IAC Closed Loop Question

Tuning concepts, methods, tips etc.

Moderators: jsmcortina, muythaibxr

ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

billr wrote:Try raising the last (highest) temp in your "closed-loop idle target" up higher. It is presently about 143F, under where you are idling, and I'm not sure if MS sets the target flat at that last value from there on, or extrapolates the target to a lower rpm based on the preceding shape of the curve.
After raising it up above 170, the engine did cool off to about 163 while I was making the adjustments. I did notice that after restarting at 163 it did keep the idle slightly higher then my target 675 idle. I also noticed that shifting into gear the idle would drop low and very slowly came back up to about 650 (very rough though). However, I would shift back into neutral and the rpms would go even higher and NOT come down. I took a datalog of this:
2017-04-22_16.37.21_Billr.msl
Will
SwedCharger-67
Experienced MS/Extra'er
Posts: 264
Joined: Sun Dec 23, 2012 2:20 pm
Location: Nordmaling, Sweden

Re: IAC Closed Loop Question

Post by SwedCharger-67 »

Your stepper motor control doesn't work properly.
It's extremely slow to open on dropping rpms, and in the end, once open it doesn't close on increasing rpms.
Martin, Sweden
Mopar 512 cui, 8.4 liters, high compression, fuel E85, full sequential control by MS3X, burning rubber just blipping the throttle... :D
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

SwedCharger-67 wrote:Your stepper motor control doesn't work properly.
It's extremely slow to open on dropping rpms, and in the end, once open it doesn't close on increasing rpms.
I figured something like that, how does one make the stepper motor work faster?

Will
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: IAC Closed Loop Question

Post by billr »

That isn't the whole issue. Going into "idle" is being inhibited by some threshold settings. I think I noticed MAP and rpmdot when I peeked at your log last night. What MSQ applies to those most recent logs? I think you were trying an MSQ provided here by "turbo conversion", but I wasn't sure whether it applied to both logs, or only the "david" one. I'll try to study what you have posted and formulate a more logical reply, but it won't be until tonight.
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

billr wrote:that isn't the whole issue. Going into "idle" ismaing inhibited by some threshold settings. I think I noticed MAP and rpmdot when I peeked at your log last night. What MSQ applies to those most recent logs? I think you were trying an MSQ provided here by "turbo conversion", but I wasn't sure whether it applied to both logs, or only the "david" one. I'll try to study what you have posted and formulate a more logical reply, but it won't be until tonight.

I used the modified msq (closed loop again 3) by David (turbo conversion) and got the msl that I titled David. However the msl that I entitled billr i was using my original msq posted "closed loop again". This is the one that you looked at and suggested that I raise the highest temp on the target idle.

Will
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: IAC Closed Loop Question

Post by billr »

Well, I have been staring at those files for a while now (including the "sammy" series); I'm not going to be of any help right now. I am especially confused by the "CL_mystery.msl" that (I think) goes with the "closed loop_again.msq" of 4/18. That log shows the idle rpm target as being a steady zero, I can see nothing in the msq to account for that...
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

billr wrote:Well, I have been staring at those files for a while now (including the "sammy" series); I'm not going to be of any help right now. I am especially confused by the "CL_mystery.msl" that (I think) goes with the "closed loop_again.msq" of 4/18. That log shows the idle rpm target as being a steady zero, I can see nothing in the msq to account for that...

Yes you are right about the files associated with each other. Thank you for your time and looking. Closed look tuning is pretty much a mystery to me at this point. It is all for getting it to idle straight when put into gear. I dont know why them target idle was zero in that log. I was wondering a lot about the advance because when the the kpa goes up (putting it into gear) even if i get the idle where i want it, it is still very rough. At this point, I may want to consider a manual transmission.

Will
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

Tried increasing the control loop gain. It does make a noticeable difference; however, I it doesn't seem to get lined up correctly. When I start the engine it will go to the target idle (675 rpm), then i shift into gear, the rpms drop but then come back up slowly and stay around the target idle. When I shift back into neutral, the rpms rise, to just under 1100 and stay there. Put it back in gear again and the rpms go down to target idle; then back up to 1100 into neutral. The only thing that stops this cycle is if it shut it off and start it back up again. Then it will start the whole process over again.

The control loop gain was at 500, I increased it to 2000, then 2500, then 2800, then 3100 and the results were the same, it was just getting faster at moving the idle.

Control loop gain 2000:
2017-04-26_21.31.55_CL2000.msl
Control loop gain 2500:
2017-04-26_21.34.48_CL_2500.msl
Even taking the control loop gain up to 3100 the idle never oscillated.

Are there any variables that specifically recognize that the rpms increased too far above the target idle?

Will
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: IAC Closed Loop Question

Post by billr »

I think it is the "integral" factor you want to increase, but you will have to change from "basic" CL tuning to "advanced" to play with that.
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

billr wrote:I think it is the "integral" factor you want to increase, but you will have to change from "basic" CL tuning to "advanced" to play with that.
I changed around the "integral factor a lot and took some datalogs. I did get to see that it stays closer to the target then before but it still the same thing. Idle at target, put it into gear and the idle drops and comes back to the target. Shift out of gear and the idle goes up to 1100 and stays there. One time it dropped the idle back down to the target, but it was at a random point and I could not repeat it. Honestly, I think it happened after I made a change in tunerstudio and "burned" it to the ecu. It was as if the ecu woke up and realized it was not doing what it should.

I did notice that while warming up, the CL target idle will slowly come down and will keep the rpms at whatever number it is at.

Here is a log of it if your interested (not much that is different):
2017-04-27_20.24.08_CL_Iterm.msl
Will
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: IAC Closed Loop Question

Post by billr »

Post the MSQ that goes with that most recent MSL.
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: IAC Closed Loop Question

Post by turbo conversion »

Give this a try.

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

billr wrote:Post the MSQ that goes with that most recent MSL.
This is the MSQ:
2017-04-28_16.34.18_CL_I-term.msq
Going to see if I can figure out some more stuff.

Will
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: IAC Closed Loop Question

Post by turbo conversion »

The reason it is hanging at 1100 rpm is the Load Threshold % was set to high, I lowered it in the last msq I posted.

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

turbo conversion wrote:The reason it is hanging at 1100 rpm is the Load Threshold % was set to high, I lowered it in the last msq I posted.

David
SUCCESS!!

It is successfully raised the idle after the idle drop into gear and lowers it after the idle rise when going into neutral. I am under the impression that increasing the closed loop gain will speed up the process of it recognizing the increased and then decreased load (or the response time getting back to the target)?

Here is the log if your are interested:
2017-04-28_16.51.54_CL_idleup.msl
Will
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: IAC Closed Loop Question

Post by turbo conversion »

Now that the IACV is working good enough for now you need to tune the VE where it is going lean once in gear.

Once you get the VE table dialed in through the idle cells you can then start fine tuning IACV.

David

EDIT: note the difference between the afr at 36.580s and 45.343s
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

turbo conversion wrote:Now that the IACV is working good enough for now you need to tune the VE where it is going lean once in gear.

Once you get the VE table dialed in through the idle cells you can then start fine tuning IACV.

David

EDIT: note the difference between the afr at 36.580s and 45.343s
I did notice that it is shooting up around 15 afr, and was wondering if that is the problem. I will be working on that tomorrow.

Will
turbo conversion
Super MS/Extra'er
Posts: 1281
Joined: Thu Aug 19, 2010 6:22 pm
Location: White House, TN USA

Re: IAC Closed Loop Question

Post by turbo conversion »

Once you get the afr stable in the idle cells you can start increasing the PID numbers,
this will make the valve respond much quicker.

I would start by raising the I by 10 then start increasing the P until you start to get oscillation then back it off until it stops.

P is proportional (how fast the valve moves to target).
I is integral (how accurately the valve maintains target).
D is derivative (to keep the valve from over shooting target)

David
1976 Datsun 280Z L28ET Garrett GT35R T3-T04E stage3 50 trim 63 A/R housing custom grind cam 2000-6000 rpm 440cc injectors intercooled 18 lbs. boost
3" exhaust turbo back LC-1 o2 sensor Hallman manual boost controller EDIS 6 ignition batch fire 60mm throttle body 5 spd T5 borg warner 3.54 lsd
ironwill11
Experienced MS/Extra'er
Posts: 199
Joined: Sat Sep 17, 2016 9:22 am

Re: IAC Closed Loop Question

Post by ironwill11 »

I have found that I am not exactly the best at adjusting the cells to dial in afr. I assume more practice makes this better; however, I did have a couple questions.

Is it advantageous to use the tools like interpolate and smooth cells (I don't really have a good idea on how these work)?

Can and should you use decimals in the cells? If I try to increase a number by 0.5 it rounds it up to the next number in the cell.

Will
billr
Super MS/Extra'er
Posts: 6828
Joined: Sun May 15, 2011 11:41 am
Location: Walnut Creek, Calif. USA

Re: IAC Closed Loop Question

Post by billr »

I use the smooth and interpolate tools. they help get rid of "spikes" that real engines don't have in their breathing or mixture requirements. That being said, I am also not the "best at tuning", and my idle may be the worst of anybody's!

I wouldn't worry about decimals in the VE table. If you VE values are "real", like in the 50-150% range, then that .5 is only a 1% step at the most. I assume you have "multiply" and "use AFR target" both on and the "req_fuel" set to the calculated value.
Post Reply