Any ideas on what is making this car go lean?
 

I tuned this car and I'm getting some strange behavior. You'll notice in both images that there are 10-15* sweeps in CLT temp (195-208*), TP is static, RPM is static, but AFR is going lean. Any thoughts on what to adjust?
https://www.miataturbo.net/attachmen...1&d=1364658254
https://www.miataturbo.net/attachmen...1&d=1364658254

vacuum leak?

Highly unlikely.

The only variables that do not stay constant during the lean spike are the AFR's and coolant temp. So either excess air is getting in (vacuum leak, even a very minor one can cause this....ask me how I know) or fuel that should be going in is either leaking or blocked but I have no idea how that would happen randomly.

When these sensors have a short, they typically give crazy readings like 500*, right? Ever see a bad sensor act like this? I haven't but I'm running out of ideas and I want this guy to be able to track his car sow I can show him the error in his supercharger ways. lol

Actually he missed the MSR-C track weekend because of the funky EFI behavior like this and he deserves better.

VVT angle?

1999 head. The car also has an MP62.

I must be going blind, but TP looks like it is varying and the other parameters seem to match as expected. TP is the yellow line in the second block of each group, right??

Most of that is static but filtered. That slight amount of TP, even if not filtered, should not create a 1ms PW adjustment. If you look throughout the log you'll find that small amount of TPaccel does not cause AFR to go lean.

Send me a log through email.

Is overrun cutoff turned on? You're looking at 1% and 0% TP respectively at the center points.

Lack of fuel.

/thread

Can I please get some thoughts on this log regarding CLT, MAT, and BattV behavior on this supercharged car? I'm not sure if I'm chasing bad sensors, bad wiring, or something else. Please also review TP vs. TP dot and it's peculiar behavior. TPSdot threshold is set to 220%/s so I'm not sure why it's triggerting from 894-902s in the log.

Thanks.

why did the CLT temp raise up like that, that seems to have some effect on it...

I'm bothered by the gradual increase, rather than a sudden spike which would tell me there is a short or something. I can't figure out what's triggering the CLT fluctuation like that...I'm not really sure what the MAT should look like on a supercharged car.

I really want to help this guy out, he's been struggling for a while. Time to start replacing sensors or what?

Can the coolant sensor heat soak?

When the clt sensor dies it almost always reads-34*F. Did you try zeroing out all the correction tables? In MS is it possible to display the base fuel and the corrections live. IE base is x and the total correction factor is x in a log?

10-15* is a pretty big swing everything else being constant.

either bad sensor or something tripping the sensor out

kinda sorta looks like its a hardware (or basically non tune related) issue

Also, how quickly its changing. These are closed brass element temp sensors, they have a thermal damping built in because of their thermal mass. IE the sensor has to heat sink before it reads properly.

air bubble in the cooling system? bad sensor?

something

Loose connection maybe?

That's what I thought too.
I don't know the values to put back in.
Not sure, this is MS2 for those in the know.

Doubtfull and air bubbles don't make AFR do that.

It's been unplugged and put back a few times, I doubt it.

bubbles tripping out the clt that is calibrated to pull fuel when super hot? I dunno

that's why I was kinda wondering about what leafy said about zeroing out the table or making it go pig rich with higher temps: it will tell you for sure if jump in clt is causing the spike in afrs

Well to rule out the clt sensor. Take the value from the current temp the ~200 cell, and paste it into the -34* cell (or the lowest one), and un-plug the sensor This will eliminate any effect the sensor could be having so we can cross it off the list of suspects.

yeah something liek that

He's got so som sort of issue with the MS or wiering look at 82 and 95 seconds in the log. Se the way the coolant sensor goes up in the same instance the map and tps goes down, also the indicated battery voltage goes down. At 128 he reduses throttle, map behaves normally, but colant and volt behaves odd. Grounding issue?

Tommy

Hustler's not wondering about those instances. A sudden CLT jump related to closing the throttle and changing alternator voltage is an understood phenomena. What he's wondering about is the slow, cyclical change in CLT throughout the log . . . looks like a sine wave.

Honestly, it looks like the sensor is good but the CLT control is lousy. It seems to rise and fall in a delayed, smoothed response to engine load. I also don't see that AFR has a relationship to CLT in these curves.

Is the cooling configuration stock (no reroute)? Older T'stat? Don't forget that in this configuration, CLT is sensed at the back of the head by cylinder #4 while the T-stat is trying to control (and cool water is being ingested by the mixing manifold) at the front. Also, sometimes the little bleed line from the T'stat housing gets clogged leading to delayed control.

If rerouted, is the T'stat in the head or remote? A remote T'stat (M-tuned) could exhibit poor control in certain circumstances.

that slight amount of TP looks like going from 6% to 0% in the second graph. when i'm at 0% throttle, my afr's go lean too, cos i'm not putting any fuel in. nor are you, PW goes to 0.

first graph has throttle position at 1% when you go lean.

how does it feel to drive, what's happening through the pedal?

i'd +1 to whoever said grounding btw.

"how does it feel to drive, what's happening through the pedal?"

When driving until warm it drives normally. When warm is alternates between normal behavior and falling on its face. Basically I am driving along it stops pulling. The AFR gauge heads toward 18 - 22. Moving the gas pedal has no effect. If I back off it will recover and pull again. The last time I drove the car it was doing this relatively often.

thermostat sticking or sluggish ?