Evenings are starting to get a bit chilly now, so I logged my car warming up after I started it after class. As you can see the idle is reasonably stable until about ~1600 rows into the log, where it starts getting big fluctuations out of nowhere.

Anyone have any ideas of what's going on? It's almost like the ECU suddenly starts looking up a different ISCV table after a little while.

The fluctuations start almost exactly 2 minutes after the car was started.

 

I get the same thing but my hardware is significantly modified. I've been reading the disassembly thread but some of the tables do no seem to have the desired effect.

Have you referenced that thread lately?

What cams and throttle body?

 

Cosworth M2 cams
Stock throttle body

I've been regularly following the ISCV related threads.

 

I had the similar issue when I disabled closed loop in periphery bits.

 

Any update raptord?

I started by zeroing all iscv demand tables to 0 deman( whoa) . Still got the fluctuation . Then zeroed the iscv step vs demand table. Flat idle. Hrmm Jamie said something about a missing link..

 

No updates, I haven't messed around with it yet. Was waiting to see if anyone else would chime in.

 

What can cause this? Let's say that the physical ISCV is not broken or sticking..
Knowing that you're already pretty much good with ISCV stuff, I'm gonna share the ideas that comes on my mind, just in case that it would help you:

To me, it seems like a PID with the setpoint too far from process value. (in this case, target idle too far from actual idle). This cause the controller to overshoot up and down and get out of control.

• It seems to happen when the ECU starts interpolating with the 122F value, in whatever table is causing that. It seems to taper off where the ECU starts to interpolate with the 170F value. So maybe trying to see if you could set another demand value for 122F at main demand tables (or another one)
• Too agressive idle error table ?
• I'd try to reset the BISS screw at any other opening to see if it will get the ECU to keep better control of the setpoint vs process value at this new BISS flow.
• Last but not least, getting ALL the ISCV tables and target RPM tables to stock and resetting the BISS could bring you back closer to a good idle tune. Last year I happened to tweak too much tables ending up all f*c* up, having no choice to restart at the beginning.

Keep updating I wanna know the outcome of your research.

 

Repeat the warmup log, but this time also log the fuel trims. The setup could be drifting lean and the ecu is over-compensating.

 

whats your timing look like.. it jumping around all over the place?

 

When the RPM starts fluctuating, yes. Anywhere from mid teens to high twenties.

 

fix it … it'll help but you've probably got a vacuum leak somewhere.

 

definitely start with a boost leak test and include ign timing, afr, stft, and idle fuel trim in the log. what is your mivec timing in the idle cells. if you want a steady idle, make sure its set to -1.2 deg.

 

There's nothing to "fix". Timing goes up as RPM goes up....

Boost leak test is on my to-do list this weekend, I'm due for one anyways.

 

Any update about what was causing it Raptord?

 

Nothing major yet, I've been too busy with work to have time to tinker around with it very much. I did notice a boost leak last weekend which I fixed, but it didn't have a very big effect on improving idle.

I'll update this thread when I've had a chance to try some of the things suggested here