Closed loop idle tuning (ISSUE RESOLVED!!!)

[clongo]

I started playing with closed loop idle tonight. I have read the megamanual on tuning closed loop and while I understand a good bit of it, a lot of it is over my head.

Here are the issues i'm having.

When i first start the car it rev's up to 2000 RPM and stays there for a few second's and then the car dies unless I give it some throttle.

It either idle's really rich in the 11's or it oscillate's really badly and eventually stall's. I have the timing map and fuel map pretty even in the idle area so im not sure what's causing this since it does it even if the sensor isn't heat soaked.

Here is the msq
2012-10-29_18.42.00CL.msq

and the datalog's. I got it when it was running really rich and oscillating badly.
2012-10-29_18.37.25.msl
2012-10-29_18.38.43.msl
2012-10-29_18.39.58.msl

[Jeremy]

Does it idle ok open loop?

[clongo]

It idled pretty decent at 12.5 - 13.5 afr. If I sat in traffic and the MAT heat soaked it would start to oscillate to the point where it would want to stall. I have not been able to get close to a 14.0 AFR at idle that didnt oscillate. But from what I have read these engine's really like to idle rich. I have the sensor as far from the engine as possible and am still having some heat soak issues with the sensor. I'm not really sure what to do about that issue.

[GI jonas]

Exact same issue here.Having to keep quite rich to idle smooth and after heatsoak oscillation.Perhaps we both missed the same thing.

[Danielni]

I am using "gslender" MS2 extra version of the code.
it includes a lot of idle improvement like timing, VE, PID adaptation to idle rpm error. Strongly suggested.

Even without those versions of the code, I had a stable closed loop idle on my stock B34. They key was to use lower PID setting :
P 20
I 40
D 10
65 Ms PID control interval
RPM Valve close 500, duty 35%
RPM valve open 2500, duty 72%

Note that I have the 3 wires valve, driven by opposite logic transistor (One straight, the other inverted)

Dan

[clongo]

I am using "gslender" MS2 extra version of the code.
it includes a lot of idle improvement like timing, VE, PID adaptation to idle rpm error. Strongly suggested.

Even without those versions of the code, I had a stable closed loop idle on my stock B34. They key was to use lower PID setting :
P 20
I 40
D 10
65 Ms PID control interval
RPM Valve close 500, duty 35%
RPM valve open 2500, duty 72%

Note that I have the 3 wires valve, driven by opposite logic transistor (One straight, the other inverted)

Dan

Thanks for the tip! I switched back to PWM warmup for the time being because the idle was so bad. It still idles odd even in PWM warmup. Tonight while driving home I stopped at a light and it started oscillating really bad at idle. I have been playing with MAT correction and I had to add a 25+% of correction in order to get it to not lean out and die. I captured a datalog when it was happening. My MAT was only 10 degree's above ambient.

2012-10-31_19.19.10.msl

It seems like the VE isn't rich enough when this happen's. But if i try to richen the VE it idles way too rich most of the time. I'm almost tempted to richen the mixture up and take out fuel based on temp with MAT correction. Anyone have any idea's? Does everyone have to use a lot of MAT correction to get their AFR's right? This crap is really starting to drive me crazy. I feel like everytime I start to make progress the car all of a sudden starts to idle like crap for no reason.

[clongo]

IDLE ISSUE FINALLY FIXED!!!!

I was discussing my issue with M.holtmeier and he advised me to try setting "squirts per engine cycle" under "engine and sequential settings" to 2 instead of 6 since im running 47lb injector's. I made that change and as soon as I fired it up my warm restart issue was gone. It idled a bit rich so I adjusted the VE table and viola! No more hunting and rock solid smooth idle!

While driving around the AFR's were really lean. I bumped my req fuel from 10 to 11 and turned VEAL on to get the rest of my VE table where it need's to be. I drove around for about 40 min with VEAL on and this thing run's amazing!

I am still running PMW warmup at the moment. I plan on making the switch to closed loop now that it idles well in PWM warmup. With 1 point of adjustment on the VE table at idle im getting .3-.4 AFR change at idle. I leaned it out to 14.7 and it idled really well with very slight hunting between 14.5-15.0 AFR. I turned it down to 13.8 AFR and it was rock solid. I should be able to get this thing dialed in much better now and look forward to hitting up the dyno soon!

[Nebraska_e28]

Holtmeier is the man!!!

[Brad D.]

Glad you got it sorted out. I am really sorry I haven't got in touch with you yet. Some craziness came up that I had to deal with and haven't had a whole lot of free time.

[mooseheadm5]

You were running 6 squirts per cycle? Those things couldn't have been open for more than a couple of ms at a time, and they do take a set amount of time to pull open.

[clongo]

Holtmeier is the man!!!

Yes he is!

Glad you got it sorted out. I am really sorry I haven't got in touch with you yet. Some craziness came up that I had to deal with and haven't had a whole lot of free time.

No worries man. My life has been pretty hectic lately as well!

You were running 6 squirts per cycle? Those things couldn't have been open for more than a couple of ms at a time, and they do take a set amount of time to pull open.

I'm still learning as I go with MS. It's been a big learning curve for me. Not having number's to reference makes the learning curve even harder.

[winfred]

Peter Florance set mine to 6 squirts when he debugged the unit, and figured out the instructions told me to build the plug and play adapter wrong for my e30, but it only has 32s, i think its like a msd unit and above a certain rpm it drops back to one

You were running 6 squirts per cycle? Those things couldn't have been open for more than a couple of ms at a time, and they do take a set amount of time to pull open.

[mooseheadm5]

I'm still learning as I go with MS. It's been a big learning curve for me. Not having number's to reference makes the learning curve even harder.

I actually don't know squat about it, but if you look at your req fuel and see how long the pulse width will be at idle you should see that with 6 squirts it will be really short with big injectors. Because of the time required to open the injectors, you'll not actually get the full fuelling time set if the duty cycle is very short.

[clongo]

I'm still learning as I go with MS. It's been a big learning curve for me. Not having number's to reference makes the learning curve even harder.

I actually don't know squat about it, but if you look at your req fuel and see how long the pulse width will be at idle you should see that with 6 squirts it will be really short with big injectors. Because of the time required to open the injectors, you'll not actually get the full fuelling time set if the duty cycle is very short.

It made sense after I talked to Holtmeier about it. I have heard of guys being able to run 42lb injector's at 6 squirts without issues. I guess 42lb injector's are the largest you could run at 6 squirts a second. Overall the car is driving much much better. I'm going to do some fine tuning tonight and go through another tank of methanol.

[Patrick McHugh]

I am having the exact same problem as we speak but I was afraid of over using my "tell me how to fix my tune because I don't know anything" card. I'm going to play with it tomorrow.

[Patrick McHugh]

Hey, what are your advanced settings in VEAL set at? I am set at
min rpm: 1300
max rpm: 15000
min fuel: 48
max fuel: 400
Min clt: 160

Do you let it correct at idle?

[Patrick McHugh]

The reason I ask is that my fuel table is tuned very but I get some oscillations when the engine runs right at the transition point between where VEAL has constantly tweaked the table and where I have manually tweaked it.

[Shadow]

I don't use autotune at idle. it may or may not be
a good idea.

For example: The rpm boxes go from 0/1100/1200.
Then you want to start it from 1300 so it doesn't touch your
idle area that you have already tuned by hand.

Autotune will change everything from the 1,200 box and up
and interpolate from whatever value you have at 1200.
Should be smooth.

Turn on your ego corrections too.

[Shadow]

What's so good about using 6 squirts?
I like my 2 squirts

[Patrick McHugh]

I don't use autotune at idle. it may or may not be
a good idea.

For example: The rpm boxes go from 0/1100/1200.
Then you want to start it from 1300 so it doesn't touch your
idle area that you have already tuned by hand.

Autotune will change everything from the 1,300 box and up
and interpolate from whatever value you have at 1200.
Should be smooth.

Turn on your ego corrections too.

I've got them on, but the min RPM set at 1,400. Have you had luck with EGO on at idle?

[Scottinva]

What's so good about using 6 squirts?
I like my 2 squirts

Better throttle response, and response in general.