DBW position translation and input calibration

[Blu302]

I am having problems with PID sensitivity when controlling the idle of my DBW.

I am looking at a way to decrease the sensitivity of the DBW control because the PID settings are at the minimum and there is still an "aggressive" control of idle(some hunting under various conditions).

I have got the position translation table set well to reduce the sensitivity when driving but I need a similar thing done to the idle control.

Will inverting the standard non-linear calibration map reduce the DBW movement at low throttle while still giving 100% at full throttle?

[MarkMc]

Hello,
The PID settings should be fine with the range they have unless your throttle body is way oversized for your engine. Lowering the ignition timing will make the engine less responsive as well. You always need to understand that the tuning is always a combination of fuel, air and spark so all must be considered.

If you turn the P,I and D to 0 does it still hunt? If the tuning around the idle area (fuel and ignition) is not accurate it will fall into areas of good tuning and then bad tuning which will affect the idle. I would start with P, I and D at 0 and use the normal position table to move the engine around and tune the fuel and ignition, especially the areas just below and just above your idle aim.

[Blu302]

The throttle body is for the 3.0L version of the engine(M54B30) and I currently have it fitted to a 2.5L. The factory setup uses an IAC valve to do most of the work at idle(mixes IAC and DBW for idle).

Idle DBW.jpg (130.86 KiB) Viewed 13235 times

The fuel and spark tables are flat and tuned. It does have a fuel delay due to MAF based tune though. When P,I,D is all 0.0 idle is stable.
I am finding that IAT is causing a variation in the "optimum" PID parameters.

[Scott@FP]

Your integral might be a bit high and hyper-active, set up a strip chart showing idle aim, idle position and see if it jumps around too much, with a separate strip chart for idle error, when idle error goes high watch the reaction, adjust I accordingly. P and D will usually be very low values.

Make sure your BOV doesn't leak to atmo at idle/low throttle and upset the true MAF reading, may need to change BOV spring, or recirc the BOV to a point downstream of the MAF and before turbo inlet if it isn't already. Once the air is measured the engine needs to see it or fueling is off.

[Blu302]

The car is n/a. I have just deleted the factory IAC with a plug.

The high "I" still gives a 3-5 second reaction time for the rpm drop. I have the normal position table set deliberately high to reduce the chance of stalling when it starts hunting. "I" gives it a factory like rpm drop when coming off throttle.

[MarkMc]

How much ignition timing at idle. Can you give me some idea of how unstable the idle is?

You also have the fuel and ignition tables for idle control to adjust for the reaction of the idle.

[Blu302]

The timing is flat at 13deg advance, 500rpm each side of idle.

I am starting to find that it might be from high intake temps.
When the car idles at the lights for some time it starts to happen.

It shows as a smooth drop down towards the idle setpoint(normal set a little high), then a sharp drop once there(~200rpm), then hesitating for a second or so where it shoots up again to start the cycle again.