always-on low-beam, no more fog lights (front & rear)

[RJVB]

Hi,

 

As related elsewhere I had an injector fail while driving back from holidays in Italy. It was then "reconditioned" yesterday by a workshop that happened to be close-by. After the initial "visit workshop" messages and limp-mode episode I think I already had an instant where the low-beam came on as soon as I turned the ignition switch. That sorted itself out, but the behaviour returned after the shop's (Bosch) diag tool had been connected.

 

When I mentioned it the mechanic considered it normal because apparently obligatory nowadays in Italy, so I presumed that something had been reprogrammed automatically. It was only 24h later that I noticed that I no longer have any fog lights at all - very annoying because the rear fog light is obligatory where I live, and the front fog lights are a real plus on unlit curvy country roads.

 

Any idea if this can indeed be related to the fact that an Italian diag tool was connected? I haven't checked my fuses, but I kind of expect that a blown fuse for a light gives the same kind of error on the dash as a blown bulb does?

I should add that the diag tool found 2 stored errors: one for the broken injector, and one for the stop lights contact switch. I could imagine that the light selector could break, but not really in this fashion.

 

Is there any kind of magic command combination I could try to get out of this set-up (like the combination to activate the DRLs)?

[xpower]

The fact it was in Italy should make no difference.

The BOSCH KTS (Diag tool) has the options to activate DRL as low beam etc, but this has to be run by the user - it's not automatic.

Your first start is back on a Diag tool to check for fault codes, and check the outputs from the switches are working.

After that it must be a settings issue on the BCM

(that's unless all your fog light bulbs have blown?????)

[RJVB]

It'd be really weird if all fog bulbs blew but none of the others, no? I thought about the switchgear too, but again that would mean that at least 2 switches blew at the same time. Maybe some animal ate through cabling, but that also seems far fetched.

There are 2 other errors on the print-out made just after I arrived at the workshop: 3FE1 and 3FE3. Putting that off to the stop light contact switch (linked to the brake pedal) must have been a communications error. They're marked "non cancellabe" so I don't even know how long they've been there. Power steering works fine.

As to the DRL reprog: isn't it possible that the tool put up a question to activate the appropriate country mode, which the mechanic accepted automatically because that's the usual thing to do for his clients?

[RJVB]

btw tje lights selector really doesn't do anything ATM; the dash indicators remain off too. Reprogramming of the low-beams combined with breakage of the front and front+rear switches seems a bit too much of a coincidence.

Unless the rear fog lights are cabled in series with the front?

[RJVB]

Yeah, I'd certainly hope that nothing is being wired in true series (but in practice it's impossible to activate the rear foglights without first activating the front foglight).

 

Evidently the Italian workshop claims that they "didn't touch all that", so it remains a mystery. I'll try to find a moment to pull one of the front foglight clusters and see if the bulb tells me anything. In the meantime, are there any reports of configuration loss or changes due to electric gremlins or the like?

[RJVB]

FWIW, both front fog bulb filaments look intact to the eye. As an experiment, I disconnected one and switched on the contact. When I then set the light selector to "low-beam + front foglights", the bulb warning light came on on the dash. Apparently you have to turn off the whole car to clear that state; the light didn't go out when I reconnected the bulb with the car switched on (I did that on purpose hoping it'd act as a sort of wake-up/reset for the control unit).

 

In other words:

1) the switchgear works

2) the system knows about the lights and will warn if a bulb is broken or disconnected. A bit counterintuitive to do that for lights you don't allow to be turned on, but well, Skoda is only "simply clever"

[RJVB]

So, I just spent 3h at the workshop trying to sort this out. And that's after we discovered yesterday that we "couldn't get in" with the diag tool they have there, so they asked someone with a VCDS to come over.

Sadly not someone trained in its use so it took a while to get around the "no reply from calculator" message we got on the PQ35. 

 

When we did get in, we could do just about everything except reprogram the damn thing. Tests of the bulbs all succeeded, so the PQ35 and everything downstream to the lights is supposedly OK. The only possible explanation for my issue is the non-erasable error that the rotational switch gives an "implausible state". So the continuous on of the low beams could be a security fallback, but if so

 

- why is there no error signalled to the user

- why do I get a bulb error when I disconnect a fog light and then try to switch them on

 

We tried disconnecting the switch completely and even connect one from an A3; no change (but the A3 selector doesn't have the "Auto" position so it could well not be recognised by the PQ35).

 

Next step will be a visit to an official dealership to make an appointment (and prepare myself for another very expensive repair) :-/

[xpower]

You could also visit a garage with a J2534 device, this has the dealers ODIS tool and can do anything the dealer can.

Costs 9euro an hour for them to access and use it (once they are all setup)

I used it the other day to correct an issue with my ABS module - an adaptation was required which VCDS wasn't able to do.

Just another option

Edited 23 August, 2016 by xpower

[RJVB]

If your car has a PQ35 BCM controller, the section for programming the lights cannot be accessed with VCDS. Byte 18 is hidden under a 00 mask, and needs a deeper acting tool such as VCP or odis-E  to access. Opening the byte 18 with VCDS will lead to all sorts of issues that cannot be corrected with VCDS alone. Take it to someone who know this, and who knows what they are doing.

 

 

Hmmm, nice to know ... 

We got into the PQ35 after letting the VCDS do an exhaustive scan of all available controllers and modules (which took at least 30min).

 

What kind of "all sorts of issues" are we talking about? And can the Bosch tool (ESI[tronic] V 10.1.1986) access this controller, and could its scan of recorded errors have led to the issue I'm dealing with now?

 

Scanning the Bosch printout again I now see that it also shows an error for the "Elettronica centr. ZE 3.3", "0x0708 interruttore luci segnale non plausibile". So that must be the error that the mechanic explained to me as having to do with the brake light switch ... which works fine. It's the same error as the one the VCDS showed me.

 

All of which suggests that something weird happened around the time I encountered the injector break-down. Where in the car does this PQ35 controller live? Could this be the result of water infiltration, for instance? My car has never been particularly water tight and has been subjected to a couple of unexpected Italian showers with bits of tissue held in the door opening (as sun shields) which acted as conducts. A priori only the rubber mats and the seats received water, though.

Or can there have been an electric gremlin, which would also explain the fried injector electronics? I don't see how it could be related, but I did drive a bit fast given the particularly bad state of repair of the SuperStrada we were on (not over the speed limit though; 110kph).

Edited 23 August, 2016 by RJVB

[RJVB]

You could also visit a garage with a J2534 device, this has the dealers ODIS tool and can do anything the dealer can.

 

Is that something one would find in a multi-brand garage? Much as I'd prefer to avoid the VW garage in the neighbouring town I fear I don't really have a choice. I'll have to go there anyway for the emissions scandal recall, and it might be clever to have this issue sorted out at the same time. Might save me the 80€ or so they charge just for hooking up their tool.

 

So I've been hearing more about EU rules for always-on low beams. What's up with that? And why don't VW turn the lights on only after the engine is running? Even my Guzzi does that ...

[RJVB]

We certainly didn't try to reprogram what looked like random bytes. But what about the lighting test and state read-out functions, are they supposed to be accessible through the VCDS? They were clearly identified at least...

 

About that 2nd rear fog light: does that correspond to the missing 4th bulb in my right rear optic cluster? And since we're talking about those lights: is it normal that the left rear fog light is always on at the same intensity as the other 3 bulbs (15% according to the VCDS)? I never noticed that asymmetry, but then again I never paid real attention to this.

[RJVB]

I'll have another look at the right rear cluster then to see if there is indeed a bulb missing. Sorry I didn't catch this particular implication of what you were trying to tell.

 

As to the VCDS messing up the programming: that cannot be the cause here. Maybe the ESI[tronic] had that effect while scanning for recorded error messages (which would be really bad) but I'm now pretty sure I already noticed my low beams were on before any device had been connected to search for the reason of the engine failure I'd had. That episode sorted itself out I think while I was fiddling with the radio or climate control, otherwise I'd have signalled the issue at the workshop.

 

Can one perform some kind of overall reset by disconnecting the battery?

[RJVB]

Sorry, didn't see your possible solution. I'll leave that to the VW garage as, ahem, the VCDS has now left the building. The VCP should be able to (de)activate permanent running lights too, no? The VCDS I saw was partly translated to French, and I didn't see the function you mention - I have a FL model too.

 

BTW, this is the recorded error:

http://wiki.ross-tech.com/wiki/index.php/01800

 

Funny it mentions the rearview mirror dimming as a possible symptom: I'd been wondering for a while already if that feature still worked. The other symptom surprises me: there is no parking light status indicator on my car (the dash backlight serves that purpose).

[RJVB]

There is only one way that the programming could change - and that is to physically plug in diagnostics and change it.

 

I truly hope so. If these values are stored the same way BIOS/UEFI settings are stored on PCs it's not technically impossible that they're changed through unofficial means (a bit like the side-effect of the VCDS you found) ... and it can also be a result of failure of the nonvolatile memory. Fortunately I see no reason why that would also lead to an 1800 error.

[RJVB]

Actually, I did mention the error before, but as reported in hex and with the improbable message ("implausible signal").

 

I was indeed thinking the permanent-on low beams could be a safety fallback in case of a defective switchgear or cabling. It'll remain a mystery why I could get a bulb warning by disconnecting one of the front fog lights, but only *after* I tried to switch on those lights. Apparently the BCM programming doesn't shunt all switch-related features when a switch (or cabling) failure is detected. 

I'd have to double-check about the dash lighting. I think it's on. And I do have the usual front fog light indicator on the dash; no lights in the switch.

 

Crossing fingers it's the switch and not the cabling (or else that that cabling isn't part of the entire loom!)

[RJVB]

Are we talking about an indicator light (green) showing that the fog lights are switched on, or some kind of warning (orange) that something is amiss with them? I'm positive that I have the former, and also that there's none of the latter.

I've still seen the green one yesterday, IIRC correctly it sits right next to the high beam indicator light. 

 

My car came with the foglights, presumably factory-fitted (in sept. 2009, some 100k km before I got her). What I don't get is why they'd put different instrument clusters in cars with different finishing. Esp. if they're prewired anyway to receive fog lights.

 

I guess we'll see what kind of new switch I get, if it gets replaced.

[RJVB]

One more question: why would the ESI[tronic] talk about an "Elettronica centr. ZE 3.3" while the VCDS clearly talked about communicating with a PQ35? Is there a known alternative to the PQ35 that's called something like a ZE 3.3?

 

 

PS: I've been told better not to use my car model's official name tag "FAP" (particle filter) on english sites ... PQ is the colloquial abbreviation for toilet paper here in France  

Edited 24 August, 2016 by RJVB

[langers2k]

The green symbol between the blue main beam and yellow rear fog lights warning is for dipped or low beam headlight,  not front fog.  As said the symbol for front fogs lights up in the switch itself.  

 

On the FL it's part of the instrument cluster

 

Along the top it is:

- left indicator

- front fog

- main beam

- dipped beam

- rear fog

- cruise control

- right indicator

 

I think on the pre-FL, the front fog symbol is on the switch.

[RJVB]

Heh. Thanks for confirming that that green indicator hasn't been lighting up by accident and for some other reason each of the numerous times I've used my front foglights over the past 3 years

I live in a rural area with unlit twisty roads where those lights give a real additional security at night, so I use them very often.

[KenONeill]

Actually, I did mention the error before, but as reported in hex and with the improbable message ("implausible signal").

 

I was indeed thinking the permanent-on low beams could be a safety fallback in case of a defective switchgear or cabling. It'll remain a mystery why I could get a bulb warning by disconnecting one of the front fog lights, but only *after* I tried to switch on those lights. Apparently the BCM programming doesn't shunt all switch-related features when a switch (or cabling) failure is detected. 

I'd have to double-check about the dash lighting. I think it's on. And I do have the usual front fog light indicator on the dash; no lights in the switch.

 

Crossing fingers it's the switch and not the cabling (or else that that cabling isn't part of the entire loom!)

Actually VCDS often returns "$item: implausible signal" messages when it can tell that $item is generating values outside a likely working range.

[RJVB]

Heh, that still doesn't mean that the signal is implausible (as in cannot be explained according to known physics laws)

 

The Bosch device gave the same error message, translated to Italian. So I'm presuming it's either the message returned by the controller, or else what VAG say it should read (I'd be curious to know the German original, though).

[RJVB]

I certainly hope the same, but we'll only know by replacing the switch. It can also be the cabling, and that could well be a lot more costly to replace.

 

I've been thinking, how (un)likely is it that I caused this myself, by forcing the little storage compartment closed with too much inside (or something at the wrong angle)? Not that I use any real force as the only thing that tends to prevent the lid from closing is a pair of sunglasses with a flexible plastic frame, which I'd rather not damage either ...

[langers2k]

Happy to stand corrected, and to hold my hand up and say I'm wrong, but I truly can't see any evidence of where the front fog and cruise warning lights should be on my factory maxidot cluster on my FL.

 

Just took a quick picture for you Mike, cruise is in the gap between rear fogs and right indicator

 

[RJVB]

Mine is yet different; the parking brake indicator doesn't have a P inside, and sits in the right dial next to the battery indicator.

 

Langers2k: did you take that picture during the lights show after switching on the contact? I just checked: I no longer get that show, except for the red error indicators. I get that for the indicators that depend on the failed switchgear, but I don't see why the others don't light briefly. Or, I'm confused with my Guzzi, which lights all indicators briefly (and lets the dials go all the way) when I turn her on.

 

Mike: no, I wasn't even aware that those fog light symbols were indicators. In the beginning they confused me, and it took me a while to figure out how to light the fog lights other than by turning the switch towards those symbols.

Edited 24 August, 2016 by RJVB

[RJVB]

So the indicators on the top row don't come on briefly when you switch on the car? Good for me (except for the part where I don't know what made me think that )

 

(And of course I knew why the picture was taken, I was just mistaken how it was obtained.)