Jaguar Prototype Testing

[Dan]

DeuceEFI - might I suggest leaving the opto's out for the moment, until we chase up another part that is fast enough. I created a thread in the RavAGE section primarily for this, as it relates directly to RavAGE schematics.

[DeuceEFI]

ok, 1K on CPU side then ! I forgot not all people have cpus to destructively test!

there are only 4 pins, but he added extra resistors and stuff.

Ok so they are working, but just not fast enough???

I don't mind abusing one of my low cost Freescale sample MCUs for testing

I originally placed the optos on the PCB and ran them thru the reflow cycle and installed the resistors exactly like the Ravage schematic. In this configuration all I saw as output to cutecom was 0x00 repeating forever.

I have only been testing with the TX from the MCU as the streams out data at a good clip to cutecom or MTX. Originally I had the RX opto installed, but I was unable to flash the firmware with it installed so I removed it to verify it would work without the optos in the circuit. At this point the RX is directly connected from the MCU to the FT232RL chip.

I originally tested the circuit exactly as drawn in Ravage, even using optos that had not gone thru the reflow soldering process on my solder less breadboard (I soldered them to a 0.100" 4 pin header strip), but when I saw it was not working I tried lower value resistors in steps down to 220ohms. At that point I began testing with the test circuit that was in the data sheet and at least it did not invert the data as the Ravage circuit had done, but even with the best combination of resistance values that I tested it still corrupted the data stream.

With that being said, since I have this setup for testing I will test any combination of resistances and configurations on my prototype board for the sake of research, just let me know what values you would like to try and I will save the LA trace and upload it for review. I want this to work as much as others do

I will continue my testing of the other circuits without the optos installed, just let me know if there is something that you would like for me to test and I'll try it...

[Fred]

The rav circuit shouldn't have inverted the data afaik. Or is it wrong?

As for research, this has become a parts selection problem now :-)

[Dan]

I am also concerned that it was "inverting" the data?

If you have tried lowering the resistance on all resistors and had no luck, I am more thinking that Fred is correct in saying that they are just simply too slow.

[DeuceEFI]

OK, I have a correction to make, the Ravage opto circuit did NOT "invert" the data, it just didn't work in that configuration as seen by my LA trace. Sorry for the confusion, when I first looked at the data it appeared to be "inverted" until I investigated further... Lowering the resistance values of the pull up resistors did not improve the way the circuit functioned.

Once I switched over to the CEL/NEC test circuit configuration without the pull up resistors the output from the opto showed data, but the data was incorrect since the opto was unable to switch from ON to OFF as quickly as it could switch from OFF to ON.

[DeuceEFI]

Here are the circuit configurations that I used for my testing...

[Fred]

Andy, all you did was change the default state, which would prevent the engine from running without the FTDI powered up. All other changes that you saw were from resistor value or current capacity in the CPU varying from push to pull. Make sure you leave the circuit as in RavAGE or it will not work.

[TonyS]

Note that the FTDI Input Switching Threshold spec is ~ 1.2V. I don't know if you can change the LA threshold to match, but it might provide some insight as to when the FTDI RXD pin "sees" valid highs and lows.

[TonyS]

Andy, all you did was change the default state, which would prevent the engine from running without the FTDI powered up.

Hey Fred, I just want to make sure that I understand what is needed -
The default state / no data going into the RX pin of the uC should be a steady logic high -yes?
What is the targeted maximum baud rate?
Thanks,
Huff

[Fred]

Standard baud is 115200, though the firmware would be happy going much faster, no one will need it really.

Default state without any current through the opto should be logic high on the CPU RX pin. Yes! :-)

Fred.