PWM Idle - No reaction. How to test?

[nlabrad]

So I'm trying to drive the PWM Idle solenoid of a 90 Miata. I have replaced Q4 for a TIP120, removed D8, removed Q12, jumpered R39 and I have the diode between S12 and FIDLE with the banded side towards S12.
First of all I want to say that I purchased 10 TIP120/122 and have already tested replacing the one I initially had and I think the problem persists.

No matter what I do there is no reaction from the valve, I have tried disabling the idle and using programable outputs to activate FIDLE manually but no go.

So what I have is the following:


-The CPU activates 5v or 0v on the PIN34, so that side is fine.
-I get 5v between GND and the bottom lead of R19
-Top side of R19 against GND, I get 0v. I guess this is because the Base conducts towards the E which is grounded? Not sure here.

So what I've been doing to test it is put one lead of my multimeter in the collector (or the lead on the right of D8), and the other end to 5V, if the TIP120 was allowing flow between the C and the E then the C would be grounded right and I should have 5v between VCC and the C of the TIP120 no?

If this is correct, I'm not getting anything. No matter if pin 34 of the CPU is at 5v or at 0v I get no difference at FIDLE/C of the TIP120.

I've also tried the conductivity test, the one that beeps.
Between GND and the collector I was expecting to get very low R when the base was powered. But I get 600R having the COM in FIDLE and the red pin of my multimeter in GND, so if you ask me I'd say it's flowing from GND to the C.

Please let me know if there is a better way to test this, I'm at loss.

I know this might be a confusing way of showing what I've done, but I think I have my whole circuit done right and it's still not working.

The PWM valve works because the stock ECU controls the idle, poorly, but it does.
Thanks!

[Matt Cramer]

Can it turn on the FIDLE light on a Stim?

[nlabrad]

Can it turn on the FIDLE light on a Stim?

Unfortunately my stim is 8000km away. Can I make a circuit on a protoboard with an LED somehow?

[Blown88GT]

Your test circuit won't work without a load, PWM Idle solenoid OR R39.
If you don't limit the current between C & E, you'll blow the blow the transistor immediately.
The diode serves as a voltage clamp for the solenoid when the field collapses.
Google "open collector circuit".

[nlabrad]

Your test circuit won't work without a load, PWM Idle solenoid OR R39.
If you don't limit the current between C & E, you'll blow the blow the transistor immediately.
The diode serves as a voltage clamp for the solenoid when the field collapses.
Google "open collector circuit".

Oh, so you're saying that I should KEEP R39 to prevent the TIP from being killed then?

I was following this, perhaps I mis read it, but I think it says to JUMPER R39, no?



Taken out of here:

http://www.megamanual.com/ms2/IAC.htm

[slow_hemi6]

Your MS has MSextra firmware on it so why are you not following and using the MSextra manuals?

[Blown88GT]

...Oh, so you're saying that I should KEEP R39 to prevent the TIP from being killed then?...

I didn't say that.
I said this: "Your test circuit won't work without a load, PWM Idle solenoid OR R39."

[nlabrad]

Your MS has MSextra firmware on it so why are you not following and using the MSextra manuals?

You're right, I started this project long time ago and I guess I haven't updated my sources. I have re-installed R39 because the msextra manual doesn't ask to take it out, I'll test this in the car in 15 min, hopefully it will work

[nlabrad]

...Oh, so you're saying that I should KEEP R39 to prevent the TIP from being killed then?...

I didn't say that.
I said this: "Your test circuit won't work without a load, PWM Idle solenoid OR R39."

My bad, I read it properly now. I'm going to test it in the car right now with R39 installed.

[nlabrad]

Well that still didn't work.
I'm lost.

I even thought well perhaps I killed the TIP120 when soldering it, so I changed it again and did a quick solder so it wouldn't get too hot, no dice.

The valve doesn't react AT ALL. I'm pretty sure it works with the stock ECU because when I turn the AC on I have idle up, and there doesn't seem to be another ECU controlled valve in the car.
I'll try to check again if I'm sending the signal to the correct wire, but I'm pretty sure I am.

Any other ideas I may try?

[nlabrad]

Well I've been doing some more testing.
I've put an LED with one leg to 5v through a 330R and the other leg to the PIN30 of the MS.

I've removed the CPU and I'm jumpering 5V to the PIN34 of the CPU socket.

I still get nothing. There has to be something wrong here. I re-wired the TIP120 to the right side of R39 and the right side of D8 to make sure the leads weren't touching in the Q4 location.
I still don't know where else to look. Seems like perhaps the problem is in another component. I'm almost thinking of building this circuit in the protoboard and move on, but I rather not.

[nlabrad]

We might have gotten lucky! I just measured and got the pin30 of the DB37 to go to ground when I put 5v at the pin34 of the CPU!

So that means that this is working.
I have two suspects.
One is the pins in Q4.
The other one is the TIP itself against the heatsink, I was using a metal screw with the spacer, but I left the TIP floating and that's when it worked (not that this was the cause, but it was floating when I got it to work).

So now I've put the TIP back on the heatsink and everything seems good still!.

So we'll see, I can't go out now and put this in the car or the neighbors will murder me, but tomorrow at 8am I'll be there testing the PWM IAC. If this thing works I'm approaching 90% of the controls the stock ECU had. Then I add flatshifting and all those fun things and I'm good.

[nlabrad]

It's working now!
I think re-wiring it did it. Not that the Q4 locations don't work, but there might have been some contact between the wires somehow.