4ddiction

Have a turbod 95 miata, with a megasquirt ms3 built by braineack. Im currently having a problem leaning out on idle.

AEM wideband reading 14.8 upon starting, the idle is steady at 900-1000. After a few seconds the car leans out and stays at 18.3ish. I checked the timing with a light and it was on point.

Boost gauge reads -20, and here are my tune and my data log.

When i give it some gas, the car doesnt really richen up. Im thinking it might be a slipped tooth on the belt. Not sure

Any ideas?

noname4me

Sorry, I am not in a position to check out your msg or logs at the moment, but here are some questions to help get the discussion going:

Is this on cold start? Hot start? Or after deceleration and going back into idle?

Is this in open or closed loop mode?

Have you tuned the VE values around the idle region?

Check that you don't have Idle VE or Idle Advance enabled with bogus values in the associated tables, especially Idle VE.

4ddiction

Still having problems, really boggles my mind why when it starts its perfect 14.3 for a few seconds then creeps it way up to 18.3. Any ideas or could anyone see if my tune is messed up?

4ddiction

Just checked to see if my aem wideband sensor was faulty, seemed to be okay.

Took it out, gauge said lean, then i placed it into a carb spray soaked rag and it riched up and stay there. Only returned full lean when i removed the rag.

Sprayed carb cleaner all the over to check for vac leaks and found none. This is getting frustrating lol

DNMakinson

iTrader: (3)

You are idling at 18 because that is what you are tuned for. When you first crank, you have enough ASE and ASE taper to enrich for a few seconds and then it goes to steady state.

Always tune in steady state when possible. Idle is certainly a situation when that is possible.

I cannot address your last method of testing the sensor. Seems rather imprecise to me.

Search on the many good threads, many recent, on how to tune idle. Then tune it.

4ddiction

Followed what was described how to test those o2 sensors.

No matter how many values i add to VE table the car doesnt richen up. Continues idling lean.

Dlaitini

Is everything else set up correctly? You have base timing set, req fuel, etc?

DNMakinson

iTrader: (3)

Your msq shows that you have Idle VE Settings turned on. This is the last line in the drop down under Startup/Idle. Since this is on, that table is controlling VE when CL idle is engaged.

4ddiction

I turned off idle VE, and the only way i could get her to idle at 14.7 was to massively change my MAT air density table.

Here is the current tune.

4ddiction

Yup, timing is on point. Req fuel i am not sure but i have not changed it.

Braineack

iTrader: (62)

I assume it idles like crap? I can't imagine the car running well at all. Typically idle gets twitchy at 15-16:1.

If you look at the PW the numbers don't really make sense compared to the AFR reading. During ASE, he's injecting around 2.5ms of fuel and getting AFRs in the 13s. Just a few moments later he's injecting the same but it's showing 18AFR. Then there's a dip in the AFR that shows it hitting 16, but the VE% and PW hasn't changed at all.



turning VE idle off is not the solution*, turn it back on. *Unless your fuel table needs tuning.
tuning the MAT corrections table is not the solution, revert it. There are no MAT corrections leaning out the car those logs.


actual questions that can help figure out a solution:

Have you always been idling at 18:1?
If not, do you have any comparative logs that show it idling around 14:1?
this is only happening at idle? AFRs are fine in cruise while driving? can you log this?
What other changes have you made to your tune recently?


If the idle quality isn't shitty as ****, I'm going to assume the WB is on its way out--especially when I see the second log and when you rev it up to 6000rpm.

DNMakinson

iTrader: (3)

Not sure why none of the elites has not come in, maybe seeing if you / we can figure it out.

EDIT: Oh, good, Braineak came in for you.

I see where you turned off idle VE and even went open loop idle; and raised the VE values in all of the idle cells. Surely, that should have made the change occur. I also see that you did increase the MAT correction.

I must admit, this is beyond me. The VE changes should have raised the fueling. The MAT change is not what you want to do, as you know.

So, with these changes, is the AFR at idle reading at your target?

Braineack

iTrader: (62)

The easiest way to tell would be to double the req_fuel.

if the AFRs are still 18, then the WB02 is dead--I highly suspect it is.

DNMakinson

iTrader: (3)

I never saw a log that showed the values you are quoting of 2.5 mS. Only ones I saw show around 1.7 or so.

Thanks for joining in.

Braineack

iTrader: (62)

I'm looking at the logs in excel right now, so it's tough, but AFRs just don't seem to match the PW in a logical manner. But it seems to me like at 1.88ms it reports 18.3, but at ~2ms it's reporting anywhe from 13-16AFR.

On my setup I know I would idle at 1.8ms at 800RPM on 460cc inectors. I dunno what injectors OP is running since I can't view the msq, but i see at teh end of the log he's at 115% warmup still and 1000RPM at that pw, so might really just be tuned lean.

hard to say without more deeeeeeets.

4ddiction

Thanks for chiming in braineack much appreciated. Currently at work so i cant check much, however, this is the latest data log i have.

The car idles surprisingly well, no hesitation really. As i was driving it, as i got off the gas between gear it would drop down and richen up.

No dont have any data logs from previous runs.

Running stock injectors, 95

Where do you change the req_fuel?

Braineack

iTrader: (62)

under engine constants; literally double the current req_fuel, that will double the pw.

I think your sensor needs replacing.

look at all the spots is hits 18.3:1 on that log, then compare it to the amount of fuel injected (pulse width).

at the same rpm and same pulse width, your AFRs will randomly show super lean; I still highly suspect the sensor.

4ddiction

Ah found it. It's at 11.5 now, when I get home I'll double it and see what happens.

When I ran those tests on the sensor it was fine. I'll test it again and if anything I'll order a new sensor

DNMakinson

iTrader: (3)

I suspect he was giving an example, not a suggestion. If you double Req-Fuel, that will have the same effect as doubling every cell in your VE table

You can do that as an idle trial, but be aware of the total effect.

How close to the end of your exhaust is the sensor? Is it possible you are getting air mixing with the exhaust under some conditions?

4ddiction

Thanks for the help!! The Req_fuel was 11.4 and i changed it to 20. Turned back on Idle VE and adjusted those numbers and it finally idles right!!

Going to continue fine tuning these values and make sure my VE table makes sense as well!