Amazingly sane idle settings with PWM board
 

I finally installed the jbperf pwm multiplier board. Now 0% idle DC is fully open and 100% idle DC is fully closed. Previously it was closed at 8% and open at 30%. (The old values were on a 1990 engine; those settings didn't work at all with the 1996, which is what motivated me to switch to the PWM board.) I'm using a 16 multiplier and running it inverted.

I could never get closed loop idle to work right with ms2extra, but I'm going to give it another try now that there is a more meaningful DC range.

The only weird thing is that at 0 RPM the MS2 is outputting garbage on the idle pin -- the valve is very noisy when the engine is off because of this. Known bug?

At 200 RPM on the stim it looks good:
http://www.boost-instruments.com/miata/200rpm.bmp

At 0 RPM it's weird:
http://www.boost-instruments.com/miata/0rpm-1.bmp
http://www.boost-instruments.com/miata/0rpm-3.bmp
http://www.boost-instruments.com/miata/0rpm-2.bmp

It looks like the MS2 is outputting garbage and confusing the PWM board.

Probably why is seems everyone's valves hum with key ON. I haven't been having an trouble with idle valve on 90-00 cars. I know y8s just installed his MS3 with the PWM board, I should be able to bear witness to it in the next day or two.

On my 2002, the idle valve would hum even with the stock ECU. The hum is the result of the high-frequency operation, as the idle valve works even when the engine is off (but the ignition is on).

That was my first thought, but the signal coming out of the ms2 shouldn't be different at 0 and 200 RPM -- both should be the crank duty setting. You can see in the traces that the frequency is way higher than the ~500 Hz expected during those solid black periods. It may be that even if the ms2+pwmboard is working perfectly the valve will hum, but it isn't a hum that I'm getting -- it's a very odd warbling/rattling noise.

Did you ground the A_GND pad to a good, noise-free ground?

Hmm. I ran separate wires from GND and A_GND to a common ground point, but I can't say for sure that it is good and noise free. There are two other microcontrollers on that ground plane that work fine at 0 RPM.

I guess the question then would be what makes the ground noisy at 0 RPM and fine at 200 RPM?

Looks like the input signal itself is fucked up when RPM==0. The high frequency noise is making the PWM board think that THAT is in fact the signal, so it multiplies it and that's probably what you see on the scope.

Does it make the humming noise and do you get the HF noise on the input signal when the car is idling?

I get a hum at the PWM frequency... I'm using 390Hz though I have two of the PWM boards. I just got lazy and it works pretty well. I never noticed a hum when the stock ECU was doing the work. Maybe there's a gentler way to control the valve?

There's no evident noise when idling. But I the car is pretty loud right now, so a steady hum wouldn't be obvious. The clacking and banging that happens at 0 RPM (audible from 20+ feet away) is definitely not present at idle.

Yeah, looking at that scope trace, that looks like a bug. I'll have to look into it.

The good news for ms3 users is that they are most likely using one of the ms3's hardware channels. They don't need that extra board, and it is unlikely that the output will do anything odd at 0 RPM.

Ken

Thanks Ken. Before you spend any time on it let me try the release firmware, as James suggested over on msextra. I'm still running a beta from Feb '09 (2.1.0m)

Alright. I'll keep an eye on this and the thread on msextra.

Ken