Volkswagen Eos Forum banner

VW EOS Roof fault

28K views 16 replies 7 participants last post by  voxmagna 
#1 · (Edited)
Hi Guys,

I have had my eos for over a year now with no problems with the roof.
I open and close everyday.
I had the roof open and when I got to work and tried to close it. It got stuck with the side member only half why up.
I closed it manually as per manual.

Now it won't open.
It opens all windows, opens the boot, opens the sunroof, opens the c segment, opens the flaps, starts to open the comes away from the A pillars about 1" and stops.

I have had vag com on it which says 03289 - unknown error code 000
and 03057 shareware version to decode all DTC's 008

I have found 03057 as sensor c-segment unlocked
I have also found that sensor to be G562 (1Q0959121B)
Also when testing on vag com when reading roof bit codes g562 has a G instead of a 1 or 0
Has anyone had this same problem.?

Is it likely that the sensor has failed?
 
See less See more
#2 ·
You are going about your problem solving the right way.

I can only comment on what I have seen from my fully paid up version of vcds from Ross-Tech - I've never seen that message! However, I have seen Chinese clones with ripped off firmware doing strange things and wouldn't trust them. Most cheap ELM 327 based tools are pretty dumb and you would be unlikely to do damage using one. But vcds uniquely gets inside control units and writes back code which is scary if it's an untrusted clone.

I've watched the sensor bits change as my roof opens and closes and they do exactly what they say on the Ross-Tech video - although MY07 doesn't use bit 13. In fact, to make the fault finding easier I grabbed still frame sequences from their vid. and created a set of individual photos of the roof status.

'Unknown error code' or communication messages usually means a dongle interface problem (clone?) or there is an issue with the roof controller itself not responding to diagnostics requests through the comfort/convenience module.

It probably won't do anything now because a trouble code has been latched and hasn't been cleared? Clearing codes requires the genuine vcds dongle to write back to the controller module. The EOS roof system is unique and most cheap diagnostics tools and clones will not even see it and report faults. Unless you are using a genuine full paid for version of vcds, you are playing with fire!

I suspect your dongle is only partially working, sees a trouble code it doesn't know about and won't clear it. A paid for vcds ships with and updates 'label files' bespoke to the EOS roof system which explain each error code.

The windows and sunroof must be working error free for the roof to work. Operate them and check you are getting no errors. The sunroof has a unique operation used only for the roof. Normally on the sunroof manual switch it opens most of the way, but for a roof op. it moves back further. Close the roof and try clearing error codes (if you can trust your dongle!). If no error codes are reported at ignition on, you can assume the controller and sensors pass the first static self check and are o.k. If you then operate the roof and a sensor fault is latched, that suggests the sensor is not reacting to the position of the roof at that point in the cycle. The roof system uses Hall effect sensors and some of them are triggered by a small magnet fixed to a part of the hinge which then passes over it.

Remember the EOS roof system faults and locks out operation deliberately to avoid expensive damage!
.
 
#4 ·
Sorry for the necroposting. I see there's just one post from the OP so the chances of him seeing this are remote, but if he managed to fix his problem it would be great to know how he did it.

I'm getting exactly the same error on my '09. I checked the status of the sensors while opening the roof using OBDeleven, and found the same thing the OP did: the 5th bit, corresponding to G562, follows the sequence mentioned at RossTech (although there IS a slight difference in other bits... perhaps the sequence in the '11 they used is different?).
Anyway... everything is cool and dandy until the roof pillars start moving back. At that point bit 5 goes from 1 (roof locked up) to "G" (circuit open or shorted to ground). I think it has to do with the wire for G562 getting stretched or rubbing against something, but I have not found where or how.

Any hints on what to look for would be great.
Thanks!
 

Attachments

#5 ·
You seem to know what you are doing. Your attached roof sequence thumbnail is easier to understand than those I have because it adds the sequence timings. Hall sensors output a voltage (usually 5V) or zero or sometimes 0.7 volt. VW often use 0.7-5 volts for sensor outputs because the system can then check if the output low is valid (0.7V) or if it is zero there's a short or open circuit. Same with the output high. 5V is expected when a sensor changes but if it stays stuck at 5V all the time, then the sensor is bad or the ground wire return side of the sensor is broken.

I've attached my copy of the sensor wires as connected to the roof module controller behind the trunk left panel. You should be able to measure the sensor voltages directly on the connector and compare your result with the others. You said the sequence wasn't quite the same, but the important thing is the sensor state is actually changing. If you disconnect the plug you may also be able to compare the resistances of the sensors (test probes both ways) . You may get strange results because Hall effect devices output through a semiconductor, but the resistance comparisons should be similar.

When you get to the roof controller module you may find as I did that water has got on the top and corroded pins on the board and plugs. This issue can be the cause of unexplained random sensor faults because electronic Hall sensors are generally very reliable. Unplug each connector and clean the pins with switch cleaner and a new 1/2" paint brush. As for trapped wires, I can't see all the sensors on MY07 but those wires I can see are fixed solid to the roof hinge parts and shouldn't move unless flaps or something else gets trapped?
.
 

Attachments

#6 ·
Hi there!
Thank you for the pinouts. I'll take a look when/if I have some time this weekend. I think I managed to change the value of bit 5 back and forth by wiggling the wire bundle behind the left hinge in a certain way, but I couldn't reproduce it later on.

In the meantime: I found a VW technical tip ("TT 61-12-04 - 2007-2015 Eos – Convertible Top Inoperative With Faults 03053, 03055, and/or 03057 Stored"). Apparently sensors G558, G560 and G562 are spliced into a single wire somewhere near the left hinge (splice "M35"), and the wire(s) can break at the splice.
According to the TT, the splice is located near the left rear roof drain.

I'll try to take pictures when I "go in".

Cheers!
 
#8 ·
The photos and write up most useful to us will be those showing how you get into the hinge area.

Like many other car manufacturers, VW wiring looms and wire sizes are probably calculated by computer to use the least amount of copper as possible over a given length for the current. Sensor wires will not take much current and if you check at the far end connector you will find the wire guage pretty small as are the small crimp connector pins the wires fit.

This is always bad if there will be movement or wire splices. In places where there is movement or vibration, wires should have many more wire strands or be woven copper braid. The computer design algorithm doesn't always consider this. You might consider either because a repair could then give you a cheaper and more reliable solution than buying an expensive replacement VW loom part. I have used copper pcb 'solder wick' inside silicone sleeve to fix similar problems.

These problems also occur on wiring that runs into a hinged trunk lid and through doors.
 
#9 ·
Ok peeps. Today the planets lined up and the temperature was warm enough to let me dig into the roof harness (encouraged from reading how someone else here fixed his car last month). This is what I found:
  • 1st and foremost: you absolutely MUST take pictures of how things look BEFORE starting messing with the harness. If the harness is not routed correctly, bad things can (and most probably will) happen.
  • 2nd: the problem in my car was NOT where the three cables are spliced into one, as mentioned on the TSB. The sensor has a yellow wire, that is spliced with two red-and-white into another red-and-white wire The splice was not in the best shape so I decided to "fix it". Remember the "if it's not broken don't fix it"? I didn't, and I regretted it for hours - and most probably will regret it in the not so distant future. In any case: after soldering the wires together things looked fine (checking the bits using OBD-eleven), but as soon as I moved the wires the 5th bit went back to "G".
  • 3rd: there was a very hard kink on the 2nd wire from G562 (brown) right net to the 3-to-1 splice. I thought the wire was broken there, but it wasn't.
  • 4th: The problem in my car was with the brown wire. This wire is spliced into a white wire that goes into the harness that goes to the booth ECU. The splice in my car was located right next to the end of the woven wire protector, right next to where the drain from the roof is clamped to one of the mechanism bars (see arrow). This splice is located in a place where it's *much* easier to work (compared to the 3-to-1).

To summarize: if your roof stops working, use a tool like OBD-11 to see the status of the 2nd location on the roof ECU. If bit 5 (red arrow) shows problems -and before doing anything else- check to see if applying pressure to the wire harness near the location pointed by the arrow makes bit 5 to go back to the correct value. If so, your problem is the 1-to-1 splice.
If that doesn't work, try applying pressure to the harness close to the hinge below the drain connection to the roof gasket. If this makes bit 5 to go to a valid value, the problem is in the 3-to-1 splice. That splice has the wires for two other sensors: G558 and G560 (green arrows) so, if any of these two bits is showing problems as well, there's a good chance this splice is the problem.
After you identify the location, take pictures of the harness and surrounding bits from different angles, to make sure you know how things should end up when you are done. Go in, fix the problem, and go out.

PS: you may want to get a good quality electrical tape before you start messing with the harness...
 

Attachments

#10 ·
Ok peeps. Today the planets lined up and the temperature was warm enough to let me dig into the roof harness (encouraged from reading how someone else here fixed his car last month). This is what I found:
  • 1st and foremost: you absolutely MUST take pictures of how things look BEFORE starting messing with the harness. If the harness is not routed correctly, bad things can (and most probably will) happen.
  • 2nd: the problem in my car was NOT where the three cables are spliced into one, as mentioned on the TSB. The sensor has a yellow wire, that is spliced with two red-and-white into another red-and-white wire The splice was not in the best shape so I decided to "fix it". Remember the "if it's not broken don't fix it"? I didn't, and I regretted it for hours - and most probably will regret it in the not so distant future. In any case: after soldering the wires together things looked fine (checking the bits using OBD-eleven), but as soon as I moved the wires the 5th bit went back to "G".
  • 3rd: there was a very hard kink on the 2nd wire from G562 (brown) right net to the 3-to-1 splice. I thought the wire was broken there, but it wasn't.
  • 4th: The problem in my car was with the brown wire. This wire is spliced into a white wire that goes into the harness that goes to the booth ECU. The splice in my car was located right next to the end of the woven wire protector, right next to where the drain from the roof is clamped to one of the mechanism bars (see arrow). This splice is located in a place where it's much easier to work (compared to the 3-to-1).
To summarize: if your roof stops working, use a tool like OBD-11 to see the status of the 2nd location on the roof ECU. If bit 5 (red arrow) shows problems -and before doing anything else- check to see if applying pressure to the wire harness near the location pointed by the arrow makes bit 5 to go back to the correct value. If so, your problem is the 1-to-1 splice.
If that doesn't work, try applying pressure to the harness close to the hinge below the drain connection to the roof gasket. If this makes bit 5 to go to a valid value, the problem is in the 3-to-1 splice. That splice has the wires for two other sensors: G558 and G560 (green arrows) so, if any of these two bits is showing problems as well, there's a good chance this splice is the problem.
After you identify the location, take pictures of the harness and surrounding bits from different angles, to make sure you know how things should end up when you are done. Go in, fix the problem, and go out.

PS: you may want to get a good quality electrical tape before you start messing with the harness...
Hi Jay ,
Some great information , where are the splices located ? my roof does not move at all , will I be able to access the area to fix the splice ? cheers Gary
 
#11 ·
I spent 3 weeks on and off finding and rewiring every single splice and tap in MY07 EOS roof loom wiring. There are lots with some in hard to find places. I used plenty of solder, heatshrink sleeve, cloth tape and acrylic spray cable water proofer. I don't expect any more of these random failure events. :)

 
#13 ·
There is no such thing as 'easy' for this work: To get to ALL the splices I had roof parts in different part open positions needed to access the taps. All the trunk and hinge cover trims had to be removed, at times the rear glass was sitting on the top of the roof and at others the front section was pulled back. The trickiest taps are behind the headlining at the rear and high up on the hinges. Most of the taps and splices start on the left side and cross over to the right but there are taps for two front sensors behind the front 'A' pillar covers. There are plenty of caveats moving roof parts into positions and keeping them there and whilst the roof ECU and battery is disconnected great potential for 'trunk lockout' and roof parts getting out of sync.

There are two nasty 2 pin micro conectors hidden and taped partly under the head lining at the back and partly behind the rear side section of trim. I replaced both with hard wired solder splices and sleeves. They are there for replacing the front lock sensors which are supplied as OE parts on very long wires with these connectors on the end. But since wiring is taped and tied to the top side member internal parts making it uber hard to remove, you would chop off their long wire to about 200mm from a new sensor and solder splice and sleeve it on the wire pair left inside.

My car cover was keeping the rain off, but you should be able to see the approximate position of roof parts underneath. The sunroof was open with a sheet of plywood across the opening to stop the cover collapsing inwards. All roof locks were off. Both top side members were pushed back about 50mm to fit 2 custom wooden spacer blocks on the 'A' pillar alignment pegs. This spacing makes it easier to remove the top side member trim covers. Some wiring splices are in the loom behind this at the back.

22712
 
#15 ·
It's a workshop procedure. Have you thought about buying an online EOS workshop manual - search here for info? If you really want to go deep into these cars you should have one. It avoids asking basic questions or repeating answers with more questions to follow. Wiring diagrams are included in the manual. Your first question if you are on the starting grid and not in pole position should be how do I remove trim? Which is covered in the manual.
 
Top