OK: The Common Issues | Posts 32+

BERT wrote:

. . . If the power transfer function is too low, the output transistor of the OK circuit will switch too slowly or not completely. A voltage greater than 0.3V should be measured across the output transistor of the OK circuit. In extreme cases, the PC may detect an incorrect input level. Solution: Increase the resistance at the DTR pin, and decrease the pre-existing resistor for the OK-LED. A very large power transmission capability, on the other hand, can lead to a fully saturated switching output transistor, which can result in a possible delay in the OK signal being turned off. Solution: Increase the current-limiting resistor for the OK-LED.

(This may result in a slower "OK" indicator light activation.)

Hello Bert.

What you are describing exactly matches my measurements.
Therefore, I have included trimmers in series with the LED inputs in section 2.3, to allow for adjustment of the optocoupler behavior for each specific PCB.
With the narrow optimal setting (adjusted using a function generator and oscilloscope), the results are quite good - likely much better than what VAGCOM requires.
I hope that the troubleshooting tips in the description of version 2.3 are sufficient to establish a stable connection after some experimentation.

Quote:

Alternatively, use an OK with an exposed base and solder a Schottky diode from the base of the OK transistor to the collector. This prevents the saturation of the OK transistors, regardless of the control current. (Principle of 74LSxx circuits)

Hmm, I didn't know that trick yet... Could you send me a circuit diagram?

Quote:

Measurements from any experiment: OK = IL74 with basic connection; Load 1k at 12V (12mA). Tax signal: 0 -> 12mA. Output delay: 2 microseconds. Control signal: 12mA to 0mA. Output delay: 16 microseconds. Tax signal: 12mA -> 0mA. Output delay: 6 microseconds. Schottky Diode Base / Collector

That sounds quite good and could make the circuit much less sensitive regarding its adjustment.

Moin allerseits, (spez. ULF)

Here's the requested addition regarding the Schottky diode:


|-->I--| C |
| / |
| |/ |
B------| |
|\ |
\ E NPN

The image shows a transistor with a Schottky diode.
Examples: OK with basic connection of the phototransistor IL1,2,5,74; A4n25....

Function:
- The transistor is driven with more base current than is strictly necessary for the collector current, in order to ensure a faster switching speed. However, this leads to saturation when the transistor is in the 'on' state, which needs to be discharged when the transistor is switched off.
- Now the diode comes into play. When the transistor is switched on, its collector voltage is less than 0.2V. The Schottky diode has a threshold voltage of 0.3V. Any excess base current that is no longer needed to switch the transistor on flows through the diode.

! Depending on the operating conditions, T1 and T2 may also be subject to significant overdriving (RC > RB).

To avoid misunderstandings: The diode solution is not a miracle cure. It can only eliminate the rise time delay, but not a slowly rising slope. The only solution for that is a smaller collector resistor or a faster comparator.

! Generally: The higher the CTR (Current Transfer Ratio), the slower the OK (Overall Knockdown). Exceptions confirm the rule.

'However, the OK-compatible adapter is likely to cause fewer problems here.'

If there is indeed a need for a potential-isolated and faster adapter, we should consider developing a new version that is easy to replicate and does not require individual tuning.

best regards

bert

BERT wrote:

Moin allerseits, (spez. ULF) Here's the requested addition regarding the Schottky diode: |-->I--| C | | / | | |/ | B------| | |\ | \ E NPN

Hi Bert.

Nicely drawn... you can only see what's meant in the "quote" window, but I understood it.

Quote:

! Depending on the OK status, T1 and T2 may also be subject to significant oversteering (RC > RB)

.

...which, however, are peanuts in terms of runtime compared to the pace of completed OKs.

Quote:

! To avoid misunderstandings. The diode story is not a miracle cure. You can only eliminate the rise time delay, but not a slowly rising slope. The only solution is a smaller collector resistor or a faster OK.

Understood.

The alignment of adapter 2.3 without an oscilloscope and a function generator is still somewhat uncertain.

I'm going to try connecting the control LEDs from the OK1 and 2 inputs to the output circuit (in series with R10 and R13, respectively).
When you adjust the trimmers in the LED test menu of VAGCOM so that each LED is slightly dimmer than at the zero point, you should be just leaving the saturation range. This ensures that the shutdown delay is short enough and the startup delay is still short enough to avoid communication problems caused by runtime.

What do you think about it?

Quote:

If there is actually a need for a potential-separated and faster adapter, one would need to consider a design that is easy to replicate and does not require individual tuning.

If you can develop something there, feel free .

Hi everyone.

ULF ->

I'm going to try connecting the control LEDs from the OK1 and 2 inputs to the output circuit (in series with R10 and R13, respectively).
When you adjust the trimmers in the LED test menu of VAGCOM so that each LED glows slightly darker than its zero point, it should...
<-ULF ( how does the quotation mark - thing work)


Yes, that might work. If I remember correctly, it's quite a complicated process.
It would be beneficial to write a program that can control the COM port either statically or dynamically (replacing the function generator).
Once things calm down again (after the year-end job rush), I'll take a look.

Another option without extensive adjustments:
Use parts with a CTR (Current Transfer Ratio) of 50% or less for OK1 and OK2. (IL74 4N26 4N27)

-> Stom OK in = 10mA (thumb)
-> Maximum CTR (current transfer ratio) = 5mA on the automotive side; no saturation.
T1, T2: Current gain of BC547B: approximately 200 at Ic=10-20mA (C type: approximately 450).
-> With a base current of 5mA, it is still possible to drive a collector current of 1A.
To ensure it still works even with a CTR of 10%, the base bid should be increased to 6.8.

! A check/comparison might still be necessary, as the 'OK' status can be very inconsistent.

! The PC817 has a CTR of approximately 200%.

! With the current configuration, the OK3 requires a CTR (Current Transfer Ratio) greater than 100% to achieve the required voltage level on the PC side. Alternatively, the resistors need to be adjusted.
It doesn't have a post-amplifier transistor; instead, it uses overdriving and a diode.

best regards

BERT.

BERT wrote:

( how does the quote - thing work)

Hi Bert.

Don't you see the "QUOTE" button at the top of each post?
"This will copy the entire post as a quote into the reply."
What is not needed --> delete.

Okay, hier ist die Übersetzung:

"Translate the following text from German to English. Provide ONLY the translation, no explanations."

"Or copy and paste individual passages into the answer, highlight them, and click on the inconspicuous quote box at the top. Then, quote brackets will appear around the text that will be displayed as a quote in the post."

Quote:

Yes, that might work. If I remember correctly, it's quite a bit of flashing.

Unfortunately, it's not quite that simple... I once checked with an oscilloscope to see how much the LED in the opto-output circuit needs to be dimmed at a Utxd voltage of 12 volts in order to achieve a "symmetrical" signal delay.
With "slightly darker" (meaning not a fully saturated OK output), asymmetries of approximately 10 microseconds were still present .
Only at approximately 2/3 of the maximum brightness were the + and - flanks uniformly delayed by 4 usec .

Quote:

-> Stom OK in = 10mA (thumb)

Here we go again with the major uncertainty: the significantly different impedances and output voltages of the various COM ports likely make such rough estimations very unreliable .

I think trimmers are essential in the input stage if you want to achieve reasonable operating times.
And with the trimmers, it is possible to simultaneously "compensate" for all possible combinations of COM impedances and OK types.
Unfortunately, the "fine-tuning" will likely look like this in practice: it's acceptable if VAGCOM is running .