AFR fluctuates wildly
 

Is this typical behavior for the wideband AFR reading on Tunerstudio?

It's hitting the right high and low for closed loop but just seems ridiculously erratic in between.

http://www.youtube.com/watch?feature...&v=Cy6VRJWeqGs

No, not normal.

What kind of car are you in?

what wideband do you have? I don't see you showing a wideband display so either you have one without a display or you are using a narrowband output.

It's an LC1 in a Mazdaspeed Protege. I've used the LC1 with a AEM FIC piggyback for a few months last year and it worked fine. Since I sold the AEM and decided to go with the MS, it's been just unplugged for the last six months or so.

Could need to be recalibrated.

I suppose I could also run the LC1 into my PC and use its software to make sure it is working properly.

datalog and tune as well, upload if you have them.

Just in the first stages of getting this running. Since this is my DD, my plan is to run it in parallel with the stock PCM so I keep my OBDII for emissions testing.

Plan is to get all sensors working properly, then injectors, then spark, then idle, then boost.

I am using a patch harness to quickly swap back and forth between MS and stock. I have it out of the car right now, because I noticed the MS was causing a slight drop in voltage on the IAC and CLT causing my car to turn on its fans a little later than normal and the temp gauge to sit slightly below half way. I just put resistors on those two lines to see if it solves that problem.

Ah crap! I just took out the LC-1 manual and right on the front, "Warning! The oxygen sensor must be connected and operating with the LC-1 whenever the car is running. An unpowered oxygen sensor will be quickly damaged."

I remember reading that when I bought it last year but it slipped my mind when I sold my AEM. It's been just sitting in my down pipe unplugged for at least six months. So that's probably past the "quickly" time period.

SOB.

OK there are 3 places in the Tunerstudio software that I can set narrowband or wideband.

Project properties -> Settings, Basic Setup-> EGO control, and Tools-> Calibrate AFR.

I had missed that last one. Despite being in my exhaust unplugged for the past six months, I think the O2 is still fine.

I am not sure what would actually damage it anyway. Perhaps contaminants that would coat it when the engine is cold without the heater to burn them off.

Anyway, the new one should be here today if something is wrong with it.

even if you did damage the o2, it's not THAT expensive just to get a new o2 sensor for the lc-1 anyway. relatively cheap compared to the rest of the stuff you have. That is if nothing else you do helps you resolve the problem. something like 50 bucks or less?

Yeah it was $65, and I paid the $4 to get one day shipping. So it didn't break the bank, but it does add up. Definitely not a cheap hobby.

Try calibrating the sensor before you replace it.

I am fairly certain the sensor if fine. I'll just have a backup now. Or I can send it back with Amazon's painless return policy and just be out the $4.

Ok so after calibrating I decided to see what the MS said my AFR was in open air.

It seems to be showing about half what it should be. I then calibrated the new O2 sensor and it was exactly the same. If I start the car, I see it bounce between 7 and 10, about half what it is at idle.

Is this because I scaled my AFR bins to 200%?

I've seen a lot of tables use the AFR and fuel load%, and others use straight MAP in kPa, what is best for a turbocharged engine? I have an MSP which has a lot of similarities to the miata which is why I am on this forum.

https://www.miataturbo.net/attachmen...ine=1328384844

That AFR target table is too lean for my tastes. I run 15.0 in cruise, 14.0 at 80kpa, 12.8 at 100kpa, 12.0 at 128kpa, 11.7 at 180kpa+.

seems to rich to me at low rpm\low boost. I would add more fuel in high rpms in boost.

Fuel load% is what the control alogthrim is set to under basic setup > engine constants.
If its set to speed density that means it is using a map sensor on the Y axis for the fuel load%

If its set to MAF it means its a calculated load% based off the MAF sensor under the Y axis for fuel load%

stuff that has map in kpa on the y axis just means that its basing its y axis on the map sensor.

Most people will suggest using speed density for turbocharged engines, personally I say either or for a daily driver road car, with more toward a maf sensor.

The car isn't tuned yet, I still have my PCM controlling everything, just want to get all the sensors working properly before I do that, and clearly something is still not right with the AFR.

I uploaded my msq file. I guess I'll just play around and see what works best, speed density or MAP.

https://docs.google.com/leaf?id=0B1e...3YjI2&hl=en_US

As I expected, changing the bins on the afr load back to a 100% scale did not affect the gauge.

I wonder if I have something setup weird in the LC1 software, I'll load that up and check.

are you sure you have a wideband wired in and not the stock narrowband wired in instead?

OK, I am going to lose all credibility here, but I had the two analog outputs of the LC1 backwards. I thought for sure output 1 was the 0-5 and output 2 was the 1.1v output. Had it backwards this whole time.

So two simple problems = big headache.