500A Controller troubleshooting

[Lectrik]

thingstodo

Thanks for looking at this.

I intended to post a screen shot of RTD Explorer that showed I connected but never did. I should have stated that I got

"Cougar OS controller firmware v1.1b".

I dropped the baud rate to 9600 to see if that would help but no change.

Ne response from command line, throttle control etc. All values above the terminal window are at zero.

[MPaulHolmes]

While at zero throttle, Put a volt meter from any of the 10 gate resistors (either side) to B-. Then give it full throttle and see if the voltage climbs to 12v.

[Lectrik]

Hummm

I checked all gate resisters to B- and get zero volts. Checked them from zero throttle to full throttle and no change.

There is no traction pack connected. Just for the hell of it I measured across B+ and B- and I read 12.81 volts. That is more than my supply battery. Maybe it is from the caps from when 12v battery pack was connected.

Connected a 12v traction pack and checked all 10 pull-up resisters again. Zero volts between resister and B- in any position of the pot.

[MPaulHolmes]

Ok so now we move up the pipeline until the badness reveals itself. If both leds are on solid, it is likely not a fault like current sensor not hooked up, or throttle out of range. I'll write more in a little bit. Going for walk with wife to get laundry off the line.

[Lectrik]

I agree. I have had the LEM disconnected and the yellow LED will flash. Also maxex throttle at power up causing the same.

[MPaulHolmes]

OK, measure the voltage from the top of R11 to the back metal face of U1 (that's a nice accessible ground). By the top of R11, I mean the side of R11 that's the furthest away from the 10 resistors GR1, ..., GR10. That's the PWM output. Measure the voltage of that at zero throttle and at full throttle. It should go from 0v up to 5v. If that's OK, the next measure the voltage from the left side of R15 to the back metal face thing of U1 at zero and full throttle. If that goes from 0v to 5v, then measure the voltage from the bottom of R16 to PH1 or PH2 (or B- bus bar).

[Lectrik]

R11 reads almost zero volts (.003) at no throttle and 5 volts at max. Actually it is not linear. There is a slight gradual rise and then jumps up.

R15 seems to react opposite of R11. At minimum throttle it reads 5 volts and then drops ..338 at max throttle. The leg closest to U6 reads 4.97 at min throttle and 4.91 at max throttle.

I measured from the bottom of R16 to PH1, The PH2 The B- and got 11.82 volts with no change from throttle control. Checking from either side of R16 has same result.

Is this implying U6 is bad? I did check to make sure it isn't installed backwords.

I will isolate both u6 and u7 and check for shorts ie; U6 pins 8&6, 5&6, U7 pin 1&2

[MPaulHolmes]

OK good. So it's definitely not a fault from the NAND gates. The signal is making it to U6, but isn't making it to U7. Yes, I think U6 is bad. If you get a replacement, I would get the hcpl-4506 instead of hcpl-4504. That's a newer part that has a better current transfer ratio, and works a lot better.

[Lectrik]

I just ordered the hcpl-4504 ten minutes ago.

It seems my soldering skills are't bad, my de-soldering skills are terrible. I forgot how difficult it is even with solder wick and a sucker. Broke a leg off.

I'll post as soon as I have a result after parts come in.

A big thank you to you Paul and everyone else who contributed. I really appreciate all the help.

John

[Lectrik]

I received my parts yesterday and got everything put back together this morning.

I replaced U6 with HCPL-4506 and also replaced U7. If anyone needs some 4504's I have two.

After connecting to RTD explorer and powering up I get ...
Cougar OS controller firmware v1.11b

There is no data displayed and no response from command line.

Back through the troubleshooting steps we covered earlier everything appears to be working. I can see throttle changes all the way to pins 6-7 of U7.

My next steps are to reverse my transmit and receive wire on the RS232 connector and put it in the car and see what happens.