Idle stabilisation

[palonte]

Hi.
I try to find proper values for the idle and there are lots of troubles. Situation:
- VE and Ign. tables are like "flat" (on that rpm area)
- EGO feedback is off
- Idle valve is closed ( there is a cap in hose ), air flow adjusting is handled on screw
- All injector calibrations are like "flat"
Still there is variations on value "log group fueling/injectors PW__ms" . This variation response (cause) rpm variation.

So, there has to be something else what I have't found, what?

[Jan32]

Do you have an logfile? It would be easier.

[palonte]

Do you have an logfile? It would be easier.

Yes, I get one. This was new features for me, seems to be very good tool.
I enclosed also my application, can laugh freely, my process is just beginning.

I get "The extension emulog is not allowed." when i try to upload it in the forum??
Do i have to send them straight to to some mail?

[Jan32]

You can change the ending of the file. Xxx.log to xxx.doc or something else. Or you can zip it.

[palonte]

Offcourse, stupid me

[Jan32]

The problem is that MAP is also includet in PW calculation look at the manual:

(PW = INJ_CONST * VE(map,rpm) * MAP * AirDensity * Corrections +AccEnrich + InjOpeningTime)

And why do you set the injector calibration values to zero? That is wrong you have to use the values of the injectors. Or did you do this only for calibrate idle?

What is now your actual problem with idle?

[palonte]

OK, thanks now I get it.
Yes calibrations are "flat" because I try to find stabil idle.

Is there some perticular calibration value you are worried about?
- battery correction (0 ms) is not valid at the moment
- Same for the Barometric correction (100%)
- IAT correction (94-92% on that area) is little
- DFPR corr. is 100% (there are no fuel pressure sensor)

I assume that "Speed density" is right fueling type for this kind of engine (nearly std.engine) or should I try to use some other?

Actual broblem in idle is that when I try to lower and lower engine rpm (closing air hole by finger) and when rpm goes under 1100 engine is very unstabil. If there is even small load change (ex. fan starts), then it starts to jumping, even the AFR-value is OK. Then offcourse MAP is jumping and then injectors PW start to jumping allso whitch increace the engine reaction etc. etc.

Is there any way to order that MAP is not so strong factor in idle situation?
TPS vs MAP correction is not the right tool for this, I assume?

Tero

[Jan32]

The values of battery correction can not stay at 0 that is not useful for the injectors, you have to fill them from the data sheet of the injectors.

And it is normal that the idle begin to jumping at load changing, you have to react on it with additional air or with the ignition angle or both. (Additional air is slowly, Ignition angle is fast)

The next step in your case is to activate and tune the ignition control at idle. Fill folowing tables:
Ignition control (Enable it; Use corr. table; set max advance and retard value up to 15)
Idle target RPM
Idle ign. corr. (play with the values)

Or you use the ignition control without correction table.

Your AFR is at idle at 14,7 if you go more rich (13,2) it would be easier. (the engine is more stable)

"TPS vs MAP correction is not the right tool for this, I assume?" Yes that woldn`t help in this case.

[Jadzwin]

What injectors do you use ?
Is your lambda sensor well calibrated ?

[palonte]

Injectors are new. USCAR EV14 (copy bosch 0280158066) 270cc/min
Also lambda is new wide band Bosch LSU4,2 and I haven't done anything for those parameters ( there is warning in the manual ).

Thanks for the very good advises. First time i get situation when engine is not "stump" and there was possibilities to see what is happening.
Also it helps that I change idle valve for 3-wire 0280140537 to 2-wire 0280140516.

There are some difficulties in program (ver.1.062) . When I activated "enable ignition control" in first time, it works but when I did it second time it didn't. I saw in the feedback (and also in ign.angle) that it didn't work. When I boot my pc it start to work again.

Anyway here you have one log/application again, there is one mystery for me. Why ignition angle took all limits for use. Table ask +5 and ignition control "max ign. retarder" gives 15 so final angle is 5-15= -10 and it is stabil even rpm live +/- 200rpm target ?

P.S. Should I use PWM control for that idle valve?

[palonte]

there is one mystery for me. Why ignition angle took all limits for use. Table ask +5 and ignition control "max ign. retarder" gives 15 so final angle is 5-15= -10 and it is stabil even rpm live +/- 200rpm target ?

I found the mystery, target idle and actual idle were too far each other.

Would it be possible to add one feature in program? It would be nice if there can be a curve tool for adjusting sensitive of injection versus MAP on the idle area.
E.g. X-axis has rpm (max value has taken in idle parameters,"idle control max rpm") and Y-axis has 0-100% (default 100%).

In my case there is huge intake manifold and it has very large volume (separate canal for low rpm and high rpm+flaps to change) that's why it is very difficult to react when engine start to jump. Jumping affect too easy large amount injection.

[gregpe]

HI
You CAN'T get the stable idle when you run -15 deg ign idle compensation and final ignition angle is -10 DEG you are firing spark 10 deg AFTER TDC!! no suprise that is a problem..
so try to shut throttle first or by-pas screw if exist , look for air leaks thak go back with ignition in reasonable area and i'm sure idle will be lot better.

[palonte]

HI
You CAN'T get the stable idle when you run -15 deg ign idle compensation and final ignition angle is -10 DEG you are firing spark 10 deg AFTER TDC!! no suprise that is a problem...

Yes I notice that value was too big but that was starting point like Jan32 told.
Situation at the moment is that I found more help for "idle ign.corr." table (after I get the activation selection) but still it's jumping under 1000rpm. I have a feel that idle valve doesn't work propely.
Like I wrote earlyer I have two version of the idle valves. 3-wire 0280140537 and 2-wire 0280140516.
Have you experience which one works more sensitive?

Allso I get the scoop measurement for the valve control and I didn't like the picture. There are 70V spikes (explanation for the USB problems?) as you can see. Is it allowed to use diode over the valve?

[Jan32]

Read the manual. Of course it is allowed to use a flyback diode, they must even be used. I hope for you that the EMU output is not damaged. I use also an flyback diode at my boost valve just to be save.

And it is clear that -15 deg are too mutch if your ignition angle at idle from table is so low. I cant understood your problem with idle, i do the complete idle control withe the ignition angle, at every situation, cold and hot engine. Normally an idle air valve is not needed.

[Kisielisodd]

For boost valve you do not need to use FB diode. For PWM idle valve FB diode should be used.

[palonte]

Yes now I found it also in the manual. Probably the reason why I miss this, was that I use the original wire loom and there wasn't any FB diodes, they had to be in original cpu.
Anyhow this helps also in valve functionally, now it is more sensitive. Still I couldn't find stabil idle under 1100 rpm (try to use both valves). Is there some clearly wrong in these logs?

[Jan32]

Did you use PID or only DC error correction?

[palonte]

Did you use PID or only DC error correction?

Yes I use also PID control. I tryed to use P= 6 or 4, I = 3 or 2 . Quite small values, should I use bigger values ?

[Jan32]

From my experience, i think it is easier to use at first the DC error correction.

[Jadzwin]

Palatone:

You need to disable any correction of DC for the beginning. Set fixed DC and try to tune your VE table to get stable idle.
If your VE is wrong in idle region you will never get good result.
Then use ignition correction table for idle stabilization (for small changes in idle speed).
Then use DC error correction table.