VAGCOM/OBD2: Optocoupler Adapter for Old 1Z

Quote:

Passat Variant, year 1994, 1Z.

It's working quite well (with Maik's adapter). https://community.dieselschrauber.org/en/viewtopic.php?t=1978 However, lately, and rarely, it sometimes shows "too many communication errors".

From JörgS at .

Hi Jörg,

Your engine control unit might be one of the rare cases where communication between an older 1Z model and VAGCOM isn't working correctly.
In my opinion, the fact that communication behavior is changing suggests that certain parameters are within the error margin and are shifting back and forth depending on factors like temperature, voltage, or something else entirely.

If you have the time and inclination for some "basic research," you could try building and calibrating the Opto-Adapter 2.3, or even better, the 2.4, and see how it performs.

It's possible that the sluggishness of the optocouplers and the resulting signal delay could be helpful here, as they provide the motor controller with a brief pause, not foreseen in the program, between its response and the arrival of the next message from the PC.

Perhaps the engine computer is also requiring or sending "incorrect" bit lengths?
They can be independently extended and shortened in both directions (PC to car and vice versa) using VR 2 and 3, and somewhere in between, the length errors are also eliminated, which is actually the intended purpose of the trimmers.
Somewhere in the upper adjustment range, the communication will likely break down if the reduced input current no longer allows for usable signals at the OK output. In that case, the adapter will also no longer be recognized.

Unfortunately, aligning based solely on the brightness of the LEDs only approximates the correct bit length. Achieving a truly precise alignment of the adapter would require the impossible software that measures the signal propagation time between TXD and RXD.

The trimmers in the pull-up circuit of the K-line and for the H-L detection threshold, which Maik had incorporated from my previous design suggestions into his adapter, offer :oops:only relatively minimal or, at the end of the adjustment range, sometimes unreliable adjustment options:wink: for signal delays or bit length distortions, because even small interference pulses can disrupt the timing in those areas.

If I have inadvertently posted some digital nonsense here, please excuse me - unfortunately, my electrochemical brain processor only works analog }.
Relevant (helpful }) suggestions are, of course, welcome.

@ulf

I found an ECU (Engine Control Unit) with the part number 028 906 021 C while I was working abroad. I have no idea which vehicle it came from. It's built exactly like my own 028 906 021 B. The engine in my Passat runs perfectly with it. However, when I try to connect with VAG-COM, I encounter the same problems! The other ECU that works flawlessly with VAG-COM is the successor to this C-type ECU, and it has a completely different design. Since I'm experiencing the same problems with two different ECUs, I'm starting to doubt that it's just a coincidence. I used to think that either the interface in the ECU itself or the ECU's software was the problem. As you mentioned, it could very well be an adapter problem. I'll build another one soon. This time, I'll use version 2.4. I think it's likely a combination of both ECU and adapter issues. We'll see when I get around to it! By the way, do you already have a PCB (Printed Circuit Board) design for this?

@maik
Okay, so, your adapter works quite well. The problem here seems to be with the older control units. ROS-Tech is experiencing similar issues. They even warn about potential malfunctions in 1Z vehicles!
And because of these problems, the new adapter from Ross-Tech, which is supposed to solve these issues, is probably being released in the first quarter now!

joergs wrote:

I'll probably build another one soon. But then it will be version 2.4. I think it's really a combination of maybe STG and adapter problems. Let's see when I have time for it! By the way, do you already have a PCB design for that?

Hi Jörg,

No, I haven't been dealing with PCB layouts yet, mainly due to time constraints.
I build my adapters using universal circuit boards.

Hi Jörg,

Have you already checked the sizes C 5, 10, 12, and 13 for your May adapter?

/viewtopic.php?t=1978
Post from HGK, January 16th.

If your problems are not caused by this and even a slow 2.4 GHz adapter doesn't provide a solution, you could try the opposite extreme: using the shortest possible connection times between the car and the PC.

Maik's adapter is likely significantly faster than the 2.4 version, but the path of the bits through the MAX chip, a logic inverter, and a transistor stage probably still takes a few microseconds (although I haven't built the adapter, let alone measured its performance).

A minimal circuit could be even faster, consisting of only one transistor stage for each direction, PC to car and vice versa. In terms of speed, it would be similar to Maik's adapter, but without the MAX and 7404 chips.

A corresponding circuit diagram, in my opinion, is ready for practical testing, but I don't know if or when I will have the time to do so.
If you're interested, I could email you the JPG file, and you might be able to take on the role of a test pilot.

This offer is valid for all 1Z drivers whose engine control units cause problems when connected to a VAGCOM diagnostic tool using older adapters, as well as for anyone else who may be interested.
However, as is my usual practice with these designs, there is no PCB layout included, only the bare schematic diagram with a few explanations and a calibration guide.
It's significantly simpler with only one trimmer compared to the 2.4 version, and it should automatically adjust to the correct bit length and full speed once the adapter (hopefully, as it hasn't been tested yet) is working correctly.

hello ulf,

Just send it over. I really want to know now! I've already invested so much time! There's no such thing as impossible!

Please send me the JPG. Do you still have my email address?

I haven't changed the C's yet. I'll do it sometime in the next few days! Maybe it will also work with the Dino control units then!

joergs wrote:

Please send me the JPG.

Do you still have my email address? }

Hi Jörg,

Yes, the one from the other day.

I'll send it right away.

I think I have a solution for my problem! See also the thread mentioned above!

However, I will still be a test pilot for Ulf's ultra-fast adapter! I now know how to trigger the error.

Certainly, adapter runtimes also play a role. But one thing I know for 100% since tonight . Ross-Tech has significant timing difficulties in VAG-COM! Ultimately, the entire project stands or falls based on the quality of the adapter. But the adapter is just one piece of the whole mosaic.