Bad idle after start, only with hot engine bay

[Tapi]

I have been tuning idle for some years now with this car (yes, years). I have got it idling nice in just about every situation except this one:

I shut down the car with warm engine. I let it be for 15 to 30 minutes and start it: Idle jumps up and down for a while and it's better to stay parked and wait for it to stabilize. Stabilizing takes somewhere 20 sec to a minute, which in some cases is annoying..

Today I finally logged one this kind of start. The only thing I would change is to ad fuel there but only with these high IAT:s which are present for just a short while. Can it be done?

Mods included: Magnacharger MP122, longtubes, 220/224 117LSA cam, 102 mm T/B.

[Tapi]

Another point of view: I made several logs by standing still and keeping steady throttle. With low rpm the run isn't steady and I can see AFR making it go up and down all the time. I tried to correct VE (this is a SD tune) on these areas to what it asks for according to AFR error but it didn't go any better. When I'm driving and there's even a slight amount of resistance in the driveline, there's no problem.

This got me thinking about my idle again especially when I see the AFR being too lean at startup and going too rich at some point.

My idle runs in open loop. period. I have played with it more than enough to see that with this s/c, this cam and these headers I can not get the WB to show correct reading on idle. Now the idle AFR is tuned with exhaust analyzer so it passes sniffing test (just barely when all the stars point into right direction). So there's no fuel trims to correct mixture on idle.

So what controls the mixture then? I attach here a log of the situation. After start it's lean until narrow band oxygen sensors start to move, then it goes rich. Why do those oxygen sensors affect, after all open loop shouldn't care or should it?

[Pur_SSyn]

Try reducing the Overspeed spark correction.

If that does not help, I would smooth out the timing tables (high, low, idle/park)


Your scan is showing some huge jumps in timing (upwards of ten degrees).

[Tapi]

I divided overspeed spark correction in half. Corrections seem slower now but the problem didn't go away. Yesterday I parked the car after 100 mile trip. I started it again after 5 minutes and I had to keep it running with my right foot. After 30 sec it would idle but it's a long wait if you're in a gas station queue..

[blownbluez06]

Your short pulse adder is hacked up and your low vacuum fueling is way off, partly because of it. Put your overspeed correction back. Your idle timing is too high for your underspeed to work properly. Carry your underspeed values out further. You see when it first starts to dip in RPM, it passes the threshold to the point where it gets to a place in your underspeed table where it's populated with 0's, no longer trying to prop up your engine speed. I see your throttle blade following the downward RPM move, causing me to wonder how your airflow is set. Your throttle blade closes down to 6.45% on its own and it should be at ~7.45% for proper idle in your specific case. Disable LTIT's and log STIT's, making sure they're close to zero. Adjust with AC off, turn AC on and off and cooling fans on and off, making sure they don't affect your STIT's negatively and if they do, work on them in their respective compensation tables and NOT the airflow table. When you hit the throttle during its routine, it's going lean in various places, causing further unstable idle. Also raise your ST idle airflow MAX table at the higher ECT's because you're limiting its memory threshold so every time it gets too far out, it's forced to adjust at that moment as opposed to from memory. Once you get the other items more stable, you can lower those values back down if you choose. While I prefer stabilizing idle with spark over throttle body, you can also use proportional, integral and derivative tables in the idle tab to help control it, but be careful, it's easy to get lost in those tables.

[Tapi]

Thanks, Our friend Pekka didn't praise you for nothing I will study your post tomorrow and go from there.

One notice: I have boost referenced fuel system so does that make any difference when you say my low vacuum fueling is way off?

[blownbluez06]

Thanks, Our friend Pekka didn't praise you for nothing I will study your post tomorrow and go from there.

One notice: I have boost referenced fuel system so does that make any difference when you say my low vacuum fueling is way off?

Ahh, Pekka is good people.

boost referenced has nothing to do with it. Cutting the short pulse adder as you have creates instability in the low pulsewidth areas because it is used to prop up your injector pulsewidth. Start with the easy things first.

[blownbluez06]

Here.

I didn't have much time, but try something like this and email me the results because I may forget to check back with you on you here.

[Tapi]

I think I'm still confused with my injector data. I thought I need to have all vacuum-referenced columns the same because of my 1:1 boost-referenced fuel pressure. Now I understand that might not be correct way of thinking. I modified short pulse adder and short pulse limit but I'm unsure if those are right either.. My injectors are FI114962 (short Siemens 60) so if you see how I messed the injector data up, please advice.

[Tapi]

Stupid me

I think Mr BBZ06 was referring to my VE table in top rows instead of my injector data when he wrote my low vac fueling is off..

But still: How does my injector data (for FI114962 injectors) look to you with more experienced ones?