94 to 97 1.8L Idle Control
 

The 94 to 97 1.8L IAC has two controls. 1 is the electronic signal produced by the stock computer or your megasquirt. The other is a mechanical valve similar to a thermostat where the opening is controlled by the coolant temperature. I'm assuming that the valve closes as the coolant temperature increases. This would aide idle control, if there is a way to account for it. Unfortunately, that is not the case with megasquirt. It adds another variable that the PID controller has to deal with.

Some users have noted that the same RPM takes a different number of idle steps each time idle resettles. This could be the source unless a lot of us have inaccurate IACs (from the electronic side).

I have a 96 1.8L w/ 99 cylinder head. My throttle body is fully closed at idle. I could remove the idle screw if I wanted and no it is not stuck. So, the mechanical portion of my IAC is allowing about 650rpm worth of air flow into the engine at idle. With the electronic portion of the IAC fully open, the idle is 3000rpm!

Here is my thought on a potential solution for closed loop idle: 1. Disconnect the coolant line to/from the IAC and bridge them. 2. Fill in the air passage for the mechanical (temp) portion of the IAC with JB Weld. 3. Use my idle screw to effectively set the baseline idle along with CL idle control.

Anyone have thoughts? I've seen that the basemap for the 99/00 uses CL idle control. Can anyone confirm if these IACs have coolant connections and a mechanical portion for idle control. I suspect that they do not. Any other suggestions?

Is anyone with a 94 to 97 1.8L running CL idle with the stock IAC setup? I really doubt it.

Thanks

Did Brain and the other regulars take the weekend off?......

my idle on my 96 has been all screwy lately. Cold it wont idle less than 2,000 RPM ... warm it eventually will hang out around 1400. Megasquirt closed loop changes seem to make little difference. No major vac leaks anywhere.

I am going to diagnose this a bit more this week. It seems like something is stuck open.

Do you have a CL Idle indicator on your dash? Are you verifying that it's illuminated when you think you're at idle? It will take a couple seconds for your car to realize it needs to idle, but if the CL Idle indicator isn't lit, you're not entering the PID control for CL Idle. Are you using Warmup or CL?

you post doesn't make sense is why, the air valve is only open in coolant temps under 70°F and shouldn't matter as far as teh MS is concerned.

What is it that you want?

I am using CL idle. I should be able to investigate it more tomorrow.

Check to make sure your AFR Targets make sense if you're using EGO control. Also, make sure your fuel map is pretty well tuned, or it will search naturally at idle. Make sure your spark is pretty good too. There is a really good map in one of the topics of the Super Important thread in this Forum.

Brain - Didn't realize it's only active below 70F. Mine might be stuck open in that case. I'll try unplugging my IAC when the car warms to above 180. It shouldn't be able to start or should stumble off right?

if you unplug the IAC, it just wont open and you should idle at your lowest rpm.

Brain - are you implying that the IAC uses the last signal as a standing position, so once the wire is pulled it will retain it's last open value? Or, are you implying that the throttle screw will hold the throttle open at it's lowest value. If the later, my throttle screw is recessed so far that the bracket doesn't even touch it when the throttle plate closes. And I still get about 650rpm with the IAC unplugged. That's why I was thinking some part of my IAC was sticking open. Any thoughts?

Ended up figuring out that my EGO was fighting with Autotune to properly tune my VE Table - please see separate thread. I'd really appreciate your input about the AFR Target table.

Not to but in but my idle isses were my own stupidity:
Problem A)
The closed DC value was set higher than what I needed
The MIN DC value was set higher than I needed.

Lowered these down and I got my idle from 2500 to 1500.

Problem B) (this one is real dumb!)
my throttle cable was slightly too tight so the butterfly was sometimes still slightly cracked open.

Fixed that and now I am idling at a steady and smooth 1000 RPM.

I recently just got rid of the p/s pump and replaced it with an idler pulley. New belts on with proper tension and I was set to start messing with closed loop idle with a/c.

I have cleaned the IAC valve as recommened through several write ups. Ohms are within spec(13ohms measured with no power) and steady 12volts from the connector.

Problem I am having is that the idle test shows no fluction with changes to dc %. Idle pwm % gauge reflects the input in "idle valve closed duty %" under idle valve settings. I have no high rpm idle, with the idle screw all the way out will give at max 1200 rpm and stall out if screwed in all the way. A/c on my idle is around 850 rpm and without is around 1000 rpms.

I have a datalog, however it fails to upload as an attatchment.

msq below for reference:

So has anyone tried blocking the thermostatic part of the idle valve?

I machined off the coolant ports but I think under hood temperature is affecting the idle valve behavior. The same duty cycle produces different idle speeds even on a fully warmed up engine at different times. Sitting in traffic (heat soak) seems to make the idle droop.

I'm still running PWM warmup because I haven't had time to futz around with the PID idle yet.

The idle on my 97 is experiencing a similar situation. However, I am using the stock computer to control the IAC valve. On my car, the car starts and warms up correctly. Soon as I start driving the rpms would hang at 1400 and take 3 - 4 seconds to return to idle every time. After the RPM's settle to normal idle, if I jerk the steering wheel back and forth real fast, it will cause the RPM's to increase and hang again. Its driving me up the wall.

Any ideas?

Mines been blocked forever, but it still shouldn't cause issues unless the thing has stuck open.

This air valve we speak of is located on the IM (has two coolant lines running off it). There's a machined port on the IM flange that allows air passage up through Idle valve, into the galley, and then into the air valve. If the coolant is below 70°F, the wax valve will open and allow air to pass the valve and dump into the IM bypassing the throttle plate. When the coolant breaches 70°F it closes.

So the only effect this should have is slightly increased RPMs when it's really cold out.


The coolant ports inside the TB simply prevent frosting inside the valve.



see: http://www.boostedmiata.com/pdfs/servicemanual.pdf page 251. I'm pretty sure all 90-97 miatas have it. In '99 they use VICs.

The 94-97 valve isn't on the IM like the 90-93 one. You probably know that. It's under the throttle body on my 96.

I haven't driven mine since I removed the thermostatic valve entirely but prior to this I was seeing the same idle valve pulse width produce different idle RPMs even when the coolant and oil temperatures were the same. Drastically different -- like sometimes 600 and sometimes 1100 RPM. If it's not the thermostatic valve then I don't know what it is...

As long as you guys have seen fit to hijack my thread, I'll chime in too.

I'm also still having difficulty with my IAC still. It's gotten about 50% better since I dropped my 800rpm ign advance to 14deg and 500rpm to 17deg (like Brain has) and I brought my idle AFRs down to 14.5. I might try lowering idle AFR to 14.0 or 13.5 if that doesn't work. I still get really low rpms sometimes when enganging the clutch to move forward a bit in traffic and then letting off slightly after to brake.

However, my car no longer tries to die or take forever to settle when I throw the lights on. Oh, putting the turbo on actually helped idle settle out a bit to - He he! It's like a big damper on the exhaust and intake side. So, everybody should get a turbo justified by better idle...

Brain, can you post your ign map, your idle fuel afr targets, and your IAC settings all at once? I understand fully that my numbers will vary, but they should still be relatively similar. Thanks.