Crank ISC duty cycle

[CJK440]

As I continue to try to tune cold starts so it will fire and come to an idle without throttle input I find all that is left is to try to tune out an RPM flare at first fire that causes RPM to immediately fall low enough to stall. Second crank fires ok and doesn't drop too low and it stabilizes.

I came accross "Crank ISC duty cycle" in SCT Advantage. I assume this is a commanded ISC position during crank. I looked at a cold idle log and saw that in fuel source 4 (open loop cold base fuel) my ISC was around .660 at open loop idle when my crank ISC duty cycle looked like it was commanding somewhere around .850.

For a test I dropped my entire Crank ISC duty cycle column by .2 and tested. Flare seems to be gone.

I'm wondering if crank ISC duty cycle should be dialed in using logged ISC duty cycle at the various ECT points either to match or be modified by a certain percentage to allow the engine to start and not have such a change in ISC position.

Am I on the right track?

 

[SHAUN351]

You may find this thread useful.

http://www.svtperformance.com/forums/showthread.php?1109136-ISC-duty-cycle

 

[oxfordgt]

LOL that is his other thread.

ISC must be dialed in at operating temp, then you can adjust the startup multiplier.

 

[CJK440]

LOL that is his other thread.

ISC must be dialed in at operating temp, then you can adjust the startup multiplier.

I believe it is dialed in. I needed to load the ISC functions from the 03/04 cobra since that the valve I am using now. Now at hot idle ISC duty is at .29 and itegrator is near zero. Idle in closed loop is fine, MAF xfer is correct.

Startup multiplier multiplies what? I assume this comes into play following initial fire?

I have a crank ISC duty cycle which I assume will move the ISC pintle to a predetermined position based on ECT with KOEO. Once the engine fires and there are sensors providing feedback, I also assume that the ISC is free to adjust to maintain idle RPM.

In my case, I get an RPM flare to maybe 2000 at initial fire, the RPM cuts usually stalling the car. If I give it a little blip to keep it from stalling it settles into a decent idle immediately. I cut back the crank ISC duty cycle table to reduce the duty cycle across the board by .3 and tested. The RPM flare is less and on a dead cold engine, still drops RPM but does not stall. I am not sure if I am on the right track here.

My question is is there a procedure to adjust crank ISC duty cycle based on logged duty cycle. For example if I log immediately after fire and find logged ISC duty cycle to be .660 at X ECT would I want Crank ISC duty cycle to match or be some percentage + or -??

 

[oxfordgt]

What is your spark after crank?

 

[CJK440]

What is your spark after crank?

about 15 once my livelink starts logging

 

[oxfordgt]

There is a table for spark when exiting crank. Set it to 10 should get rid of the flare

 

[CJK440]

There is a table for spark when exiting crank. Set it to 10 should get rid of the flare

This?? Cold temp lately has been around 50 degrees. Seems to jive with the table.

 

[oxfordgt]

No, but that table should be all 50.

 

[CJK440]

No, but that table should be all 50.

That table is set to stock settings for the strategy. Why 50?

I did find the "spark when exiting crank" table. Its set to 10 for 100 ECT and lower. Its 5 for 150 ECT and higher.

 

[03Steve]

First, I would get the logging working during crank condition. Without data it can be a shot in the dark. Fuel source, spark source, commanded lambda, spark, desired idle, ISC duty cycle, relative throttle, absolute throttle, fuel pressure, load, etc. You seem very familiar with the controls, so I'm going to assume you know what you are looking for based upon the information you have provided thus far.

I am not sure what you are using for logging. I am pretty sure that an X2 and Livelink 4.11 and/or Livelink 6.5 will work during a cranking state. I am not sure about other SCT devices or software versions. If by chance you are using Binary Editor 5.76 or similar it will definitely work during cranking.

From there, you have an engine that flares and dies. There are a few things that can cause this. If you are looking to minimize the flare, and you have cut cranking ISC down to the 30% level, it sounds like the closed throttle angle is high. You can try reducing it by wrenching on the throttle stop to close the blade a bit. On aftermarket throttle bodies and engine combos that usually requires some tinkering.

We can make guesses on what spark is doing, but the guesswork will be eliminated when spark source is identified during a cranking state. It doesn't always come from the table shown above during crank. Sometimes it will actually come from MBT. Load is typically above 50% during crank and RPM is around 150-200, so there are cheats you can do there when needed.

For air through ISC, there are a few possible RPM adders. Time based, ECT, ACT, A/C, etc.

For fueling, if the engine flares and dies sometimes it can be due to fueling, but not always. Capturing fuel pressure and commanded lambda is useful, along with the fuel injector DMR of your choice.

In summary, more information is needed during cranking.

 

[CJK440]

I am using an X3 and have both versions of livelink but default to the older version.

To be honest, I just assumed as soon as I hit the key, the X3 would shutoff and would never log. I know my wideband will be not reading for a while no matter what.

I assume I start logging KOEO and once the PIDS report give the key a twist??? Sorry for the elementary question.

 

[03Steve]

20 second countdown on power up for the wideband I assume? You can move forward without it for now. Have a look at what is being commanded, fuel pressure, and injector output to quantify fueling during crank.

Yes on the last question. That will do it. If you want to get fancy unplug the fuel pump. You can get a few different reports, and analyze the data before firing and temperature increases takes place. Battery charger is useful.

 

[CJK440]

Well you learn something new every day. I started logging with livelink 6.5 with KOEO and cranked it and no issue. I kept logging until it went into closed loop.

What is interesting is my most recent tune fired up and went into an idle perfectly. Ambient/ECT was 52 degrees, 24 hours since last run.

The things I fooled with are.

1-Added .3 to a block of base fuel table cells (circled in green) and feathered each adjacent cell surrounding by reduced amounts. I did this because I was having a major rich condition until it went into closed loop. 10.5 rich. This seemed to solve it even though I must say I'd rather find a true culprit. Now after I did this, once it fires it goes quickly to 13.5 and no more black smoke on fire.



2- I subtracted .3 from the entire crank ISC duty cycle table. This minimized the flare. ISC duty cycle during crank is .531 which appears to match what this parameter would be commanding when I eyeball interpolate.

3- I changed idle speed adder from 200rpm for 30 sec to 300 for 10. Perhaps this "caught" the RPM drop to keep it from stalling.

It seems to me a bunch of bandaids but since its just cold start and if I get it to behave I suppose i can live with it.

So on to my crank datalog......

During crank....

ISC duty cycle .531
Spark advance 8 degrees
Spark Source 0
Fuel source is 15 up until it fires than goes to 4

If you are looking to minimize the flare, and you have cut cranking ISC down to the 30% level, it sounds like the closed throttle angle is high. You can try reducing it by wrenching on the throttle stop to close the blade a bit. On aftermarket throttle bodies and engine combos that usually requires some tinkering.

At hot idle my ISC duty cycle is .300. If I close the blade, this will be higher and I don't think I want it higher right?? The engine RPM drops ever so slightly when I unplug the ISC when hot.

 

[oxfordgt]

I have my base fuel cold set to all 1 then use the startup enleanment table to adjust fueling.

 

[CJK440]

I have my base fuel cold set to all 1 then use the startup enleanment table to adjust fueling.

I tried altering the enleanment table first. What is the relationship of the enleanment table? Is it a multiplier for teh cold base fuel table? Mine was all 0's by default so unless 0 means its not referenced I assumed it was an adder or subtractor. It has a min value of 0 and max of 1.990.

What I did was leave the base cold table stock and made the entire enleanment table .2 and noticed no difference during open loop after cold start.

 

[CJK440]

One other thing I didn't pay attention to yesterday was spark advance when transitioning from crank to idle.

It starts as a flat 8 during crank but once it fires, spark jumps to 24.5 degrees peak in about .4 seconds. It then immediately drops to 15. All of this happens within 1 second of firing. Perhaps the culprit of my RPM flare and stall at first fire?? Or perhaps this is normal. See below.

Yellow is spark adv
Blue is RPM
Purple is ISC DC
Maroon is fuel source


What I don't understand is the idle (cold and hot) spark relationship in advantage vs what I log. There are discrepancies.

Crank spark should be 10 for all ECT under 150. (but I logged 8)
My exit crank spark is stock at 10 for ECT under 100 (but I get the 24.5 degree flare.
Desired idle spark neutral is 63.75??
Min idle spark is 10 (but hot idle fluctuates between 8-10)


As of this AM when cold starting the RPM flares to 1400 or so and stalls. Next attempt does the same but stays running where it goes to a decent idle. At least now I can start the car cold without fogging up the place with unburnt fuel while I pedal it for 10-15 seconds to keep it alive. But still something seems off.

 

[CJK440]

It starts as a flat 8 during crank but once it fires, spark jumps to 24.5 degrees peak in about .4 seconds. It then immediately drops to 15. All of this happens within 1 second of firing.

Quoting myself here. One other thing I noticed. The RPM flare is accompanied by 25 or degrees of timing for a fraction of a second. I also noticed that high timing is during spark source 1, base timing.

So on a crank datalog my spark source starts at 0 (crank) then when it fires goes to 1 (base) for a 1/10th of a second then goes to 9 (idle feedback) for a tiny bit then goes to 7 (cold engine low load).

Should spark source be going to base right after crank???

 

[03Steve]

Fourth paragraph of post 11

 

[CJK440]

Fourth paragraph of post 11

Ahh, I see, makes sense. It sure looks like the brief time it jumps to base timing (immediately after crank), it seems to apply an amount of advance that MBT is showing at the same load/RPM.

Is it just a matter of me nerfing the MBT table in those cells to kill the flare?? Perhaps match the BSK data in the cells affected?