I ask becuase my brother has been having problems with the idle on his 87 GTA with Minirammed 409. All the usual goodies, still run by stock 87 ECM with TPIS custom-burned chip. No check engine codes, runs like a raped ape the rest of the time.

Here's what we've diagnosed (or THINK we're diagnosed). After tracking down a few vacuum leaks and other minor problems we're down to ECM/programming/sensor related stuff. All the sensors are operational and everything is giving "sane" readings hot and cold.

Except the O2 (which is new) ad only at idle.

The block learns at idle/warm are in the 160 range and the thing is clearly running fat, stumbling and choking to stay alive. The O2 sensor seems to be dropping to 0 Volts at idle sometimes, like it's not staying hot enough to give good readings. Reved up and under load it's doing OK. The car has Ceramic coated headers in place of the original manifolds. We think it's getting SO fat at idle that it is snuffing the O2. And yet it remains in closed loop AND reads "lean" about 75% of the time in the "rich/lean" field on our scaner while this is happening. It reads "lean" 100% of the time when the O2 voltages go to zero at idle.

Now here's the kicker.......... pull the batt cable and reset the ECM (back to 128 block learns) and fire it up it idles so smooth you could sit a glass of water on the intake. Drive it around for an hour and the idle goes to he11 again, block learns back up in the 150 range.

The question is....... WHY? Is the ECM not aware the O2 is getting snuffed at idle sometimes? Why is it adjusting the fuel mix so rich it is clearly killing the idle? I thought it was supposed to learn to run it BETTER, not worse.

My brother wants to get a heated O2 sensor so that the ECM is getting correct readings even at idle but I'm not so sure. Rumor/speculation/cohjecture is welcome. Real answers would be better, though. Thanks in advance.

 

the o2 sensor will start outputting voltage to the ecm when sensor temp is approx. 600 degrees f. ecm is made aware of closed loop readiness (at least where the o2 sensor is concerned) by the amount of crossovers made above and below .450 mv in a given time period. obviously several other factors are monitored by the engine management system to prove out closed loop readiness-coolant temp, monitored sensor voltages within spec.etc-but as far as o2 goes the crossover count is used by the ecm to decide its ready to go closed loop.

 

There was a guy over in the TPI board that had this issue...Turns out that he had a small exhaust leak around the area that he had his O2 sensor (non-heated) and it was impacting the readings. He couldn't hear it, unless he was under the car next to it. Might be something to think about.

I decided to use a heated sensor (uncoated 1 5/8" headers) as it is a bit downstream and I don't want it potentially cycling in and out of closed loop.

Any codes being thrown?

 

Either the sensor isn't staying hot enough to give a good reading or your calibration is very lean at idle >128blm is rich and 160 is very lean.

Brendan