OBD MPGuino gauge

[mattjreilly]

Got it Froggy, thanks. I'm going to try jumpering the L line tonight and see what that does.

[Froggy]

Hopefully it will be enough to bring it to life... I still have not made up a dual 33290 interface (where the L line stops), so my L line still blathers away all 10400 outgoing data...

If it does connect, check also that turning off the car and back on will resume communications. Some ISO 9141 will happily resume talking, but my car needs to re-init.

[dcb]

FYI, the elm 323 interface has the L line controlled by a separate pin

[Froggy]

Yes, but the 323 is a more expensive solution.

Indeed, I thought I would have to go that route, but splitting the K line signal to include the L line worked for me.

[dcb]

I meant that they are doing more than just splitting the L line, perhaps for a reason.

[Froggy]

Yes, accordingly, the L line is used just for initialization. After that it should not be used. By splitting K into K and L, we do start the init ok, but then we continue transmitting 10400 baud data on it. This could on some cars pose a problem (if they take it as an error state, or as an attempt to reconnect or whatever...) Jumping K to L to drive L is a quick fix, and it may work, but the solution goes against the standards which could leave a car doing strange stuff.

If the two ISO chips work (one for K and one for L) the extra cost is very small compaired to going the ELM 323 or ELM 327 route just for the L line. I just need two resistors before I begin constructing it and testing it.

[dcb]

not two chips, just another pin and just change the 5 baud init code to twiddle both K and L pins then set the L pin high right after init.

[Froggy]

I think you would need the chip to drive the line to 12 volts which is specified as ON.

The arduino would drive it high only to 5 volts. The chip would drive it to 12 volts.

[mattjreilly]

I tried splitting the K & L lines but initialization still fails. Froggy, did you get it to work with the code in the repository or only after you modified it?

Matt

[Froggy]

I got it working right away without modifications. I had to modify it as once the car turned off, I would have to reset the arduino to reinit, which is impractical to do...

Have you verified the chip functions?

With a volt meter you should see the K line high, and drop down to zero volts. Powered by the computers USB, high will only be 5 volts, but that's ok. It will tell you the car will get 12 volts once it's hooked up to the car.

That tells you the TX is working. Next, add some code to the iso_init to read back the RX line. Before you change the K line state (HIGH or LOW) read the K_in line with digitalRead(k_in); do the same after, and the results should differ. If not, you won't hear the car. So check that wiring.

If all the wiring is good, and the chip is working, then your car is either expecting something else before it talks to you, or, you are ignoring what it says.

Try checking for ANY feedback from the car. Change the code to be successful if the car sends any data at all back. See if you still time out or not. If you still time out, then the car is still waiting for what it thinks is the correct startup sequence (whatever that is).