P2564 and VW wiring from hell

[ianfar]

So my 04 PD gets a MIL, P2564. I search this great site and finally get around to checking the wiring. Sure enough, the loom wore right through two of three wires at the connector to the Vane position sensor.

I get the new connector and the repair kit, solder, heat shrink, toss the loom, and plug her in.

I still get the code. Clean connections, use dielectric grease, no help there.

I try to take off the wiper arms to get at the ECU. A PULLER you say??? To remove wiper arms??? Only VW.... I make a puller and get to the ECU.

The centre wire, green is about 80 Ohms from connector to connector and roaming all over the place. The others are good.

All that to ask you this:

The Bentley shows a connector in the harness between the sensor and the ECU, anybody ever find this connector?

It would make the fix easier...

 

[ianfar]

I rechecked the wiring and all tests OK.

This leaves me with the sensor/actuator itself. I've never heard of anyone needing one. Any guru's have experience?

VW won't sell it, of course, but for a couple grand extra, they'll throw in a turbo and manifold!

Any insight would be appreciated.

Thanks

 

[MayorDJQ]

Have you used VagCom to reset the code? Most codes op a VW don't magically clear themselves.

If it turns out you do need the actuator, check with tdiparts.com, worldimpex.com, autohausaz.com. There are probably others that sell the actuator as well.

 

[DanG144]

You might try electrical contact cleaner on the position sensor.

I am not sure there is yet a source for the actuator and position sensor...except used?

Dan

 

[Dimitri16V]

mine failed @ 70K . Replaced with a used unit.
It is made by Tyco electronics

 

[ianfar]

Thanks for the replies. I have cleaned it, but with WD40. I then used dielectric grease. I should get some deoxit...

I will try those vendors, but was planning to go to the wreckers for a used one. It seems they don't fail often.

When you say clean the sensor, do you mean the contacts, or is there a way to clean the unit itself?

Yes, VAGCOM clears the code but it returns within a minute of starting.

Thanks

 

[DanG144]

The sensor has some moving parts, right? Squirt in there liberally.

Deoxit is good stuff, then Deoxit Gold.

Keep your belly pan or a skid plate in place.

Dan

 

[DanG144]

mine failed @ 70K . Replaced with a used unit.
It is made by Tyco electronics

Did it have any manufacturing part #s on it?

Dan

 

[ianfar]

According to VCDS, the vanes are moving to match requested boost. The comfirmation signal from the sensor isn't getting to the ECU. At least this is my understanding of it.

I think that the moving parts work fine, but it is the electronics that is at fault.

 

[DanG144]

Could you be specific at what you are looking at with VCDS?

For many PD's block 43 field 2 is turbo actuator position feedback. Most other fields for the turbo are N75 control signal.

Dan

 

[ianfar]

Yeah, Dan, that is the block I looked at. When at idle or with the engine off, it read the same ~95 if I remember right.

I don't seem to have any power loss and mileage is great Requested boost and MAP levels follow each other well.

I was able to move the actuator arm 1/2" with the motor off.

I don't have a mityvac to do the rest of the test I've read about, but it seems clear that the vacuum/N75 system is working fine.

There is just no signal getting to the ECU.

If you can think of any test for me to do I am game, but I am new to VCDS.

 

[DanG144]

With the engine off, key on, it should read about 0% (-7 to +10%) and at idle it should go to about 100% (90 to 110% or so) with a difference of about 95 to 105 between the two. (Block 43 field 2)

At least that has been my experience.

For some reason the displays vary a good bit from ECU to ECU, but on mine, I can do a BASIC SETTINGS test (block 13 I think it was) that would raise my idle to 1400 rpm, then display the position feedback in one field the idle speed in another, the N75 control signal in another, and MAP in the last field. You should see MAP change at least 30 units. Mine was more like 70 mbar change.

At least once I was able to save a log of the basic settings data; at other times I could not record a log during the basic settings test.

On some other ECUs the position feedback is not displayed.

 

[DanG144]

From everything you have said, you still have a bad wire, not a bad sensor.

At least if I understood your 80 ohms and varying all over the place comment to be from the ECU end of the conductor to the sensor end of the conductor.

 

[ianfar]

Well the last time I tested the wiring I got no grounds, shorts or open wires on any of the three wires. All showed 0 Ohms.

I will run the test you describe and let you know. It's been a week since I did it and I may be wrong on the numbers.

 

[ianfar]

Block 43 gave me a lambda of 90.4% running or not.

Basic settings group 13 gave me a error reading, probably because the ECU isn't getting a signal.

I looked at meas. block 11 again and requested boost follows actual MAP, and the EGR changes accordingly.

Do you or does anyone else know any resistance measurements for the sensor?

If anyone has a stock BEW turbo laying around and could take measurements, I'd be awful grateful!


Another thought I have is this: One of the wires to the sensor should be ground, but I get no continuity to ground on any wire. I may have to check this again as I only used a screw head mounting something to the fender as a ground reference. Going to the battery negative may be a better idea.

 

[Dimitri16V]

all three wires go to ECU, if I remember correctly. you can check the resistance of the sensor from the ECU connector, just find the corresponding pins.

 

[ianfar]

Yep, they all go to the ECU. Two of them also go to the EGR.

I was just thinking one must be a ground connection.

I could check the resistance of the sensor, but without knowing what it should be, how would I know if the readings I get are good?

Thanks for the help, guys. Anybody wanna crawl under there and measure the resistance of theirs???

 

[PDJetta]

Maybe the ground is through the turbocharger housing, so you may not have a ground wire.

Before replacing the turbo, try running new "test wires" to the propper ECU pins if you can. These will parallel the existing wiring and if it works, its a wiring issue.

If you end up needing a turbo, you can get the same new one for about $900 from vendor(s) on this list.

--Nate

 

[ianfar]

I had no intention of replacing the whole turbo, just the sensor. Just because VW won't sell it on its own, doesn't mean it cant be replaced by itself. The wrecker was my plan.

I had thought about running temp wires, but two of them tee off to the EGR circuit. Besides that, the wiring tests fine after replacing the connector and a section of wire.

What I'm searching for now is the resistance readings off a working sensor so I can prove it is the problem before buying a replacement from the source.

 

[ianfar]

Just a note for anyone searching this problem later.

Resistance values for a good sensor are:

1-2 4.3kohm
2-3 115kohm
1-3 110kohm

These values are for the shaft in the fully extended position. I didn't think to check fully compressed position.

My bad sensor had values in the megohms.

If you find one of these things and are going to change it out, don't move the locking nut position on the shaft. This way there will be no need to adjust it later. This is assuming a stock setup.

 

[DanG144]

Thank you, sir.

If I get any more data (and I might next week) I will report it here.

Dan

 

[DanG144]

I did not have the resistance readings in my notes, but I do have that VCDS seems to be able to read from -7% to +128%.

Most of the ones I have checked (more than a dozen) were very close to 0 and 100%.

You can read this in Measuring block 43 field 2 for most PD's.

 

[ianfar]

I gladly share what little I know. I have used other peoples' experiences to help me with my car, and I am happy to add to the wealth of info on this site.

I did my first timing belt ever due to the info on this board. I'll let you know in about 80Kkm if it was a good idea or not!

Ian