VCDS recode for Interior light

arclight99

Registered User
Joined
Jan 14, 2016
Messages
23
Reaction score
0
Points
1
Hi,

A3 8V 2.0 TDI Sportback.

I recently retrofitted the LED interior dome/reading lights to my car, as it did not have the Lighting Pack, so had only the cheapo halogen units!

I followed a 'How to' on the A1 forum, and the whole set up works fine, with the full functionality you would expect - until, that is, you start the engine.
Then after 5 seconds, the front and rear dome/reading lights all begin to pulse/flicker at about 2 per second, continuously, until you switch the engine off.

Do I need to go into the VCDS menu to enable/disable something?

Thanks in advance,

Arclight.
 
Hi,

A3 8V 2.0 TDI Sportback.

I recently retrofitted the LED interior dome/reading lights to my car, as it did not have the Lighting Pack, so had only the cheapo halogen units!

I followed a 'How to' on the A1 forum, and the whole set up works fine, with the full functionality you would expect - until, that is, you start the engine.
Then after 5 seconds, the front and rear dome/reading lights all begin to pulse/flicker at about 2 per second, continuously, until you switch the engine off.

Do I need to go into the VCDS menu to enable/disable something?

Thanks in advance,

Arclight.
I've got the parts for mine as well along with a harness whose location under the dash I haven't located yet. Did you get the dedicated harness as well or did you hack the existing harness?
 
No, I followed the instructions on the A1 forum here

http://a1-forum.co.uk/a1forum/viewtopic.php?f=28&t=13473

I made my own wiring loom which connects the front unit to the rear unit, through the headliner, then you just swap over a few pins on the 'supply' feed to the front unit.

I used genuine connectors from Audi, some pins from China (Ebay!) and some 0.5mm wire.

Trouble is, it doesn't work with the engine on!
 
No, I followed the instructions on the A1 forum here

http://a1-forum.co.uk/a1forum/viewtopic.php?f=28&t=13473

I made my own wiring loom which connects the front unit to the rear unit, through the headliner, then you just swap over a few pins on the 'supply' feed to the front unit.

I used genuine connectors from Audi, some pins from China (Ebay!) and some 0.5mm wire.

Trouble is, it doesn't work with the engine on!
Wondering if you've found any joy in this before I venture down the same path? How messy was it to run your custom loom from the front to the rear light? I've got the sunroof option so it's a bit of a path.
 
I used a 1m steel ruler, attached the new cable to that and pushed it from the back to the front, under the head liner.
Worked fine.
A sunroof would be difficult though.
I guess you might have to run it down one side, above the doors?
No idea really.

I don't even know if the front module would be the same as a non-sunroof car.

Have downloaded the wiring diagrams for the car, and plan to do some tests to try and resolve the flickering, this week.

I'll post again if I find anything out.

arclight
 
I just did this upgrade on my 2015 A3 (8V) US-spec sedan. I also have the issue with the flickering map/dome lights while the engine is running. Before doing the installation I took measurements of all the pins in the harness engine running, off, doors open, doors closed. If you're at all familiar with charging systems, you already know that you'll see slightly more than 12 volts on a charged battery and ~14 volts with the car running and all things normal. Since the flickering happens when the car is running and the alternator is [correctly] producing more than about 13.5 volts, the LED assembly clearly is looking for 12 volts and is responding to the over-voltage by flickering. Not sure if this has ramifications or not.

So, to correct it I see one of two possibilities: re-pin the power lead for the lights to something that produces strictly 12 volts, car running or not (and I don't know if that exists in the car), or install a voltage regulator of some sort in-line that will produce a consistent 12 or so volts. I've used LEDs in lots of automotive/RV environments with no adverse results based on 12 versus 14 volts so I think it's most likely the design of the light fixture assembly's internal electronics. I'm going to research this some more. I'll probably try to source an inexpensive 12 volt regulator of some sort and tuck it up above the light.
 
I did wonder about this at first myself, but didn't think this was likely with modern car design, especially with so many on board computers.
They wouldn't like a fluctuating voltage, surely?

Also, nobody on the A1 forum seems to have had this problem, it would seem.

Before I did this mod, I inserted LED wedge bulbs into the 'standard' unit, and would have thought they would flicker, if anything would.

I contacted someone who had had this problem, and he solved it by running a feed from the fuse box up the A-pillar, but didn't give any detail.

I'm going to run a wire from fuse 8 to pin 1 on the light unit (the missing wire, which is bridged from pin 5) which according to the wiring diagram, should have everything in its correct place, with no bridges.

Failing that, I was going to try a ballast resistor somewhere.

arclight
 
Just to test the fuse 8 idea, I ran out and hooked up my voltmeter to it and with the engine running (whilst my interior light attempted to do a strobe effect), I measured 14+ volts so I don't think connecting to fuse 8 is going to help--because it's still putting out more than ~12.5, the circuitry inside the light assembly still going to do the flickering/strobe thing. I'm going to see if I can source a small voltage regulator in the next couple of days and I'll report my findings. Perhaps that's something that's built into the cars with the ILP and the extra pin that we have bridged between pins 5 & 1 and maybe the regulator exists on just one of those circuits. I don't have access to a detailed wiring schematic so I can't say for sure but in theory that should work.

I'd be hesitant to do it with resistors for a couple of reasons--unless it's big, it's could get warm, secondly, you'll see the voltage drop when the car isn't running so if you had a resistor kit that dropped you from, say, 14.4 volts to 12.4 volts while running, you'd see approximately 10.4 volts when the engine isn't running, and that could add new problems.
 
And as to fluctuating voltages, to correct for that for enough current that a car draws, a regulated power supply would be huge and has its own inefficiencies as well, thus requiring a larger alternator and thicker cables, thereby adding weight. Most circuits are designed to accommodate a pretty big swing in voltages but clearly this light assembly is not one of them. Prior to installing the ILP map/dome light kit, I too had installed Ziza LED bulbs and they handled it fine, but that's because the base lighting kit is 'dumb' with no processing in-between the supply pins and the bulbs. Because there is an actual circuit board and processors inside the ILP lights, those are likely the bits that are sensitive to the voltage fluctuations and it's obviously not internally regulated. At least that's my theory so far. ;-)
 
I see your point, however, the red LED's which provide the down lighter and illuminate the buttons, do not flicker at all.

From my interpretation of the wiring diagrams (and I'm no expert with electronics!) the connections to the halogen unit are identical to the LED unit, with the exception of pin 1, which on the LED unit, comes from fuse 8.

Fuse 8 is on the headlight switch circuit, so that when you turn on side/headlights, the red LED's come on - and don't flicker!

Surely if there was a voltage fluctuation problem, it would affect these as well?

Unless, as you say, there's something in the module, which doesn't like the 24 volts.

Arclight
 
Also, if it is an over voltage problem, why do the lights only start flickering, after the engine has been running for 5 seconds or so?
Surely it would start right away?

My theory is that bridging pin 1 and 5 is causing a conflict in the modules microprocessor
 
The backlighting LEDs are definitely on a different sub-circuit, that's why they're pinned out separately and most LEDs going directly to an unregulated power source will handle it fine if it's within their tolerances. Older motorcycles demonstrate this well: at idle the headlight would be somewhat dim but as revs climb, output increases to a maximum around 5,000 RPM which is where most make maximum voltage up to redline. Cars have alternators as opposed to generators but the core operating principle is the same. I also put my voltmeter across pin 6 and the voltage varies based on the illumination knob that pops out of the headlight switch cluster, which as you know adjusts many of the interior backlit switches and indicators. Since that is on a separate sub-circuit from the actual dome/map lights, that one is unaffected.

As to the lights flickering after ~5 seconds, I can confirm that on my car as well. The reason is because the alternator doesn't start producing output (the ~14 volts) until the engine is fully running (I had a meter across the terminals and it stays around 12-ish volts until about 5 seconds when it quickly climbs to 13 and then low 14's). It's likely a circuit that switches on once the engine is truly started to prevent unintended stalling if the alternator's drag on the engine would overwhelm a cold engine and/or to ease cranking of the engine. Most automotive alternators are capable of producing at least 60 amps or more of current. To generate that much current from an automotive alternator can take a few horsepower which could make or prevent the ability of a cold car to start and/or idle reliably.

I don't have access to any of 8V factory wiring diagrams/schematics but most cars will come off the fuse to the device(s) they protect. In the case of the backlighting, power goes to fuse #8 first, then splits out to the modules it protects. I'm very curious about the 2nd power circuit for the ILP fixture, what power source is it 'supposed' to go to? Is there a regulated circuit somewhere?

As to other devices not flickering or changing with engine off versus running, I'm sure they are internally regulated or are designed to accommodate voltage fluctuations. In working through this, the more that this whole thing with the fact that ILP cars use the same connector to the light fixture but have different pin assignments make some sort of odd sense. I suspect they put in a different harness for the ILP cars that has a regulator built into the harness, particularly since one can be put inline and doesn't need to connect to anything else. That's just my theory and not having access to an ILP car, I can't confirm it but it would make sense.

My local, albeit fiscally stressed, Radio Shack does have appropriately sized regulators in stock ($2.00 US/each) so I'm going to pick up a couple this afternoon and hopefully try it out tonight after the kids go to bed. I'll let you know how that plays out. I'm really hoping that the current draw is low enough that I don't have to improvise a heat sink to provide cooling for the regulator (probably not since the current draw is pretty low for most LEDs). If I do I might see if there's a part of the underside of the roof's bracing that already has a screw because the roof would be a very large heatsink. ;-)

The part that has me thinking now is why they have two separate power circuits for the ILP lamp. Is one direct to the #8 fuse?, are they both? If both are, there must be something inline in the ILP harness to regulate the voltage.... I don't think this will be difficult to solve, I'm mostly curious as to how the factory does it. I am surprised that they don't have an internal regulator in the light module for the lamps, I'm sure it would simplify manufacturing in that they wouldn't need to have so many different harnesses for slight product variations. Maybe it's a manufacturing limitation...
 
Image
Image


Here are the wiring diagrams for the right hand drive models.
Sorry about the quality, don't know if you'll even be able to zoom in.

I downloaded them from Audi Erwin, but only have these paper copies with me at the moment, hence dodgy photos!

The wire from pin 5 on the module, goes to fuse 29 and that is fed from pin 12 on the J519 computer.

Fuse 29 is for the interior lights, according to the handbook.

Also, I disconnected different wires from the input to the module, one by one, to try and eliminate some things.
With pin 1 disconnected, nothing on the module came on at all!
 
I get what you're saying about the old motorcycle alternators, I remember the set up well.

My question relating to this problem is, however, why is the 'flicker' rate of the LED's constant, across the RPM spectrum?
 
My theory on the flicker rate is that it's because the internal circuitry for the lamps themselves is freaking out with anything over ~12.5 volts and that's how it reacts, maybe it's starting to shut down when it gets over 13. Hard to say. If I were to fully disassemble the assembly and look at the components inside and lookup their specs, I'm sure we could get the specifics, but I didn't see any way to do that without likely breaking the assembly, and it was expensive enough as it were! Heading out now to run some errands but picking up a couple of voltage regulators is on my list for this afternoon.

Arclight99, thanks for posting the diagrams. I'll try to have a look at those too.

Aron
 
Well, better, but still some noticeable flickering. I added a couple 10-microFarad capacitors across the input and output of the voltage regulator and definitely saw further reduction, but not enitrely visibly eliminated. The capacitors help to stabilize the voltage while the regulator adjusts output to the level you want. I am going to get some larger capacitors and I think I'll be good enough at that point. I may also buy an inexpensive pre-built regulator that has capacitors and a potentiometer for adjusting output. This isn't the most elegant way to get it done but the end result should be close. I'm still curious if there's an additional/different pin-out that's pre-regulated for the ILP map light...
 
So which pin/wire/wires did you put the voltage reg across?
 
I cut the purple/white supply lead. The first leg of a voltage regulator chip is input, the middle is ground, and the third is the regulated output. The only size Radio Shack had was a 1 amp rated regulator and you generally don't want to be more than about 75% of the capacity. I suspect that when all 5 lights are on, it might be getting close to 1 amp so a 1.5 or 2A regulator would be ideal. I then added capacitors that connect to the purple/white before and after the regulator and the other leg goes to any good and convenient ground. It's not perfect as to reducing flicker, but it's dramatically improved. My next step, maybe tonight after the kids go to bed, is to try larger capacitors. The capacitors smooth the power. I'm also wondering if the lamp is sensitive to AC voltage. Alternators generate AC electricity and then the alternator's rectifier/regulator effectively turns the output to DC, however it's common to have some leftover AC and perhaps the light fixture is sensitive to that as well. Obviously the LEDs themselves are fine but it could be the circuitry in-between. Audi almost certainly does something different with the ILP cars because this is more that what you would likely build into a harness. A single regulator possibly, but adding capacitors and what not is going to make for a small circuit board that has to be stuffed somewhere. I didn't solder/weld anything in yet because I'm still experimenting, I've just used alligator clip test leads to attach everything so my interior looks like it was hit by string cheese at the moment. I'll be driving the other car today....
 
Arclight99, since we're retrofitting our cars, we split off a single supply line. Do you know where the two separate leads on the factory ILP cars connect the two different supply lines? Thanks, Aron.
 
You might be able to see my notes on the photos if you zoom in, but here's how I interpret the wiring diagram.
This is for right hand drive Europe, but I think yours should be the same really.

Halogen

Pin 1 - brown, earth/ground
Pin 2 - green, goes to pin 67 on J519
Pin 3 - Empty, or red/blue for auto anti dimming rearview mirror.
Pin 4 - violet/white, goes to fuse 29, which is fed from pin 12 on J519
Pin 5 - grey/white, goes to to pin 62 on J519 ( also on the hazard warning light and light switch circuit - for red ----------backlights)
Pin 6 - Empty

LED Module

Pin 1 - grey/violet, goes to fuse 8, fuse holder C, also on light switch circuit
Pin 2 - green, goes to pin 67 on J519
Pin 3 - brown, earth/ground
Pin 4 - Empty, or blue/red to auto dimming rearview mirror.
Pin 5 - violet/white, goes to fuse 29 (interior lights) fed from pin 12 on J519.
Pin 6 - grey/white, goes to pin 62 on J519 ( also on the hazard warning lights circuit and light switch, for red
............backlights

Hope this helps.

As you can see, even though the pin assignments into the LED unit are different from the halogen, the functions are identical.

I've looked at where I can source an LDO that would be appropriate, so will give that a try perhaps!
 
Well, I think I've sorted it out, at least to my satisfaction and I think it will be to yours if you go this route. I picked up some assorted capacitors, 35 volt specification, more than ample for this project considering that maximum voltage on these lines will be less than 15 volts. I initially tried a 220 microFarad cap but ended up soldering in a 1,000 microFarad capacitor. No voltage regulator necessary, just the capacitor. The gist of it is this, capacitors are polarity sensitive so you will SUPPLEMENT the existing circuit by putting the positive side of the cap in parallel with the supply (purple w/ white stripe) and put the side that the arrow points to any good ground. I opened up the sheathing on the dome/map light ground harness (without splitting the cable) and soldering in the negative lead. Be sure to fully insulate the leads from the capacitor since the legs are long and if you go with the 1,000 microFarad unit, it is kind of big (about 1" long). I used heat shrink tubing on the legs, leaving just enough of the wire exposed to solder into the existing circuits. I taped it up and secured the capacitor (to prevent it from making any noise in the back of the light area).

I'm attaching a picture of the capacitor I used (and it cost $1.49 at Radio Shack). I'm exceedingly satisfied with the results, it took a while to get there but it was worth it. I like these lights so much more and now that they don't flicker, they're perfect. YMMV. Link also to the video I made during the trials. The regulator is in the video but it is bypassed so you can disregard the regulator that you see in the video.

I'm actually having trouble uploading photos via the forum so here are links on my server and YouTube:
Final Capacitor Used:
photo_53686172656430302f323031352041756469204133203856_32303136303132345f417564695f41335f38565f4361702e4a5047

Video Demonstration:
 
That looks like a great solution!

So to summarise, you used a 35v 1000 microfarad capacitor, attached the positive end to the grey/white wire (pin 6) without cutting it and the negative end of the cap to the ground (brown wire) also without cutting it.

Did you still leave the bridging wire between pins 5 and 1, or is that now redundant?

What about heat?
Do capacitors get hot?

Thanks for your perseverance with this!

arclight
 
That looks like a great solution!

So to summarise, you used a 35v 1000 microfarad capacitor, attached the positive end to the grey/white wire (pin 6) without cutting it and the negative end of the cap to the ground (brown wire) also without cutting it.

Did you still leave the bridging wire between pins 5 and 1, or is that now redundant?

What about heat?
Do capacitors get hot?

Thanks for your perseverance with this!

arclight
All retrofit wiring that we have done stays in place, no change there. Put the positive (feed) side of the cap on the 1/5 supply bridge, the negative to ground. Typically they don't get hot (unless wired backwards, then they die quickly too).

It's been a long time since my last electronics course so if I had thought back to that, this solution would have come back to me faster, but honestly it was kind of a fun discovery process. I can't see any downside to this either. It slightly stabilizes power ahead of the circuit too, though the capacity isn't large enough to make a big difference.
 
Ah right, so the positive side of the capacitor goes to the violet/white wire, which is the pin 1 pin 5 bridge.
I thought you'd put it on the grey/white wire, pin 6.

Sorry, but some more questions.......
Don't want to wake up in the middle of the night to find the fire department hosing down the smoking wreckage of my burnt out A3!

Been reading up on capacitors, but my small brain can't quite get to grips with all the tech!
Does there need to be any resistors or anything added?
How did you come to use 30V 1000 microfarads?
Been reading about capacitors needing to discharge etc., or will it just discharge through the LED module anyway?

arclight
 
Capacitors are pretty benign, they simply store and release energy. They are much faster acting than batteries, however they have very little capacity. Correct, it will discharge while powering the lights in the overhead fixture. If that light is off, there are enough circuits in our cars drawing small amounts of power when switched off, the cap will discharge through those in probably a few seconds without supply from the running car.

The number of farads is the capacity so that's why a 10 microFarad cap made a tiny difference. The 220 pretty much took it out and the 1,000 was above and beyond. Voltage spec, you want to go beyond the maximum voyage of the circuit. 35 and 50 volt specs are common. You can also get different accuracy specs. 20% is the baseline and you can spend more for 10 or 5% versions but since we are just looking to stabilize voltage that's jumping around and the Farad capacity is so much greater than what's actually required, I'm not worried about the accuracy grade.

No resistors necessary.

I suggest testing first using the test leads, then you can feel the capacitor after it has been connected for some time, and turn on and off. I was not able to discern any heat increase.
 
Yeah, keep me posted if you do it and how it goes! Hope I might have helped, your information was huge too. Good times on the InterWebs. ;-)
 
I finally got the time to get onto this.
I'd bought the capacitor, ready to go in, but thought I'd just give my theory a try.

So I ran a wire from fuse 8 to pin 1 on the front module (disconnecting the bridge wire from pin 5) and guess what?
It all worked perfectly!

I thought the same as you, regarding the volts out from fuse 8, but hey, it just works.
 
So you went straight to the fuse, and not the BCM? Just tapped the fuse? So there must be noise or something on the purple/white wire then from the sounds of it....
 
Yep.
Just ran a wire straight to fuse 8, as per the LED wiring diagram.
Just got to do a permanent install now.
 
Yep.
Just ran a wire straight to fuse 8, as per the LED wiring diagram.
Just got to do a permanent install now.
I had a couple of hours free today and went back in. I removed the capacitor and attached a jumper to the fuse 8 under the dash and it worked as you mentioned. I also had an empty spot two or three fuses over for the non-existent alarm and that also has power with the ignition off. I tested it and the lights worked as expected so I built a fused terminal to go into it and patched in there without the capacitor. All seems well, thanks for the tips! Ideally I'd like to go straight to the BCM but this will work fine for now.

Next project will probably be the puddle lamps and door warning lights. Have you done those yet?
 
No, I don't think I'm going to bother with those.
Too much hassle taking the door cards off etc.

I did find a good thread on it somewhere, and it doesn't look too difficult, but does require some recoding on VCDS.

I remember seeing some kits out there, somewhere, with pre-built looms and light units ready to go.

Good luck!
 
Well I don't know what's happened now.
Went through all the possible settings yesterday and all worked perfectly.

Today, went to finish off the permanent install, and at some point everything went off.
Then the alarm went off (I was sitting inside at the time, with the doors closed) I cancelled/unlocked with the key fob.
Now, everything works ok, until you try to turn on any of the individual lights manually, then that light/lights only stay on for two seconds, then fade to off!

No flickering anywhere, and all the door open/closing, lock/unlock functions work as you'd expected.

This is really starting to annoy me!

I've disconnected the pin 1 feed now at the fuse box and will leave it for a while, in case the computer needs to reset or something.
Think I must have upset something when the alarm went off........
 
Later...........
Still the same, but only when engine running!
 
The behavior of the quick off sounds like it would normally when a door is opened then shut while the engine is running. Does it fade off slowly (theater style) or just shuts off quickly? Mine is still operating correctly and I've driven it for 3 days since finalizing my installation with the pin 8 getting its power from the empty fuse socket to the left of Fuse 8 (Fuse 5 or 6).

The only thing left for me on this project is that I am awaiting fuse block terminals and I'll move all of my add-ons (permanent 12 volt power to rear power port, ignition-switched radar detector power, and the 2nd dome/map light feed) to the back-side of the fuse block for more of a stock integration. I don't like having all these fuse taps on top of the factory fuses. I'll probably add a couple extras for future use as well, just in case.
 
When you push any of the buttons, doors shut, engine running, the corresponding light comes on for 2 seconds then fades off (theatre style). Everything else works as it should.

Wondering if I might have blown something inside the light module.

Will try one of the other fuse outlets perhaps.
As I say, really annoying after finally getting it working (or so I thought!)
 
Last edited:
That is very odd, it's like it's briefly following instruction as it should. There are some processors in the light module (they're super small) and it might be possible to take them out but I'd think it to be an all or nothing proposition, not partial as you described. The only thing I can recommend at this point is measuring voltage at the 6 pin terminal to confirm that you're getting power on the supply lines (and by power, it should be +12 - 14-ish volts on pin 6 and the one that purple/white goes to. Use the brown ground wire on the terminal when making measurements as well, just so you are testing it all in the same spot. I forget the color of the door trigger wire but that one should go to ground when the door is open and then there's the backlighting/LED halo light supply, which is irrelevant for this purpose. You may want to test the pins while it's connected to the fixture and NOT connected to the light. If something shorted in the light module, it's possible that you could see a big voltage drop due to a component shorting, resulting in the light going out. I'll try to think it through some more too and will post more if I do think of something.
 
Had another look at things today, but still no change.
Checked the voltages coming out of the 6 pin plug.

Engine off:
1 - 12.25
2 - blip to -0.4 then 0 (less than a second)
5 - 12.25
6 - headlights off 0, headlights on, 12.73

Engine running
1 - 14.33
2 - blip to -0.4 then 0
5 - 13.87
6 - headlights off, 0, then headlight on, 12.73

Still think somethings failed inside the light module itself.
My car doesn't have any empty fuses next to 8 (6 is alarm, 7 is auto climate control)

Will have to try and get hold of another module, just to check, but that won't be cheap!
 
Got hold of another front light module - still the same problem :worried: