Problems with VAG COM adapter

Hello,

I hope someone here can help me, if it's still possible.
I'm having a problem with the airbag warning light in my Audi 80 B4 Avant, manufactured in October 1994. It blinks for about 2 minutes and then stays on continuously. The problem first appeared after I went through a car wash, and then my speedometer and all the instruments went out. It turned out to be a blown fuse. After replacing the fuse, everything started working normally again, except for the airbag.
I now wanted to create an adapter to read out any potential error messages. I decided to go with Maik's adapter, for which I also made a PCB (single-sided copper).
I'm now unable to connect to the ECU.
Under Options, the response I get for the Com Port Test is:
COM Port Status: OK, Adapter Status: 0 - Found, Ready, Fast-Synch Compatible: NO.
I've also tried various settings, including setting the start baud rate to 2400, reducing the block interval (down to 55), and trying character intervals of 3.2, 1, and 0, but none of it has made a difference.
During the LED test, sometimes all the diodes light up, and sometimes only one or two. However, when two LEDs are lit, it's never just one of them that's lit.
I have 12V on both the K and L lines.
If I measure the K-Line with respect to ground, I get 0.5 V at the output of the MAX when the K-Line is 12V, and -10V when the K-Line is 12V. When no PC is connected, it's -8.5V, which should be okay.
At pin 8 of the MAX, I'm measuring -5.7V. When I apply +5V to this pin, the LED lights up, and the K line goes to ground.
The L-line has a voltage of 12V, and when I apply 5V to pin 13 of the MAX, the L-line also goes to ground.
Everything should be fine then.
I also connected an oscilloscope to the K and L lines. When I try to connect, I can see level changes between zero and V_batt in both signals, and both signals, K and L, are synchronized.
I use a Siemens Mobile 510 AGP laptop as my personal computer.
Can anyone tell me what else I can do to revive my adapter and, more importantly, my airbag?

THX
Nick.

Hello,
I've read on this forum that there are often problems with the runtime environment.
I took some measurements on my adapter:
K-Line to L-Line:
Rising slope: 33-40 us.
Falling flank: 31-36us.
At 2400 baud and a bit time of approximately 416 microseconds, this should absolutely not be a problem.

I don't know what else to test anymore.

Sure, here's the translation:

'Hello' or 'Greetings'
Nick.

Which VCDS (VAG-COM) version are you using? Are you able to access the engine control unit (ECU)?

Hello,

I am using VAG COM version 208.1.
I can't access the engine control unit.

Regards,
Nick.

Hi.

Quote:

I took some measurements on my adapter: K-Line to L-Line: Rising slope: 33-40 us. Falling flank: 31-36us. With a baud rate of 2400 and a bit time of approximately 416 microseconds,

this absolutely shouldn't be a problem.

Hmm????

There shouldn't be any connection between the K-line and the L-line, in my opinion.

Or are you referring to the runtimedifferences between RTS --> L and TXD --> K when there is a synchronous level change on RTS and TXD?

If your adapter already has a difference of almost 40 microseconds , how long are the actual runtimes then, specifically for TXD --> K and RTS --> L ?
On the other hand, even 20 usec with opto-adapters is considered a very slow response time .

Or did you measure in nanoseconds but write microseconds?

Nicklas wrote:

I use VAG COM Release 208.1. I cannot access the engine control unit.

Hi Nick,

Is there actually any speech recognition system that works with this dialogue?

If not: Does a workshop have a connection for a 5051 / 5052 interface?

hello,

I may have expressed myself poorly when I talked about runtime.
'I mean, if I try to establish a connection with VAG-COM without an ECU, just for testing, I can measure the signals mentioned with these delays on the K and L lines.'
So, the K and L lines are moving almost simultaneously.

You've made me suspicious again, so I tested it again, and I can pull the K and L lines separately to GND using the +5V on the SubD connector.

Nick.

Nicklas wrote:

I mean, if I try to establish a connection with VAG-COM without an ECU, just for testing, I can measure the signals mentioned on the K and L lines with these delays.

Hi Nick,

So, that means that when TXD (or RTS) transitions from low to high, K (or L) changes from 12 to 0 volts approximately 35 microseconds later?

(If so: I wouldn't have expected Maik's adapter to be so "slow".)

What rise and fall times can you measure at the K, L, and RS232 outputs of the MAX (in volts/usec)?

Addendum: If you are powering your laptop directly from the vehicle's electrical system, this could lead to problems with your non-galvanically isolated adapter.
Then try it with a PC that is independent of the vehicle's onboard network.

Have you ever tried adjusting the trimmers for the threshold level? Are the correct resistors installed, especially around the K-Line?

Hi,

I mean that VAG Com (software) switches to the L line after the K line, with a delay of 35 microseconds.
Okay, the 35us is software-based.

Rise time of K Line: 1.1 us.
Rise time L line: 1.0 us
K Line data retention time: 100ns.
Fall time for L line: 60ns.

I think the recovery time is much faster because the transistor actively pulls the signal low, and only requires a 4.7k resistor to pull it high.
Measured values are 10% - 90%.

The days of the serial interface:
Rise time: 500ns.
Waste 620ns.
10% again - 90%

Total processing time for adapter:
(RS232 - K Line)
10% increase in RS232 signal to 90% decrease in K-Line signal: 900ns (decrease actively driven).
10% RS232 signal degradation to 90% K-Line signal increase: 5.18us (pull-up).
'Do you think $5.18 is a bit too expensive?'

I hope this helps a little.

Nick.

Somewhere here in this forum (I can't find it anymore), another Audi 80 1Z owner reported that they were only able to establish communication at 240 baud. Normally, Maiks' adapter should work. I use it myself. What hardware (laptop) and operating system are you using? What processes are running in the background?
Your VCDS version is fine. With it, I can access all control units in my 1Z, which didn't always work with the previous versions!

What specific error message are you getting? 'Too many communication errors'?

Hi Nick,

The adapter's performance data seems to be okay.

Hi Nick,
I personally use the adapter on a Golf manufactured in June 1994, with the latest VAG-Com test version, and I haven't had any problems with it. Perhaps you made a mistake when assembling the circuit board? Maybe you swapped some wires or component values. The function test using VAG-Com seems to be working perfectly.

Best regards,

Maik wrote:

Hi nick, I personally use the adapter on a Golf manufactured in June 1994, with the latest VAG-Com test version, and I haven't had any problems with it. Perhaps you made a mistake when assembling the circuit board? Maybe you swapped some wires or component values. The function test via VAG-Com seems to be working perfectly.

Hi Maik,

I was also thinking about that, but if manually connecting the K-line (button to ground) according to his description raises the voltage at RXD from approximately -7 volts to +7 volts, then the adapter seems to be fundamentally okay, doesn't it?

Unless the button is located on the circuit board and there is a break or a reversed connection in the K-cable to the car, for example, K <--> L.

Hello,

The button is not mounted on the circuit board; it was only for testing purposes.
I'm going to swap the K and L pins; nothing should happen since they're all just open-collector drivers.
I also tried setting the 'Start Baut' value to 240, which is possible, but it still didn't establish a connection.
However, when I try to set VAG COM to 230, I get an error message stating that only baud rates between 1200 and 19200 are possible, but it doesn't complain when I set it to 240. Is that a bug in VAGCOM? Or does VAGCOM stop at build number 240 2400?
'By the way, I'm using a P-II 400 processor with Windows NT 4.0 SP6.'
I'm currently in the process of installing Windows 98 SE.

Nick.

Hi,

I also have an Audi B4, model year 1994, with a 1.8L engine, and I can connect to the engine control unit without any problems at a baud rate of 240. However, I can only read the measurement block 00.
However, with other tax devices, it doesn't work with 240, but rather with a baud rate of 1200.

Hello,

I'm now using my old Toshiba T4900 laptop (P100) with Windows 98SE. When I set the baud rate to 240, I immediately got connected, but an error message appeared. The airbag control unit could also be read, and the registered error could be reset without any problems. It appeared to be a short circuit to ground at the indicator light.
I wanted to thank everyone here who helped me.
I will provide a picture of my adapter tomorrow, if I receive my webspace.
Even though Maik might not mind, I'd like to revise his control panel. I was thinking of using optocouplers like the 6N135, which can easily handle up to 1MBaud, and also adding a DC/DC converter.
'In all the adapters I've seen so far, there was also a 100-ohm resistor in the GND line, which makes the optocouplers completely pointless.'
I'll let you know when I have more information.

Regards,

Nick.