Idle control on specific conditions

[AVP]

which gauges should i be looking at?

are you sure the integral has nothing to do with the I value? if you check the help file it says that it is related to the I value you use. This is something common for all PID settings.

i have to try and see if i have any log file, although it would be best to try something with the current settings rather than old settings.

[AVP]

although i settled for the PID settings that seemed to work for most if not all occasions, today it started happening again.

here is what happens: when im in gear and i take my foot off the accelerator, the car goes to full vaccum (15-18kpa if rpm> than 2500) , if at that point i go into neutral,----> although lamda should go from 20.1 (fuel cut) to lamda=1.01 at idle, it stays 20.1 all the way to the lower rpm point, until the rpm dips in and then jumps up again as it should stay on the idle area.

i have: fuel cut until 1600rpm,

fuel resume: 1500rpm

and i use fuel cut below 30kpa.

WHY vems does not resume the fuel as the rpm drops below 1500rpm although the map kpa is 30kpa?  Is that supposed to happen?

i thought that the fuel was resuming below 1500rpm regardless of the kpa value that i set!!!

Even so, on various occasions im below 1500rpm and kpa above 30 and still i see lamda= 18.x something!!!


To try and see a difference, i set fuel cut 1700 and fuel resume 1600rpm and also the fuel cut below 24kpa.

it seems to be a bit better now, so ill recheck and report.


BUT im wondering why those figures dont always comply.

i also have the EGO correction Freeze enabled. Does that have anything to do with it??

[gunni]

post your log of this occurance

[AVP]

unfortunately i was out for a drive, i didnt make any logs.

all this is described by observing the LCD vems display of the lamda and map on the fly

[mattias]

i thought that the fuel was resuming below 1500rpm regardless of the kpa value that i set!!!

No, why would you assume that?
I find that it's usually conflicting to use both, so I use mainly rpm based overrun fuel cut.

Also remember, when you have used fuel cut the intake is very dry, it can take a while for the engine to play catch with the intake wall wetting.

[AVP]

so, what do you use? you set map = 0?  and go by rpm only?

[mattias]

so, what do you use? you set map = 0?  and go by rpm only?

Of course.

which gauges should i be looking at?

are you sure the integral has nothing to do with the I value? if you check the help file it says that it is related to the I value you use. This is something common for all PID settings.

i have to try and see if i have any log file, although it would be best to try something with the current settings rather than old settings.

Why are you asking the obvious? The IAC gauges are what you should be looking at, bring them up. IAC int, duty cycle and target rpm are the main ones of interest.

What do you mean that I'm sure the integral has nothing to do with the I value?  Of course it's related.
If you minimize the increase/decrease the PID has no impact on the idle control. If you check the target and actual rpm, and the IAC int value you should be able to draw a conclusion pretty quick on what's happening.

Share that log that I was asking about before.

[AVP]

i see
i will try and see if i have a log that has some kind of these circumstances and post it up.

i wasnt aware of these gauges so this is why i havent been paying attention to them. I was tuning the idle the same way i did on megatune, byt change and feel.

i will also see if i do the same with the fuel cut issue, maybe it will be better.

[mattias]

About 90% of all datalogged variables don't have a gauge, you have to bring it up - nothing new here.

Some gauge group screens have more or less gauges of interest, depending on what the intended use for that group is.
Example VE tune is for VE tuning, not idle tuning - but I use that view most of the time and just right click on a few gauges and change them into whatever I need to monitor.

[AVP]

http://www.box.net/shared/nuskyfkxl7

here is a link to a log that i have done a few days ago, and the PID settings i have there are the same as the ones i have today.

only difference today is the fuelcut and resume rpm and the map kpa to 24 for fuel cut.

please explain to me the iac int, and rpm target as it goes through the log.

i noticed that the rpm target somewhat follows the actual rpm when not on idle. why is that?

[AVP]

has anyone seen the log??

[mattias]

I checked it out, your "IAC int" value is railing at -20 trying to do it's best to lower the RPM down to the target RPM.

Your ignition based idle control is what's doing most of the work, but it's clearly not enough.
20 deg advance and 14 deg retard is a lot of power to this function.
I would limit it to +/- 5 degrees at most unless the setup has no idle air valve.

The IAC close delay closes the  valve completely after 4 sec of the throttle being above the idle threshold. I usually set it to a much higher value, like 20 secs which means I won't notice it much when it's closing. On a boosted car it could be important to allow the pedal to completely control the air flow, not that I've given it much thought..

Your Idle air reference table looks the opposite of what I would expect. Less duty at low temperature, and more at high temp. Are you sure it's wired in correctly? Because you haven't given it any power to do much about the idle via the integral increase/decrease so that you may not have noticed this fact. In your case the reference duty cycle table is what dictates the idle speed the most  + the almighty idle ignition advance control.

[mattias]

Listen.. set integral increase/decrease to 0.
Disable the ignition advance idle control.
Bring up gauges for target rpm and idle air duty cycle.
Open the dialog for the idle reference duty cycle table.

Cold start the car. Change the cells to have duty cycles so that your idle speed is 300 rpm above the target rpm in the gauge.
Watch the target and actual interpolated (from the table) duty cycle and what it does to your current engine speed.

It's very important to get the correlation that a higher duty cycle = higher RPM. If this isn't true, you must set the idle solenoid channel to it's inverted function to make any kind of idle tuning possible and make sense.
Make sure that you see that the ignition advance during idle is always what you have set in the ignition table as the advance greatly affects idle, just what the idle ignition advance control does when activated.  Hopefully this will be replaced by a proper table just for the ignition  advance at idle.

When the reference table is set, use the warm idle duty cycle for the cranking values for both hot and cold engine - that usually works out well for me.

Now you can enable the idle ignition advance control if you like, good values would be 5/5/25/4 (top to bottom). Then it's time to play with the PID and make it behave.

[AVP]

as i was using in megatune, i had improvised a system where the idle was ignition controlled mostly during warm up to keep the revs up, and then the ISV came into play to some good extent.

regarding wiring, since you have already mapped an S2 you should know that there is nothing that needs done with wiring on the audi 5cyl engine, as the VEMS is plug and play. So i havent touched any wiring to make it become opposite in the ISV reaction.

as i mentioned before, if on these excact settings i use integral limit higher than 20-25 the idle rises, it does not fall.


i will try to give your idea a go to see the reaction i get.

[mattias]

I have done mostly S4, the only S2 cars I have played with have had custom intakes with no ICV. 
Compared to an S4, you have half the duty cycle in your reference table. It should be more like minimum 35-40% and 45-50% when cold.

Edit: it could be that your TPS is adjusted to have a higher idle position..