CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.

TPI Install Troubleshooting

BCCHowdy

Registered Member
Joined
Dec 27, 2019
Posts
11
Reaction score
5
Location
Tucson, AZ
Hello!

I just finished a TPI swap on my '85 Chevy Blazer and would definitely appreciate help troubleshooting the gremlins I have. I had this same running setup in an '84 GMC pickup which sat for 10+ years. I pulled the motor and hardware, ECM was trashed (mice), rebuilt the motor and got a new Painless wiring #60102 for the install. Everything is in, wired, connected and I got an 1227165 ECM from a scrap yard out of a corvette.

When I turn the key to the 'on' position the fuel pump comes on and stays on. It should only cycle for 2 seconds then off, but it just stays on the entire time. Jumping A & B on the ALDL connector does nothing, no codes, not even 12. I also remember when keying on and the pump cycles for 2 seconds the SES light should come on until engine start, but nothing. I tried connecting TunerPro RT with my laptop and ALDU1 adapter... nothing. Engine will crank but no fire. I figured with the ECM not even giving me code 12 when I jumper it something else is going on. Since the ECM was a scrapyard pull I swapped it with a reman with the same result. (wasted $$$)

One thing that's irritating about the Painless Wiring kits is you don't get an actual schematic. It's nice that everything is labeled for install but once it's installed, loomed, and tucked away where it's supposed to be it's difficult to troubleshoot. There are two capped fuses in the Painless Wiring harness, one with red wires with a 30 amp which has power, the other is a 15 amp with purple wires and no power. I think this might be the crank wire from the starter solenoid so that makes sense. I didn't check for power on the 15 amp when cranking.

B1 connector at the ECM has power
A6 connector has power when key is in the 'on' position
I did not check for power at C16

I did not connect the oil pressure switch wires to anything, they're tied off.
I removed the fuel pump relay and have power at on the green/white wire when key is on as expected, but it stays on which is unexpected.
I checked for power at on the green/white wire and also the tan/white wire with the ECM removed per another thread to look for voltage that shouldn't be there.

It's like the ECM is only half powered. I don't get it. Grounds are good. I cleared off the paint on the firewall and bolted the ground eyelet there. I measured no voltage drop from there to the negative terminal on the battery. I even ran a wire from the eyelet on the firewall direct to the battery to eliminate any other potential ground issues. No change.

I'm not sure what else to try, but this is super weird behavior. Any suggestions would be greatly appreciated.

(I posted this same message over at ThirdGen.org as well looking for help)

-Howdy
 
It's my EEPROM! Not sure what, but something with my chip is no good. Either the chip itself is bad or I programmed it incorrectly. Wrong definition or wrong offset maybe? I'm using a 27SF512 with an adapter. Finally the light bulb went off that maybe I should take my chip out and try the bone stock memcal. The SES light comes on like it should with key-on and the pump primes for 2 seconds then off again.

Now it turns over like there's no spark. I have an MSD 6A box and I'll move on to troubleshooting that next. Getting closer!

-Howdy
 
Ha! I've had a .bin become corrupted before and that had me chasing my tail too. When I see the check engine bulb light fail to blink, I know I'll hear the electric fan come on immediately afterward, meaning there is a PROM issue.

I've noticed that burning the chips I get a failure to erase (failed blank check) or bin not matching what was uploaded. I cant tell if its hardware or software, but usually re-sesting the chip in the burner solves it, if not, shut TP down and restart, plus unplug the APU1. Also had it where emulation wouldnt work at all
Have not tried that again to see if it was a fluke or something with the hardware that went bad.

Glad you got it!
 
I got it! I had a number of problems:

1. Tested the MSD box per the manual and it seems to work fine.
2. The A and B wires on the 2 pin distributor connector were reversed. Fixed that.
3. I attempted to start with the stock memcal which was pulled from a corvette but I don't remember what year. Still no start. I was able to hook up tunerpro though and the dashboard read all kind of crazy values like rpm 6,500 while not running, tps voltage 4.87. (Guessing my definition file doesn't match the data feed) So I went back to troubleshooting my tune and chip adapter. I'm using the 27SF512 chips with GP1 adapter. Verified the file I loaded matches what's currently on the chip. The instructions for the BURN1 programmer states that the chip divet must be towards the handle. Surely this holds true for the GP1 adapter right? No... I had the chip in backwards.

I put the chip in correctly, keyed on, pump cycled for 2 seconds like it should and ses light on like it should be. Tunerpro reported legit values for the sensors. Fired right up. Did a 30 minute break in and a short test drive and it seems to run pretty good. I'll do some datalogging and tuning next weekend.

-Howdy
 
Yep, I actually etched the location of the chip notch into the ECM housing and the APU1 so I didnt make that mistake again!

I just had a write error to one of my chips, it appears to be a chip issue. I'd blank it, then the blank check would fail. Reseat the chip in the burner, it would pass that, but fail verifying the uploaded file with what was burned to the chip. After playing with the chips location in the socket, it worked fine. I learned after a few frustrating trips out to swap chips on the ECM to always blank, verify the blank, then verify the upload before removing it from the socket. Wont fix a corrupted bin issue, but keeps from making unnecessary trips into the house.

The legs on the chips are so soft, I think perhaps a little misalignment means some sort of substandard connection that corrupts the data transfer. Just a something to look out for I guess.
 

Latest Posts

Top Bottom