[Zirconocene]

Hi all-

I'm looking for some help after a failed repair this weekend. What follows is the background, what's been done, some questions about how things work technically, and a request for advice on how to proceed. Here goes.

I performed a clutch and associated parts replacement due to some of the standard slipping, banging into drive symptoms that indicate a drivetrain issue.

The throwout bearing was just about completely stripped, which is what was likely the cause of the bad shifting/drive engagement. Here are the new parts I replaced:

• New throwout bearing
• New clutch fork
• New clutch pivot thing (plastic, OEM BMW part)
• Refurbished clutch: sent the old one out to Kentucky Clutch and had them install Kevlar friction material

Work I did while the transmission was out:

• I swapped the bolts from the slave cylinder into the transmission as some insurance for later.
• I inspected the slave cylinder but didn't see any signs of leaking so it was not replaced.
• I replaced all the o rings on the solenoids
• I unpinned and resealed all the wires to the sensor strip, as they were cracking and copper was exposed

When I went to start the adaptation, it never completed and I ended up with a red cog and a constant chime. I was able to drive that way without any performance issue and the new clutch and components gave me fast, smooth shifts (after breaking in the new clutch gently).

It seemed like I needed to revisit the components that I thought were OK. As such, I recently did the following work:

- Rebuilt the slave cylinder (new seals, kit from SMG Society MLR)

After many, many rounds of bleeding the system, the adaptation failed again, so I moved on to the last component that I thought might be the issue:

- Replaced the PLCD sensor/harness

Again, the adaptation failed in INPA. I tried to use DIS, as well, but that did not work. I've never been able to understand how to use ISTA, so I was not able to try the adaptation with that software.

Here's the technical questions:

• Besides the seals in the slave cylinder, what else could wear out? The other important thing in there is the magnet ring that slides along the rod does that wear out? It is possible that it has slid out of position somehow?
• Is there any other way to test the PLCD itself? I'm not sure, now, that my old PLCD was faulty.
• The error seems to be that the zero point of the clutch position is out of range. Would this be caused by a lower general thickness of the friction material (which is what I observed with the Kevlar clutch material, as compared with the standard/remaining organic clutch material)

What's my next step?
Can I somehow trick the software into thinking that the clutch engagement point is in range, somehow?
Is it possible this is a result of my "fix" of the sensor strip wires?

Let me just say, the car has been driving great, just driving me crazy with the incessant chime.

Thanks

 

[Zirconocene]

@flacoramos : That's a bummer, if I need to buy a new clutch disc, but it was always going to be a bit of an experiment with the Kevlar material. I didn't realize that it would be so much thinner than standard. I'll start pricing things out.

Cheers

 

[Zirconocene]

OK, after some discussion with local car folks, some more questions.

• When the slave cylinder is not engaged, meaning when the clutch is against the pressure plate, is the rod in the slave cylinder compressed, or extended? Just as a thought experiment, which way would one want to change the slave cylinder rod to compensate for the thinner clutch material: shorten it, or extend it?
• If the idea about the thinner clutch material is correct, it implies that what the car is reading is a clutch disk that is too worn. Is that right?
• What is the minimum clutch thickness for the car? Meaning, if I can figure out what the thickness was of the material that was used to refurbish the clutch disk I can sort of verify this whole idea of that as the root cause.
• Finally, is it the PLCD that ultimately tells the car whether the clutch disk is too thin and is what is reporting the clutch zero point out of tolerance?

Many thanks

 

[Zirconocene]

Hi @platii , I was hoping you'd jump in.

I actually drove the car for about a year, over 5000 miles, with this issue, and it never went away. I kept hoping that it was air that needed to make its way through the system but that never seemed to work out.

The car is currently not driveable, as the driveshaft is disconnected and the exhaust is not attached (not that I couldn't drive the car without the exhaust, but I'm sure my neighbors would hate me for it).

I have a new error currently, I think as a result of my messing around in DIS, where the pump won't build pressure, or activate when I open the door. I believe that I need to clear the current errors then start the adaptation again to get the SMG to a place where I can get you the errors that come up. Currently, the error is 4F67. Prior to this round of work, the error was always something like "...(something) out of tolerance", aber in Deutsch.

I don't remember the thickness of the Kevlar material, just being very surprised at how thin it is. I am resolved to replace the clutch, again, and can measure everything once I get the old clutch disk out. This time, it's getting a proper Sachs disk.

For the 5-6 mm that you talk about for thickness, is that the total thickness of the plate? Assuming equal wear, what does that mean for the friction material? Or is that the minimum thickness of the friction material?

As always, many thanks

 

[Zirconocene]

@dbenkert : I should be able to pull those numbers; they're in the Digital and Analog Values part of INPA, right?

Here are the numbers I have from the last adaptation attempt, before I messed things up with DIS.

Zero current clutch value: 1050.0
Clutch engagement position: 650.0
Clutch bearing control adaptation value P0: 50
Clutch bearing control adaptation value P2: 100
Zero point of clutch position: (!) 111
Valve cover clutch: 20.0
Auskuppelstelllung: (!) 100.0
Clutch bearing control adaptation value P1: 80
Clutch bearing control adaptation value P3: 100

I don't know if that helps; I can try to get those numbers if I've given you a nonsense response.

Thanks

 

[Zirconocene]

Yes the total plate thickness, about 2.5mm on each friction side.

Was this kevlar clutch used since a year ago?.

Correct, this is the clutch I have been using for the past year. The driveability was great, but it was after some really terrible times with the degraded release bearing, so everything seemed like a huge improvement.

Cheers

 

[Zirconocene]

Here is a picture I took of the Analog 3 values, which is what I think was requested.

The car currently has a 4f67 error and the pump won't prime, so I'm hoping that disconnecting the battery will help clear that and I can get any other information folks might need to help me out.

If this isn't the right screen, please let me know and I'll get the right one.

Thanks

 

[Zirconocene]

@Alan Arnesen : I thought I knew how to activate the pump in INPA, but I can't seem to find it. Can you help a brother out?

@platii : I'll see what I can do to get that data. I went ahead and ordered a new clutch disc, so I'm a little on the fence about just trying to pull the transmission today, to be ready to install the new clutch next weekend. I have other projects that are acting up, so it might be a bit of a decision.

Thanks all, I really appreciate the help. I love this car, and I love this transmission, but I definitely don't love working on this transmission!

Cheers

 

[Zirconocene]

OK, quick update. I ended up removing the transmission (Again. If I'm not careful, I might get good at this) and discovered the my rebuild of the slave cylinder had failed. The internal lock ring that holds the boot down had come out, which may have been playing a role in some of the errors.

At any rate, I have now ordered a new, standard clutch disk, as well as a new E46 slave cylinder (I can't help myself from experimenting!), and I'll update this thread once I get everything installed. With the bolts swapped from the E60 slave cylinder, if the SMG2 slave cylinder works, it's half the price or less than the E60 version. Which is somewhat suspect in and of itself, admittedly.

Cheers