Alarm Horn Not Sounding

[marshallnoise]

Hello All!

I just got a 2011 A4 Avant and replaced the alarm horn thinking it was broken or had some kind of internal fault, but it does not make any noises or chirps when arming or the panic button is pressed. I know when the alarm was removed and I ran OBD11 faults, the car recognized the alarm was missing. When I plugged the new one in, the code went away.

Now, I am familiar with the VAG alarm systems and when I do arm it, the red light on the driver's door does stay on for about 30 seconds like there is a fault somewhere. I walked around and attempted to open up each door while it was armed and they were all locked.

Any ideas what the problem could be? Let me know if I need to do an adaptation or something.

Thanks,

Paul

[RPMtech147]

Short test. I see the driver in the BCM/Comfort module/ whatever Audi wants to call it depending on the year, crap out from time to time as the siren gets soaked with water over the years.

[Smac770]

If you don't have an error code but you get no chirp, then the system thinks something is still open most likely. You have the hood, the rear lid, and the four doors. They might be physically locked, but does the vehicle know that?

You can check the blocks under 46-central convenience and see what they say. 002 and 016 for the rear lid, 006 through 009 for the doors, 026 second field for the hood. Not sure if the status of the fuel filler door lock is considered, but it is managed by central locking.

For the H12 alarm horn, there's 3 pins:
1 - control from J393 BCM2 [46-central convenience] (it's more likely the horn control line is part of the J518 [05-auth] function block, but it's all in the same physical box now)
2 - ground
3 - 12v power from fuse 12 in column red on driver's end of dash
but if it was the fuse, then the interior monitoring unit G578 would also be dead, and you'd have an error code about that in either 05 or 46.

In the J393 measuring blocks, alarm status is under 020. Second field, can be active, inactive, or no communication. So if there was a problem interacting with the H12 unit, you'd likely have an error state.

[marshallnoise]

Thank you guys! I will work at it and report back.

[marshallnoise]

Looks like the Rear Window Breakage sensor is tripped which probably is causing the issue. Hmmmmm. Must have been broken and replaced but the wire never fixed or something.

[Theiceman]

wonder how a window rear breakage sensor works ?? does your rear defogger work ? or does it have broken elements ?

[marshallnoise]

wonder how a window rear breakage sensor works ?? does your rear defogger work ? or does it have broken elements ?

Haven't had the car long enough to need to use the defogger (also live in SoCal) so I don't know. But from what I have read is that it is literally a wire embedded into the glass that if its broken, it triggers the alarm. Simple circuit. If I can locate the connector, I will just jump it and see if it goes away.

[Smac770]

Avant, you have one on all three rear windows (G304 rear, G183 rear left, G184 rear right). It's just a wire, but the wire path passes through all three glass panels to ground.

J393 T32d (gray) pin 21 > gray/white > G304 > G184 > G183 > chassis ground in harness wiring that eventually connects to ground post at luggage left.

If I remove the T32d and test resistance on pin 21 to ground (find the metal lug on the big brown wires just below the J393), I appear to get about 9Ω. I wouldn't short pin 21 direct to ground without some form of load on the line.

[marshallnoise]

Avant, you have one on all three rear windows (G304 rear, G183 rear left, G184 rear right). It's just a wire, but the wire path passes through all three glass panels to ground.

J393 T32d (gray) pin 21 > gray/white > G304 > G184 > G183 > chassis ground in harness wiring that eventually connects to ground post at luggage left.

If I remove the T32d and test resistance on pin 21 to ground (find the metal lug on the big brown wires just below the J393), I appear to get about 9Ω. I wouldn't short pin 21 direct to ground without some form of load on the line.

Thank you sir!

[marshallnoise]

Well, I finally got to checking out the rear glass break sensor in the right rear of the car. It is absolutely buried under trim, but pretty obvious when you get to it.

PXL_20240423_014223433 by Paul Abbott, on Flickr

Whoops, looks like a pad came unsoldered.
PXL_20240423_014306533.PORTRAIT.ORIGINAL by Paul Abbott, on Flickr

PXL_20240423_014459289 by Paul Abbott, on Flickr

PXL_20240423_014504569 by Paul Abbott, on Flickr

I did the absolutely correct thing and just stripped the wires and connected them together. I am not terribly concerned in my neighborhood. Besides, I am not sure what the hell someone can steal reaching into the window in that spot.

I have not checked to see if the error went away, but I can confirm that it does not fix the alarm issue no matter what. So I am still going to be hunting that down when time permits.

[marshallnoise]

Still working through this. The alarm is still not working so I am going to have to start looking at actual wiring and perhaps the J393 module for water damage or something.

EDIT: Just keeping a log of this so hopefully it can help someone somewhere down the line.

[marshallnoise]

Well, I pulled the J393 and it looks flawless. No signs of water intrusion and no burnt components. So I moved on to the G578 alarm module and it is in good condition too. It is located where the map lights are.

I read that the G578 communicates to the H12 alarm horn using the "LIN Bus" and only has two connections: To the J393 and the G578 and the latter is just a 12v independent supply.

So my question is, could there be a coding issue that is stopping this communication between the J393 and the H12? The J393 is clearly trying to talk to the H12 and reporting no communication. But what if the J393 is looking somewhere else on the LIN Bus and it is not finding it?

I am going to keep looking at the errors on the car.

One weird thing that MIGHT be triggering an issue is that the driver's interior lock switch functions perfect, but the LED that lights up red on the switch (indicating the door is locked) is intermittent. Would this stupid LED cause this kind of issue?

[Smac770]

Post the actual DTC from the J393 so the rest of us can see what the car is actually saying. VCDS would be significantly preferred here, since it provides a useful level of architecture. From my VCDS scan of 46:

...
Subsystem 1 - Part No: 1K0 951 605 C
Component: LIN BACKUP HO H03 1301
...

If the J393 was not coded to expect the alarm system, it would not be looking for it and would not be producing an error about it.

The alarm horn H12, in the plenum chamber, has a three-pin plug. pin 3 to pin 2 should read battery voltage. Voltage is supplied to pin 3 by fuse 12 in column red on the driver's end of the dash. Ground is pin 2 connected to the chassis. Pin 1 is the LIN bus pin. You'd need a scope to sufficiently view the LIN communications waveform. With a voltmeter, all you can tell is if the line is constant 0v or 12v, neither of which would be correct for an active LIN bus. Would expect somewhere between 1v and 11v depending on the traffic level.

The G578 shares 12v power source with the H12, but it does not share the LIN bus line. The G578 is on the LIN bus with the sunroof (J245) and sunshade (J394) control motors. That bus is J393 T32c pin 32, violet/yellow wire. The H12 is on a dedicated LIN line, J393 T32c pin 16, violet/white wire. I don't know why the G578 is not listed as a LIN bus subsystem like the other four (alarm horn, homelink controls, sunroof motor, sunshade motor).

[marshallnoise]

I am sorry I don't have VCDS. But I think I can get something similar to what you are asking for. Will have to go out and get that.

I checked all the doors and they showed closed when they were closed and locked when they were locked. However, I went through the channels I came across a few curious things. Part of it is that I haven't looked for a full list of channels so I could answer my own questions there.

46-51
Screenshot_20240521-174752 by Paul Abbott, on Flickr

46-52
Screenshot_20240521-174804 by Paul Abbott, on Flickr

46-54
Screenshot_20240521-174819 by Paul Abbott, on Flickr

46-20 this is the no communication thing on the alarm.
Screenshot_20240521-175019 by Paul Abbott, on Flickr

I haven't begun looking for solutions for these yet.
Screenshot_20240521-175651 by Paul Abbott, on Flickr

Screenshot_20240521-175755 by Paul Abbott, on Flickr

EDIT: Resized the pics.

[marshallnoise]

Post the actual DTC from the J393 so the rest of us can see what the car is actually saying. VCDS would be significantly preferred here, since it provides a useful level of architecture. From my VCDS scan of 46:

...
Subsystem 1 - Part No: 1K0 951 605 C
Component: LIN BACKUP HO H03 1301
...

If the J393 was not coded to expect the alarm system, it would not be looking for it and would not be producing an error about it.

The alarm horn H12, in the plenum chamber, has a three-pin plug. pin 3 to pin 2 should read battery voltage. Voltage is supplied to pin 3 by fuse 12 in column red on the driver's end of the dash. Ground is pin 2 connected to the chassis. Pin 1 is the LIN bus pin. You'd need a scope to sufficiently view the LIN communications waveform. With a voltmeter, all you can tell is if the line is constant 0v or 12v, neither of which would be correct for an active LIN bus. Would expect somewhere between 1v and 11v depending on the traffic level.

The G578 shares 12v power source with the H12, but it does not share the LIN bus line. The G578 is on the LIN bus with the sunroof (J245) and sunshade (J394) control motors. That bus is J393 T32c pin 32, violet/yellow wire. The H12 is on a dedicated LIN line, J393 T32c pin 16, violet/white wire. I don't know why the G578 is not listed as a LIN bus subsystem like the other four (alarm horn, homelink controls, sunroof motor, sunshade motor).

Ok, here is what I could get. I assume that 46 = J393 because under subsystems, there were three subsystems but J393 was not one of them.

All errors under 46 Comfort System
Screenshot_20240522-142549 by Paul Abbott, on Flickr

Alarm horn error under 46
Screenshot_20240522-142601 by Paul Abbott, on Flickr

GTO BF Subsystem
Screenshot_20240522-142618 by Paul Abbott, on Flickr

J245 PS133 Subsystem
Screenshot_20240522-142631 by Paul Abbott, on Flickr

J394 PS229 Subsystem
Screenshot_20240522-142640 by Paul Abbott, on Flickr

[Smac770]

Yes, J393 is the control module identifier for the BCM2 aka CCM. In control module diagnostics, it's at address 46 (labeled central convenience / comfort system / etc).

So you have a DTC for cannot communicate, yet OBDeleven is still presenting the INFO for the H12 LIN bus member. I wonder if OBDeleven is caching that info. Or if the H12 is registering on power up just long enough for the J393 to get the member info, but then falling off after.

If you disconnect the three pin plug on the H12, do you measure 12v from pin 3 to pin 2? And if you measure the voltage from pin 1 to pin 2, do you see anything when you do something like open/close the sunroof or sunshade? If those measurements are not as expected, try again with the negative jump start post instead of pin 2. Maybe there's an issue with the ground line. If there remains a lack of 12v at pin 3, check the fuse. If there's no voltage noted on pin 1, there's a wiring issue. If all the voltages look good, best guess is the H12 is flaky. I don't know if a dead internal battery in the H12 would come across in this manner.

Better yet, if you hook 12v and ground to the H12 but not the LIN bus line, do you measure a voltage on the H12 LIN bus pin (use a pigtail wire in the harness connector in place of the proper LIN bus wire)? A LIN bus slave should make some voltage activity I'd expect at power on.

OBDeleven, you can just run the scan, and the results should be in the history log. That text log is usually simpler to parse than a bunch of screen caps. Maybe one day VoltasIT will unsuck themselves and make their scan output actually useful in a community environment.

[marshallnoise]

Yes, J393 is the control module identifier for the BCM2 aka CCM. In control module diagnostics, it's at address 46 (labeled central convenience / comfort system / etc).

So you have a DTC for cannot communicate, yet OBDeleven is still presenting the INFO for the H12 LIN bus member. I wonder if OBDeleven is caching that info. Or if the H12 is registering on power up just long enough for the J393 to get the member info, but then falling off after.

If you disconnect the three pin plug on the H12, do you measure 12v from pin 3 to pin 2? Yes!

And if you measure the voltage from pin 1 to pin 2, do you see anything when you do something like open/close the sunroof or sunshade? No.

If those measurements are not as expected, try again with the negative jump start post instead of pin 2. No. Maybe there's an issue with the ground line.

If there remains a lack of 12v at pin 3, check the fuse. Fuse is good.

If there's no voltage noted on pin 1, there's a wiring issue. [B]Yessir. This is where we are at./B]

If all the voltages look good, best guess is the H12 is flaky. I don't know if a dead internal battery in the H12 would come across in this manner.

Better yet, if you hook 12v and ground to the H12 but not the LIN bus line, do you measure a voltage on the H12 LIN bus pin (use a pigtail wire in the harness connector in place of the proper LIN bus wire)? A LIN bus slave should make some voltage activity I'd expect at power on.

OBDeleven, you can just run the scan, and the results should be in the history log. That text log is usually simpler to parse than a bunch of screen caps. Maybe one day VoltasIT will unsuck themselves and make their scan output actually useful in a community environment.

Thank you for the great help. I am sure we have a brake in the line somewhere. I cut the purple/white wire about 6 inches down from the plug to check continuity between the plug and the end of the wire I cut and the copper looks good AND there were no problems at the plug.

I guess I am going to have to find a wiring diagram and find out where the heck that sucker goes!

[Smac770]

What I would do as a counter test is splice a new wire from the H12 pin 1 lead that you have to the wire at the J393 end. I would just put a terminal pin on the temporary wire and "backprobe" it into the slot in the 32-pin plug (black 32-pin, pin 16). No need to actually break the wire on the J393 end. This at least would confirm it's all good if the wire path is all good. Or if there's something more.

Plus, my error on the voltage stuff. The sunroof stuff is on that other LIN with the interior alarm sensor, not the alarm horn. To invoke LIN activity on the alarm horn bus, you'd need to do something like lock the car / unlock the car. But you'd need the hood to appear closed for the lock to occur. To accomplish that while the hood is open, you can disconnect the hood detect plug which is above the left head lamp, next to the union for the release cable. You'd also want to check both ends, the end at the H12, and backprobe the pin at the 32-pin on the J393. If nothing is happening at the J393 end of the wire, nothing will be happening at the H12 end of the wire. Sorry about that error.

[marshallnoise]

What I would do as a counter test is splice a new wire from the H12 pin 1 lead that you have to the wire at the J393 end. I would just put a terminal pin on the temporary wire and "backprobe" it into the slot in the 32-pin plug (black 32-pin, pin 16). No need to actually break the wire on the J393 end. This at least would confirm it's all good if the wire path is all good. Or if there's something more.

Plus, my error on the voltage stuff. The sunroof stuff is on that other LIN with the interior alarm sensor, not the alarm horn. To invoke LIN activity on the alarm horn bus, you'd need to do something like lock the car / unlock the car. But you'd need the hood to appear closed for the lock to occur. To accomplish that while the hood is open, you can disconnect the hood detect plug which is above the left head lamp, next to the union for the release cable. You'd also want to check both ends, the end at the H12, and backprobe the pin at the 32-pin on the J393. If nothing is happening at the J393 end of the wire, nothing will be happening at the H12 end of the wire. Sorry about that error.

I appreciate your help! I think it does make sense to just run some wire from the front to the back and see if things work. Don't know when I will get to it, but I will and I will report back.

[marshallnoise]

Sure enough, there is a break in the purple/white line going from pin 1 of the alarm horn to the black 32-pin plug connected to J393.



So the question is this: Do I attempt to find the break in the line and patch it all up or just run a new line all the way to the back. Some how the wire gets from the alarm horn to the left hand side of the engine bay to the back right of the trunk area.

[Smac770]

Yeah, that's a long run. Unfortunately, the wire does not show as passing through any connector along the way. Not the 17-pins in the ECM box, not the 17-pins at the base of the left a-pillar, not the 17-pins at the base of the right a-pillar.

We know the wiring runs through the left fender.

interior harness - engine bay: https://www.ilcats.ru/audi/?function...81&language=en
item 7 is the H12 alarm horn plug
item 8 is the grommet at the fender
union (40) is to the interior harness left: https://www.ilcats.ru/audi/?function...82&language=en

interior harness right: https://www.ilcats.ru/audi/?function...83&language=en
item 54 are the 32-pin plugs into the J393

How does the wire get from interior harness left to interior harness right? Via the interior harness center tunnel: https://www.ilcats.ru/audi/?function...84&language=en
upper (50) is the left harness connection, (51) is the right harness connection.

Unfortunately, the diagrams show the connection from the left harness to the roof harness; (91) in the left diagram = the lower (50) in the tunnel & roof diagram. But neither the left harness nor the right harness actually diagram the connection point to the center tunnel harness (the 50 and 51 on the upper harness in the tunnel & rood diagram).

Suffice to say, however it gets there, you're not finding it. Because it's also not the only violet/white wire running around in the interior harness.

I've never found a good static pic of the fender pass-through (that grommet 8 in the engine bay harness diagram), so I just reference this video from Naptown. This is of the right side, but same business on the left side. https://youtu.be/C_-MfYokbMc?t=3

[marshallnoise]

Yeah, that's a long run. Unfortunately, the wire does not show as passing through any connector along the way. Not the 17-pins in the ECM box, not the 17-pins at the base of the left a-pillar, not the 17-pins at the base of the right a-pillar.

We know the wiring runs through the left fender.

interior harness - engine bay: https://www.ilcats.ru/audi/?function...81&language=en
item 7 is the H12 alarm horn plug
item 8 is the grommet at the fender
union (40) is to the interior harness left: https://www.ilcats.ru/audi/?function...82&language=en

interior harness right: https://www.ilcats.ru/audi/?function...83&language=en
item 54 are the 32-pin plugs into the J393

How does the wire get from interior harness left to interior harness right? Via the interior harness center tunnel: https://www.ilcats.ru/audi/?function...84&language=en
upper (50) is the left harness connection, (51) is the right harness connection.

Unfortunately, the diagrams show the connection from the left harness to the roof harness; (91) in the left diagram = the lower (50) in the tunnel & roof diagram. But neither the left harness nor the right harness actually diagram the connection point to the center tunnel harness (the 50 and 51 on the upper harness in the tunnel & rood diagram).

Suffice to say, however it gets there, you're not finding it. Because it's also not the only violet/white wire running around in the interior harness.

I've never found a good static pic of the fender pass-through (that grommet 8 in the engine bay harness diagram), so I just reference this video from Naptown. This is of the right side, but same business on the left side. https://youtu.be/C_-MfYokbMc?t=3

Speaker wire it is! Just kidding. I will get something and run it in some fashion. Probably just go down the passenger side and along kick panels if I can find a way.

Weird thing to have break!

[marshallnoise]

Got her fixed up last night. I mainly want the damned confirmation I locked the door so yes, this process was worth it to me.

[Theiceman]

from oil pressure expert to Alarm wiring expert in one easy step lol....

[marshallnoise]

from oil pressure expert to Alarm wiring expert in one easy step lol....

They are nice cars with some seriously under implemented engineering. I couldn't have done this part without Smac770 for sure though!

[Smac770]

Well done. Clear, informative video too. Always nice when people contribute quality visuals for the uninitiated. You went full surgery on it more than I would have. Nice find on the harness path along the left. Harness pathing is one thing the wiring diagrams do not document (and hell, they don't even document the harness itself starting with B8.5). But once it's fixed and back together, it's a great feeling.

I was looking at the wrong plug (7) in the parts diagram. The upper (7) that passes through the left fender is for the G395. The lower (7) for the H12 passes through that grommet in the side of the ECM box. The other violet/white passing through that grommet (1) [seen at 0:17 at upper left] is to plug (3), for the windshield wiper motor, on pin 4. Comes from the J519 16-pin pin 2 there under the dash behind the headlamp switch.

B8.0 A4 engine bay interior harness.jpg

[Smac770]

Haha, saw you had an earlier one too. Did you fix the rear lid issue? Or is that somewhere down the list?

[marshallnoise]

Haha, saw you had an earlier one too. Did you fix the rear lid issue? Or is that somewhere down the list?

LOL, I am planning on another thread to discuss that damned thing. I hear relays clicking when locking and unlocking and when exercising the latch, it is clearly sending signals to the computer otherwise the car wouldn't arm. So I am not sure what the hell I did to cause this issue now. If it's not one thing...