Occasional High Idle Speed - $8D 383 fully modified TPI

[zaut]

I am writing/asking this because of two reasons. One my son has his laptop at school so I can’t data log and two I am looking for confirmation of my theory. Occasionally my TPI engine will start idling very high. I do not have a vacuum leak. One time while it was idling at about 1600 rpm, I unplugged the connector to the TPS sensor. When I re-plugged it, the idle went back to normal. It has not gone into the high idle since then. I was never getting a check engine light. (Except when I unplugged the TPS). I am thinking that I was not getting a good ground to the TPS. This in turn was giving the ECM a 5 volt signal to the TPS signal input (Circuit 417). Because the ECM was seeing a wide open throttle signal, the IAC follower function was causing the IAC to open fully allowing a lot of idle air in. Does this make sense? I would think if the ECM was always seeing a WOT condition I would have other issues during the problem but I wasn’t seeing any during. The car definitely has better acceleration now. I am thinking that the acceleration enrichment function was not working with the TPS ground problem. Anyhow, I just wanted pick some other brains on this issue. I haven’t been here for awhile because I got the car running great thanks to everyone’s help. Thanks.

[RobertISaar]

it's a possibility. i recently had to replace a TPS due to it acting badly after the engine warmed up.... when it did, after closing the throttle, the voltage dropped too far below the normal TPS closed voltage position and the ECM learned that position as 0%, so when the throttle was actually closed, the ECM thought it was open by 3% or so and caused a ~1000 RPM idle.

[EagleMark]

Most ECM will set to 0% when the engine is started. Or does some have another learn?

You seem to have found an issue with the TPS, but it could be the sensor itself or a ground. Can check with voltage meter at idle around .54 volts and swing slowly and smoothly to around 4.5 volts. Any glitch or sticking and you found the issue.

[RobertISaar]

with the MPFI and SFI 60V6 applications i'm familiar with....

they check voltage at key-on, and the lowest voltage it sees before cranking is considered the 0% TPS point. with the SFI stuff, this is done with every key-on, but with the MPFI, it only does it if the NVRAM failed(battery disconnected).
while running, if at a high enough RPM and low enough MAP, the TPS 0% point can be changed in a downward direction, but not upwards.

[zaut]

I need to keep my mouth shut. I think I jinx myself. The high idle started again. I checked the TPS and it is working fine. 0.54 volts a closed about 4.5 volts at WOT with a linear increase as I open the throttle. I will have a laptop this weekend to dig deeper. What is funny is it is not consistent. I shut the car off and it gets back to normal. I hate intermittent problems. They are the hardest to diagnose.

[EagleMark]

That sounds more like an IAC issue. Dave and I both found bad IAC valve this week, checked wiring, Ohm sensor etc... IAC would hang... it resets when turned off and mine anyway would work it's way back up in IAC counts and stay there. Replaced and fixed.

[zaut]

EagleMark, You may be correct with the IAC Valve. I finally got a PC connected to the car. The target idle was 600 rpm and the engine was running at 850 - 900. The IAC count was 59. I shut the car off and restarted and the idle went down to 600 and counts were 37. When the IAC hangs, does the counts hang as well? I would think the ECM would pulse the IAC to change and count those pulses, regardles if the valve moves or not. I did not think the IAC gave any feedback of its actual position.

[RobertISaar]

the IAC doesn't give feedback.... the counts displayed are what the ECM is commanding it to be and can be incorrect if the IAC isn't acting as expected.

[EagleMark]

EagleMark, You may be correct with the IAC Valve. I finally got a PC connected to the car. The target idle was 600 rpm and the engine was running at 850 - 900. The IAC count was 59. I shut the car off and restarted and the idle went down to 600 and counts were 37. When the IAC hangs, does the counts hang as well? I would think the ECM would pulse the IAC to change and count those pulses, regardles if the valve moves or not. I did not think the IAC gave any feedback of its actual position.Like Robert said... seems we are finding more and more IAC issues lately or guys are just beginning to see idle issues and notice IAC counts? I think the count is going up because the ECM is commanding more but the valve is not moving or getting stuck in the move?

The one I did hands on last week did the same as yours. I still have the bad IAC here and going to tear it aprt to see if it is stripped? Or if I can see what is happening... I think 30 year old IAC are just wearing out!

[zaut]

I can agree that if the ECM commanded a move, it would assume it was made (no feedback). The count would be what the ECM thinks the IAC is at. However, (please correct me if I am wrong) the counts are not absolute. The ECM counts up and down. What I have seen is low counts meant the IAC was closed or low idle RPM, and high count meant the IAC was open. I would think that I should see a very low count when the IAC was stuck open and the ECM was trying to close it. The data I am missing is what are the "normal" counts for the IAC at any one RPM. Maybe 59 and 37 are very low counts, below what should be normal. I guess I have to make a $50 gamble on a new IAC or keep looking to see if it is anything else. Since my engine is highly modified, I am guessing no one can say what the normal count should be for my configuration. And YES my IAC is 27 years old so it could just be tried. One thing I have noticed since I completed the rebuild was that at cold startup, until a fixed amount of time, the IAC would stay at a fixed number, I think 150. Is that normal? Is there a way to adjust that?

[EagleMark]

There's a max count of usually 145. Then a park count and depending on ECM the IAC will close all the way and go to that position. Well not if the IAC is stripped.

[zaut]

I replaced the IAC last weekend. I am still getting the high idle. I am now wondering if the canister purge valve is sticking on. I will unplug it for a few days to see if that is the problem. Unfortunately I haven't had the high idle when I had the computer hooked up to watch the IAC counts.

I did notice another issue when I had the computer hooked up. I fall out of closed loop occasionally. The O2 ready goes away at the same time. Is this indicating a bad O2 sensor?

On a up note, I finally got the car to the drags Saturday. It run a 13.70 quarter mile with street tires. I was hoping for low 13sec to high 12sec. Since this is my first chance at getting good log data, I am very happy with the 1/4 ET.

I was seeing some knock retard occurring at high MAP in the 3200 to 4000 rpm range right after shifts. I need to make adjustments. But I want to remind EagleMark that the file I put together for using the EST module instead of the memcal knock module works. I would hope you would consider adding that file to somewhere on the site as a reference. I uploaded it again with this post.

[zaut]

Does anyone know what "IAC Command speed open loop offset" does? This bin I started with has that value at 1000 rpm. I am not sure what it is for so I wonder if that may have something to do with my high idle speed. Also how does "IAC Learn" affect things, what does it do?

[JeepsAndGuns]

Where is your o2? Is it in the stock location, is it the original one, do you have headers? I dont know anything about the chevy cars, is it a heated sensor? If you have headers and its not a heated sensor, that might be your problem with it dropping out of closed loop. The o2 is cooling off. You could try wiring up a heated o2 sensor.

[zaut]

I have Edelbrok headers designed to replace the stock manifold. I was thinking that the O2 was cooling off too. Is that common? It happened when the car was idling at normal speeds. Can you recommend a heated replacement? (year, model, etc) I was thinking about doing that anyways because I can get the car in closed loop quicker. I understand there different types of O2 sensors, one I have heard of is a wide band O2 sensor. Not knowing the difference, I have been hesitant to change it. Thanks.

[EagleMark]

I do have your pdf file and it is very nice work thank you! Just not sure where to put it right now since it covers Memcal type ECM/PCM...

Sure sounds like "IAC Command speed open loop offset" is an idle RPM setting for open loop. You can easily see if system is going Open Loop when the idle issue happens in data.

O2 sensors cooling off in headers is a common problem. Most narrow band O2 sesnors work the same whether heated or not, just the heating element is added. For GM a heated replacement can be from many vehicles, the only difference would be the connector. IIRC a 1995 Chevy truck and 1995 LT1 engine are the same and a good place to get one ordered from. Then add a rely and wiring for the heat circuit.

[zaut]

I did a little research an now know the difference between a wide band and narrow band O2 sensor. A wide band has a 0 - 5v output range, a narrow band has a 0 to 1 v output. My ECM uses a narrow band O2 sensor. What I need to know is what vehicles used a narrow band heated O2 sensor.

[zaut]

Thanks EagleMark. Do you know how many amps the heater circuit takes? If not I can bench test and measure the current.

The "IAC Command speed open loop offset" is set at 1000 RPM. I never changed it. I don't see where or what it affects. In open loop my idle (when working normal) follows the target idle rpm versus coolant temp table.

[RobertISaar]

I did a little research an now know the difference between a wide band and narrow band O2 sensor. A wide band has a 0 - 5v output range, a narrow band has a 0 to 1 v output. My ECM uses a narrow band O2 sensor. What I need to know is what vehicles used a narrow band heated O2 sensor.

actually, to get more in-depth the wideband controller outputs a 0-5 volt signal.... the wideband sensor itself is likely different.

your car uses a narrowband.


Thanks EagleMark. Do you know how many amps the heater circuit takes? If not I can bench test and measure the current.

most heater circuits i've seen peak at ~50 watts or so. at 12 volts, that's just over 4 amps. they quickly drop the amount of current used as it heats up though.

[EagleMark]

It may be for start up?

I don't know the amps of heated O2? For wiring I have used another fuel pump relay.

[zaut]

I think I found the root cause of my high idle. My Charcoal Canister Purge is on all the time. Unfortunately it appears that the ECM is at fault. The ECM turns the solenoid valve when it does not want to purge. The valve is normally open. I traced the circuits and I get 12 volts and continuity to pin F7 on the ECM. I never get a low resistance on that EMC circuit. It’s always about 12kohm. What my service manual says is that you will lose four outputs at a time. But that is for the original 85 ECM. This one is upgraded to a 1990 7730 ECM, which I do not have the service manual for. Does anyone know it I should have lost three other circuits and if this is repairable? Another thought I have is did GM change the logic of how the 90 canister purge worked? I was surprised to read that the valve was a normally open valve. Could by chance the 90 version use a normally closed valve and I am getting the purge at idle and not during driving? I will test this tomorrow.

[RobertISaar]

GM seemed to use normally open valves with setups that used ported vacuum, so that if the valve were stuck open, it wouldn't effect idle. when they transitioned to normally closed, they generally used a constant vacuum source. or at least this is what i've found to be true of most of the FWD GM stuff.

[zaut]

I hooked up a light to connector at the purge solenoid this morning. I went for a short drive, an behold, the light comes one when the laptop says it’s purging. This is the opposite logic used in the original 85 ECM. My original EMC was a 1985 165 1F$ mass air flow system. I converted to the 1227730 $8D system. So anyone else out there doing this change needs to know the purge system works differently. I am off the junk yard to get another valve/canister. No more high idle issues!!! I hope. Plugging the purge line is working until I get a new valve.

Does anyone have the wiring diagram for a three wire heated O2 sensor? I will be adding that soon.

[RobertISaar]

doesn't 8D have the option to use normally open or normally closed CCP solenoids? it should. if not, it would actually be a really easy patch.

[EagleMark]

Does anyone have the wiring diagram for a three wire heated O2 sensor? I will be adding that soon.To easy, just a sensor wire and the other 2 are for heating element, pos and neg and you can reverse them.

[JeepsAndGuns]

doesn't 8D have the option to use normally open or normally closed CCP solenoids? it should. if not, it would actually be a really easy patch.

Just looked on my 8D xdf and there is a flag at 0x18 mask 80, thats called "canister purge when output energized" and then at 0x16 mask 2, there is a flag called "16 Hz ccp in use". Whatever that means. Either one of those flags anything usefull?
I'm thinking change the flag at 0x18 mask 80? On a stock AUJP bin, this flag is checked. So maybe try looking at your bin and see if its checked, if it is, try unchecking it and see if it works how you need it, going by the flag title alone, you would think its what you might be looking for. But I cant promise you anything, as know very little to nothing about most stock chevy vehicles.

[RobertISaar]

the 18, 80 sounds like the right one. the other option it what chooses the frequency at which it will drive the solenoid. generally, 16, 32 or 64Hz.

[zaut]

Yes, now that you mention it, I see a flag for it. It is called "canister purge when ouput energized". Thanks a lot.

[JeepsAndGuns]

Let us know if it fixes the problem.

[zaut]

Lol... I would, but you called me out. I already replaced the valve with a normally closed version before I read the post last night. After testing yesterday, I went junk yard hunting for that and some other misc parts. I feel foolish for not looking to see if the output was programmable. But the canister is much newer and the existing one had a problem of leaking charcoal. There is a filter media on the bottom that was deteriorated. So it worked out for the best anyways.


Eaglemark, I understood that the two extra wires are for the heater. What I am not sure of is which two of the three are for the heater and which is the O2 signal. I haven't got the senor yet, it is on its way. Maybe it will be evident with a meter once I get it. Thanks.

[JeepsAndGuns]

I can not speak for all of them, but on the 4 wire bosch sensor I used, the heater element wires were both the same color. In my case, both were white.

[zaut]

I got the three wire O2 yesterday. Two wires were black an the other was blue. The two black wires read 15 ohm between each and open to the blue and the O2 case. The blue read open to everything, like the old single wire O2 I was replacing. I connected the two black wires to the heater circuit and the blue to the O2 input. Everything is working fine. I have not been droping out of closed loop and have ajusted my closed loop timer to very short times and so far all is good.

[zaut]

On another area related to the O2 sensor. I kept the air injection system on the car. I see a scalar for Air Divert O2 rich/lean reduction in millivolts. Does this offset the point where the EMC thinks the AFR is when air is being pumped into the exhaust headers? If so, is there a way to determine if the value in the scalar is correct?

[RobertISaar]

yes, that's what it is for, essentially a correction since the O2 sensor is mounted after the AIR introduction point, so it will read the true AFR plus the external air being injected.

to correct it, if it's needed, i would tune the fueling with it off, then when you have that done, turn on the AIR system and then adjust that value until the fuel trims are back inline.

[zaut]

I did tune with the AIR, EGR,CCP etc. off. Thanks to the advice from EagleMark. Now that all the emissions equipment is turned on, I do see high and low BLMs compaired to what I had tuned to. Now that I have a heated O2 sensor, I am looking into that closer. By the way, the heated O2 is great. On hot restart, I am in closed loop in two seconds and about 30 seconds on a cold starts. What I am wondering is if there is a way to see the sensor output shift when the air is turned on. Once in closed loop, I would think that the ECM will adjust the AFR as soon as it sees the shift occure. My thought is to set the closed loop timer to a very long time and the scalar set to zero; start the car until the O2 is ready then cycle the air on and off and see the voltage shift. Would keeping the engine in open loop stop the AFR adjustment that you get in closed loop? Does this sound lake a good approch?

[RobertISaar]

i would assume staying in open loop would make the ECM stop accounting for it, but i haven't checked the hack to confirm it.

think of it this way, if doing it in closed loop, you can just watch the BLM/INT react to it and then just move the AIR value up or down to get them to return to their normal values.

[EagleMark]

AIR Pump is not used that often and only open loop, like cold starts for emmissions. Some were plumbed in after O2 sensor to add air to exhaust before the CAT for added heat, not sure if this type is open loop only? So what type does your car have? What type does the new ECM and bin have? I ask because I vagely remember an AIr pump system that did pump air during decel?

In Open Loop there is no fueling adjustment. Watch your BLM and INT should be 128. It runs strictly off VE table with no adjustment.

High and Low BLM changes will occur after a tune is done do to weather changes.

[RobertISaar]

having the AIR pump run before the O2 will actually help heat it up faster...

now that i think about it, that offset may be for entering closed loop, since to do so, the O2 voltage has to go so high or so low compared to 450mV.... perhaps it takes ~100mV away from that requirement?

[EagleMark]

Yes it would, but that was never the purpose of it. Just emmissions passed on to EFI from carbs and then phased out as not needed. Even 1995 LT1 came with Air Pump and was later removed as a factory service bulliten that caused issues but issues were not fixed because it was not needed. SO they got a plug in air inlet, relay removed and PCM re flashed with a TSB sticker.

Which would be an easier fix to disable it and leave it on if needed for astetics or emmissions visual check?

I've only done limited testing and research because of this. Testing at the time found it only working cold start. So it would never be found in an emmissions test.

[RobertISaar]

knowing what i've seen GM do....

is it generally active at coolant temps(startup or otherwise) from the ~65 to ~80 *F range? because that would allow it to run during the EPA city MPG test loop to get into closed loop faster. i've seen a lot of little things like this that were implimented to game the system.

[zaut]

My air system has two valves, one pumps air to the converter the other to both exhaust manifolds. If the only time it pumps air the exhaust manifolds is in open loop, I an not going to worry about its affect on closed loop AFR. I will disconnect the hoses tomorrow and run to see when I get air to the manifold. Since I am trying to keep it looking stock, I kept the system intact. Thanks guys for the great help.

[zaut]

OK, I tested the air valve to see if only applies air to the exhaust manifold only during open loop. It turns out to not be functioning correctly at all. The only port that air doesn’t get pumped through is the dump port. The AIR valve is bad, and most likely was when I tuned my VE tables. I plugged the line to the exhaust manifolds and now my BLM’s are running high. I think the air pumped into the exhaust was always skewing the EMC to run richer than it should. Now that the leak is plugged, the EMC has to adjust much leaner causing the BLM’s to go high. Does this make sense?

[EagleMark]

Yup! Hence my understanding and limited testing of the air injection is only used Open Loop. Just plug it and turn it off.

[zaut]

Yea, I read the service manual and it said it was only active in OPEN loop. But in my case, with a bad valve, it was pumping air into the exhaust manifold all the time. It is now plugged and turned off. Can I assume it is good to pump air into the converter all the time? I left that plumping intact.

[EagleMark]

Hmmm? I don't know? Does the manual tell when it usually works?

[RobertISaar]

i don't know about all of the time..... but with the 60V6 stuff that has an AIR setup, it's generally only cold start and when in PE for an extended period of time.

[zaut]

That is what it said in the air system service section of my GM Service manual. I agree with you, I don't need it to work, I am just tying to keep it look stock under the hood. Actually I plugged the hose with a cork in the middle of the hose so you cannot tell that it is not working.

[zaut]

This is a thread I stated last year when I was having a high idle problem. I thought I got it worked out, but I thought wrong. When I start the car, it idles correctly. After driving it, it starts to idle very high. If I turn the car off and back on, it will idle correctly until it take it for a drive.. When it is idling correctly the IAC position will be in the high 30's-low 40's. When the engine is idling high the IAC will be in the 80's and 90's. Does anyone know why the EMC will not lower the IAC counts to slow the engine down?

[zaut]

Well, I think I finally got to the real root cause of my high idle. I noticed that when the engine was idling high that the TPS percent was not zero. After close examination of my TPS on the BBK throttle body, it turns out that the follower arm of the TPS was dragging against the throttle body. So the TPS was not returning to zero every time. This was then making the IAC follower function think the throttle was open and therefore opening the IAC motor. Amazing. 10 seconds on the bench grinder and the TPS now has plenty of clearance. One thing I don't understand is why the IAC motor does not move for a set period of time when the engine is cold started? Is that the IAC learn timer setting? And does anyone know what the maximum count the IAC go to? On cold start, the 160 counts is not enough to keep the engine running. And lastly, what does IAC open loop command offset do?

[EagleMark]

with the MPFI and SFI 60V6 applications i'm familiar with....

they check voltage at key-on, and the lowest voltage it sees before cranking is considered the 0% TPS point. with the SFI stuff, this is done with every key-on, but with the MPFI, it only does it if the NVRAM failed(battery disconnected).
while running, if at a high enough RPM and low enough MAP, the TPS 0% point can be changed in a downward direction, but not upwards.Since you found the throttle hanging up you may want to disconnect battery for awhile and let things reset... Robert quote above says it won't reset to TPS 0% without! Then see how it does.

For a modified engine sometimes IAC is not enough for your cold start issue? Do a data log and when warmed up see what IAC counts are doing? If like 40 or so, well you have plenty of room to open throttle blades so more air can get in at cold start. Do want to end up with some IAC counts for idle control at end.

[RobertISaar]

it may not be necessary with 8D, but id rather err on the side of caution.

[zaut]

The car is running the best it ever has since the rebuild and tune. The TPS % is set to zero on every restart with the 8D. I did a little playing around with restarts with partial open throttle and it set TPS % to zero. I am still perplexed as to why when I cold start, the IAC stays on the 160 park position for a very long time. EagleMark, are you saying that a IAC down as low as say 10 at hot idle is ok? I have the throttle blades set at 100 RPM below my normal idle set point with the IAC port blocked. I can open then up to close to the idle set point and see what happens with IAC counts.

[EagleMark]

I don't know $8D code well enough to say what the IAC should do at cold start? But 160 is full open and seems to need that air a long time so throttle blades may not be open enough.

I know the GM procedure says how to do this at 100 RPM below Idle speed. But if the engine is not fully warmed up like after a drive on a warm summer day? This can be way off! TBI is way more sensitive then MPFI, but I still like to see IAC data at end of log, a minute in drive and a minute in Park. As long as you have some IAC counts, like 10 in park at that point it's OK, may be 20 or so in drive.

I forget what all you did to this motor? But if getting the throttle blades set and IAC counts in order does not help cold start issue then it's time to look at some bin changes. BUT!!! Before you mess with any cold start issues you have to make sure your VE tune is in order first. Then look at Open Loop AFR tables or $8D equivelent...

[zaut]

Thanks again EagleMark. I will start with opening the throttle blades and monitor the IAC counts as you suggest to see if that is why the IAC counts stay at 160 so long.