What's in the workshop this week, Pelikanblau edition - Page 12

Djinny · #**111** 20th April 2018, 09:43 PM

Ohh... mesmerizing perfection!

All those details and fixes!

I feel so relieved that those redunant wires have left the body of the blue one. They never really belonged there.

Thank you for repairing that lower dash. It's been a pain to see it like it was, but it looked somewhat beyond remedy.

Looking very much forward to new and unknown levels of perfection!

Davey Blueeyes · #**112** 20th April 2018, 09:44 PM

Oh MJ, this is just awesome! Thank you so much for this, lapping this up over a cuppa tonight!

Norretal · #**113** 20th April 2018, 09:47 PM

Quote:

Originally Posted by Djinny

Ohh... mesmerizing perfection!

All those details and fixes!

I feel so relieved that those redunant wires have left the body of the blue one. They never really belonged there.

Thank you for repairing that lower dash. It's been a pain to see it like it was, but it looked somewhat beyond remedy.

Looking very much forward to new and unknown levels of perfection!

Oh to be you very soon, Vera.

Not jealous at all

27litres · #**114** 21st April 2018, 03:45 AM

Seeings as the control module has been replaced, I'd be looking for a bad earth.
Was any of that butchered wire near the central locking module? Looked mostly under the dash... Any taps into fuse wiring?
When I bought my B5 it did not have remote entry. I got an audio shop to wire in an aftermarket remote entry kit.
He made a complete botch of the wiring and the car lost its soft-off interior lighting and one or two other niggles.
I undid his work and rewired it myself into the factory position (under back seat) and fixed all the problems that way

MikkiJayne · #**115** 21st April 2018, 10:29 AM

Quote:

Originally Posted by 27litres

Seeings as the control module has been replaced, I'd be looking for a bad earth.
Was any of that butchered wire near the central locking module? Looked mostly under the dash... Any taps into fuse wiring?

The door switches are being bypassed by the interior light bulbs, so door open = 0Ω, door closed = 6Ω. As far as the cluster is concerned, 6Ω is the same as 0 so it thinks the other doors are open when they're not. The challenge is finding the stimulus that causes this to happen as it is not consistent.

There were wire taps all over the place under the dash. None near the central locking module, but there is one in the relay box which I'm going to investigate today.

There was some aftermarket footwell lighting which would come on and off seemingly at random which used a Raspberry Pi-grade latching circuit and a foglight switch The Pi-grade board is not automotive rated and so didn't cope well with fluctuating voltages etc and I found with the engine not running it would turn on and off if you moved the passenger seat backrest

Not the driver's seat mind, or any of the other seat controls. Just the passenger seat backrest. No idea what weird effects that was having on the rest of the car's electronics, but it was also wired in with scotchlocks so its also gone.

moltuae · #**116** 21st April 2018, 01:19 PM

Quote:

Originally Posted by MikkiJayne

The door switches are being bypassed by the interior light bulbs, so door open = 0Ω, door closed = 6Ω. As far as the cluster is concerned, 6Ω is the same as 0 so it thinks the other doors are open when they're not. The challenge is finding the stimulus that causes this to happen as it is not consistent.

There were wire taps all over the place under the dash. None near the central locking module, but there is one in the relay box which I'm going to investigate today.

There was some aftermarket footwell lighting which would come on and off seemingly at random which used a Raspberry Pi-grade latching circuit and a foglight switch The Pi-grade board is not automotive rated and so didn't cope well with fluctuating voltages etc and I found with the engine not running it would turn on and off if you moved the passenger seat backrest

Not the driver's seat mind, or any of the other seat controls. Just the passenger seat backrest. No idea what weird effects that was having on the rest of the car's electronics, but it was also wired in with scotchlocks so its also gone.

Judging by that info and Vera's description of how the fault developed, it sounds to me like some after-market device has been tapped into the interior lighting circuit (to detect an open door) and that device has failed (or perhaps just lost power), causing the circuit to be pulled to earth at that junction.

My guess would be the 6Ω you are seeing is the combined resistance of the bulbs between the door switch and the unwanted path to earth. Are you getting 6Ω at every door switch? Any variation between them might give a clue as to where the fault lies.

MikkiJayne · #**117** 21st April 2018, 02:10 PM

Quote:

Originally Posted by moltuae

My guess would be the 6Ω you are seeing is the combined resistance of the bulbs between the door switch and the unwanted path to earth. Are you getting 6Ω at every door switch? Any variation between them might give a clue as to where the fault lies.

That's exactly what it is. Open any door (which itself will show near 0Ω) and all the others show 6Ω through the various bulbs in the doors and under the seats etc which confuses the cluster. The reason this is detectable is that I have the interior lighting relay out, so instead of turning on a light, the circuit escapes to ground through the bulbs. I've changed the relay as the original had a curious habit of turning off with the doors open (which it shouldn't).

I have a theory that the original rear left door switch chattering perturbed the relay somehow so when the door was fixed, the relay was then bad and the problem extended to all the doors. Vera pulled the interior light fuse to stop the light coming on, which ironically also causes the cluster to do the all-doors trick as above!

Good idea about the aftermarket stuff too. The random footwell light circuit was connected to the dash lighting harness. I'll have a check of the diagrams to see if the dash and interior lights are connected. Its entirely possible that relay board was causing some odd effects.

I've also found a scotchlock in a big ground wire in the relay box...

tc4332 · #**118** 21st April 2018, 03:15 PM

I like the theory that Mark and MJ have come up with, but could be a bugger to locate.
Back to the old fashioned way, cut the cct. in half, then in half again following the symptom.
Offhand, do either of you know what the resistance is on a working car?
To my mind only 6 ohms could be, like Mark said, dragging the voltage level down to a point that triggers the action.
Again it would take a lot of bulbs in parallel to drop the overall resitance down to 6 ohms.
Good luck with this one Micheala

MikkiJayne · #**119** 21st April 2018, 03:45 PM

Quote:

Originally Posted by tc4332

I like the theory that Mark and MJ have come up with, but could be a bugger to locate.
Back to the old fashioned way, cut the cct. in half, then in half again following the symptom.
Offhand, do either of you know what the resistance is on a working car?
To my mind only 6 ohms could be, like Mark said, dragging the voltage level down to a point that triggers the action.
Again it would take a lot of bulbs in parallel to drop the overall resitance down to 6 ohms.
Good luck with this one Micheala

Thats exactly what I've been doing, and it was that which led to my 'lightbulb moment'

All the door wiring except the window motors goes through a junction box at the bottom of the right A-pillar (on both RHD and LHD!). Splitting it there isolates all the electronics and all you're left with is switches and bulbs.

In this pic, with power applied by the relay current sinks through the door switch to ground (red line), lighting the bulbs and the cluster sees a clear 12V or 0V on the sense line.

With no power to the circuit and T10 disconnected, the door thats open measures 0Ω across its contacts as expected. However, if you measure at the blue junction on any of the other doors the current (from the multimeter) takes the path in blue through the interior bulbs to ground via the open door.

The cluster inputs obviously float high with no connection so 0Ω or 6Ω via bulbs are effectively the same thing to the logic circuit and it thinks all the doors are open.

The central locking motor controls the relay which powers all this stuff. I have discovered that when you disconnect from it in VCDS it shuts down, powers off the relay and the cluster thinks all the doors are open!

It has to be locked and unlocked again to 'reboot'. This alone has caused me to think its broken itself a couple of times until I spotted the pattern

I'm not touching the electronics today!

tc4332 · #**120** 21st April 2018, 06:13 PM

I can only wish you good hunting with this one MJ.
I understand where you are at, but very interesting.
I think you are correct thinking that this is down to some amateur messing with something not properly understood.
I'd love to get my teeth into this with you. But then you don't need an old fogey peering over your shoulder.

Compliment time.
I've been looking all my life for a good looking young lady with mechanical and electrical skills along with a good sense of humour, don't bring your wedding dress to the Annual Meet I'd be tempted.