Diagnostic connector

Hello!

I have assembled the diagnostic connector for the Ulf's version (after the Ulf's 200-4000rpm test, now also the Ulf's diagnostic connector ). It is correctly recognized by vag-com 0.78b, and the climate control STG can also be read out with it (at 4800 baud).

With the motor control unit, the error message "can't sync baud" always appears. Also, adjusting the baud rate to all possible 1200-step increments up to 19200 baud did not improve anything.

What could be going on there?

Best regards, Georg

Hello Georg,

My friend and I also built the circuit independently of each other. VAG-Com only recognizes the circuit when the VR12 potentiometer (i.e., 0Ohm) is at its end position?
Unfortunately, we both have the same problem: we are unable to read the data. The error message is: No Response From Controller

Did you perhaps experience the same problem and therefore had to set the baud rate to 4800? Have you made any changes to the circuit?

Best regards, Achim

Hello,


Also, bei mir funktioniert die Schaltung nur, wenn der Potentiometer auf 0 Ohm steht. Aber ich erhalte keinen "No Response"-Fehler. He only says "can't synch baud".

Did you turn the ignition on? Is your car one with a K or with the older L-Line?


Best regards, Georg

Hello everyone

Sorry that you are having problems with my circuit

I, of course, didn't have the opportunity to test it in a large-scale trial – but it's working perfectly for me.

The response "Cant synch Baud" appears when the timing between the question and the answer is not correct. Conversely, this means that the response bits are at least fundamentally recognized.
This can be self-induced if the adapter is supplied with only 12 volts on the side and the K-line is manually connected to ground in sequence during the connection process.

How long is your cable between the adapter and the car socket? Shielded or unshielded?

Have you definitely adhered to all the specified component values?

If the adapter is not recognized, this indicates in 1. Line for construction defect.
Example: Incorrect resistor values, especially for R8 (black-green-yellow-gold) and R9 (red-red-orange-gold).
With 15k and 220k, it would be plausible that the thing would, at best, run with great difficulty, if VR 1 is at zero.

"However, since you both need to set VR1 to zero, it seems unlikely that identical false values would appear."
Then simplify the circuit by bridging D3, replacing R9 with a 15kOhm resistor, and completely bridging VR1, also replacing it with a wire bridge.
The adapter remains functional, but the calculated switching threshold for responses towards the PC is then fixed at approximately 6 volts.
It is possible that the potentiometer is causing problems with the transmission clock.


Furthermore, you could try the following changes:

1. R10 to bypass (this increases the high-level by a maximum of 2%)

2. Connect LED3 (and the parallel R4) with a piece of wire.
If the behavior should improve as a result (but not perfectly), you can additionally reduce R3 to 2.2 to 1.5 kΩ (1.5 k = original design).

3. Reduce R7 to 1.5 or 1.0 kOhm.

I had planned to use the higher resistors to reduce the output load of the OK's, and to make the circuit less sensitive to its coupling factor.
Unfortunately, this also makes the adapter "slower", which, as I said, didn't cause any problems for me.

If you try this, I'll be eager to see your results.

Hello,
The circuit works if you replace the resistor R3 with a 1.8kOhm resistor. Also, the potentiometer range is also okay, but only when the switch (100kOhm) is in the off position, the adapter is not recognized.

The real purpose of the DIY project was that I couldn't log into my original adapter from DRV - The starting amount should be increased. After entering the login code 12233, the error message appeared: 'Too Many Communications Errors To Continue!'
I have tried three different computers in total - no luck! My friend's car works perfectly.

Since the login with the homemade adapter still doesn't work, I still believe that the login code is incorrect - this has also been written in this forum, that Audi supposedly changed the code!!
If anyone has connections to Audi, please check the manual first! The part number for the engine control unit is 4B0 907 401 C2, the software is 01012, and the engine is an AKN.

Greetings and thanks
Achim

Problem solved - PC instead of laptop connected and it works.

Georg_G wrote:

Problem solved - PC instead of notebook and it works.

Hi Georg & all,

I conducted some dry runs with the adapter on my PC and made oscilloscope measurements: adapter detection, LED test. Quite revealing...

Apparently, the Signal pins have quite "soft" impedances, so that the adapter must be specifically designed for each computer. Reducing R3 is one of the possible measures, but from another perspective, it could "backfire".

Furthermore, slight blocking currents flow through the opto-outputs, which further distorts the levels.

This means that a deliberately low signal can sometimes end up in the undefined range between +3 and -3 volts (if I have correctly understood the threshold voltages on the COM port).

All of this makes it quite difficult to find component values that work flawlessly on every PC and laptop.

As a possible solution, I am currently, for example, adding additional diodes to the schematic, which have made the signal levels "more distinct" in my adapter and PC.

I will send Rainer an updated plan soon.

For those who want to try it out in advance:

1. Insert an additional diode (cathode to R1) between pin 7 of the PC connector and R1

2. Insert an additional diode (cathode towards the PC connector) between OK 3 (Emitter pin 3) and Pin 7 of the PC connector

3. In parallel to the LED input of OK 3 (Pins 1 + 2), insert an additional resistor of approximately 2.2 kOhm.

This would mean that R3 must remain at 2.7 kOhm.

I don't know yet if this is already all the possible changes

Hello Ulf,

I have tried the following changes that you suggested:

Quote:

For those who want to try it out in advance: 1. Insert an additional diode (cathode to R1) between pin 7 of the PC connector and R1 2. Insert an additional diode (cathode towards the PC connector) between OK 3 (Emitter pin 3) and Pin 7 of the PC connector 3. In parallel to the LED input of OK 3 (Pins 1 + 2), insert an additional resistor of approximately 2.2 kOhm.

R3 was also set to 1.6kOhm.
Now, the interface has been successfully recognized by all VAG-COM versions on the laptop. Previously, this was only the case with version 0.78beta.
I was also able to communicate with all the STGs. However, only when the engine was stationary. While the engine was running, there were always "too many communication errors to continue".

Then I tried the following change, which you suggested a little earlier, to make the interface faster:

Quote:

2. Connect LED3 (and the parallel R4) with a piece of wire. If the behavior should improve as a result (but not perfectly), you can additionally reduce R3 to 2.2 to 1.5 kΩ (1.5 k = original design). 3. Reduce R7 to 1.5 or 1.0 kOhm.

As previously mentioned, I have reduced the value of R3 to 1.6kOhm.

That way, it also worked while the engine was running.


Question: The remaining two LEDs on the optocouplers should also contribute to reducing the speed. In the original design, there is also a direct connection from R1/R2 to pin 1 of the OK. "However, LEDs switch on and off relatively slowly compared to, for example, a 1N4148."
Should I perhaps remove them as well?



Thank you for your help, Ulf.

Greetings, Georg

Hello

I also replicated the adapter, but it didn't work.
Then I reduced the resistor R3 to 1.5k and turned VR1 to 0 -> it didn't work with my laptop, but it did work with my friend's laptop. So, it must have been a problem with my laptop. Then I replaced R2 with a potentiometer, and at a setting of 140 Ohms, I was getting signal everywhere except for the internal room monitoring.

Okay, now here's my actual question:
On my Seat Leon Cupra TDI (Engine Code: ARL), I want to achieve an average fuel consumption, but the BC (on-board computer) shows more than the car actually consumes. Where can I find this? (Like on Channel 3, it's not the same as in the Golf).

Thank you in advance.
Bartolomäus

Georg_G wrote:

Question: The remaining two LEDs on the optocouplers should also contribute to reducing the speed. In the original design, there is also a direct connection from R1/R2 to pin 1 of the OK. "However, LEDs switch on and off relatively slowly compared to, for example, a 1N4148." Should I also remove it?

Hi Georg

Hm, do you think the LEDs 1 and 2 are significantly "slower" than the transmitting LEDs in the optocouplers?
(I don't know, just asking)

If you want to bypass the inherent inertia of the control LEDs in every case, simply connect 1 capacitor in parallel.
"I don't really know how big they need to be, but I wouldn't start with anything under 1 µF, just a guess."

My oscilloscope measurements on the adapter have led me to a rather different conclusion:
I now have 4 more diodes and 1 resistor more in the circuit to partially relieve the COM port and make the signals cleaner = more distinct, which also works according to the oscilloscope.
This version works perfectly for me - however, the first one didn't. Draft: Also, I can't provide a reasoned forecast about the success of on other PCs

I just sent an email to Rainer with the new schematic; I'm also sending it to you again.

BartW wrote:

Then I replaced R2 with a potentiometer, and at a setting of 140 Ohms, I got a signal everywhere except for the internal monitoring.

Hi

140 Ohm for R2 seems too low.
Depending on the current supplied by the TXD output of the PC, the transmit LED in the optocoupler can receive up to approximately 60 mA – that's quite a lot, and I don't know how long the component can withstand that.

Also, the control LED probably needs sunglasses too??

I'm also sending you the new version of the adapter.
With a bit of luck, it will also work with 1k for R2 - before your optocoupler is fried...

Hi

It works for me, but only with 140 ohms. If I choose a higher value, the adapter doesn't connect to any control unit.
The control LED has normal brightness.

Hi Ulf
Thank you for the post, I will try it and then report back.

Greetings
Bartolomäus

BartW wrote:

However, it only works for me with 140 ohms, if I choose a higher value, the adapter will not connect to any control device. The control LED has normal brightness.

Hi Bart
then the COM port might be a bit weak in terms of power delivery, so that 140 ohms might work...

Have you really used the specified 2.2 kOhm (red - red - red) resistors and not, for example, red - red - brown = 220 Ohm?
Sometimes, red and brown are almost indistinguishable.

Hi Ulf
have red-red-black-brown-gold = 2.2k taken.