Tuning Sanity Check
 

I'm tuning a local's 97 (with a GT2560, 99 head and IM).

I upgraded his ms2 parallel to ms3x. I have it running very well but I'm noticing sometime odd over the last couple days of driving it.

After startup and for the first 10-20 minutes of driving or so everything is kosher.

Looking over logs I'm used to seeing:

900RPM
31-32 kPa
14.5-14.7AFR
45% GVE
1.737ms PW
30-31% idle duty


once the car gets warm and I start seeing CLT temps get closer to 190-200°F I start noticing that I'm getting idle droop and rich AFRs, then I notice the normal conditions have change where I'm seeing this:

900RPM
31-32 kPa
12.8-13.2AFR
45% GVE
1.73xms PW
36-38% idle duty

I'm using the VE idle table and Idle Duty Initials Values table.

I'm seeing a large increase in the amount of idle duty needed at idle to maintain 900RPM and the AFRs start going very rich. I setup my initial idle duties table to go back to 31% + dashpot and I was getting perfect idle recovery, since 31 + 4% = 35%, this is lower than the 36-38% needed to now maintain my target RPM. So I've had to turn it back to "last known value" which I dont like as much.

But I cant really figure out what would cause this sort of air:fuel changes...it's becoming a nuance to deal with because I'll get the idle AFRs perfect for most of a drive, then I'm rich as hell. But if I try to lean out the AFRs the next time I drive the car I'm WAY too lean.

I suposse I can turn on EGO at idle, but I'm not too found of doing that...but it seems that might be the only option unless I can figure it out. Otherwise the tune is coming together well and everything else is functioning as expected.

My log files are huge cause I'm logging 30 minutes or so at a time, so I'll just attach the current MSQ.

Why aren't you fond of using ego correction at idle?

Probably cause when I used it on MSI years ago it never worked well and I've never really needed to use it since.

EGO correction at idle is awesomesauce on MS3x.

also the initial values thing doesn't work as well as last known IME.

Initial values works great during warmup, but going from warmed up to heatsoaked (and heatsoaked hot start), the duty is way off.

I've got around the poor warmup behaviour, buy reducing the delay to PID idle, and reducing the cranking duty.

AFR's do that on mine too, btw is it due to the change in the ideal gas law correction curve implementation in the last few code releases?

Dorothy,
We're not in Kansas any more

Ok. I dont have these sorts of issues on my personal setup with open loop idle and initial duties.

I'll turn on EGO and leave initial duties off. Doing the later already solved the one problem. care to post your EGO settings?

I have the MAT corrections pretty much zeroed out, GammaE has been at 100% during these conditions. I just worry that I won't be able to reduce the PW much lower than 1.7ms to keep the AFRs in check. But now that you mention it, maybe I just need to go back and tune this table a bit to keep things in check, his AIT was around 85-90 after a while when I started noticing the change, but only on a 70°F day.

Give Ed my regards

yeah, see if I get EGO great at idle, it sucks on highway, what settings are you using richyyyyyyyyyyy?

Today I start the car on a 38°F morning and it idles higher than normal.

Eventually after PID kicks in, it's idling at 19.9% PWM Idle Duty, which is my valve closed value, and it's idling at 1200RPM.

As I drive for a good twenty minutes and the AIT warmed up to about 45°F, I nocited the idle duty is now 34% and holding 900RPM.

This doesn't make any sense to me. If I increase the duty, the idle speed increases, so this doesnt make sense at all...

?? Increasing idle valve duty should increase the idle speed..... You're needing more duty with the car warmed up as the air is thinner. Does sound like the bleed valve needs tweaking slightly to get the idle valve to not be at minimum when in WUE.

If it idles too high straight after cranking, tweak the cranking duty down a small amount. I found that helped immensely. Also reduce the delay to PID.

Sorry I'll post my settings tonight, I completely forgot. EGO seems to work fine at both cruise and at idle with them.

I can't wait for your settings screenshot so I can actually visualize and hopefully comprehend this idle tuning crap you guys are talking about. lol

I'm saying I needed signifcantly less duty today; which was significantly colder out then last week's tuning.

The as the car warmed up, my idle duty was pegged at 19.9% as the idle slowly worked down from 1400RPM. 19.9% is my minimum PID duty value; so I can't bring it down any more.

Eventually after a good twenty minutes of driving I saw the duties up to about 34-35% to maintain a 900RPM idle speed. So as the car warmed, it required 14% more idle duty to idle at LESS RPM. So the behavior I was seeing doesnt make any sense to me, and this is untypical of any other miata I've tuned.

I didn't perform the idle valve test mode on this car to determine the min/mxa duties or the min/max idle speeds, so I'll back back out at lunch today and fool around and see if maybe the bleed valve is too far out or something.

Leak at one runner of intake manifold... motor warms up, swells, closes leak?

I'll give it a once over with brake cleaner tomorrow AM.

What about the EGR, that's still in place, I was wondering if that was doing something like what you're suggesting. It needs vacuum to open correct? so if I pull the vacuum line to it, there's no way it should open...even though there's nothing controling the solenoids to it....just trying to rule out things here.

Otherwise the car is running great and spinning RA1s in 2nd gear (6sp) at 15psi once it hits about 4K.

Why is richy so full of hate?

[quote=Braineack;1003957]I'm saying I needed signifcantly less duty today; which was significantly colder out then last week's tuning.
[//quote]

Sorry, i'm still confused, if it's colder out you'll need less duty?

Ah, yeah seems a bit odd now, possible leak somewhere that closes up as the car warms up?

Definitely worth a go.

Don't hate a lazy person :(

Seriously though I forgot and I don't have a copy of my map here at work!

Have you tried the IAC valve in test mode? At least mine is closed around 30% going below that makes the rpm go up.

that's a flyback diode issue on your setup. Are you running idle control on the mainboard of a v3.0? I bet you still have d8 installed. 90-93 valve should be closed at ~19% and fully open at ~60% if you've built it correctly using a 1n444 diode to 12v for a flyback.


I havent tried to test mode or check for leaks yet cause I'm lazy. It's not even a big deal cause it's running very well, I just dont like knowing that it's doing something odd behind the scenes, if I didnt have my laptop hooked up and I was monitoring gauges I'd have no idea.

I don't know actually, as the ECU is built by Dimitris (reverant). Anyways, is this something i should be conserned of? My current adjustment range 30-60% is in fact way more than i'd actually need even though my car sees cold starts from -20'C in the winter.