When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
DFI and ECMDiscuss all aspects of DFI (Digital Fuel Injection), ECMs (Electronic Control Module), scanners, and diagnostic equipment. Fine tune your Third Gen computer system for top performance.
I am curious about the idle speed logic for my '85 305 TPI. Idle is about 1050-1100 (hot) in park and 925-950 in gear with about +/- 50 RPM drift or instability. Engine is in closed loop. When the engine is cold, the speed is only slightly higher and more stable. I have been through most of the mechanical and electrical reasons, and was looking for a clue for whether there is an idle stability algorithm or something similar that could be causing the idle to be driven high. That might give me a path to diagnose.
I have an Actron / Bosch scan tool that shows the IAC at about 50-60 in park and about 110 in gear (observed operating range seems to be from 145 to about 20). It appears to be controlling to a set point. If it went to 60 in gear, the idle speed would be much lower. I'm trying to understand why it might be targeted that high as I expected that somewhere around 650-700 would be normal. Engine has about 150 miles on rebuild and runs great, except for idle speed.
Background:
Car was parked for 25 years and engine was rusted internally.
Engine is fresh rebuild of original 305. Completely stock except +0.030 with mildest Lunati Voodo cam (207/213).
18 inches of vacuum in park. Base timing set to 6 BTDC, and ESC appears to be managing spark normally when viewed with a timing light. Checked for vacuum leaks at all hose connections and intake system with smoke generator.
Because the car sat for so long, I have rebuilt or replaced nearly all the ignition components, entire fuel and intake system, and most control systems sensors with only a few exceptions:
ECM / PROM original
TPS sensor original, but voltages are right on-both with scan tool and DVM
MAF and burnoff module original
ESC module original (has set a 43 code recently, - that gave me a 400-500 RPM idle speed, but cleaned module connections and it hasn't returned yet)
Distributor housing and shaft original
TPS is typically .51-.53. My manual says .54+/- 0.75. so that's fine.
Haven't touched idle air because I don't have a reason yet. In default modes with DTCs when IAC can't control, I get 400-500 RPM- close enough- so not going to tamper with that for now. If IAC was showing signs of running near limits, I would adjust. It appears to be consistently running to the set points mentioned--IAC of roughly 110 in gear and 60 in neutral. Something is telling the control system to go there.
So when you say default mode, you mean in test mode which drives the IAC fully closed? Shorting terminals A&B on the OBD connector and turning the key on? Have you verified that it is fully closed? It could be getting stuck not going fully closed. The IAC position is a only a "count" that the ECM assumes where the IAC position is- there is no feedback of the actual IAC position to the ECM. It's just a pulse driven worm-drive type motor and threaded shaft plunger. Each pulse moves the plunger's threaded shaft in or out one pulse. When the IAC resets it drives all the way to closed then counts back open to proper idle. The ECM assumes the IAC moves to the position when it is directed. That is how the ECM compensates loads that affect idle (AC, auto trans in gear, vehicle moving or not).
There is another process you can try. With the ignition off, depress the pedal slightly so that the TPS is off idle. Start the car with the pedal in this position. The manual says to then let it run for five seconds. (you need to release the pedal once it starts but the manual isn't clear about this). You should then hear the ECM adjusting the idle to it's specified rpm.
Last edited by Tootie Pang; 07-27-2020 at 05:53 PM.
In the process of debugging and shaking down the car I have had several cases where a DTC / Check engine light was set and idle speed went low until the DTC was corrected. Idle settled around 500. Perhaps it wasn't IAC, but seemed to behave as if it was on base idle. I can check it more closely, but I don't think that's my problem. It appears still factory sealed. I know the cam could require a slight bump one way or the other, but system behavior doesn't lead me there.
With codes clear, idle has been very consistently high for over 200 miles now. It did seem that IAC may have been sticking or sluggish at times, so I replaced it last week and cleaned passages. The new one is definitely an improvement in stability and starting. The old one was rather carboned up and I often needed to assist it with accelerator on starts. But, the idle speed set points between the two are identical. The position is a few counts different, but speed is the same and now it will start without pedal assist. I guess I have been accidentally running your suggested procedure on starts by necessity.
Basic behavior using round numbers: In gear, at 950 or so, shift to park, momentary (few seconds) speed flare to 1200-1300 or so as the load comes off, watch the IAC drop back from 110 to 60 and drop unloaded idle back to 1050 or so. Drop it into drive, speed drops to 950 IAC cranks up to 110. Very repeatable. I have worked on other engine control systems and this looks like a commanded set point.
I opened up the ECM and checked the PROM and it appears to be factory "FTM". It could have been re-burned I suppose, but I don't think the previous / original owner was into that. On other systems I have been involved with, we had stall savers and other types of overrides that would elevate idle speed based on sensor inputs.
That triggers a thought--What does the A/C idle override look like? The A/C system is connected, but has no refrigerant and I am running in vent mode until I can repair it, so there should be no head pressure switch override affecting idle.
I was able to dial in my idle yesterday. I forgot about the timing connector. I might redo it later but the car idles perfectly.
I will verify the base idle is within range per the manual at my next opportunity. I am not sure what you mean about dialing in the idle speed because my understanding of the intent of that adjustment is not to dial in the idle speed. On the contrary, the intent of that setting is to center up the IAC so that it can meet the required idle speeds without running out of travel. If you use it to actually modify the idle speed and run the IAC to the ends of its travel, you can cause wear or wear out of the parts. If my cam has changed engine torque and airflow enough to change the base idle substantially and push the IAC out of range, then I'll need to adjust it. Otherwise I think I still need to find what is driving the control system to command a high idle. It still appears to me that since my IAC is not anywhere close to its limits, it should have authority to bring the idle speed down.
When checking for vacuum leaks, did you check the throttle body shaft? If it's getting unmetered air from the shaft, it will raise the idle. The reason i say that is, at one point in the history of my Chamaro, I had a piece of junk Holly throttle body that leaked like a sieve. I replaced it with a B&M and everything went back to normal. I have heard on stock throttle body shafts, they wear a little and create an air leak. Any way, something else to check. Good luck.
The base idle is set so that the engine will still idle around 500 RPM with the IAC fully closed and only base timing. The ECM then takes it from there and sets RPM and timing accordingly.
What I meant by dial in is that the ECM maintains perfect idle in all conditions. To do that I set my base idle and then reset the IAC.
The base idle is set so that the engine will still idle around 500 RPM with the IAC fully closed and only base timing. The ECM then takes it from there and sets RPM and timing accordingly.
What I meant by dial in is that the ECM maintains perfect idle in all conditions. To do that I set my base idle and then reset the IAC.
Ok Tootie, we're on the same page and as I said, I'll check it. But given my understanding of vehicle electronic throttle system controls, I don't think it is the likely root cause. As you said, the ECM should be in control. There should be two targets for the base throttle setting. First, as you said, enough air for the engine to idle in a control system failure - limp home mode. Second is for the air flow authority of the IAC to be in range so the IAC doesn't run to it's stops in normal operation. Those considerations drive the specification in shop manual. Since I have a scan tool reading values on a new IAC and can see that it is actively controlling and not limited out, the base throttle setting should not matter because the ECM is in control as intended. It will be a couple of weeks before I can actually get back to the car and work on it again, so I am trying to line up a set of possibilities.
What I was hoping to gain from this section of the forum was someone who understood the code and the calibrations for idle. There are a only a few sensors on this system, I have replaced all but the MAF. If I was back working with the controls engineers on a 2020 small block, we have throttle algorithms with multiple level override logic that will elevate idle speed from the normal target if idle speed is unstable, if it too rich or lean or if the air is on, if AIR is running, if it's too hot, too cold, high altitude, and so on. I don't know what algorithms these early cars have, so it is harder to diagnose, because I don't have the tools or knowledge to go into the controller and check. I don't even know what the intended target speed is, maybe 950 drive / 1050 in park is "normal" for an early 1985 TPI engine.
I was finally able to get back to the car today and I think I have it resolved. I picked up a clue from another forum post on the ECM programming section and I confirmed with a section in my service manual. Some of these early ECMs had a diagnostic mode that modifies the idle and is triggered by the scan tool.
Since this was a major project on a long parked car, I have been using my scan tool every time out to monitor parameters as I did my shakedown drives. The answer is that the scan tool causes elevated idle speed in park of 1050 along with elevated ignition timing by something like 8 - 10 degrees. This explains why the system was showing the IAC controlling to a high idle. The apparent roughness and random misfire was caused by excessive spark advance.
I ran it today without the tool. Cold idle in park was about 1200 and as the engine warmed up, it progressively dropped to about 650 on the dash tach and about 550 in gear. The idle quality is much improved, no audible misfire in the exhaust any longer. It sounds healthy now. It has a very light idle vibration and shake, but I did go to a little hotter cam and it's probably about right for that.
I don't understand why GM programmed the system to respond like this. They must have had a reason, but it's a mystery. I have an Actron CP9600 scan tool that is triggering this mode. It's one of the few tools out there that will go back to OBD-I.
I was finally able to get back to the car today and I think I have it resolved. I picked up a clue from another forum post on the ECM programming section and I confirmed with a section in my service manual. Some of these early ECMs had a diagnostic mode that modifies the idle and is triggered by the scan tool.
Since this was a major project on a long parked car, I have been using my scan tool every time out to monitor parameters as I did my shakedown drives. The answer is that the scan tool causes elevated idle speed in park of 1050 along with elevated ignition timing by something like 8 - 10 degrees. This explains why the system was showing the IAC controlling to a high idle. The apparent roughness and random misfire was caused by excessive spark advance.
I ran it today without the tool. Cold idle in park was about 1200 and as the engine warmed up, it progressively dropped to about 650 on the dash tach and about 550 in gear. The idle quality is much improved, no audible misfire in the exhaust any longer. It sounds healthy now. It has a very light idle vibration and shake, but I did go to a little hotter cam and it's probably about right for that.
I don't understand why GM programmed the system to respond like this. They must have had a reason, but it's a mystery. I have an Actron CP9600 scan tool that is triggering this mode. It's one of the few tools out there that will go back to OBD-I.
Does your car still run good? I adjust the idle / tps / changed all the sensors but it will idle nicely for a while, then the computer takes control and bumps the idle way back up to 1200 - 1500 making all my work for nothing... :/ ugh
Does your car still run good? I adjust the idle / tps / changed all the sensors but it will idle nicely for a while, then the computer takes control and bumps the idle way back up to 1200 - 1500 making all my work for nothing... :/ ugh
Yes, my car is running well. I have about another 600 miles or so on it. I did have an anomaly due to a problem with a new IAC. During troubleshooting I installed a new IAC since the original one looked to be factory and didn't look very clean. The new part was good for a couple of weeks, then began to stick or fail closed after a hot soak. Symptom- Drive about an hour on an 80 degree day, park 15 minutes and it wouldn't start or idle on restart without right foot throttle assist (probably means I ought to bump the base idle- it was trying to run at about 400 RPM). When the car cooled for a couple of hours, it would return to normal. I reinstalled the original GM factory IAC and it's running fine now while I warranty the new IAC.
Yes, my car is running well. I have about another 600 miles or so on it. I did have an anomaly due to a problem with a new IAC. During troubleshooting I installed a new IAC since the original one looked to be factory and didn't look very clean. The new part was good for a couple of weeks, then began to stick or fail closed after a hot soak. Symptom- Drive about an hour on an 80 degree day, park 15 minutes and it wouldn't start or idle on restart without right foot throttle assist (probably means I ought to bump the base idle- it was trying to run at about 400 RPM). When the car cooled for a couple of hours, it would return to normal. I reinstalled the original GM factory IAC and it's running fine now while I warranty the new IAC.
that's good!
I'm still trying to figure mine out...I want to assume injectors as they appear to be original and the car is always water your eyes rich! ugh
I'm still trying to figure mine out...I want to assume injectors as they appear to be original and the car is always water your eyes rich! ugh
I wouldn't necessarily jump to injectors. For speed to be high, the engine is getting additional air or making more power for the same amount of air. If it is rich, it is more likely to be controlling rich for some reason. High idle most often is excess air, frequently due to vacuum leaks or as in my case, there was reason the IAC was controlling idle speed high. Check vacuum hose connections, purge system, PCV, etc. The best that I can suggest is that you have someone check the engine with a good scan tool if you don't have one. You don't mention any check engine lights, but it's hard to diagnose without data. There ought to be clues in the data stream.
I was finally able to get back to the car today and I think I have it resolved. I picked up a clue from another forum post on the ECM programming section and I confirmed with a section in my service manual. Some of these early ECMs had a diagnostic mode that modifies the idle and is triggered by the scan tool.
Since this was a major project on a long parked car, I have been using my scan tool every time out to monitor parameters as I did my shakedown drives. The answer is that the scan tool causes elevated idle speed in park of 1050 along with elevated ignition timing by something like 8 - 10 degrees. This explains why the system was showing the IAC controlling to a high idle. The apparent roughness and random misfire was caused by excessive spark advance.
I ran it today without the tool. Cold idle in park was about 1200 and as the engine warmed up, it progressively dropped to about 650 on the dash tach and about 550 in gear. The idle quality is much improved, no audible misfire in the exhaust any longer. It sounds healthy now. It has a very light idle vibration and shake, but I did go to a little hotter cam and it's probably about right for that.
I don't understand why GM programmed the system to respond like this. They must have had a reason, but it's a mystery. I have an Actron CP9600 scan tool that is triggering this mode. It's one of the few tools out there that will go back to OBD-I.
In the scan tool options there should be a test drive function or some such - in the factory scan tool there is an option once it connected to run in a passive mode that doesn't modify the idle and timing, etc. Perhaps if you look more in your scanner options you will find this. This is just one of those things about older OBD-I ALDL protocols that the dealer techs knew and understood from their training and no one outside the factory or the dealer was supposed to be using that stuff. It's not a "problem" once you know how to use the tool properly.
I wouldn't necessarily jump to injectors. For speed to be high, the engine is getting additional air or making more power for the same amount of air. If it is rich, it is more likely to be controlling rich for some reason. High idle most often is excess air, frequently due to vacuum leaks or as in my case, there was reason the IAC was controlling idle speed high. Check vacuum hose connections, purge system, PCV, etc. The best that I can suggest is that you have someone check the engine with a good scan tool if you don't have one. You don't mention any check engine lights, but it's hard to diagnose without data. There ought to be clues in the data stream.
No check engine lights and the car runs good minus the high idle and RICH exhaust....I just got a BOSCH 1300 in the mail, BUT in my other thread I noticed MAF reading was like 2000 plus? GD above said that I might have to check / update the software on the tool since it shouldn't read that high at all....so far other readings seem ok though, but im seeing how to update the scanner.
I have a smoke machine coming in to test for any vacuum leaks too! Im open to any and al lsuggestions though!
In the scan tool options there should be a test drive function or some such - in the factory scan tool there is an option once it connected to run in a passive mode that doesn't modify the idle and timing, etc. Perhaps if you look more in your scanner options you will find this. This is just one of those things about older OBD-I ALDL protocols that the dealer techs knew and understood from their training and no one outside the factory or the dealer was supposed to be using that stuff. It's not a "problem" once you know how to use the tool properly.
GD
For some reason I can't respond to my other thread regarding the scan tool ibought, but that was the one with the MAF reading at over 2000.... I am taking your advice and seeing if I can update the scanner!
In the scan tool options there should be a test drive function or some such - in the factory scan tool there is an option once it connected to run in a passive mode that doesn't modify the idle and timing, etc. Perhaps if you look more in your scanner options you will find this. This is just one of those things about older OBD-I ALDL protocols that the dealer techs knew and understood from their training and no one outside the factory or the dealer was supposed to be using that stuff. It's not a "problem" once you know how to use the tool properly.
GD
I've been through my tool and the manual and it doesn't appear to have a selection for this, but I'll keep looking. I'm tempted to go for the laptop cable and software to see if that's better. I think that OBD1 support is so old, that the newer tools are just not as capable as before. Back in the day, I did have access to and use the Tech 1 tools, but I never had an instance where I encountered the need for this feature. It was probably just due to the specific vehicles I was working on. I do recall my controls colleagues talking about "10k mode", but I never had a need to dig into what it was all about.
No check engine lights and the car runs good minus the high idle and RICH exhaust....I just got a BOSCH 1300 in the mail, BUT in my other thread I noticed MAF reading was like 2000 plus? GD above said that I might have to check / update the software on the tool since it shouldn't read that high at all....so far other readings seem ok though, but im seeing how to update the scanner.
I have a smoke machine coming in to test for any vacuum leaks too! Im open to any and al lsuggestions though!
My Actron 9690 tool actually is a Bosch system based on the software splash screen. It also gives a mass air flow reading of about 2000-3500 around the high idle speed (1000 rpm) area for my V8, depending on load conditions. It is labeled as g/s, but that seems way high. I think there is a scale factor issue. It has the latest software.
No check engine lights and the car runs good minus the high idle and RICH exhaust....I just got a BOSCH 1300 in the mail, BUT in my other thread I noticed MAF reading was like 2000 plus? GD above said that I might have to check / update the software on the tool since it shouldn't read that high at all....so far other readings seem ok though, but im seeing how to update the scanner.
I have a smoke machine coming in to test for any vacuum leaks too! Im open to any and al lsuggestions though!
An additional thought for you. As you diagnose, consider that there is only an O2 sensor on one bank, plus many times reactions of these systems are counter intuitive, meaning that you need to think carefully about reactions. For example, a misfiring cylinder will look rich and the oxygen sensor on that bank will pull the system hard lean to compensate. I've seen this with a cracked spark plug insulator. On the other hand, if you have a lean condition from a restricted injector or vacuum leak on the bank with the O2 sensor, it will try to drive the system richer, so the opposite bank will be too rich once the computer compensates.
My Actron 9690 tool actually is a Bosch system based on the software splash screen. It also gives a mass air flow reading of about 2000-3500 around the high idle speed (1000 rpm) area for my V8, depending on load conditions. It is labeled as g/s, but that seems way high. I think there is a scale factor issue. It has the latest software.
If you divide the gms/sec by 256 the result looks to be close to what it should be. The reason for this is that the airflow value is 16 bits. The upper 8 bits is the integer portion of the airflow. While the lower 8 bits is the fractional portion of the airflow.
OK, just checked the data stream spec and it is incorrect. As I mentioned above divide the gms/sec value by 256 for the correct value. This is from the data sheet spec, which is why the scan tool gms/sec value is wrong:
22 AIR FLOW RATE (MSB)
23 AIR FLOW RATE (LSB) G/S = (MSB)*256+(LSB)
My Actron 9690 tool actually is a Bosch system based on the software splash screen. It also gives a mass air flow reading of about 2000-3500 around the high idle speed (1000 rpm) area for my V8, depending on load conditions. It is labeled as g/s, but that seems way high. I think there is a scale factor issue. It has the latest software.
thats good to know! So its not only my scan tool showing weird values.! I guess they don;t really check the stuff for OBD1 as everything is al lOBD2 these days.
An additional thought for you. As you diagnose, consider that there is only an O2 sensor on one bank, plus many times reactions of these systems are counter intuitive, meaning that you need to think carefully about reactions. For example, a misfiring cylinder will look rich and the oxygen sensor on that bank will pull the system hard lean to compensate. I've seen this with a cracked spark plug insulator. On the other hand, if you have a lean condition from a restricted injector or vacuum leak on the bank with the O2 sensor, it will try to drive the system richer, so the opposite bank will be too rich once the computer compensates.
I always wondered that logic of having the o2 in one side only..... my 2.8 has the O2 after all 6 cylinders, as well as the 4.3 in my S-10.... but the TPI has it in only one manifold monitoring just the driver side bank.
OK, just checked the data stream spec and it is incorrect. As I mentioned above divide the gms/sec value by 256 for the correct value. This is from the data sheet spec, which is why the scan tool gms/sec value is wrong:
22 AIR FLOW RATE (MSB)
23 AIR FLOW RATE (LSB) G/S = (MSB)*256+(LSB)
It should be: G/S = (MSB) + (LSB) / 256
RBob.
So our scan tools require us to do math?!?!? I paid for it to do all the calculations for me!!! lol
Thats good to know though. It comes out to 8-9 for me so that seems corrrect!
I always wondered that logic of having the o2 in one side only..... my 2.8 has the O2 after all 6 cylinders, as well as the 4.3 in my S-10.... but the TPI has it in only one manifold monitoring just the driver side bank.
Remember that ECM controls were still new in the '80s. The first widespread use of O2 sensors for GM was in 1981 and the engineers were still learning. The sensor needs to heat quickly and so needs to be close to the engine for best performance. Each application is different. If you have an engine with a crossover pipe under the engine, you can get all the cylinders through one sensor more easily. The TPI's downpipes likely come together too far from the engine for an unheated sensor. Ultimately the regulators pushed OBD2 and incremental improvements to OBD2 in order to force-fix many of these compromises. Today, the diagnostics must detect misfire in any cylinder and take action, and a V8 usually has 4 O2 sensors.
Remember that ECM controls were still new in the '80s. The first widespread use of O2 sensors for GM was in 1981 and the engineers were still learning. The sensor needs to heat quickly and so needs to be close to the engine for best performance. Each application is different. If you have an engine with a crossover pipe under the engine, you can get all the cylinders through one sensor more easily. The TPI's downpipes likely come together too far from the engine for an unheated sensor. Ultimately the regulators pushed OBD2 and incremental improvements to OBD2 in order to force-fix many of these compromises. Today, the diagnostics must detect misfire in any cylinder and take action, and a V8 usually has 4 O2 sensors.
I wish OBD1 had the capabilities / easy use of OBD2.
Hi, dan5, please tell me about your Actron 9690. Does it show live data of all the sensors? Or are there any restrictions? I'm looking for such a scan tool. Than you!
The tool picks up almost all of the sensors on my 1985 LB9. There may be some differences in the available parameters by model year and ECM. The only omission on my '85 is that there is no knock sensor and timing information. I will provide some screen shots from the tool for my engine. These shots overlap, but contain all the available parameters. This is the full list and is just a random shot of a frame of recorded data at idle. The tool will let you view live or you can record data sets. The tool will collect 64 frames of data over 143 seconds on this ECM. It also shows what DTCs are set.
I think the IAC counts should be much lower at idle. Around 5-10.
Tootie, that is what started this thread. The idle was being driven up to 1050-1100 RPM by the tool interacting with the ECM. Depending on the engine temperature, the IAC was running 50-70 counts at 1050 to 1100. I can't tell where the IAC is running at 500 or 600 RPM, because the tool won't allow that, but it should drop substantially.
The tool picks up almost all of the sensors on my 1985 LB9. There may be some differences in the available parameters by model year and ECM. The only omission on my '85 is that there is no knock sensor and timing information. I will provide some screen shots from the tool for my engine. These shots overlap, but contain all the available parameters. This is the full list and is just a random shot of a frame of recorded data at idle. The tool will let you view live or you can record data sets. The tool will collect 64 frames of data over 143 seconds on this ECM. It also shows what DTCs are set.
Thank you very much! Wow, there's quite a lot of information. This tool is very interesting!
If I know the parameters of a stable condition, it will absolutely help solve the problem when something happens.
In fact, I was not sure which OBD2 scan tool works for the thirdgen's, so your information really helps me. Thanks a lot!
Thank you very much! Wow, there's quite a lot of information. This tool is very interesting!
If I know the parameters of a stable condition, it will absolutely help solve the problem when something happens.
In fact, I was not sure which OBD2 scan tool works for the thirdgen's, so your information really helps me. Thanks a lot!
You're welcome. I am glad to be able to help. When I bought the tool, I was not sure how good it would be, even after reading the online manual. But it is very good and it has proven to be a great help as I was working to put my car back on the road. I had a number of issues that were much easier to resolve with the tool. Sometimes it is just good to know that the sensors are reading correctly. I had two brand new sensors fail within a short time after installation (IAC and O2). This tool helped me get those diagnosed quickly.
07-27-2020, 11:21 AM
