IAC Valve Power Problems

[MattsMSP]

I've had an idle problem with my MSP for awhile, but only on warm starts or when it's warm outside (70s and above). It fires, drops to idle, and then drops way down, stalling on occasion. I usually hold my foot on the pedal so it 'runs' around 1k rpm until the symptoms go away (about 20 seconds or so). Then it runs great until the next time around.

I've cleaned the IACV twice, hunted for vac leaks, replaced the plugs and coils with no luck. Last night I decided to take the IAC solenoid out and test the thing to double check that it was working. I set it up on the bench with a 12V source and all is good - valve pulls in when voltage is applied. I then went back to the car and checked that voltage was present across the two contacts on the voltage source connector for the IACV; it reads 12.15V when the ignition is 'on.' I plugged the IAC in (not installed) and turned the ignition to the 'on' position and the plunger didn't budge. I cleaned the contacts on both sides, made sure there was a solid connection, double checked for voltage, and the IAC still wouldn't move. I wiggled the wiring harness around with my multimeter attached to the connector to see if there were any breaks in the wiring, but it held a steady 12.15V the entire time.

All this had me second guessing what I was doing(detect), so I tested the IAC solenoid to discover that my initial thought pans out. One contact is +V, one is -V. I wasn't sure if the solenoid chassis was the ground and there were two input signals to reverse the direction or not. That led me to measuring both the connector terminals on the wiring harness relative to ground. Turns out the negative terminal has a steady 0.06V showing, keeping the solenoid from functioning.

I looked at the FS wiring diagram, and it shows the IACV being fed by brown (BR) and gray (GY) wires. Mine has brown and white. I didn't see any connectors near that location that have brown and gray wires, so it wasn't switched with something else. It shows the brown and white (gray) wires going to pins 54 and 83 respectively on the PCM. I haven't checked to PCM to make sure the same color wires on in those pins yet.

That being said, do any of you have suggestions on what to look at next?

The car has been modified, but nothing has been done to the original electrical system. Thanks in advance for the help!

 

[Dr0]

what sort of mods

 

[MattsMSP]

Sorry for the delay, every time I sit down to write I get called out for an ambulance run.

At this point, I've got a 2.5" TBE with High-Flow cat and Apexi muffler, SMIC, built motor with 0.020" overbore, Exedy Stage 1 Clutch, SRI, and pushing 8.5psi. Upgrades coming this June will be a tuner and new gauges, including a wideband. But, before I add anything else on, I'd like it to at least idle correctly.

I stopped in to a dealer last Friday and Mazda states the IAC works correctly, but may have an internal fault under specific conditions, such as when the engine bay is warmed up. I haven't been able to source a used solenoid yet to test it, and sure don't want to buy an entire IAC kit just to find out that doesn't fix it. Nobody has been able to find any vacuum leaks and resetting the PCM doesn't fix it, even temporarily.

 

[protege3690]

could also be a sticking EGR valve. i had similar symptoms and i would have to feather the gas pedal to get it to run without stalling. would happen on occasion when coming to a almost sudden stop as well.

 

[MattsMSP]

could also be a sticking EGR valve. i had similar symptoms and i would have to feather the gas pedal to get it to run without stalling. would happen on occasion when coming to a almost sudden stop as well.

I forgot to mention that the Canadian EGR was put in while the engine was out. This problem started a few months after the motor was dropped back in, and has only gotten worse as summer has approached - has started doing this on 'cold' starts as the ambient temps have increased. Once the idle corrects itself, the symptoms never return unless the car is shut down and restarted later. It can't hurt looking into the EGR though.

 

[jumaaneface]

Do you have a boost and/or wideband? Mine is currently doing this exact same thing. Warm starts my afr will go 17.2 or higher and vacuum will read -10inhg or less until I blip the throttle a few times and it corrects itsself. Shop friend of mine suggested the iac valve or tps. He stated it might also be my head gasket, though I show no other symptoms. Also have a forged motor that was put in about 10k miles ago.

 

[jscott]

I would say IAC is acting up...Mine was doing the same thing, I unplugged my IAC (although this made my idle too low so I adjusted the idle) it idles fine now....although it sets SES light but I can live with that until I get a new IAC....hope this helps

 

[MattsMSP]

Do you have a boost and/or wideband? Mine is currently doing this exact same thing. Warm starts my afr will go 17.2 or higher and vacuum will read -10inhg or less until I blip the throttle a few times and it corrects itsself. Shop friend of mine suggested the iac valve or tps. He stated it might also be my head gasket, though I show no other symptoms. Also have a forged motor that was put in about 10k miles ago.

I do have a boost/vac gauge, but only a lovely narrowband at the moment (wideband will be here soon). The car fires up, drops to idle for a couple seconds with vac at ~20inHg, then it begins to act up with the vac dropping to 10ish. It almost always lasts for ~20sec. The afr on the other hand shows no signs of it running lean - it dances like usual. As crazy as it sounds, it reads accurately on certain occasions because it reads lean in boost and the shop computer agrees with the afr (sad1), but that's another story. I'll roll my motor over to 10k miles this weekend. The idle and the lean boost condition did not start right when the motor was dropped back in, however.

I thought about the tps as well. Once in awhile, while cruising, I can feel the car slow down a bit (no misfire), but it never accelerates unexpectedly which to my understanding is the most common symptom. Unfortunately, I can't find a tps or complete throttle body locally either. The impedance reading on the tps falls within Mazda's specs, but that doesn't mean it's not the culprit.

 

[MattsMSP]

I would say IAC is acting up...Mine was doing the same thing, I unplugged my IAC (although this made my idle too low so I adjusted the idle) it idles fine now....although it sets SES light but I can live with that until I get a new IAC....hope this helps

To be honest, I never considered trying that. I'll try to give that a shot this weekend once all the storms roll through (cars outside this weekend). I'm not too concerned about a CEL temporarily.

 

[jumaaneface]

I do have a boost/vac gauge, but only a lovely narrowband at the moment (wideband will be here soon). The car fires up, drops to idle for a couple seconds with vac at ~20inHg, then it begins to act up with the vac dropping to 10ish. It almost always lasts for ~20sec. The afr on the other hand shows no signs of it running lean - it dances like usual. As crazy as it sounds, it reads accurately on certain occasions because it reads lean in boost and the shop computer agrees with the afr (sad1), but that's another story. I'll roll my motor over to 10k miles this weekend. The idle and the lean boost condition did not start right when the motor was dropped back in, however.

I thought about the tps as well. Once in awhile, while cruising, I can feel the car slow down a bit (no misfire), but it never accelerates unexpectedly which to my understanding is the most common symptom. Unfortunately, I can't find a tps or complete throttle body locally either. The impedance reading on the tps falls within Mazda's specs, but that doesn't mean it's not the culprit.

It shouldn't be running lean in boost at all. The stock tune on our cars is pig rich. Check car-parts.com you can find used parts for relatively cheap on there. Or pm scratchnsniff on the forums. He has an absurd amount of protege parts for great prices. It is however most likely the iac. I can confirm that was my issue as I replaced it today.

 

[MattsMSP]

It shouldn't be running lean in boost at all. The stock tune on our cars is pig rich. Check car-parts.com you can find used parts for relatively cheap on there. Or pm scratchnsniff on the forums. He has an absurd amount of protege parts for great prices. It is however most likely the iac. I can confirm that was my issue as I replaced it today.

You're absolutely correct, it shouldn't be running lean. It starts out rich and over the course of a few seconds slides lean. Like I said, no vac leaks and there are no leaks in the intake piping. Fuel trims show the PCM adding a ridiculous amount of fuel, but the PCM still shows a lean condition. It seems to occur right about when it switches to open loop mode. If you have any ideas, let me know.

I'll check out the site and shoot scratchnsniff a message. Thanks!

 

[jscott]

Exhaust leak before O2 would cause PCM to add fuel and show lean fuel trim...I had this happening to me...I replace mani gasket turbo to mani gasket and turbo to down pipe gasket...that took care of the lean condition

 

[MattsMSP]

Exhaust leak before O2 would cause PCM to add fuel and show lean fuel trim...I had this happening to me...I replace mani gasket turbo to mani gasket and turbo to down pipe gasket...that took care of the lean condition

Yep, makes total sense, but when it enters open loop it takes the 02 out of the loop, correct? That, to me, would indicate something else is causing it to go lean. Fuel pressures are within spec.

 

[Installshield 2]

to clarify, the stock MSP ecu runs very lean under full load and WOT under closed loop...it doesn't get stupid rich until 4,200 rpm or so, where it switches to open loop...the closed loop does give a shot of rich mixture on initial throttle opening, but it quickly leans out...dangerously so...that is 'normal' for the msp computer, sadly...

other than replacing your IAC...you can try the adjustment screw on the top of the throttle body, on the firewall side of the IAC...its just a small philips screw...this allows permanent idle speed changes (until the next reset) if you follow the FSM procedure and jump some terminals in the engine bay diagnostics panel...but adjusting the screw without doing that will still noticeably change the IAC's stepping...i had to do exactly this to stop from stalling with some very lopey cams...

be aware though...there are two adjustment screws on the throttle body...one is the throttle stop screw, which is under the cable linkage on the TB, which controls how closed the throttle plate is with no accelerator input...the other is the IAC by-pass port, which is the one you want...don't mess around with the throttle stop, that is set by Mazda on a flow bench...adjusting that will automatically change you're throttle position sensor outputs, and you'll quickly make things messy...

try the correct one though...it won't hurt anything...again, it won't change idle speed...but it will change the steps the IAC makes to control idle speed...i was able to back mine out about 2 full turns, which made the rpm drop smoothly and safely from throttle lift...rather than plummeting down to 300 rpm before the IAC unwinds and brings idle speed back up (and this often just resulted in a stall)...tightening the screw in will decrease deceleration vacuum, so will make revs fall like a brick...opening it will do the opposite...and its pretty easy to put back to 'normal' if it doesn't help anything...just keep track of how much you turn it...small adjustments don't make wild changes or anything...

 

[MattsMSP]

to clarify, the stock MSP ecu runs very lean under full load and WOT under closed loop...it doesn't get stupid rich until 4,200 rpm or so, where it switches to open loop...the closed loop does give a shot of rich mixture on initial throttle opening, but it quickly leans out...dangerously so...that is 'normal' for the msp computer, sadly...

other than replacing your IAC...you can try the adjustment screw on the top of the throttle body, on the firewall side of the IAC...its just a small philips screw...this allows permanent idle speed changes (until the next reset) if you follow the FSM procedure and jump some terminals in the engine bay diagnostics panel...but adjusting the screw without doing that will still noticeably change the IAC's stepping...i had to do exactly this to stop from stalling with some very lopey cams...

be aware though...there are two adjustment screws on the throttle body...one is the throttle stop screw, which is under the cable linkage on the TB, which controls how closed the throttle plate is with no accelerator input...the other is the IAC by-pass port, which is the one you want...don't mess around with the throttle stop, that is set by Mazda on a flow bench...adjusting that will automatically change you're throttle position sensor outputs, and you'll quickly make things messy...

try the correct one though...it won't hurt anything...again, it won't change idle speed...but it will change the steps the IAC makes to control idle speed...i was able to back mine out about 2 full turns, which made the rpm drop smoothly and safely from throttle lift...rather than plummeting down to 300 rpm before the IAC unwinds and brings idle speed back up (and this often just resulted in a stall)...tightening the screw in will decrease deceleration vacuum, so will make revs fall like a brick...opening it will do the opposite...and its pretty easy to put back to 'normal' if it doesn't help anything...just keep track of how much you turn it...small adjustments don't make wild changes or anything...

Thanks for the clarification. It goes lean when it enters open loop mode (~4k rpm). It doesn't matter how far your in on the throttle, as long as it sees boost it goes lean - from around 11.0 to 18.0.

I'll definitely give the IAC screw a shot and let you know how it goes. I have my truck for work all week and won't see my car until Sunday sometime, so I'll try it then.

 

[Installshield 2]

^oh wow...something else may be up then man...sorry, i don't own a MSP or a car with an MSP computer, so my info runs short there...i just know a normal MSP will run very lean through closed loop, then very rich in open loop...

the TAS screw may help, but i'm pretty sure something else is up if you're that lean in open loop...

 

[MattsMSP]

^oh wow...something else may be up then man...sorry, i don't own a MSP or a car with an MSP computer, so my info runs short there...i just know a normal MSP will run very lean through closed loop, then very rich in open loop...

the TAS screw may help, but i'm pretty sure something else is up if you're that lean in open loop...

Yea...that's what I was worried about. The IACV shouldn't have any effect on how the car runs when not at idle, should it? I still haven't gotten hold of an IAC and have been away for work, so not much has been done yet.

 

[MattsMSP]

It's been a month, and I finally got my hands on an IACV to hopefully rid the idle problem. Unfortunately, I've had no luck in that department. The idle acts the same with either IACV installed. I'm open to suggestions if any are out there

 

[MattsMSP]

Not to revive this, but I figured if anyone has this problem come up in the future and stumbles upon this thread, there will be an answer. After getting home from touring all summer, I was able to install my Wideband and discovered that it read opposite of the Narrowband when in open loop mode - wideband = rich; narrowband = lean. I replaced the primary O2 sensor and the lean condition went away I haven't had the idle problem either, but it could be a fluke and reappear. If it does rear its head again, I'll make sure to post it. There were no CELs through this whole process.

 

[31R]

Great to hear. I've been experiencing the same problem just recently. I've never replaces my O2 sensors, which led me to believe it was that. When you say primary, do you mean the upstream O2?