Page 1 of 1

m20 turbo

Posted: Sun Feb 20, 2011 12:20 am
by Mike_e30_325i
hi i am currently turboing my e30 325, i am wireing up a megasquirt 2 ecu but can i use the standed icv, has any one allready done this if so how do i wire it up as i think the icv has a live and 2 signal wires.

Re: m20 turbo

Posted: Sun Feb 20, 2011 8:22 am
by GeoffBob
I am not familiar with Megasquirt, but provided you have at least one spare low-current PWM (pulse width modulated) output available on your Megasquirt, then it is possible to control a 3-wire ICV. I did this on my own engine to great effect.

Most important is that your ECU provides you the opportunity to set-up (in the Megasquirt software/firmware) a proportional-integral-differential (PID) controller in order to control your engine idle speed as a function of engine (coolant) temperature. For this you’ll need a spare PWM output on your ECU which will connect to your ICV via a small circuit (that I will explain in due course). This way you can implement closed loop control so that your idle-speed will self regulate. The idea being that you define the "set-point" (the RPM that you want your engine to idle at) as a function of engine temperature - Slightly higher when cold, progressively dropping down to a comfortable idle speed as the engine warms up.

Be warned though, setting up closed loop idle control can be tricky. If you set your proportional gain parameter too high, for example, the idle speed will become erratic and unstable, get it too low and your idle speed will be badly regulated. If you don’t know what a PID controller is, and why you’ll need to implement one in your Megasquirt, then read here. You'll need to learn to adjust the proportional gain, integral gain and the differential gain values in order to get your idle-speed properly controlled.

Now, to the hardware: The 3-wire ICV comprises two small ”aelectromagnets”a. One electromagnet works to close the valve, while the other works to open it. The two electromagnets are connected to each other to form a common central point that connects to a 12V supply. The idea is then to connect the remaining two points (opposite ends of the electromagnets) alternately to 0V via a transistor each. Turning one of the transistors ”aon”a will energise that particular electromagnet, causing the ICV to want to either open or close. Now, if we were to rapidly turn each transistor ”aon”a and ”aoff”a alternately at a rate of a few hundred to few thousand times per second, then we may control the degree to which the ICV is open by controlling the duty-cycle of the switching signals applied to the two transistors. If the ”aon”a times of each transistor are identical then we have a 50% duty-cycle, and the ICV will be held exactly half open/closed. Shift the duty-cycle to say 75% and the ICV will be held 75% open, and so on so forth.

If you fancy tackling this problem and are electrically inclined then ask and I’ll post up a diagram of the circuit I used to interface the PWM output from my ECU to the three connections on the ICV. Don’t underestimate what you are taking on though. The circuit is simple enough, but tuning the PID values can be a PITA. Also, be absolutely sure to get your ECU mapped with the throttle under manual control only! Closed-loop idle speed control is something that can only be added after you have a stable open-loop idle in the first place. Closed-loop idle speed control absolutely CANNOT cure a bad idle (hunting for example) due to poor tuning in the first place.

HTH
Geoff