Volkswagen Eos Forum banner

41 - 60 of 104 Posts

·
Registered
Joined
·
5,994 Posts
That's excellent. I can now see that making the complete part isn't viable because it includes the gear. Even though your Mole grips roughed it up a bit, it shows the metal is poor and I'm not sure I want to even get a tap down it or use my TIG welder. My next idea is to use an Allen bolt with a hex head, drill out their hole leaving as much metal as possible, insert the plain section of bolt as far as possible set in with industrial super glue then side pin with a 2mm pin.

thought about it I think that’s why people get issues when manually cranking and needing the motor readaption. As the motor position and roof position are no longer in sync.
It depends how they do it. I think if the motor controller has power when you use the manual crank, it should keep updating the revolution count to the new position which is o.k. I'll check tomorrow if the motor still remains powered with ignition off. But if there is no power or the battery is flat, or the motor is cranked or turned after removing it from the Bowden cable it will be out of sync. If it's out of sync. by just a small amount, I don't know if the re-adaptation procedure could work without dismantling anything? I like looking at the gear whilst operating the switch, because it confirms the switch works, there is control and the motor runs in both directions.

The other issue screwing things up is when the roof controller is left stuck in some intermediate position usually showing the dash confirmation but won't go anywhere due to a roof fault. Whatever you do to the sunroof motor it still won't hand back to manual control. In fact, before I got my roof and controller back to a fully closed position without errors, I'd done the sunroof reset, replaced the motor and tried the pull up switch. The sunroof came right back as 1 touch. But when I pulled up the switch again it went fully closed(?) on its own as for 1 touch then bounced back on its own to part open as it hit the pinch protection. That's when the penny dropped that the sunroof never normally does a 1 touch on the switch, only during a roof op., when activated to close by the rain sensor, or key in lock windows reset.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #42
Yeah remaking the shaft will be tricky. Maybe it’s designed like that so it’ll break first, in the instance someone tries to manually open or close a catastrophically failing sunroof, preventing anymore serious damage. As it could easily have been made more robust with a smaller channel to key the spacer bushing.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #43
Ok so I received a replacement ecu today.
22272


It’s in much better condition than the water damaged one from the car. I plugged it in (under the impression only new units need coding). But I thought I’d try to see if the boot lock mechanism works first before closing the boot itself. The pull down works when latch is locked. However the boot release off the badge or key fob does not. I also got roof error message on mfd.

So I assume I’ll need to recode this. Has anyone replaced a roof ecu before. And know what exactly needs doing to adapt a replacement ecu?
 

·
Registered
Joined
·
5,994 Posts
MY07 roof module reads back as 1Q0 959 255 B and since your 'A' version is from an 06, they are probably similar? My long coding is 0000693, but there's no guarantee yours will be the same.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #45
Yeah I’ll have to get them on vcds and see if there’s any discrepancies. Between the coding.
thanks
 

·
Registered
Joined
·
5,994 Posts
I decided as all my headlining was dropped and the motor reset was no longer a pain, I would do something about the manual turning. I can now see the big hole that most will fall into: When you are working blind to insert a hex key and pick up the motor shaft it can be a fiddle. But most important, THE SHAFT IS LOCKED SOLID TO THE WORM WHEN IT ISN'T PRESSED IN! Trying to turn the key to wind the sunroof will strip the hex until the key just turns around and does nothing.

That motor part and gears is really high precision and fixing it like I have done can easily go wrong! If it does, the motor must be replaced. After checking the depth of my stripped hex shaft I used an M5 plug tap (square end) to cut an internal thread. I'd already measured the free space between the cover and the motor and needed an M5 stainless hex screw to finish 12mm above their shaft. That height would put the screw level with the cover when fitted and make the hex much easier to see and locate with a key. The ABS panel hole is 8mm dia. which I might drill out to 8.5mm.

I used industrial strength Superglue to lock in the screw, then pinned it with some 1.5mm stainless welding wire. The standard hex screw cap size is about 8.5mm dia. and won't pass through the cover. I had to grind the head down a little to 8mm. Nearly finished, I then realised why there was a groove down the side of their shaft - although it weakens it. It's there to slide on the nylon spacer! I used a 1mm angle grinder cutting disc to cut a crude groove in the head of the hex screw to allow the nylon spacer to slide over. I'm just about done with the sunroof now and ready to get the headlining and side trims back.

22274


22275
 

·
Registered
Joined
·
5,994 Posts
Something I spotted when I was inside the motor gearbox: The large white nylon gear is the one that often fails when a sunroof binds through lack of lube. Why is it made of plastic? Because sitting over it are what look like 2 Hall sensor ICs and embedded in the nylon gear must be metal or magnets, which is how they do the revolution counting or use them for the 'stops'. I hope they haven't created a special for the EOS and your Mini sunroof motor works o.k with the roof controller.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #49
I don’t think there are any sensors. The replacement gear I have is pure nylon. It’s designed like that so the gear fails rather than causing more damage to a non lubricated roof. It’s meant to be sacrificial. There’s someone on eBay who does sell a brass replacement, but as mentioned this part is designed to fail when it sees excessive resistance to prevent damage to the more expensive roof parts.
From reading a few posts from other forums it seems many have used mini motors successfully. The consensus seems to be the same that it’s the controller electronics that differ but all the motors are of similar spec.
 

·
Registered
Joined
·
5,994 Posts
I don’t think there are any sensors.
I should have taken a photo, but when you lift off the large white nylon gear, underneath across its radius they have cut a slot in the case sitting over a thin circuit board on which are mounted 2 chips. I should have looked for some ID on them, but if I ever need to go back, there I'll investigate further and wave my boy scout compass around looking for metal or magnets. V.W do a lot of sneaky things.

When you get your roof controller working can you confirm the mini motor works o.k for the roof op.?

What I can't understand is all the window motors have over current sensing pinch protection which we can easily test. So why doesn't pinch protection protect the nylon gearbox when the sunroof friction goes up? I did think about adding my own lower rated thermal fuse because my current measurements on the window motors showed they only took about 2 amps which would shoot up with friction. You think they would set the pinch protection just below the torque level at which the gearbox would fail?

The more I look the more I think their design is based on 'anything wrong increasing the torque or friction, fail the motor' to require investigation and an expensive repair? I think I'll have to go back to a simple car like a Renault 4 or Trabant!
 

·
Registered
Joined
·
33 Posts
Discussion Starter #51
That’s a very good point. it is puzzling because the sunroof does have pinch protection. But I think as you said this value is outside a safe torque setting for the nylon gear. If I’m correct the motor runs at a certain torque setting. It’ll sense resistance and stop and then try a higher torque setting to carry on the close. If it senses resistance again it’ll just stop and open back. I think it’s when you keep repeating this operation to attempt to close a stick roof is when the gear just fails.

I’ll be in the workshop today with the vcds hooked up. I’ll update how I get on. 🤞🏽
 

·
Registered
Joined
·
5,994 Posts
I think they have to balance the pinch protection working because if the sunroof got stiff it could be left stuck open. IMHO I and others would probably rather accept that than have the gearbox parts destroyed. That's their philosophy for the cab. roof - anything wrong lock it out and take to workshop.

But look at what the shop manual says:

The sliding sunroof motor is equipped with a force limitation (anti-trap) mechanism. If the sliding roof encounters an obstruction while closing, the sliding roof motor stops and moves the glass panel in the opposite direction.
♦ If the glass panel sunroof is now closed a second time the glass panel will close again but with higher force.
If the sunroof encounters an obstruction again while closing, it will open again automatically.
If the glass panel is now closed a third time via the switch, the glass panel moves but without closing force limitation.
♦ All bolts are microencapsulated and must always be renewed.


QED in bold could be why the gearboxes self destruct!!

What I don't know or have tested is whether pinch protection is active for the entire sunroof roof travel or only during the last few inches of closing? I suspect the latter, because when window regulators fail, they don't always stop on pinch, but chew up the winder drum and wire. But now I've read how their pinch protection works, self destruction of the sunroof and a failed roof op. seems to be built in. I think the same thing happens with the windows. I've always advised against operating the window switch the first time anything unusual is encountered, because the regulator may be repairable. But if my advice is ignored 3 times then you're stuffed!

Because of these thoughts I had considered only two possible solutions. Either hack their firmware which is virtually impossible, add the equivalent of custom pinch protection to the motor externally using peak currect detection, or select a resettable thermal fuse that allowed motor operation with normal friction, but stopped it when it became excessive. I realize if this happened during a roof op. the roof would stop mid cycle.

I could design & build either of these solutions but now I'm getting lazy, I will experiment with something like this:


There's plenty of room behind the headlining & near the motor or even at the 12V feed point on the roof controller plug. The maximum current for this module is 10A, but that could be increased with a mod. using a relay. I'm pretty sure this motor doesn't run an average current more than 4 Amps and the delay setting in the module could take care of any initial peak current surge for motor start or closing. I need to check this module can always power the motor keeping the controller memory alive, but there's a way around that.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #53
Yeah I think it checks near the end too.
so I’ve tried the new roof module, but having issues fully opening the roof. It looks like the sunroof is the issue. It stops the roof operation once it gets to fully open position on sunroof. I think it’s not going that extra bit. Do you know how the roof detects sunroof position?

22276


it stops here and brings up roof error on the dash.

I checked codes on the roof module nothing stored. Telling me more it’s something to do with the sunroof motor.

I tried the measuring blocks under security pass on vcds the sunroof operation aborts near the end of fully open. It doesn’t like something there. I’ve looked online and none of the hall sensors have anything to do with the sunroof position. So I’m guessing this is done by the motor/electronics.

the roof operation gets to this position and won’t go any further.
22277


this is the position it stops on measuring blocks.
22278


it’ll work up until the point where the front roof member is supposed to disengage from the a pillar.

looks like I’ll have to get a eos sunroof motor and see if that fixes the issue.
 

·
Registered
Joined
·
5,994 Posts
I didn't want to dampen your spirits for the Mini motor but I and others have dabbled with parts and some Chinese that don't work because the EOS parts and firmware are very specific. Some parts may work fine on Golfs & Passats, but the EOS has another control layer for its roof.

The sunroof has no position sensors. The sunroof motor initialises from the fully closed position and the motor or gear pulses are counted by the motor controller. Obviously the motor controller talking to the roof controller must be EOS specific and the motor must give exactly the right number of revolution pulses corresponding to distance travelled for the roof controller to tell it to stop when it is extended further back, so it is ready for the package to drop into the fish tanks. Just like the fully closed position, I think the extended further back is where the motor could detect a stall? If you have used the EOS motor controller and added a different motor, something must be different either in its gearing or the way in which revolution pulses are counted by the controller board. I rest my case that the motor has something in it or embedded in its gears which is used to detect revolutions equivalent to distance or even revolutions X2? Or the motor could be designed to work like a stepper motor getting impulses which they count. In these cases how the motor rotation translates to linear distance must marry with the controller supplying the pulses. You only see 3 wires to the sunroof unit, but the method they use will sit between the motor electronics and the motor. Sorry but I haven't dug that deep. There are at least 2 ways translate motor movement to linear distance. You either have sensors on the sunroof sliding parts like they use for the roof, or do it cheaper by counting the motor revs.

Even with everything EOS compatible and married to correct parts, you can still get odd things happening, but at least then you know there is some adaptation or calibration work to look into.

You've got a roof controller with unknown provenance, a sunroof motor you've hacked together with non-EOS parts and possibly a roof system with sensor faults? That's plenty to try to understand and get working! You can follow my check list for the roof sensors and make sure each sensor checks out. That then leaves you with the sunroof motor and the roof control module.

There is one thing about the sunroof. Normally it only moves when you hold your finger on the button and moves to the normal open position first. But when I was playing with my wrongly adapted motor I got the sunroof to open with just a momentary click of the manual switch. I can't now remember during a roof op. if the sunroof stops dead mid opening when you release the auto roof switch, or carries on regardless to complete a fully open or close? If it's the latter, then it must be using one touch during a roof operation and we know from the windows that has to be learned.

There are plenty of YouTube side shots for the EOS roof sequence but when I needed to remind myself where all parts went I found this overhead view vid. very handy:
 

·
Registered
Joined
·
5,994 Posts
You know I've had my roof controller and sunroof at the mo. giving every imaginable error, DTC's and strange behaviour? One advantage I have is I know my roof controller and sunroof are OE and good so I'm fighting the system 'adaptation' , normalization or just compatible and working correctly together as they call it.

I marked a roof side member with a strip of masking tape at the normal manual open position for the sunroof making it easier to check the draw back during roof op. I can confirm this happens if you try a roof op. and the sunroof isn't right back in the extended position:

1. All windows dropped o.k (but some may not if not initialized).
2. The sunroof opens to normal position, but doesn't extend rearwards ready for package stowing.
3. The roof locks open o.k and the trunk locks are on ready for swing out.
4. The trunk lid swings out o.k
5. The rear window lifts over the top rear panel.
6. But the roof side members DO NOT separate from the 'A' Pillars.

If I open the sunroof part way and a front window, close the doors and hold the remote lock button > 10 seconds, the window closes, but the sunroof does not. Note: I have my adaptation settings set for convenience closing on the remote and I have fitted the RLS. Therefore, I suspect that whilst the sunroof switch works, something is wrong in communication between the roof controller and the sunroof. I already said sunroof control is via the roof control module.

I don't know if this helps you and IMHO you sometimes have to try several things in different order before the system works as it should. You also have the added uncertainty of the provenance of key parts.

All these modules with some form of single wire control require a power connection and data control. You can disconnect the power source, but the data input line can still be present with voltage being fed from somewhere else. When power is re-applied with a dataline active, I think this is why the lockout and spurious communication errors can occur. The modules seem to need both power and data lines disconnected and removed to get clean initialization. In practice disconnecting the battery and reconnecting 1/2 hour later should be the right thing to do, but this often throws up random faults on other modules, a risk you may have to take.

This is what I did for the roof controller and sunroof motor:

1. I calibrated the sunroof according to my post.
2. I removed F29 50A fuse from the engine bay fuse box. This removes power from the roof controller AND sunroof motor.
3. I removed all the plugs from the roof controller to disable any voltages present on sensor or data lines.
4. I checked and cleaned the roof connectors with contact cleaner replaced them and Fuse F29.
5. I scanned, cleared any previous DTCs and scanned again until there were no roof errors (roof and sunroof both closed).
6. I operated the driver panel window switches to check if one touch was working on ALL windows (It wasn't).
7. I operated the sunroof taking it back to half open and left a front window lowered.
8. I closed all doors and the hood then inserted my key in the drivers door lock and held it in 'lock' for about 15 Sec.
9. All windows and the sunroof closed. I held the key there for 10 seconds.
10. I tested all windows would open in the opposite key direction (but not the sunroof).
11. I half opened the sunroof and a front window, closed the door and held the remote on 'lock' The open window AND sunroof both closed.
12. I part opened a window and the sunroof, closed the doors and poured water over my windshield RLS. After about 20 Seconds the open window and sunroof closed.
13. With the engine running & a fully charged battery, I started the auto roof open. Everything worked as I wrote earlier AND the roof side members started lifting away from the 'A' pillars. I haven't completed the cycle to stow yet because I've still got the headlining and side member trims loose.

When I'm working on the car in this state, I don't move any roof parts manually AND now always remove the rear windows fuse F32 in the dash fuseboard. Why? Because the RLS or auto close1 touch can see the rear windows open after a few minutes and will try to close them when the side flaps are open. Don't ask, but I can tell you the plastic hinge parts on these flaps can then get crunched, even though the window pinch protection has reacted to the obstruction. When I do have to move roof parts, I remove the 2 chrome hoops from the trunk lid, peg the lock to fully open and have my get out of jail mods to get into a locked trunk. Be careful because you are getting too close to an EOS locked trunk situation.

It's a nice day today and there's plenty of time with social distancing to work on EOS projects. Just think of the clean fresh air you will get when yours is fixed!

I hope you get creative and solve your problems. We look forward to reading a happy conclusion.
 

·
Registered
Joined
·
33 Posts
Discussion Starter #56
Hi vox

I might try the steps you’ve taken. To see if it will all sync.
I think it’s time for me to do the boot release mod too.

I’m certain this roof ecu is fine it has no communication errors like the previous one and also all the plug contacts are intact and not corroded.
It also has no dtc codes. Through measuring blocks it also shows communication with all the hall sensors, and they are all behaving as they should, bar the front two as the roof hasn’t gone that far in its cycle. I will get the multimeter out and follow your pdf for checking all the plugs. Just to be sure.

prior to doing any of this work I’ve dried out the boot checked the pump for any water damage and what I can see of the wiring and it all seems fine.

I’m sure my issue is sunroof related. When I bought the car that nor any roof movement would work. And the ecu was full of errors mostly communication due to the water damaged board.

I’ll get it working eventually just needs some more digging and adjusting.
but I’m happy with the progress so far at least I’ve got the roof open to lube more of the rubbers and prevent more water ingress. Considering I’ve only had the car a few weeks and done a few days work on it, it’s come a long way seeing as the roof hasn’t been operational since 2017 going by the receipts it came with when it last went in for diagnostics.
If I make anymore progress I’ll update you. Hope yours is going in the right direction too.
 

·
Registered
Joined
·
5,994 Posts
6. But the roof side members DO NOT separate from the 'A' Pillars.
Read carefully. That happens if the sunroof hasn't moved right back and there are no roof DTC errors!
Hope yours is going in the right direction too.
Thanks, I'm doing so many different things in different areas that aren't documented or haven't been attempted before. Last week I didn't think it would ever get back together, but with much patience, it's slowly coming back to life which is re-assuring.

IMHO I now think the biggest leak problem is coming from the fish tanks and sunroof seals. I've got mods for those - all bundled up with what I regard as 'condensation related', but I haven't got time yet to post up my notes and photos.
 

·
Registered
Joined
·
5,994 Posts
Since you've gone dark and had a water logged roof ECU, you may find my post update helpful. MY07 is now back to life from a state of wire bundles, in progress mods of all sorts, new drain hoses here and there and roof parts every way but loose. But my EOS roof now operates like a yo-yo without faults. Unfortunately, I'm socially distanced and running out of EOS fixes and mods to do.

 

·
Registered
Joined
·
33 Posts
Discussion Starter #59
Hi Vox,

will give it a read.
I haven’t really had a chance to do any work on the eos. Since the lockdown. My workshop is not near my home so it’ll have to wait unfortunately.
I’m pretty sure my issue is the sunroof not going back far enough. As I’m getting no dtc codes for the roof. And as I previously mentioned when monitoring binary output of hall sensors they all behaved in sequence as they should up until the roof stops operation.
 

·
Registered
Joined
·
5,994 Posts
I'm in socially isolated lockdown too but managed to get MY07 finished and back together all working o.k. I drove it top down for an 'essential journey' to get shopping with my phone and Google maps turned off! That was nice. With a lot of the roof wiring loom unwrapped for investigation and re-wiring, it reminded me of studio installation work. Everything to an outsider looked in complete chaos with wires of all colors, tools, roof parts, trim panels and tools everywhere.

I’m pretty sure my issue is the sunroof not going back far enough. As I’m getting no dtc codes for the roof. And as I previously mentioned when monitoring binary output of hall sensors they all behaved in sequence as they should up until the roof stops operation
The roof system doesn't have any sensors for the sunroof (no binary output) and I wouldn't expect it to throw up dtcs. The roof controller sends the extend back command to the sunroof, there's no data acknowledgement but I think the roof controller waits for the sunroof to pinch stall before proceeding. There are only 2 positions for it to pinch stall - fully closed and fully back extended. If the sunroof sits there in the normal open position, the motor hasn't stalled and the main roof op. will just sit there waiting for it, or time out with a watchdog error.

I suspect your sunroof controller/motor isn't seeing the EOS unique sunroof 'extend back' command sent from the roof controller, or it's adaptation behaviour is for a non-EOS sunroof? If you say you haven't replaced the sunroof motor controller (?) that leaves the adaptation procedure or something connected with the motor or gear you replaced? IMHO there is often more going on inside these units than you think. After following the adaptation procedure I wrote up and repeating it to get the motor to run each way out of the sunroof first without the motor stopping, mine came good after it learned the pinch points using the key in the driver door lock. Well, it learned the fully closed point and the intermediate sunroof open first. Then after the first roof op. I think it learned the EOS extended open position at the stall point.
 
41 - 60 of 104 Posts
Top