A code 31 just means the engine is stopped. It doesn't indicate an error (19 is what indicates a failed crank sensor).

Can you see live data on your code reader?

If so, let it warm up fully and you can verify that the engine coolant temperature reading is plausible without changing any more parts.

While you're there, have a look at the MAF sensor output (typically around 12kg/h at hot idle) and make sure that Lambda control for both banks has entered closed loop and both integrators are around the mid-point (127).

Have a look to ensure all the other readings are plausible. If possible post up the complete set and we'll see if anything looks odd.

Kevin Wood wrote on 31. Aug 2011 at 11:16:


I have the live data function, and will write the readings down and post them here. But i know that they are not the same, one bank is about 10 of, iirc. But i will post them later on
Thanks for answering 

Sehen wrote on 31. Aug 2011 at 21:43:


for some reason your engine is retarding ignition..

so it must have a knock..

initial guess, either timing incorrect or problem on some valves or faulty coil

cem wrote on 31. Aug 2011 at 23:17:


Knock sensors inactive, forgot to wrigth that  down

Sehen wrote on 08. Sep 2011 at 23:53:


ECU's dont fail frequently.. there is a solenoid behind the air pipes (which causes the idle jump up down) .. and also check the air cable connections..

ps: misfiring tells me that you have a combined problem..

Right, the ignition timing is varied by the ECU to stabilise idle speed, so I'm not surprised that it's all over the place when the engine is misfiring. It's the ECU trying to compensate for the idle speed varying. It won't be knock sensor input, because they'll be inactive at idle anyway.

Everything else looks reasonable in the live data. Integrator slightly out on one bank, as you say, but I've seen worse.

Does the coolant temperature in the live data change when it starts misfiring? The fault code you have suggests that perhaps the wiring to the coolant temperature sensor has been intermittent. It may be nothing but, if the coolant temperature takes a dive due to a poor connection, the ECU could be enrichening the mixture leading to a misfire.

The other thing that strikes me as odd is that the rev counter is affected by the misfire. When the crank sensor was changed, was a genuine GM one used or a pattern part? It might be worth making sure the wiring to the crank sensor is routed well away from both the hot exhaust manifold and the coil pack and HT leads (route it up the inner wing and across to the cable tray on the 2,4,6 cam cover).