The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support RTM Racing
Please Support STM Tuned

ECMlink Dreaded idle surge after AEM wideband install?.?

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

AEM is annoying sometimes.

When my ISC went bad it still tested fine in the Ohm test, just like what you are saying.
I would not waste my time buying a used one. Get a new one while they are still available, and before the new prices go any higher.
Right now from Rock Auto you can get a new one for:
$69.79 (Standard Motor Products AC146)
or
$157.99 (WVE/AIRTEX/WELLS 2H1098)

I am using the Standard Motor Products AC146 in my car and it is good. When I bought mine the Airtex/Wells was only about $70 but their price has gone wacko since then.

The drill-down in Rockauto goes like this: Eagle, 1991, Talon, Turbo, Fuel and Air, Idle Air Control Valve.
Here's a link that should take you right there.

https://www.rockauto.com/en/catalog...,fuel+&+air,idle+air+control+(iac)+valve,6072
Unfortunately I had already had ordered a used one from performance partout, it was only $30 shipped. So I guess it's worth the gamble. I had replaced my isc once already, back in 2017. If I remember correctly it was a standard brand, not sure why it failed so fast. I did take it apart, and inspected it. Can't really see anything of concern, and I can manually turn the plunger in and out. I'm really hoping the issue isn't in my engine harness. I've had several points of failure that haven't made much sense. I think maybe the harness got damaged somewhere, and I can't find it without a major job of pulling it out. And unwrapping. So I might do an ohm test on all of the wires to the ISC. From the ECU to the harness plug. My question is I see there are 4 ISC wires coming from the ECU, but 6 at the plug. Are the other two just sensor grounds?
 
I've had several points of failure that haven't made much sense. I think maybe the harness got damaged somewhere, and I can't find it without a major job of pulling it out. And unwrapping. So I might do an ohm test on all of the wires to the ISC. From the ECU to the harness plug. My question is I see there are 4 ISC wires coming from the ECU, but 6 at the plug. Are the other two just sensor grounds?

The 6 wires at the plug, actually they do all go back to the ECU.
On the plug where the wires are labeled 1-6, the wires #2 and #5 are for +12 volts (power). They go to #102 and #107 on the ECU which are "12v power from MPI relay".

The other 4 wires at the plug: 1 and 3 go to a coil in the ISC, 4 and 6 go to the other coil in the ISC, and those 4 are the ones you are already seeing.

I haven't seen mention yet of the ISC driver chips (in the ECU) in this thread. But they can be bad. They can be visibly bad, as in visual inspection yeah that looks bad haha. ECM tuning has some pics of blown chips. http://www.dsmlink.com/wiki/driver01#blown_isc_driver

If you look at the "Photos" tab in my profile here, You'll see one good photo of ISC chips that look good and are good. Bring that photo up to full size for a good look. I had the cam focused on the ISC chips because that was my purpose for that photo. That ECU had just come back from ECM tuning.

You will also see a large circuit diagram of the whole engine harness in my photos. Be careful, that is for the 1990. Two of the ISC wires are turned around on the 1990, compared to the later years - plug positions 4 and 6, I believe.
 
Last edited:
The 6 wires at the plug, actually they do all go back to the ECU.
On the plug where the wires are labeled 1-6, the wires #2 and #5 are for +12 volts (power) but they get the 12v from the ECU. They go to #102 and #107 on the ECU which are "12v power from MPI relay".

The other 4 wires at the plug: 1 and 3 go to a coil in the ISC, 4 and 6 go to the other coil in the ISC, and those 4 are the ones you are already seeing.

I haven't seen mention yet of the ISC driver chips (in the ECU) in this thread. But they can be bad. They can be visibly bad, as in visual inspection yeah that looks bad haha. ECM tuning has some pics of blown chips. http://www.dsmlink.com/wiki/driver01#blown_isc_driver

If you look at the "Photos" tab in my profile here, You'll see one good photo of ISC chips that look good and are good. Bring that photo up to full size for a good look. I had the cam focused on the ISC chips because that was my purpose for that photo. That ECU had just come back from ECM tuning.

You will also see a large circuit diagram of the whole engine harness in my photos. Be careful, that is for the 1990. Two of the ISC wires are turned around on the 1990, compared to the later years - plug positions 4 and 6, I believe.
Perfect, that's exactly what I needed to know. Now at least I can test every wire to ISC plug. Also one thing i can probably cross off being an issue is my ECU. I just got it back from ecmtuning. I sent it off to be checked out, because of some of the issues I'm having. And he gave it clean bill of health. So that definitely leads me to believe the harness is probably an issue. For sure have a gremlin hiding somewhere. Anyway thanks again for the great info, and help.
 
Just food for thought....I fought idle surging on my 1g for years and years...one day someone mentioned the idle stop screw being out of adjustment. So I did the feeler gap check and sure enough the damn throttle plate was open too far. Adjusted it to proper spec and never had idle surging issues again
 
If you decide that you need new connectors anywhere in the harness, you might know already, Brad Sheridan is the guy. His website is Sheridan Engineering, here: http://connectors.sheridanengineering.com/index.htm
Yeah actually I wasn't aware of this, that's great to know. I've been going through ohm racing for my harness connectors. Although I've had decent luck with them, he doesn't have near the selection as that. Also they seem to be a bit more reasonably priced. Thanks again.
 
If you decide that you need new connectors anywhere in the harness, you might know already, Brad Sheridan is the guy. His website is Sheridan Engineering, here: http://connectors.sheridanengineering.com/index.htm
So I got my new/used ISC in the mail today, and quickly installed it. I thought my issues were solved, but not 100% yet. First heat cycle, it idled exactly where it use to. Starting around 1400rpm, after a minute drops down to about 1000. And down to my target idle at 850 in closed loop. But after shutting it down, and restarting in about 30min, it was idling extremely low, like 600ish. So I check the ISC position using my laptop, and it's maxed out at 120 no matter what I do. Later I check it again, and it's extremely low, like 9. So it's all over the place, not sure what to think? Also I should state that I did a ohm test on it, everything looked good. So do you think it's a bad ISC, or other issues?
 
Well, the way mine acts is basically like this:
At every cold start, when I first turn the key on and before hitting the starter, my ISC logs 120.
I usually log several seconds between key on and start, because my start is on a separate button on the center stack above the radio.
So you can see that real well in my logs, because I gotta move my hand from the ignition key to the starter button, and I'm not in a big hurry about it.
Then when I shut the engine off (warm), I get about another 10 seconds of logging, and the first thing the ISC does after shutting the ignition off is it ramps up to about 130 and then ramps right back down to 90 and stays at 90 all the way to disconnect which is about 8 seconds later.
From the cold start, ISC will stay at 120 for a minute or so and then start coming down gradually.
When fully warmed up, if I'm sitting at idle for a while someplace, the ISC will usually be at about 15 or 20. But it sometimes will go all the way to 0 for a bit, usually after taking foot off the gas.
I logged a long drive right after putting in my new ISC. Then 2 days later I logged another long drive right after changing my timing map to drastically reduce timing at low load and low RPM. The difference I think I see there in the ISC values is that before I reduced my timing, the ISC was spending more time at 0 or very low values like 9 or less. After I reduced my timing the ISC didn't hit such low values much, but was more likely to be around 15 to 20 at stable idle.

My Target Idle in ECMLink is 1100 rpm.
My Front O2 is a real, organically grown, OEM narrow band, fairly new, mounted in the O2 housing.
I do have an AEM wideband mounted on the downpipe but it doesn't control anything, it just runs a gauge (which seems to give good numbers) and it sends a hilarious inaccurate analog signal to ECMLink so don't pay much attention to my "AEM Linear Wideband" numbers in the logs.

Here is a log that is recent and short and it was a cold start. I warmed it up for 6 minutes, then shut it off. I didn't go anywhere, just sitting in neutral in the garage. For most of the first 4 minutes I have my foot on the gas just a little, because it will run a lot cleaner that way while warming up. In the last 2 minutes the ISC gets down as low as 23 and that's about it, because coolant is still only up to 160 deg F.

You probably don't want to spend the rest of your life looking at these logs LOL but what the heck, here's the long drive (about an hour) that I logged after reducing my low load timing and I'm with the new ISC.

And here is the long drive that I logged 2 days earlier with the new ISC but before reducing my low load timing. About a 45 minute drive.

You'll probably notice that my Front O2 doesn't start oscillating until coolant reaches 160 deg F.
In the 1st log it starts oscillating much sooner but that is because I have my foot on the gas. I took my foot off the gas at 143 deg F and the O2 quit oscillating. Then 70 seconds later, temp reaches 160, and it starts oscillating. I don't know how "normal" this is. It might be due to something in my setup. But it seems to be OK.

Well I see the files didn't attach where I wanted them to in the text body, but the titles and dates and size will tell you which is which!
 

Attachments

  • log.2018.08.31-01.elg
    168.2 KB · Views: 65
  • log.2017.10.24-01 first drive with reduced timing near idle, much better.elg
    1.5 MB · Views: 64
  • log.2017.10.22-03 long drive with new ISC, 116th, 124th, to Duvall to Snoq.elg
    1.2 MB · Views: 50
Last edited:
Well, the way mine acts is basically like this:
At every cold start, when I first turn the key on and before hitting the starter, my ISC logs 120.
I usually log several seconds between key on and start, because my start is on a separate button on the center stack above the radio.
So you can see that real well in my logs, because I gotta move my hand from the ignition key to the starter button, and I'm not in a big hurry about it.
Then when I shut the engine off (warm), I get about another 10 seconds of logging, and the first thing the ISC does after shutting the ignition off is it ramps up to about 130 and then ramps right back down to 90 and stays at 90 all the way to disconnect which is about 8 seconds later.
From the cold start, ISC will stay at 120 for a minute or so and then start coming down gradually.
When fully warmed up, if I'm sitting at idle for a while someplace, the ISC will usually be at about 15 or 20. But it sometimes will go all the way to 0 for a bit, usually after taking foot off the gas.
I logged a long drive right after putting in my new ISC. Then 2 days later I logged another long drive right after changing my timing map to drastically reduce timing at low load and low RPM. The difference I think I see there in the ISC values is that before I reduced my timing, the ISC was spending more time at 0 or very low values like 9 or less. After I reduced my timing the ISC didn't hit such low values much, but was more likely to be around 15 to 20 at stable idle.

My Target Idle in ECMLink is 1100 rpm.
My Front O2 is a real, organically grown, OEM narrow band, fairly new, mounted in the O2 housing.
I do have an AEM wideband mounted on the downpipe but it doesn't control anything, it just runs a gauge (which seems to give good numbers) and it sends a hilarious inaccurate analog signal to ECMLink so don't pay much attention to my "AEM Linear Wideband" numbers in the logs.

Here is a log that is recent and short and it was a cold start. I warmed it up for 6 minutes, then shut it off. I didn't go anywhere, just sitting in neutral in the garage. For most of the first 4 minutes I have my foot on the gas just a little, because it will run a lot cleaner that way while warming up. In the last 2 minutes the ISC gets down as low as 23 and that's about it, because coolant is still only up to 160 deg F.

You probably don't want to spend the rest of your life looking at these logs LOL but what the heck, here's the long drive (about an hour) that I logged after reducing my low load timing and I'm with the new ISC.

And here is the long drive that I logged 2 days earlier with the new ISC but before reducing my low load timing. About a 45 minute drive.

You'll probably notice that my Front O2 doesn't start oscillating until coolant reaches 160 deg F.
In the 1st log it starts oscillating much sooner but that is because I have my foot on the gas. I took my foot off the gas at 143 deg F and the O2 quit oscillating. Then 70 seconds later, temp reaches 160, and it starts oscillating. I don't know how "normal" this is. It might be due to something in my setup. But it seems to be OK.

Well I see the files didn't attach where I wanted them to in the text body, but the titles and dates and size will tell you which is which!
Thanks for all the input, I'm at work now, but I'll look over the logs when I get home. You brought up a few good points, and things that I really need to try. I'm getting some huge inaccuracies through my AEM wideband analog output. It's showing a healthy 14.5-14.9 on my gauge, but logging a low 13. And being that it's the only o2 sensor installed, that is bound to throw my whole tune off. So I'm convinced I need to switch back to my stock o2 for normal operations (have a brand new bosch). Then do like you and weld a bung in my downpipe, and I'll tap the analog into the EGR. Perfect little project for my new 140a welder I just bought. I know I'll sleep much better knowing that's all good.

So one question for you though. When your using link to adjust your ISC position, do you have click both the timing & diagnostics boxes? Or just one to get an accurate reading?
 
So one question for you though. When your using link to adjust your ISC position, do you have click both the timing & diagnostics boxes? Or just one to get an accurate reading?

I don't use ECMlink to adjust the ISC position, so I'm not sure what you mean by that. All I do is log ISC position, and it sounds like you are already doing that Ok.
The only ISC adjust I can think of is just setting the Target Idle in RPM/TPS.
Maybe you mean something like an initial calibration of the ISC after you put a new or used one in, like how does the ECU know where that thing is to begin with?
That is something I've thought about some because when I installed my new one I "bench tested" it for a while first and by the time I got done with that the position of the pintle was different than when I first took it out of the box. I don't know the answer but I think it has a way of calibrating itself. Maybe not all at once, maybe it takes a few cycles. That is what I thought was so interesting about post #8 in the other thread I mentioned back in post #25 in this thread. The idea there is that after some cycling, "It wound up a little more extended at the end than it was initially" (the pintle). That's when there is nothing for it to physically run into. I suppose in the car it runs into its valve seat which stops it, then it knows where it is. Anyway, I never did anything like an initial calibration on it. I think it calibrated itself somehow.

On the AEM wideband, my AEM is not the latest unit, it is the one that was sold up until 2016. That model is well known to have weird issues with ECMlink. If yours is the newer one that first became available in 2016, then I don't know maybe those are better.
One thing you could do is log the actual wideband raw voltage and display that in your logs. I don't remember what it is called but somewhere back in my logs I have one or 2 logs where I looked at those voltages and then looked at the AEM chart where they tell you what the A/F is for that voltage, and sure enough it was crap, way off. So it was as if either the AEM was putting out goofy voltages, or the ECMlink was just reading them wrong. I am just enough of an electronics nerd to know that the input impedance of the "receiving" device (ECMlink) has to be pretty high in order to not alter the signal of any weak input circuitry (the AEM analog out). And that I suppose is why some people have built "buffer circuits" to go between the AEM wideband and the ECU input. But I've heard that even that doesn't always work. On the other hand some people I guess have no trouble with it. Weird stuff this is.
Anyway, if you could log the raw voltage from your AEM wideband, you might be able to figure out if a correct narrow band simulation from it is even possible.
But going back to a regular narrow band for ECU control sounds like the right thing to do. Then the wideband would be just for the gauge, like on mine. If you want actual good wideband logs you might have to get an Innovate wideband instead of the AEM.
GSTwithPSI (user here) uses an Innovate wideband and does the narrow band simulation with it and it seems fine and his logged A/F ratio values seem fine.
 
Last edited:
I don't use ECMlink to adjust the ISC position, so I'm not sure what you mean by that. All I do is log ISC position, and it sounds like you are already doing that Ok.
The only ISC adjust I can think of is just setting the Target Idle in RPM/TPS.
Maybe you mean something like an initial calibration of the ISC after you put a new or used one in, like how does the ECU know where that thing is to begin with?
That is something I've thought about some because when I installed my new one I "bench tested" it for a while first and by the time I got done with that the position of the pintle was different than when I first took it out of the box. I don't know the answer but I think it has a way of calibrating itself. Maybe not all at once, maybe it takes a few cycles. That is what I thought was so interesting about post #8 in the other thread I mentioned back in post #25 in this thread. The idea there is that after some cycling, "It wound up a little more extended at the end than it was initially" (the pintle). That's when there is nothing for it to physically run into. I suppose in the car it runs into its valve seat which stops it, then it knows where it is. Anyway, I never did anything like an initial calibration on it. I think it calibrated itself somehow.

On the AEM wideband, my AEM is not the latest unit, it is the one that was sold up until 2016. That model is well known to have weird issues with ECMlink. If yours is the newer one that first became available in 2016, then I don't know maybe those are better.
One thing you could do is log the actual wideband raw voltage and display that in your logs. I don't remember what it is called but somewhere back in my logs I have one or 2 logs where I looked at those voltages and then looked at the AEM chart where they tell you what the A/F is for that voltage, and sure enough it was crap, way off. So it was as if either the AEM was putting out goofy voltages, or the ECMlink was just reading them wrong. I am just enough of an electronics nerd to know that the input impedance of the "receiving" device (ECMlink) has to be pretty high in order to not alter the signal of any weak input circuitry (the AEM analog out). And that I suppose is why some people have built "buffer circuits" to go between the AEM wideband and the ECU input. But I've heard that even that doesn't always work. On the other hand some people I guess have no trouble with it. Weird stuff this is.
Anyway, if you could log the raw voltage from your AEM wideband, you might be able to figure out if a correct narrow band simulation from it is even possible.
But going back to a regular narrow band for ECU control sounds like the right thing to do. Then the wideband would be just for the gauge, like on mine. If you want actual good wideband logs you might have to get an Innovate wideband instead of the AEM.
GSTwithPSI (user here) uses an Innovate wideband and does the narrow band simulation with it and it seems fine and his logged A/F ratio values seem fine.

Yeah what I'm referring to is the timing ground connector by the battery, and then the diagnostics connector on the obd connector. Just like when you set your base timing, your supposed to ground that connector. Supposed to stabilize your rpm, and timing (I think). Well anyway you can do both of those functions through link, without having to physically ground anything. And your supposed to be able to adjust biss, watching your log, to get the isc position at 30. I'm just not sure if I need to ground both, or just one of those functions through link?

But anyway I'm 99% sure I've found the issue, at least with ISC circuit. So I decided to finally test all the ISC wires with my ohm meter, from the ECU to the plug. And I got a good low resistance on all but one. #2 on the connector, I have nothing on, it's an open circuit. And looking back at your earlier posts, #2 should be to the MPI circuit, @pin #102. So I was hoping to get a confirmation on this, and I'll just supply a new wire. Solder, heat shrink, and wrap it. Then hopefully be done with this mess. I'm pretty sure now that my engine harness does have a point of failure. And instead of pulling my hair out trying to find it. I'll just bandage it up for now, but I will dig deeper later, and hopefully find the source of the issues. Thanks again, you've been a huge help.
 
Well, I probably shouldn't comment on how to set the idle because I've never had to do it!
My ECMlink was setup by ER and I have never had to mess with it. All the years before I had ECMlink my idle was always a little high which was fine with me so I just left it!

On the wiring to the ISC, it sure seems like there should be good continuity from #102 on the ECU to #2 on the ISC. But depending on where the break is in that line, power could still come around to it from the ECU pin #107 line which connects to #4 and #5 on the MPI relay and then loops around and connects to the ECU #102 line! That is, if that diagram is correct.
Anyway, it seems to me that finding the problem in that #102 line is the thing to do.

Want to make sure you have a copy of this ECU pinout diagram, the one on the ECMlink web site:
http://www.dsmlink.com/images/forums/1GECUPinout.pdf
 
When you get around to it, you can get your older AEM wideband to log accurately with ECMLink. You have to set it up as "Linear" in the drop down, and then manually enter the proper voltage values in the table, which you will find yourself. Heres a video made by another member showing how to do this:

You must be logged in to view this image or video.

You can get it very close this way. Definitely close enough for what most are up to.
 
When you get around to it, you can get your older AEM wideband to log accurately with ECMLink. You have to set it up as "Linear" in the drop down, and then manually enter the proper voltage values in the table, which you will find yourself. Heres a video made by another member showing how to do this

Thanks!
 
When you get around to it, you can get your older AEM wideband to log accurately with ECMLink. You have to set it up as "Linear" in the drop down, and then manually enter the proper voltage values in the table, which you will find yourself. Heres a video made by another member showing how to do this:

You must be logged in to view this image or video.

You can get it very close this way. Definitely close enough for what most are up to.
Yes I agree, that's awesome. Thanks for sharing.
 
Looking at this diag
 

Attachments

  • 4g63ecudiagram_1.jpg
    4g63ecudiagram_1.jpg
    296 KB · Views: 145
Looking at this diagram, shows exactly what I need to see. Definitely one of the better ones I've seen. But here's the thing, pin#102 supplies most the sensors with it's 12v source. So if I was having an issue with that circuit, I would think it would be on all the sensors. I guess depending on where the brake in the wire is. But I'm thinking I should go ahead and test from a few other plugs, see if I can get continuity back to the 102. If the ISC is the only one missing it's 102 source, then I'm going to try running a temporary wire. See if I can get normal movement from the ISC. That should really narrow things down.

You must be logged in to view this image or video.
 
Looking at this diagram, shows exactly what I need to see. Definitely one of the better ones I've seen. But here's the thing, pin#102 supplies most the sensors with it's 12v source. So if I was having an issue with that circuit, I would think it would be on all the sensors. I guess depending on where the brake in the wire is.

Ahh good you found the diag for the 91-94. Made by the same guy that made the 1990 diagram: Jeremy, Hot Rod Coffee Shop.
https://blog.hotrodcoffeeshop.com/2013/02/23/4g63-wiring-diagrams-schematics-for-engine-swaps/

I was thinking the same thing about how other sensors are on the same power circuit.
Also, the #102 and #107 lines connect to each other where I put the blue arrow in this markup. Redundancy? I don't exactly get it.

You must be logged in to view this image or video.
 
Last edited:
Ahh good you found the diag for the 91-94. Made by the same guy that made the 1990 diagram: Jeremy, Hot Rod Coffee Shop.
https://blog.hotrodcoffeeshop.com/2013/02/23/4g63-wiring-diagrams-schematics-for-engine-swaps/

I was thinking the same thing about how other sensors are on the same power circuit.
Also, the #102 and #107 lines connect to each other where I put the blue arrow in this markup. Redundancy? I don't exactly get it.

You must be logged in to view this image or video.
Let me ask you one more question, and then we can let this thread die. So today I actually manually set my idle to around 1100rpm, and took it for a quick test drive. It really seems like I should be able to run like that for now, until I get the issue sorted. Obviously a lot of people run without ISC/IAC control, and I realize it wouldn't be ideal in the colder months. Do you see any reason why I shouldn't run it without the ISC functioning?

Also just an update, the only thing I can find wrong is pin#2 at the ISC lacking it's 12v source. I'm considering just bridging pin#2&5 together a few inches back from the plug. Like you've already stated, there supposed to be bridged anyway. Still I sure would like to have an explanation of what happened, and find some closure here. Thanks again.
 
Well, having an ISC that doesn't move at all is better than having an ISC that moves erratically/wrongly.
I really like having an ISC that works properly, but it is mostly for convenience.
So here are some things I can think of:
If you still have emission checks, usually the car has to idle below 1000 RPM. But a 1991 is well over 25 years old now so you shouldn't have that problem anymore!
Cold idle. It's nice to have that faster idle after a cold start. But you can use your foot too LOL.
Idle speed increase with steering input, from that pressure switch on the power steering pump. Which I don't have anymore, just because nobody bothered to hook it back up after ER did my engine in 2016. With the basic idle already at 1100, I don't miss it. If I was trying to have an 800 rpm idle I might want that back.
If I had an automatic trans it would be nicer for the trans I suppose, to have a low idle with good control. Don't have an auto.

You might have a stalling issue when you are coasting down in neutral and you take your foot off the gas. I've had that and it was pretty miserable. It can make maneuvering around in gas stations and parking lots pretty rotten. So you might have that even with a basic idle of 1100. It gets you when you need your right foot on the brake so it can't be on the gas pedal. That's about the only thing I can think of that might really be a pain - if it happens.
But I can't think of any really killer diller reason why it's bad.

I just found a diagram of the inside of the MFI relay (MPI relay), which I had wondered about. Because I didn't know if the MPI pins 4 and 5 were on switches or coils. Turns out those 2 pins #4 and #5 on the MPI relay are from a switch in the relay, so they are putting power out to #102 and #107 of the ECU, as well as to all the sensors that are connected to that line. If ECU #102 is somehow not getting power from the MPI because of a break, that I would want to fix.
The MPI pins that are to coils are MPI #7 (from ECU #56) and MPI #8 (from ECU #63 and #66).
This diagram I'm looking at is "Diagram 8" on page 6-49 of the Chilton manual for 1990 to 1998 Eclipse.


You must be logged in to view this image or video.
 
Last edited:
Well, having an ISC that doesn't move at all is better than having an ISC that moves erratically/wrongly.
I really like having an ISC that works properly, but it is mostly for convenience.
So here are some things I can think of:
If you still have emission checks, usually the car has to idle below 1000 RPM. But a 1991 is well over 25 years old now so you shouldn't have that problem anymore!
Cold idle. It's nice to have that faster idle after a cold start. But you can use your foot too LOL.
Idle speed increase with steering input, from that pressure switch on the power steering pump. Which I don't have anymore, just because nobody bothered to hook it back up after ER did my engine in 2016. With the basic idle already at 1100, I don't miss it. If I was trying to have an 800 rpm idle I might want that back.
If I had an automatic trans it would be nicer for the trans I suppose, to have a low idle with good control. Don't have an auto.

You might have a stalling issue when you are coasting down in neutral and you take your foot off the gas. I've had that and it was pretty miserable. It can make maneuvering around in gas stations and parking lots pretty rotten. So you might have that even with a basic idle of 1100. It gets you when you need your right foot on the brake so it can't be on the gas pedal. That's about the only thing I can think of that might really be a pain - if it happens.
But I can't think of any really killer diller reason why it's bad.

I just found a diagram of the inside of the MFI relay (MPI relay), which I had wondered about. Because I didn't know if the MPI pins 4 and 5 were on switches or coils. Turns out those 2 pins #4 and #5 on the MPI relay are from a switch in the relay, so they are putting power out to #102 and #107 of the ECU, as well as to all the sensors that are connected to that line. If ECU #102 is somehow not getting power from the MPI because of a break, that I would want to fix.
The MPI pins that are to coils are MPI #7 (from ECU #56) and MPI #8 (from ECU #63 and #66).
This diagram I'm looking at is "Diagram 8" on page 6-49 of the Chilton manual for 1990 to 1998 Eclipse.


You must be logged in to view this image or video.
Have kind of a unfortunate update for you. Things went from bad to worse on my wiring conundrum. First I decided to go ahead an order a brand new ISC, just to eliminate that being a possibility. After installing it, and logging the first start up. I realized I was having more issues. I had switched back to my stock o2 sensor, but now it's not working. So I tried my spare, with the same results. So I starting testing all the wires at the O2 connector. Turns out I was missing a ground, which I decided to supply a new one. Just to see if that fixes it. For some reason I'm getting intermittent failures on several wires. That will be 3 circuits now with issues.
So anyway, I'm done deal with this. One of my biggest pet peeves is janky wiring. As much as it really pains me, I gotta pull the engine harness loose from inside. Find the failure points, and fix it right. It's very possible I may have hit the harness with a drill bit a few months ago, while drilling through the firewall. What really sucks is I just had the dash removed doing my heater core replacement. Sure would have been much easier then, but at least I know how it comes apart now. I guess in the end ill sleep better knowing it's all good, and my car won't be starting on fire LOL.
 
Man that sucks to hear things have gotten worse. You'll be doing the right thing by tearing it all back apart and fixing it right. If I'm looking at a car to buy and see someone had been splicing wires or as you said janky wiring. Good luck to you!!!
 
One of my biggest pet peeves is janky wiring. As much as it really pains me, I gotta pull the engine harness loose from inside. Find the failure points, and fix it right. It's very possible I may have hit the harness with a drill bit a few months ago, while drilling through the firewall. What really sucks is I just had the dash removed doing my heater core replacement. Sure would have been much easier then, but at least I know how it comes apart now. I guess in the end ill sleep better knowing it's all good, and my car won't be starting on fire LOL.

Yeah seems to me most "car guys" are reasonably into the mechanical stuff but when it comes to wiring they treat it like, I don't know, dead seaweed washed up on the shoreline or something. So things done by the previous owners or even shops, it all goes downhill.
The original unmolested factory wiring, and your own wiring, are usually the best!
It took me months to relocate my ECU to the passenger footwell, and I did as much as possible in the warm well lighted comfort of my apartment.

Do you know there is such a thing as engine compartment wire? GXL, SXL, and TXL are what I would call engine compartment wire, although other types of wire could work too.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top