SOLVED: High idle, increases with coolant temp - IAC?

[Bad Luck]

Mine appeared to be full closed too when I looked. [MENTION=319302]Bad Luck[/MENTION], is it possible that the screw on your throttle body (the one with the Allen head and a locknut) is misadjusted, preventing the throttle plate from closing fully?

I haven't adjusted it, and it doesn't appear to have ever been adjusted. I don't have any idle issues so I don't think it's unusual.

I just cleaned mine the other day and it sure looks like it closes completely with zero gap.

It appeared to be close, but with the throttle body removed from the upper intake I can see light at the edges when shining a pocket flash light from the back side.

 

[Spiker Engineering]

So here is how cleaning my IAC caused it to start acting up (I won't call it "fail", because it is still perfectly good, it just needed some additional resuscitation).

I used Timmy's video for guidance (IAC Valve Testing and Cleaning - YouTube) and was really careful to avoid dousing the inside of the IAC with solvent, worrying about the delicate bearings. I just cleaned the rotating barrel with a q-tip soaked in solvent.

After that, I ran the opening and closing tests as per the video (and FSM). It did open fully when commanded to open, and rotated into the closing direction when commanded to close. So everything seemed good to go.

So in my autopsy, I again looked at the rotating barrel. Here it is in the normal room temperature position:

You can see that there is a small window opening allowing air to pass through the IAC. To check for freedom of motion, I inserted a small screwdriver into the opening, and rotated the barrel towards the open position - it rotated easily:

So again, I thought everything was good.

Here's the thing though - when the IAC is assembled, it is nearly impossible to manually rotate the barrel into the fully closed position - there is nothing to push on. So I had to go by the closing test in the FSM and in Timmy's video. And just like in the video at 35:30, when I applied power in the closed direction, it started rotating in that direction (I'm using the words "started rotating" very intentionally here). But just like in the video, the opening didn't fully close, it just moved in that direction.

Once I had the IAC apart and could rotate the shaft by hand, I noticed that when trying to fully close the barrel, there was a sudden added resistance. I gave it a bit more torque, and finally saw the culprit: there was a layer of carbon on the barrel, preventing it from closing fully:

So what happened was that my q-tip cleaning dissolved the carbon, but the carbon then reconstituted itself on a different portion of the barrel, and once it dried, it prevented the barrel from closing. Without taking the IAC apart, I couldn't rotate the barrel far enough to notice this, but the actuator didn't have enough torque to overcome the resistance.

I haven't tried it yet, but I will clean off this carbon and reassemble, and I am confident that the IAC will work perfectly.

So my takeaways are:

1. Don't clean the IAC unless you have to! Mine was working perfectly fine before, and I could have left it alone.

2. If you do clean the IAC, be sure all the carbon is actually removed, not just moved around. Try to find a way to move the valve FULLY to open and closed positions, to make sure there is no carbon left anywhere.

3. Don't clean the IAC if you unless you have to!

Hope this helps others! And Timmy, in no way am I criticizing your video, it is a great service to the community. I just wanted to add the comment that it's critical to get ALL the carbon out, and to check for FULL closing and opening operation after cleaning.

 

[Bad Luck]

It's a really complex system. I spent many hours researching the IAC, and know much more than I knew earlier, but there are still parts of it that I don't fully understand. Here's my "IAC 101" understanding:

There is a bimetallic spring on one end, which rotates the shaft as the coolant temperature changes:

The spring's tang engages into a brown plastic "hat" that sits on the shaft:

On the opposite side of the shaft is the electromagnetic actuator:

Like you said, the actuator does not send any feedback to the ECM, it's just a dumb torque motor (not even a stepper motor).

But here is where things get wonky - the two actuators (spring and motor) are NOT directly connected to the shaft! In other words, if you hold the hat with one hand, you can rotate the actuator side of the shaft back and forth quite a bit with your other hand.

That was totally unexpected for me, but it makes sense once you look at the design more closely. The brown hat that the spring rotates actually has a lot of radial play with the shaft.. You can see how much wider the opening inside the hat is compared to the small tang on the shaft:

So basically, the spring rotates the brown hat as the coolant temp changes, and that brown hat acts as a window frame for the shaft - it forces the shaft to rotate, but allows a lot of wiggle room within that frame. So I think that's what the electromagnetic actuator does - it fine tunes the position of the shaft within that rotating window frame. Does that make sense?

What I don't know is this: is the electromagnet strong enough to overpower that "window frame" which is rotated by the spring, or can it only move the shaft back and forth within the confines of the window frame? In other words, do the two actuators work in a complimentary fashion, or can the electromagnet overrule the spring's position? I'd be really curious to know the answer to this.

I know I still didn't get to the root cause of my failure, I'll do that in the next post.

I think the ECM monitors engine coolant temp, engine rpm, throttle position, and maf readings and uses that data to determine if the IAC actuator needs to add additional air to what the mechanical thermostat in the IAC is already performing. It would make sense to me for the mechanical thermostat to be the primary device as it would be able to add air for cold engine starting without the need to read any data, then after starting the ECM would be able to add additional air as needed to reach the higher idle.

 

[mtbtim]

So here is how cleaning my IAC caused it to start acting up (I won't call it "fail", because it is still perfectly good, it just needed some additional resuscitation).

I used Timmy's video for guidance (IAC Valve Testing and Cleaning - YouTube) and was really careful to avoid dousing the inside of the IAC with solvent, worrying about the delicate bearings. I just cleaned the rotating barrel with a q-tip soaked in solvent.

After that, I ran the opening and closing tests as per the video (and FSM). It did open fully when commanded to open, and rotated into the closing direction when commanded to close. So everything seemed good to go.

So in my autopsy, I again looked at the rotating barrel. Here it is in the normal room temperature position:

You can see that there is a small window opening allowing air to pass through the IAC. To check for freedom of motion, I inserted a small screwdriver into the opening, and rotated the barrel towards the open position - it rotated easily:

So again, I thought everything was good.

Here's the thing though - when the IAC is assembled, it is nearly impossible to manually rotate the barrel into the fully closed position - there is nothing to push on. So I had to go by the closing test in the FSM and in Timmy's video. And just like in the video at 35:30, when I applied power in the closed direction, it started rotating in that direction (I'm using the words "started rotating" very intentionally here). But just like in the video, the opening didn't fully close, it just moved in that direction.

Once I had the IAC apart and could rotate the shaft by hand, I noticed that when trying to fully close the barrel, there was a sudden added resistance. I gave it a bit more torque, and finally saw the culprit: there was a layer of carbon on the barrel, preventing it from closing fully:

So what happened was that my q-tip cleaning dissolved the carbon, but the carbon then reconstituted itself on a different portion of the barrel, and once it dried, it prevented the barrel from closing. Without taking the IAC apart, I couldn't rotate the barrel far enough to notice this, but the actuator didn't have enough torque to overcome the resistance.

I haven't tried it yet, but I will clean off this carbon and reassemble, and I am confident that the IAC will work perfectly.

So my takeaways are:

1. Don't clean the IAC unless you have to! Mine was working perfectly fine before, and I could have left it alone.

2. If you do clean the IAC, be sure all the carbon is actually removed, not just moved around. Try to find a way to move the valve FULLY to open and closed positions, to make sure there is no carbon left anywhere.

3. Don't clean the IAC if you unless you have to!

Hope this helps others! And Timmy, in no way am I criticizing your video, it is a great service to the community. I just wanted to add the comment that it's critical to get ALL the carbon out, and to check for FULL closing and opening operation after cleaning.

Don't worry Leon, I'm not taking any offense to what you said. This is great information to share. I'm going to share a link to this thread in a pinned comment for that video so people can learn from it.

Maybe it is beneficial to clean a little more aggressively with the spray. That one guy worried about the bearings is just one guy's opinion. I think you can probably spray cleaner into the IAC without penalty. I obviously did and I imagine tons of others have as well.

My opinion is I still think cleaning the IAC is a good preventative maintenance thing to do, because even though you had a bad result after cleaning it which inspired you to purchase a new one, not cleaning it could possibly allow enough carbon to eventually cause the IAC trap door to become gummed up and not open and close properly.

I'm not sure waiting to have a problem with it is the best path. I might be wrong and cleaning the IAC isn't something all that necessary as preventative maintenance.

 

[mtbtim]

I haven't adjusted it, and it doesn't appear to have ever been adjusted. I don't have any idle issues so I don't think it's unusual.



It appeared to be close, but with the throttle body removed from the upper intake I can see light at the edges when shining a pocket flash light from the back side.

Well, obviously by shining a light you will see some light getting through. If the plate fit so tight that you couldn't see light, it would get stuck in the bore of the throttle body. The 96-00 model engines aren't designed to get the air they need to run via the throttle plate. That's the job of the IAC. For the 01 and 02 models that have a hybrid throttle body with a drive by wire with a drive by cable back-up, the throttle plate doesn't close all the way because there is no IAC providing air to the engine while idling or under deceleration with no gas pedal input.

 

[phattyduck]

The Toyota IAC of this era are a bit weird in the automotive world - there is a positive and negative input from the ECU. At rest it is in the 'mid' position, so air should bypass normally. The ECU PWM controls one or the other input (not both at the same time) to add or subtract air flowing through the IAC.

The extra bi-metallic controlled 'window' is an interesting one...

And as Tim said, the IAC has *most* of the bypass air at idle. There is also a bit that flows to the air assist injectors. The throttle plate should be basically fully sealed when closed.

-Charlie

 

[Bad Luck]

Well, obviously by shining a light you will see some light getting through. If the plate fit so tight that you couldn't see light, it would get stuck in the bore of the throttle body. The 96-00 model engines aren't designed to get the air they need to run via the throttle plate. That's the job of the IAC. For the 01 and 02 models that have a hybrid throttle body with a drive by wire with a drive by cable back-up, the throttle plate doesn't close all the way because there is no IAC providing air to the engine while idling or under deceleration with no gas pedal input.

The gaps I have seem to be a couple mm which is sufficient airflow for most vehicles I've worked on to start. When I have time I'll cover the air intake for the IAC and see if it still starts.

 

[Spiker Engineering]

Don't worry Leon, I'm not taking any offense to what you said. This is great information to share. I'm going to share a link to this thread in a pinned comment for that video so people can learn from it.

Maybe it is beneficial to clean a little more aggressively with the spray. That one guy worried about the bearings is just one guy's opinion. I think you can probably spray cleaner into the IAC without penalty. I obviously did and I imagine tons of others have as well.

My opinion is I still think cleaning the IAC is a good preventative maintenance thing to do, because even though you had a bad result after cleaning it which inspired you to purchase a new one, not cleaning it could possibly allow enough carbon to eventually cause the IAC trap door to become gummed up and not open and close properly.

I'm not sure waiting to have a problem with it is the best path. I might be wrong and cleaning the IAC isn't something all that necessary as preventative maintenance.

Awesome, I'm glad you took this the right way Tim. I think that additional info (to REALLY clean the barrel fully) would be useful to folks.

I also agree with you that perhaps dousing the IAC with cleaner is a better strategy than being careful to avoid washing out the bearings. If I had doused it, the carbon would have just washed off. Instead, by using the q-tip, I ended up just moving the carbon around, and that stuff is sticky!

BTW, in my seemingly endless research of IACs when I was trying to figure out how they work (and can fail), I came across a somewhat obscure post on Yotatech, by Dr Coffee, of all people. He had high idle after cleaning the TB, and fixed it by spraying cleaner into the hole at the bottom of the TB (above the IAC air bypass):

"What worked for me was to remove the plastic air tube and leave it off, squirt 2 quick shots of MAF cleaner down the hole in front of the throttle plate and start the engine. It fired up and then died (no MAF signal), I did it again and the same thing happened, did it one last time and it stalled again. after 3 cleanings, I reconneted the Air tube and it fired up at idled @ 750rpm."

"My belief is that the IAC valve got jammed with carbon crud after spraying the TB with carb cleaner and the carbon drained down the IAC hole and dried on the pintle. What I did above loosened the carbon and allowed the engine to suck the crud past the pintle, clearing the blockage. But whatever the reason, it worked for me and it didn't involve removing the IAC or rebuilding the TB. Maybe someone can try this on theirs the next time it comes up. Sure was an easy fix."

I think his post is talking about exactly the same issue I had, I just didn't quite grasp it because my IAC barrel did move - but not through the full range, it turns out. Hindsight is always 20/20.

 

[Spiker Engineering]

The Toyota IAC of this era are a bit weird in the automotive world - there is a positive and negative input from the ECU. At rest it is in the 'mid' position, so air should bypass normally. The ECU PWM controls one or the other input (not both at the same time) to add or subtract air flowing through the IAC.

The extra bi-metallic controlled 'window' is an interesting one...

And as Tim said, the IAC has *most* of the bypass air at idle. There is also a bit that flows to the air assist injectors. The throttle plate should be basically fully sealed when closed.

-Charlie

Charlie, you are better suited to understand this than me (you're the electrical engineer, I'm mechanical ), so perhaps you can weigh in on this.

In this video, Kevin uses a scope to monitor the IAC control signal (at the ~8:00 mark), and shows that the signal is a high frequency square wave:

Why does Toyota use this high frequency modulation instead of something simpler?

 

[mtbtim]

Awesome, I'm glad you took this the right way Tim. I think that additional info (to REALLY clean the barrel fully) would be useful to folks.

I also agree with you that perhaps dousing the IAC with cleaner is a better strategy than being careful to avoid washing out the bearings. If I had doused it, the carbon would have just washed off. Instead, by using the q-tip, I ended up just moving the carbon around, and that stuff is sticky!

BTW, in my seemingly endless research of IACs when I was trying to figure out how they work (and can fail), I came across a somewhat obscure post on Yotatech, by Dr Coffee, of all people. He had high idle after cleaning the TB, and fixed it by spraying cleaner into the hole at the bottom of the TB (above the IAC air bypass):



I think his post is talking about exactly the same issue I had, I just didn't quite grasp it because my IAC barrel did move - but not through the full range, it turns out. Hindsight is always 20/20.

Yeah, Dr. Coffee had the same result but got there by a different means. Lots of people screw up their IAC by cleaning the throttle body and letting the dirty runoff drain down into the IAC gumming it up. I always tell people they're better off spraying the throttle body cleaner onto a rag and then wiping out the carbon build-up in the throat and on the baffle of the throttle body.

 

[Devbot]

Charlie, you are better suited to understand this than me (you're the electrical engineer, I'm mechanical ), so perhaps you can weigh in on this.

In this video, Kevin uses a scope to monitor the IAC control signal (at the ~8:00 mark), and shows that the signal is a high frequency square wave:

Why does Toyota use this high frequency modulation instead of something simpler?

LOL i ask this literally every time i work on my car, Toyota insisted on taking the long and hard way for damn near everything for little to no reason
I swear the Japanese motto it to overcomplicate things by two fold for a 5-10% increase in efficiency and durability

 

[Spiker Engineering]

I'll do some experiments today to try to understand how the bimetallic spring and the electromagnetic actuator work together. Stay tuned.

 

[brillo_76]

Me on this thread.... opcorn:


Toyota does this for longevity. Plus they lots of times revision parts for use down the road. So The IAC valve has a logic signal coming out of it that is what those square waves are just modulated DC. Its a very clean signal to. The ECM reads that pattern and makes the appropriate changes.

My guess is that Toyota was switching and getting ready to the stepper moter setup like we see in the late models 2001 and 2002 and beyond.

You can see the progression of systems across this generation as they were going from analog 1996-1998 to digital 2001 and 2002.

The ABS system, the braking system. coke system all went through upgrades so that newer systems could be used. Integrated throttle by wire. Integrated master cylinder for the more complex systems of VSC and Trac.
Not to mention the electoral acutator for 4wd vs the mechanical.


I watched the same thing when they went from vacuum switching to electrical sensors in the late 1980s early 90s.

 

[jgue467]

...I always tell people they're better off spraying the throttle body cleaner onto a rag and then wiping out the carbon build-up in the throat and on the baffle of the throttle body.

That's all I've ever done on my '01, easy and only takes 5 mins, just rub well around edges in front and behind the flapper, where it connects on sides I use an old toothbrush to gently get any debris out from there and rub off with a rag... then the smooth idle is back for another 50-60K...

 

[Spiker Engineering]

So after cleaning the barrel more thoroughly, it moves very easily and smoothly. Here it is in neutral position at room temperature:

Here it is with 12V applied to the Open terminal:

And here it is with 12V to the Closed terminal, much closer to "fully closed" than before:

I don't have the equipment or the time to figure out if the actuator can rotate the spring, it's too hard to tell what's happening inside. When I hold the end of the actuator where the spring is, the actuator feels like it's providing a decent amount of torque. But when I try to flex the spring, it's pretty stiff too.

I think the way this works is that the spring rotates the "window" within which the electromagnetic actuator then fine tunes the actual angle, and I doubt that the actuator has enough torque to overpower the spring and push the barrel beyond the limits of the "window" (slop). But I don't know for sure .

 

[brillo_76]

Good enough. We know more about it than we did before

Sent from my SM-A536V using Tapatalk

 

[Bad Luck]

So after cleaning the barrel more thoroughly, it moves very easily and smoothly. Here it is in neutral position at room temperature:

Here it is with 12V applied to the Open terminal:

And here it is with 12V to the Closed terminal, much closer to "fully closed" than before:

I don't have the equipment or the time to figure out if the actuator can rotate the spring, it's too hard to tell what's happening inside. When I hold the end of the actuator where the spring is, the actuator feels like it's providing a decent amount of torque. But when I try to flex the spring, it's pretty stiff too.

I think the way this works is that the spring rotates the "window" within which the electromagnetic actuator then fine tunes the actual angle, and I doubt that the actuator has enough torque to overpower the spring and push the barrel beyond the limits of the "window" (slop). But I don't know for sure .

I love technical information and experimentation like this! When you remove power from the valve in either direction does it stay or return to a centered/neutral location?

 

[Spiker Engineering]

I love technical information and experimentation like this! When you remove power from the valve in either direction does it stay or return to a centered/neutral location?

It wants to return to the same position, with a decent amount of torque. And if I move it with a screwdriver, it again comes right back to the same position.

 

[Bad Luck]

It wants to return to the same position, with a decent amount of torque. And if I move it with a screwdriver, it again comes right back to the same position.

That might explain the signal waveform you are seeing then. If it's trying to return to the home position then it would make sense to see a pulse width modulated signal to get the varying percentage/degree/amount of open or closed from the home position.

 

[phattyduck]

Charlie, you are better suited to understand this than me (you're the electrical engineer, I'm mechanical ), so perhaps you can weigh in on this.

Why does Toyota use this high frequency modulation instead of something simpler?

It is an easy way with an electromechanical system to get an 'analog' control from a digital signal. When the signal is ~50% high and low, the mechanical system smooths out the impulse to 50% of the voltage/force/position available. Much cheaper ($, PCB area, power, reliability, etc.) to use a digital output than an analog one (which would basically be the same thing anyway, just with electrical filtering and then less efficient analog drivers)

The use of a 3-wire IAC (and the bimetallic spring) means that the ECU has to do less 'work' to get a good idle and the engine should generally idle even if the IAC is unplugged. The default (not driving either open or closed signal) should be reasonably close to the 'mid' range needed to idle around the right speed. At least until the IAC gets stuck due to build-up...

Me on this thread.... opcorn:

Toyota does this for longevity. Plus they lots of times revision parts for use down the road. So The IAC valve has a logic signal coming out of it that is what those square waves are just modulated DC. Its a very clean signal to. The ECM reads that pattern and makes the appropriate changes.

That digital signal is an OUTPUT from the ECU. The feedback mechanism is the engine idle RPM. The ratio of high and low is what the ECU is controlling in this case.

-Charlie