Corvette CCM Reverse Engineering Anyone?

[spfautsch]

All the code in the car itself uses metric measurements for temperatures, and I have confirmed via the ADX files for 1990-1996 that the equation never changed from 1990, which was "Count x 0.75 - 40 = Deg C".

Not trying to refute this as it's been a couple seasons since I looked at it, but my recollection of how the PCM handles temperatures is that it uses a proprietary unit that allows the module to deal with a wide range of temperature readings with a single byte (0-255). I referred to in in my notes as 'engineering units'. I'm fairly sure the temperature based table lookups used this unit raw and the .xdf converted it to a human readable conversion. I only recall this because I spent a considerable amount of time figuring out how to determine which ECT voltage divider / range the PCM was using based on cranking spark advance for the diy-ltcc project. I was certain enough this was the case I fired off a question to Banish (who I believe did some calibration work for GM at some point) asking if they had an official unit name for this.

There should be comments in the diy-ltcc source pertaining to this. I certainly recall the (temp_units * 0.75) - 40 = degrees celcius equation. If I get motivated to look I'll dig into the source and try to refresh my memory.

[spfautsch]

Here you go - this is from the routine that feeds internal controller data to the serial port for logging. If memory serves, the 'engineering units' was chosen for the controller's internal use to match the ranges of the stock $EE spark vs ect tables.

            case 'C':
              // coolant temp C
              Serial.print((ectValue * 0.75) - 40);
              break;
              

            case 'F':
              // temp in fahrenheit
              Serial.print((ectValue * 1.35) - 40);
              break;

[-=Jeff=-]

I think I know where I might have gone wrong. I am going to look at my code tonight

[NomakeWan]

Not trying to refute this as it's been a couple seasons since I looked at it, but my recollection of how the PCM handles temperatures is that it uses a proprietary unit that allows the module to deal with a wide range of temperature readings with a single byte (0-255). I referred to in in my notes as 'engineering units'. I'm fairly sure the temperature based table lookups used this unit raw and the .xdf converted it to a human readable conversion. I only recall this because I spent a considerable amount of time figuring out how to determine which ECT voltage divider / range the PCM was using based on cranking spark advance for the diy-ltcc project. I was certain enough this was the case I fired off a question to Banish (who I believe did some calibration work for GM at some point) asking if they had an official unit name for this.

There should be comments in the diy-ltcc source pertaining to this. I certainly recall the (temp_units * 0.75) - 40 = degrees celcius equation. If I get motivated to look I'll dig into the source and try to refresh my memory.

Apologies for my imprecise language. You are correct--it's not really that the PCM/CCM are using metric units. It's that the way the human squishy bit interfaces with those units is given, by GM, in metric units. The equations provided in technical documents about how to interpret those "engineering units" are all metric. That's what I meant, but it's not what I said.

[-=Jeff=-]

So I am not sure why, but i was doing something convoluted.. I used the above calculation now from spfautsch. Sometimes I just can't see the simple solution.

As for my comment of it not being needed, I happen to be testing at a temperature that raw was within 3 of the actual, so it 'looked' good

[-=Jeff=-]

made some display improvements today after finding that it was hard to see yesterday on my test drive

https://youtu.be/mldRp7XDwUk

[-=Jeff=-]

Thanks to all the efforts in this thread I have this:

[NomakeWan]

That is so beyond cool. Way to go!!

[-=Jeff=-]

New screen in progress.. thinking of making an page for the ALDL full stream

https://youtu.be/a_M40EqBYjo

Scott, going to look over your code again and make sure it will work for the 1990 and then give it a try

[-=Jeff=-]

I have a friends CCM, powers on them about 30-60 later shuts off. I know he cooked something but don't see any signs of a cooked part, anyone have schematics for these?

[NomakeWan]

As far as I know no schematics exist for the CCMs (outside of GM). Also, 30-60 seconds, or 30-60 minutes?

[-=Jeff=-]

Oops sorry 30-60 seconds

[-=Jeff=-]

Setup my bench to read a couple CCMs. but I keep getting the Bus is too Noisy.. I have not changed anything on my setup. Suggestions on what to check?

EDIT:
1990-1991 works but 1992-1993 keeps giving the Bus to Noisy, then is says it cannot find the Heartbeat. I have another port setup and running YAT to read the data, the CCM seems to be sending the commands as expected, but for some odd reason, the Arduino setup to fake the messages is not working. As I thought having the reply messages will help. I also tried driect connection to CCM, (No ECM present)

[-=Jeff=-]

After reading through the post and fixing some Arduino code, I got it working. I need to improve my bench setup and clean the wiring up

[In-Tech]

Hiya,
What was the secret find? Serial?