Comfortmatic gearbox.
And only one major difference and that is the RPM of the clutch, on mine any time the engine is running it shows pretty much the same rpm as the engine, (regardless of the gear selected including neutral)
On the working one it shows zero which is what i would expect as the clutch plates shouldn't start rotating until you engage a gear and start to move.
So am I looking at a new clutch or a sensor problem? Clutch problems are that they normally won’t engage, not that they stay engaged, and if it was engaged then wouldn’t I hear it?
I guess it could be that the clutch has engaged and got stuck with the gearbox in neutral and that this is why I am getting an error message which is stopping me selecting any gear, but to be honest I don’t really have a clue.

 

[MSjoberg]

Earlier this year I discovered, to my surprise, that the clutch plate in fact is rotating, i.e. the clutch is pressed, also in Neutral on my 2018. I reported this in another forum and this behaviour was later confirmed by other owners after they checked their cars.
Actually it is also confirmed by Fiat in the x250 eLearn descriptions for bleeding the clutch circuit, in which they use the rotating clutch plate in Neutral as a checkpoint that there is no air left in the system.
With this new knowledge, what surprises me is that you have actually seen a car with non rotating clutch plate in Neutral…

 

[Sparks1061]

Earlier this year I discovered, to my surprise, that the clutch plate in fact is rotating, i.e. the clutch is pressed, also in Neutral on my 2018. I reported this in another forum and this behaviour was later confirmed by other owners after they checked their cars.
Actually it is also confirmed by Fiat in the x250 eLearn descriptions for bleeding the clutch circuit, in which they use the rotating clutch plate in Neutral as a checkpoint that there is no air left in the system.
With this new knowledge, what surprises me is that you have actually seen a car with non rotating clutch plate in Neutral…

Oh great I was hoping this would give me some clue as to what is wrong with my van and this just adds to the confusion

 

[MSjoberg]

Oh great I was hoping this would give me some clue as to what is wrong with my van and this just adds to the confusion

So sorry, my brain must have slipped and got me to mess it up in reply #3!

Naturally the surprise was, as written, that the clutch is depressed also in Neutral with the engine running. So the clutch plate rpm is zero then!

In my graph below the blue line is pressure plate position, grey is engine rpm, red is gear engaged and yellow is clutch plate rpm (different scales).

When engine is started (grey line rises from 0) in Neutral you can see that the pressure plate moves and after some regulation (Kiss Point detection, yellow line?) settles at a steady position. The position changes a little bit when gear 1 is engaged and more as the accelerator is pressed, clutch rpm increases and the car moves.

 

[Sparks1061]

So sorry, my brain must have slipped and got me to mess it up in reply #3!

Naturally the surprise was, as written, that the clutch is depressed also in Neutral with the engine running. So the clutch plate rpm is zero then!

In my graph below the blue line is pressure plate position, grey is engine rpm, red is gear engaged and yellow is clutch plate rpm (different scales).

When engine is started (grey line rises from 0) in Neutral you can see that the pressure plate moves and after some regulation (Kiss Point detection, yellow line?) settles at a steady position. The position changes a little bit when gear 1 is engaged and more as the accelerator is pressed, clutch rpm increases and the car moves.

View attachment 433845

That makes more sense, so either the Comfortmatic isn’t providing the correct inputs to the clutch unit, or I have a problem with the clutch hydraulics and there is not enough pressure to disengage the clutch, or the clutch is physically damaged and won’t disengage,
do those possibilities all sound reasonable?

 

[MSjoberg]

Yes, sounds reasonable.
If the conditions are right, the ECU will send the right current to the clutch solenoid valve EV0 that will open enough to make the media converter/clutch master cylinder provide enough pressure to the slave cylinder to disengage the clutch. And this movement will be verified indirectly by the clutch position sensor in the media converter. So there are a number of units plus pipes, connectors and cables that could cause problems here.

I found a Multiecuscan log file showing the current sent to EV0 for my car. When the clutch was fully engaged and "at rest", i.e. in Neutral with engine off or in 1st gear while driving, the current was 0 mA. During clutch maneuvers (and in Neutral with engine on) it rose to 1000-1100 mA (which is in the same range as the parameter Solenoid valve current at zero flow, but I don't know if this refers to EV0 or some other valve).
These values can of course be somewhat different for other cars than mine and perhaps also depend on clutch wear, but they might at least be useful for comparison.

EV0 obviously can be maneuvred manually in Multiecuscan, that might be useful in the search for the fault.

 

[Hank25]

And so back to this subject again, today finally managed to plug my OBD reader into a working Ducato with a Comfortmatic gearbox.
And only one major difference and that is the RPM of the clutch, on mine any time the engine is running it shows pretty much the same rpm as the engine, (regardless of the gear selected including neutral)
On the working one it shows zero which is what i would expect as the clutch plates shouldn't start rotating until you engage a gear and start to move.
So am I looking at a new clutch or a sensor problem? Clutch problems are that they normally won’t engage, not that they stay engaged, and if it was engaged then wouldn’t I hear it?
I guess it could be that the clutch has engaged and got stuck with the gearbox in neutral and that this is why I am getting an error message which is stopping me selecting any gear, but to be honest I don’t really have a clue.

A really common fault for ducato clutches is the plastic spring seating bits break off the clutch plate and drop down jamming between plate and flywheel.Sounds like your problem.You will need a new clutch plate and possibly flywheel as well.

 

[Sparks1061]

A really common fault for ducato clutches is the plastic spring seating bits break off the clutch plate and drop down jamming between plate and flywheel.Sounds like your problem.You will need a new clutch plate and possibly flywheel as well.

Possibly, I have read a lot about this problem, my understanding is that although both the manual and comfort matic vans have the same gearbox and clutch it only effects the Comfortmatic, because in the manual vans the plastic just drops into the bell housing, but the comfortmatic have so many sensors it causes a problem, I read somewhere that it is possible to reset the gear box to fix this on a temp basis but havent found any information on how to do it. If anyone knows how, it would be useful just so I can move the van into a better place for towing!
however a new clutch and flywheel is an expensive option and I need to be sure it is nothing else before I splash the cash, and I have a couple of issues with that diagnostic, 1. I have not seen this particular error message mentioned by anyone else who has had the problem, in almost all cases where the obd code has been read it has mentioned the thrust bearing. And 2. no one has mentioned the clutch rpm, although that could just be no one has bothered to make a note of it.
anyway still trying to find someone to do a clutch change so in the mean time will keep looking at other options as well.

 

[MSjoberg]

This thread shows one case with the ”plastic problem” and P2919-64

Most cases I have read about for the ”plastic problem” did not require a new flywheel, just the clutch kit. I have also seen that some car owners have refused the original LUK clutch as replacement and got a SACHS kit installed instead.

Fiat Ducato Fault Code P2919-64 ( Thrust Bearing ) | FIAT Ducato

Hi 2016 Fiat Ducato 2.3 ltr Comformatic Gearbox with a fault code P2919-64. Info is Thrust bearing. Take action to fix this sensor fault. I parked the camper up overnite after reversing it into position. No problem. Next day tried to move forward but camper would only show neutral. Checked and...

www.fiatforum.com

 

[Sparks1061]

This thread shows one case with the ”plastic problem” and P2919-64

Most cases I have read about for the ”plastic problem” did not require a new flywheel, just the clutch kit. I have also seen that some car owners have refused the original LUK clutch as replacement and got a SACHS kit installed instead.

Fiat Ducato Fault Code P2919-64 ( Thrust Bearing ) | FIAT Ducato

Hi 2016 Fiat Ducato 2.3 ltr Comformatic Gearbox with a fault code P2919-64. Info is Thrust bearing. Take action to fix this sensor fault. I parked the camper up overnite after reversing it into position. No problem. Next day tried to move forward but camper would only show neutral. Checked and...

www.fiatforum.com

Is it normal to get the same fault code but different text for the same fault? As I said I have the same fault code but the message is gearbox control unit initialization failed and not anything about the Thrust bearing

​

 

[Sparks1061]

Yes, sounds reasonable.
If the conditions are right, the ECU will send the right current to the clutch solenoid valve EV0 that will open enough to make the media converter/clutch master cylinder provide enough pressure to the slave cylinder to disengage the clutch. And this movement will be verified indirectly by the clutch position sensor in the media converter. So there are a number of units plus pipes, connectors and cables that could cause problems here.

I found a Multiecuscan log file showing the current sent to EV0 for my car. When the clutch was fully engaged and "at rest", i.e. in Neutral with engine off or in 1st gear while driving, the current was 0 mA. During clutch maneuvers (and in Neutral with engine on) it rose to 1000-1100 mA (which is in the same range as the parameter Solenoid valve current at zero flow, but I don't know if this refers to EV0 or some other valve).
These values can of course be somewhat different for other cars than mine and perhaps also depend on clutch wear, but they might at least be useful for comparison.

EV0 obviously can be maneuvred manually in Multiecuscan, that might be useful in the search for the fault.

Hi again, So I had a look at the figures and compared them to my mates known working van and got more strange figures,

Engine off: his van shows 1068.6000 mA for the "Solenoid valve current at zero flow" regardless of the gear being selected and 0mA for the "Clutch solenoid valve current"

Engine running: 1045.2000 mA for the "Solenoid valve current at zero flow" regardless of the gear being selected and 1021.8 mA when in neutral or 1st increasing to 1045.2 mA when selecting reverse or 2nd for the "Clutch solenoid valve current"

And then my heap of Sh*t:

Engine off: shows 1076.4 Ma for the "Solenoid valve current at zero flow" regardless of the gear selected, but while I get zero for the "Clutch solenoid valve current" when I select a new gear I get readings of 1279 mA falling back to zero, I am also getting readings in the Odd and Even gears selection sol. valve current of up to 882 depending on the gear selected.

Engine on: 1076.4 mA for the "Solenoid valve current at zero flow" regardless of the gear being selected and between 1037 and 1131 mA for the "Clutch solenoid valve current" depending on the gear selected with a clear change as the gear is selected. I also get a reading on the Odd and Even gears selection sol. valve current as I start the engine, but nothing after that, I guess because the fault code is stopping the robotic gearbox from attempting to select any gear.


Not sure what all this means, but my next step is to try replacing the clutch and comfortmatic fluids and make sure it is bled properly,

 

[MSjoberg]

Terve,
The valve current at zero flow seems to be a reference value, it does not update very often. So those values looks normal.

Clutch solenoid valve current is analog to how much you press the clutch pedal if you have a manual gearbox. So with engine off, it should be zero in Neutral and with any gear selected, in steady state. With engine on and car at standstill the robot presses the clutch so the readings for your mates car seems normal if the car was standing still.
The readings for yor car also seems reasonable (to me as amateur).
Another parameter to check could be the clutch stroke. Unfortunately there is no sensor showing the actual position/movement of the clutch pressure plate, the position is measured indirectly at the master cylinder. I assume that some hydraulic or mechanical faults could make this sensors values non-representative.


(Measured position values for my car:

		Motor running
Gear	Mode	Clutch pressure plate travel mm	Clutch pressure plate reference mm	Clutch engaged position (self-adjusted) mm	PIS spacer closed position (self-adjusted) mm
N	Standstill	9,95	9,96*	18,19	4,944
1	Standstill	11,41	11,40*	18,19	4,944
R	Standstill	11,40	11,40*	18,19	4,944
1-6	Driving	18,20	19,08	18,20	4,944
		Motor not running
N	Standstill	18,20
1	Standstill	18,20

*) Multiecuscan comment:: The value has no meaning if the clutch is not driven

Gear	Selection position mm	Gear/engage position mm
R	5,69	9,27
N	5,68	18,25
1	13,03	9,36
2	12,92	28,59
3	20,80	9,38
4	20,71	28,31
5	28,27	9,27
6	28,19	28,12

)

 

[Sparks1061]

So finally today braved the weather and changed and bled the fluids in both the actuator and the clutch, hasn’t made any difference, however a couple of things I noticed.
i used a compressed air bleed tool to drain the actuator, and had air in the line all the time, it could of been leaking where the rubber cap for the bleed nipple joins the line to the tool, but I couldn’t find or hear any leak there, I would have expected to get a solid flow of break fluid with no air bubbles? and I also seemed to put much more fluid in than I took out, about 250 ml in and only about 170 out, as I was sucking out the fluid and topping in up as I went I would expect similar volumes, unless I had very low fluid levels to start (but the filler tank was full to about 10mm below top,) so possibly I had an air lock,
however the pressure has always shown as OK.

Thoughts?

when it came to bleeding the clutch, I started the process on Multiecuscan and afterski a couple of second the program crashed, however I could hear something happening in the. Background and when I checked the fluid level had dropped so I assume the bleed worked ok, has anyone else come across this problem?

 

[MSjoberg]

What part of the system is it you refer to whith ”actuator”? The only bleed nipple I know of and have used is the one where the tube to the clutch enters the bell housing i.e. It bleeds the clutch. But you wrote ”actuator and the clutch”.
Which ones of the fluids had a lower level after the Multiecuscan clutch bleed? If it was the hydraulic oil (the bigger container) and the system was depressurized before the operation it is logical, since the pressure build-up in the accumulator ”consumes” oil from the container.
It would be logical for the clutch fluid part also, if there was air in it before. Which it shouldn’t have been…

 

[Sparks1061]

What part of the system is it you refer to whith ”actuator”? The only bleed nipple I know of and have used is the one where the tube to the clutch enters the bell housing i.e. It bleeds the clutch. But you wrote ”actuator and the clutch”.
Which ones of the fluids had a lower level after the Multiecuscan clutch bleed? If it was the hydraulic oil (the bigger container) and the system was depressurized before the operation it is logical, since the pressure build-up in the accumulator ”consumes” oil from the container. It would be logical for the clutch fluid part also, if there was air in it before. Which it shouldn’t have been…

Hi,
sorry yes getting my self confused , I thought the break fluid was for the robotics and the hydraulic oil was for the clutch, do I have that wrong way round?
the larger container with the hydraulic oil was the one that dropped in level when I ran the clutch drain process on multiecuscan, and yes it’s what I expected, even if multiecuscan crashed, (I could hear the pump running)
the smaller tank which was bled through the nipple and topped up as the level dropped was the one that seemed to have air in the line and that I put more in than I took out.

 

[MSjoberg]

OK.
Large container: hydraulic oil. For pump, pressure accumulator, solenoid valve unit, gear actuators and ”primary side” of clutch master cylinder.
Small container: brake fluid. For the ”secondary side” of clutch master cylinder, clutch pressure line and clutch slave cylinder.

 

[Sparks1061]

OK.
Large container: hydraulic oil. For pump, pressure accumulator, solenoid valve unit, gear actuators and ”primary side” of clutch master cylinder.
Small container: brake fluid. For the ”secondary side” of clutch master cylinder, clutch pressure line and clutch slave cylinder.

Thanks, that makes sense now,
what do you make of air in the brake fluid and being able to add more fluid than I took out? Might try and bleed it again and see if I can prove if the air showing in the line is simply a bad fitting seal on the bleed nipple, or something else?

 

[MSjoberg]

I have not seen any air entering at the clutch bleed nipple (but when bleeding wheel brake cylinders) but of course it is a possibility. The nipple normally doesn’t need to be opened much, less than one turn is enough for my car.

The difference in volume, filled minus extracted, seems quite strange considering the fill volume specified by Fiat, 50 ml. Not sure if this is the total clutch system volume or just the container, but strange anyway.
Speculation: If there is a leakage, e.g. faulty o-ring, in the clutch master cylinder, hydraulic fluid from the other circuit could bypass the piston, mix with the extracted clutch fluid and add to the volume. That would be a serious fault needing repair.

See fig 8 in this document: https://www.fiatforum.com/downloads/teaching-guide-m40-robotised-gearbox.425/

The link below leads to a good procedure description in German. Google usually translates quite well from German to English. Removing the headlight for easier access as he describes is something I haven’t tried yet. Maybe it is a good idea. I have just removed the top part of the air filter housing.

Wechsel der Hydraulikflüssigkeiten der Comfortmatic - Ducatoforum.de

Hallo, nach intensiver Recherche und auch diversen Diskussionen hier im Forum habe ich nun das Hydrauliköl selber gewechselt. Es gibt mehr oder weniger detaillierte Anleitungen in den Foren. Es gab aber keine zusammenhängende, die mir alle Informationen…

www.ducatoforum.de

 

[Sparks1061]

I have not seen any air entering at the clutch bleed nipple (but when bleeding wheel brake cylinders) but of course it is a possibility. The nipple normally doesn’t need to be opened much, less than one turn is enough for my car.

View attachment 434272

The difference in volume, filled minus extracted, seems quite strange considering the fill volume specified by Fiat, 50 ml. Not sure if this is the total clutch system volume or just the container, but strange anyway.
Speculation: If there is a leakage, e.g. faulty o-ring, in the clutch master cylinder, hydraulic fluid from the other circuit could bypass the piston, mix with the extracted clutch fluid and add to the volume. That would be a serious fault needing repair.

See fig 8 in this document: https://www.fiatforum.com/downloads/teaching-guide-m40-robotised-gearbox.425/

The link below leads to a good procedure description in German. Google usually translates quite well from German to English. Removing the headlight for easier access as he describes is something I haven’t tried yet. Maybe it is a good idea. I have just removed the top part of the air filter housing.

Wechsel der Hydraulikflüssigkeiten der Comfortmatic - Ducatoforum.de

Hallo, nach intensiver Recherche und auch diversen Diskussionen hier im Forum habe ich nun das Hydrauliköl selber gewechselt. Es gibt mehr oder weniger detaillierte Anleitungen in den Foren. Es gab aber keine zusammenhängende, die mir alle Informationen…

www.ducatoforum.de

Thanks, that guide from ducatoforum.de was what I used I think first step is to bleed the dot4 fluid again and see if I again get more out than I put in, and see if I can get rid of the air in the system, both issues would seem to suggest problems with the Master/Slave system, at which point I need to find a garage