Hasn't that question been discussed already?

 

maybe 250 posts ago?

 

Yes, the car stays idling and does not die. The problem is the AFR fluctuates at idle which causes the RPMs to rise and fall. Sometimes the RPMs will fall far enough that it will stumble and die.

The LTFT is about +14 and the STFT is about +10. The engine is running lean. A lean fuel condition can be caused by:
Dirty Fuel Injectors - I had mine cleaned and flow tested when I put the turbo in.
Bad or Dirty MAF- I cleaned it again the other day.
Low Fuel Pressure (leaky fuel pressure regulator or weak fuel pump) - I replaced the fuel pump with the Mazsport fuel pump when I put the turbo in.
Vacuum leaks (intercooler piping, intake manifold, vacuum hoses, throttle body) - Since I installed everything I figure this is the problem.
Bad Gasoline - I had ~3 gallons in the car when I stored it for the winter. I added ~5 gallons of shell 93 octain and a bottle of fuel system restore before I started the car.

I have a couple of things I plan to try but I have been working 10+ hrs a day and have not had a chance. Hopefully in the next day or two I can try a couple of things to see if I can narrow the problem down further.

Thanks for everyones suggestions! They are appreciated!

 

Do you have a vacuum gauge? Are you still going to Indy? I have one I can loan you. All though if your idle is bouncing around the gauge would too.

 

also more simple stuff like.

loose plugs.
loose plug wires.
bad plug wires
bad plugs.
bad coils.

and the big thing you mentioned is bad gas.. i dont know how the filter works with the mazsport pump. but i would look at it close..

beers

 

Yeah, I have a vacuum gauge. Thanks for the offer! I am not planning on going to Indy anymore.

 

The Mazsport plug wires are new, they snap on, and I made sure to get them on right. I have new coils that I got with the Mazsport Ignition Solution. I'm waiting on some new plugs.

I smelled and looked at the gas and it didn't seem to be bad. I added the new shell 92 octain. I could fill it the rest of the way up with new gas but if I need to get into the tank to do something with the fuel pump gas will go everywhere. I have a fuel pressure gauge that I am going to try and install this weekend.

 

I fixed most of my idle issues - it was tied to my MAF housing and settings in EMU.

So that could be an issue. How many Grams per second of air are you seeing at idle; and what is your absolute load?

 

Here is a datalog at idle. The sample rate is .25sec. The OEM wideband latency is about 1sec. I have calculated load in this log. I will have to do another one with the absolute load. Anything else you would like to see? I can log up to 6 values with my Harrison CANScan.

 

tdiddy; You mind Calling me?

I'll pm my number.

 

Kane made a very good point that I had not noticed. In the Idle log I posted 2 posts back, if you sort by RPM and look at a range you will notice that the AirFlow numbers are all over the place. That means the MAF housing, MAF sensor, or the turbo could be the problem. In my setup, the MAF is about 4-5 inches away from my turbo inlet which could really be messing with my airflow readings.

 

So are you going to extend the MAF to the other side of the support behind the front bumper or get an INTx and call it a day?

 

Very good question! IDK???

 

I didn't see the log, but there should be no more than 1 g/sec variation across 150 +/- RPM at idle.

If he extends his MAF tube, he won't need an Int-X.
Not to mention the other problems that the Int-X would cause. Why trade one (solvable) set for another (intractable at that)?

Plus, when you have an issue, throwing new equipment at it is quite possible the worst idea conceivable.

 

Be a pioneer - no engine management!!!

 

Between 700 and 850 rpm at idle, the airflow can be anywhere between 7.91g/s and 3.36g/s

 

Isn't there another 8 owner that's near you that would allow you to swap thier MAF sensor in to see if the trims straighten out? It would be a super quick check and a lot easier than extending your MAF housing or switching to INTx only to find out the sensor was the problem in the first place.

 

3? Wow.

OK, I looked at the log.
Your CalcLOAD is all over the planet and the MAF is chasing the lambda.
It would be useful to see actual throttle position in this kind of log, but it is safe to say that there is a major flow/vacuum issue somewhere.

I presume you have verified that the injector "O" rings are not leaking?

 

In 150 samples at idle I have 20 readings < 4g/s

 

Now that we know what is going on with the Fuel Trims I can create a new log this evening that will show the throttle position instead of the LTFT.

I will be checking the "O" rings tonight with starter fluid.

 

I sprayed all the vacuum sources I could get to with starting fluid but didn't find a leak. If I am going to find a vacuum leak I am going to have to pressurize the engine or have a smoke test done.

Anyway, here is a new log.

 

Boy, does THIS whole experience sound familiar......

 

Go get some PVC - time for experiments.... yeah.

 

i was thinking the same thing.. what would be the odds.

the ltft and stft look the same? i think you explained them like that...

beers

 

I couldn't imagine having made the same mistake twice, especially since I now understand the reason Mazda did the injector harness the way they did. Plus, I think Tommy and I checked all that before the UIM went back on and he has checked them since then, too.