Hello, About a month ago I assembled my turbo Miata and it was running great! I was daily driving the car for a few weeks and then earlier this week we got hit with some snow and I let the car sit for the past 4 days. Yesterday morning when I went to start up the car and get it ready to install a roll bar I noticed as I started to drive to my friends shop that the car was displaying full lean on the AFR gauge. I quickly pulled over and pulled out my laptop to see if I could see if there was anything in plain sight wrong with the car, it was reading a steady 18.7 AFR with no change in the AFR reading with throttle input. I was only a minute away from home so I took it easy and parked it in the driveway. I let the car sit for a few hours and went out and tried it again. I noticed that when first starting the car it would display a 14.7 AFR with small variation for about 15 seconds and then slowly leaned out, as well as the car had a higher idle than normal. I let the car warm up a little more to see if it would change but no dice. I thought perhaps that my Bosch sensor that came with my AEM 30-4110 wideband gauge might have gone bad even though it gave the 14.7 reading in the first few seconds of starting the car. Today I pulled the sensor out of the downpipe and tested it to see if it was still capable of reading a rich value with a shop rag and some brake clean and it read full rich so I think that's not my issue. With the sensor back in the downpipe I fired up the car and tried adding more fuel at idle to see if I could get the gauge to read anything but adding more fuel did not improve the reading so I reset it back to where it was initially because the car was running good before. I'm wondering if I might have a small vacuum leak somewhere in my system, I visually inspected all my hoses and they appear to be fine but I am hoping to get my hands on a leak tester to be sure of it. I'm wondering if anyone has any input on what they think may be causing my problem, I'll include a log of my car starting up and idling as well as a video of my AFR gauge.

Link to video of physical AFR gauge while datalogging:

 

The car doesn’t drive any different than it has been, correct?

Your O2 sensor is likely bad, even if it reads rich when doused in brake clean. If your car actually leaned out that hard during idle, it would either stall or be running like garbage.

In my experience, the Bosch sensors we use on these cars are hit and miss in terms of longevity. My AEM 30-4110 came with a sensor that did what yours is doing right out of the box. I replaced it with another one and that one lasted about 6 months then **** the bed in the same fashion. Not much you can do about it. I’m currently running a wideband out of a Ford Raptor in my car (probably also a Bosch) with the same calibration and it’s been going strong for a month or so.

 

The car will still rev like normal and the idle doesn't have any hiccups in it just about 300rpm higher than normal.

EDIT: Just ordered a new sensor so once that gets here will test and see if that was the issue

 

Not sure what happened with your DOA sensor, but there is something you can do about the general unreliability of aftermarket widebands. As you said, OEM's use the same sensor but don't have reliability issues. A big killer of the sensor is heating before the engine starts, so you turn the key, sensor heats up, then start the engine and exhaust passes over the hot sensor, I believe it's the condensation that kills them faster. You can run the wideband off a relay that's triggered from the fuel pump output, or some method with generic outputs that has a delay after engine start.

 

Ah ok, that's good information regarding the elevated idle. Does it do that always regardless if the engine is cold or warmed up?

Still, a vacuum leak shouldn't pin your O2 sensor reading to full lean off the charts.

Just out of curiosity, with the car cold, turn the key on without starting the motor and see what your AFR gauge does. Every time one of my O2s went bad, it would do the same thing with the engine off as when I started the car: read around 14.5 for several seconds then work its way to full lean off the chart. If the sensor is working correctly, it should just stay pegged close to 14.5 indefinitely while cold with the key on and engine off.

 

So far the sightly higher idle is happening regardless of if the engine is warm or not. The car will usually have a slightly higher than normal idle when I cold start it but this wont go away with the car getting warm.

When I turn the key on without starting the engine I have the same thing happening where it will hang around 14.5-14.7 for about 15 seconds and then slowly lean out.

I only included the partial suspicion about a vacuum leak because I hadn't touched anything on the car for the 4 days it was cold and snowy and thought that maybe the cold caused a vacuum line to crack or something.

 

Oh dang, good to know! Thanks for the feedback, man.

It's funny you'd say that. Both times my O2 died, it was when I started my car on a cold wet morning.

 

Alright, yeah, so your sensor is definitely bad. Sounds like you might have a vacuum leak as well. If you want to check for that in the meantime, the quickest way aside from listening for one in the engine bay at idle would be to do a test with brake cleaner. Spray different areas around your vacuum lines/intake manifold at idle and pay attention to any fluctuations in idle speed.

Or maybe you already know all this and I'm just being redundant haha.

 

I'll give that test a shot when I get home from work today, and I will update the thread when I get the new O2 sensor installed when it comes in. I hope that it fixes my problem and i'll be back to boosting shortly!

 

Love to hear it, man. Good luck and keep us posted!