Audi S2 Trigger problems.
Found the cause of the trigger issue (3b/18515) : excessive voltage between (otherwise connected) GND5 - GND confirmed ... protection diode removed ... damage will likely follow
GND5 pins: mot55/10,14,19,30, measure with DVM (power off, ECU unplugged from harness)
- are they connected (shorted) in ECU ?
- are all the connections strong in the harness ? A bit difficult to test with just DVM, >1A current would be necessary for effective test. Eg. max 10mV / A ? (max 20mV drop from the current of a 2A 25W brake light ?) (inductance can still be excessive, but more difficult to measure)
- is the cylinder head grounded properly ?
- is the block grounded properly ?
'''This car has been running vems for 8 years and newer had problems and newer installed bridge or diode. Also I can see that GND and GND5 is shorted back side of pcb so maybe this cores a ground loop.
As I had pre. trigger errors I compared box 18515 with 10045.
I did not find big changes in pre. trigger only R30 was different.
So tested all kind of R30 pull up from 10K to 250K but no luck.
I noted that the computer had hard time to connect and only 2 to 8 Hz update frq.
This can only be noise.
- I investigated the pcb further and find that the GND bridge there was mounted a diode in the new 18515
- Dissoldering this diode and the S2 was spinning again.
- yes, the diode between GND is not necessary for operation. The diode does not do anything if GND-s are strongly connected (<600mV voltage drop).
- I do not know why vems.hu think this is a good ide to root that noise into the processor side, maybe the diode was mounted the wrong way?
- the diode is shorted through the GenBoard/Manual/GroundConnections so it has no effect (regardless of the direction) if the ground connections are good (it might not be that obvious in case of an old motronic55 harness, but the concept is the same)
- the GND5 ->|- GND diode (was it that way ?) provides some protection in case grounds are insufficient or overloaded
- is it possible that perhaps somehow some driven load sinks excessive current into the ECU output (finally to GND5) ? Maybe some of the wires came out, or loose ? With all good grounds, the diode certainly has no effect.
- Ground problem confirmed If the diode has any effect, it is certain that grounds are not strongly connected (simply, a diode has no effect if the 2 legs are shorted), and
- it is likely that problems will develop. It can also explain why the serial connection got worse and worse with the previous box.
- ground problem (confirmed) => more problems are expected (and with diode removed, permanent damage is just a question of time)
Also I have mysterious map readings.
Image will explain that.
End of 4 days of work and testing.
"custom Audi 3B ecu."
- 1: 5 output for logic level vag coils. 50 cm of wire from the ecu to a Econoseal 10 pin in the end of wire´s.
- 2: 2 x knock
- 3: output to carbon canister valve N80 55p pin 5 (please set it up in Vt)
New Audi 3B box v3/18515 arrived today thanks for that.
Made the config file and tried to start.
The new box have trigger errors (less pre. trigger) so NO GO.
What is changed in the box regard to box 10045?
Never had any trigger problems with box 10045.
We compare your project notes (from old and new) and measurements, and look up benchtest records.
Scope is best, but at least DVM DC measurements required.
