Lambda takes its sweet time to register

Hi, I’m seeing a trend with my Lambda sensor that I’m not loving, but am not sure if its something that I changed. I got this car on the road at the end of September, it was still quite hot outside when it was first on the road. Upon startup, I would get a good lambda reading withing a minute or so of fire up. Now that it has cooled off, it can take between 9 and 15 minutes to get an accurate reading. It’ll do its typical warm up sequence, and then peg the gauge at full lean. Sensor diagnostics all look happy though. Once the coolant temp hits about 147°F the lambda value drops to where I’d expect and all is fine. I’ve changed a million things over the last couple of months, and don’t know if any one setting made this happen. Or is this an indicator of a sensor going bad? Datalog diagnostics say all is peachy, the reading is just off. Comparing logs, the coolant “switchpoint” has moved up about 10 degrees over the last few weeks. So, whatever is happening, seems to be getting worse.

Thanks for any help you can offer!

10-29-26 more boost.emupro (33.8 KB)

2026_1103_1349.emuprolog (6.1 MB)

I would recommend enabling “Heater Without RPM.” This will speed up the O2 sensor warm-up time. If this setting is disabled, the O2 sensor won’t begin heating until the engine is running, so you’ll see little to no O2 activity beforehand. If your trying to tune for cold start this will be a must have.

That said, this looks more like a failing O2 sensor to me. Have you tried swapping it out?

Also, are you running leaded fuel, such as X85L or Pro E112? Those fuels can contaminate and damage an O2 sensor pretty quickly.

Thanks for the response. The sensor warms up nice and quick, less than a minute and it’s sitting around 1436°F, and stays there, so I don’t know that I need to get it to warm up without the engine running, yet. That doesn’t account for the long time before it provides useful data. I’m wondering about the sensor as well, but the fact that the data becomes accurate pretty consistently right around 147°F CLT(arbitrary number, not water cooled) temp, makes me think something is hitting a switch point. I tried another sensor yesterday, it wasn’t new and never got accurate, despite running for almost 17 minutes and getting past that CLT threshold. I haven’t ordered a new sensor yet, that’s a next step assuming nothing can be found in the data provided.

If you heat your O2 sensor when RPM is 0, then you increase the risk of damaging it if condensation lands on it. I think the option is only there to test lambda after a start when tuning but shouldn’t be left on after tuning.

Well I’ve ordered a new sensor. I’ll try that and report back. Hopefully it’s just a hardware issue, despite the fact that it acts like it’s reacting to a switch point of some sort.

New sensor seems to have done the trick.

how old was your previous sensor ?

Like 13 years :grimacing: though it had probably less than 2000 miles on it. It was one of 4 that I had in a per cylinder setup on a different engine. I think I’ve gone through almost all of them now, figuring that I’d use them up before buying another. But none of them has lasted very long.

Hopefully this new one works for a good long while.