Resolved 2G Rev/Coming to stop high idle issue

[Dryppical]

I have recently completed my manual swap on my 1997 Eclipse GSX and I went to take it for a drive. There's this weird issue of the rpm's like sticking but it's not the cable.

When I was idling and the rpm's got stuck when giving it some throttle, I went to the throttle body and tried to close it more and it did not move and the rpm's stayed the same. So if I'm idling, if I give it a little over 2% throttle, the car shoots up to 1600rpm and will hold for 3-5 seconds, then come back down. When I'm driving and coming to a stop, slowing down. If I press in the clutch, rolling to my stop, the throttle sticks right around 1600rpm and will fall back down after 3-5 seconds. I don't know where to start looking.

I got some logs, so hopefully that helps. I bought this DSMLink ECU used so if the settings still aren't stock, please let me know what to fix. I think I got it back to stock but I don't know.

Also, another random issue is the engine fan is constantly on. Both temp sensors are new.

My mods are: 16g turbo, cold air intake, 3" GM MAF (setup as blow through), FMIC, 2.5" intercooler piping, brand new ACT streetlight lightweight flywheel (.02 miles on it), brand new ACT 2600 clutch and street disk (.02 miles on it), MBC turned all the way down. Car has 95 JDM 4g63t in it.

I believe that's everything. Thanks

 

[SixBolt_16G]

Interesting. Im dealing with something similar if not the same overall issue. (Only difference is that mine didnt start until after the cruise control was removed and put in a non-cruise control cable, but the effect is very similar to what youre describing. ) The more gas i give it the worse it hangs and if I dare get into boost, it seems like rpms rise even more when I push the clutch in to shift..

If you put your hand on the accelerator and give it some gas, but then you pull up on the accelerator to pull it back into its resting position, do the rpms immediately drop like normal? My rpms seem to respond to doing so and im leaning towards investigating whether there is actually a problem with the cable at the pedal rather than around the throttle body.. Ive boost leak tested like like 5 times now and checked/adjusted the throttle cable just as much and still seeing the same issue.

Same flywheel, same new clutch. Almost similar set up to you except speed density.

 

[Dryppical]

Hopefully we can figure this out . No, the rpm’s will stick for about 3-5 seconds before falling back to normal

Removed the throttle cable and it does it without it being connected so I would think it would be safe to rule that out. I revved it by hand and rpms stuck. Tried to close the throttle shut more and no change. Would still stick for 3-5 seconds before falling back to idle

 

[SixBolt_16G]

Hopefully we can figure this out . No, the rpm’s will stick for about 3-5 seconds before falling back to normal

Otherwise, if it wasnt a problem with the accelerator sticking, i was going to asking about ECMLink settings, because I got screwed with the guy tuning it not being able finish. So im also dealing with a half-assed tune.

 

[Dryppical]

I bought the ECU used from a guy and went in and changed some settings a while back to get it to stock and downloaded the stock setting things from ECMLink. I think it’s back to stock or at least mostly, I have no idea. Still trying to figure it out

 

[dwb]

Everything on your DA tables are set to stock.

I see your idle target has been upped by 50 rpm, not a big deal. You're also set to use narrowband simulation, so I assume you have a wideband in the front O2 location. Can you confirm?

I'm not sure what else might be non-stock since I'm used to looking at a 1g.
You also have a check engine light code P1500 Generator Terminal Circuit Malfunction. That appears to be an alternator malfunction.

Do you have a 1g throttlebody with an idle switch or something? I thought 2g simulated idle from TPS. I also see an instance where TPS is showing 1-2% but idlesw is still indicating 0%. Again, maybe this is a 2g thing.

Something looks odd with your CoolantTemp. It's only reading from 153 to 160. Same with the IntTemp, it only shows 77 the whole time.

I don't see that your are capturing ISCposition which would be beneficial for troubleshooting idle issues.

There's some other things that you should be capturing too. Here's a list of values I found somewhere on this site that's a good start:
AFRatioEst
AirflowPerRev
Battery
BoostEst
ClosedLoop
CltchSw
CombinedFT
CoolTemp
FrontO2
HP
IdleSw
InjDuty
IntTemp
ISCPosition
KnockRet
LoadFactor
MAFComp
MAFRAW
OpenLoop
RPM
SelectedLFTF
ThrotPos
Timing
Torque
TPSVolts
WBFactor

 

[We're on Boost]

I have recently completed my manual swap on my 1997 Eclipse GSX and I went to take it for a drive. Theres this weird issue of the rpm's like sticking but it's not the cable. When I was idling and the rpm's got stuck when giving it some throttle, I went to the throttle body and tried to close it more and it did not move and the rpm's stayed the same. So if I'm idling, if I give it a little over 2% throttle, the car shoots up to 1600rpm and will hold for 3-5 seconds, then come back down. When I'm driving and coming to a stop, slowing down. If I press in the clutch, rolling to my stop, the throttle sticks right around 1600rpm and will fall back down after 3-5 seconds. I don't know where to start looking. I got some logs, so hopefully that helps. I bought this dsmlink ecu used so if the settings still aren't stock, please let me know what to fix. I think I got it back to stock but I don't know. Also, another random issue is the engine fan is constantly on. Both temp sensors are new
My mods are: 16g turbo, cold air intake, 3" gm maf(setup as blow through), fmic, 2.5" intercooler piping, brand new act streetlight lightweight flywheel (.02 miles on it), brand new act 2600 clutch and street disk (.02 miles on it), mbc turned all the way down. Car has 95 jdm 4g63t in it. I believe that's everything. Thanks

It looks like you aren't logging ISC position. Gotta do that.

Your timing goes very advanced when you give it that 2% throttle. That high timing speeds up the rpm a lot and it takes a while for it to come back down after you take your foot off the gas.
You should lower your timing in Direct Access (TimingMaxOct and TimingMinOct) at the lowest load factors 0.3 and 0.4 to something like this (from my car):

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

You could even have it that low at load factor 0.5. This isn't going to effect your power any because there is no power down there anyway. As soon as you get into the gas pedal a little bit, the load factor will shoot up closer to 1. Having high timing like stock at low loads is, as far as I know, just for slightly better fuel economy and maybe emissions. ECMtuning also has a wiki page someplace where they suggest lowering your timing at low loads for the same reason.

You probably have more going on than just this. I notice your front O2 quits cycling at idle after it settles down to the target 800-900 rpm, and the wideband reads very lean then, like low 16's. When I had a GM MAF with a DSMchips ecu, I had very accurate fuel metering everywhere except at idle, where I had to bump it up richer by quite a bit with the overrides I had in ecu+ which was a nice piggyback I had at the time. One way to bump up your idle richness is by adding some Global Deadtime on the Fuel tab. You can use that kind of like the idle mixture screw on an old carburetor small motor (think outboard motors and soforth). I like it because it is so darn easy to do, and then to undo if you don't like it. It does make a difference.

I can't find any fuel trims in your logs. STFT and CombinedFT. Gotta log those.

 

[Dryppical]

Alright, I added all those log things you guys said to and did a new log.

Dwd, I set the idle target to 750rpm and it still stays in the same range it did before. I don't know why it wont go down anymore. Yes, my wideband is in the front o2. I didn't notice I had a check engine light, thank you. Looks like I will have to find out why. It looks like its only charging to 13.6 at idle so its on its way out. I do not, its the stock 2g. Is there a setting incorrect in there? Is the Idle switch supposed to change with throttle? I'm new to the platform so I don't have a clue. It was probably like 20 degrees when I took that log and my fan stays on constantly. I don't know why its doing that so I'll have to figure that out but it does get hotter than that. I tried to do some research on the IntTemp and the GM MAF doesn't log that? I don't have an IAT sensor either.

We're on Boost, I changed those timing settings you showed me. I did a boost leak test and put it to 20 psi. I discovered I had a boost leak right before my MAF so I tightened my clamp and fixed that one. I sprayed all my couplers, j pipe, throttle body, BISS screw, injectors, and I couldn't find anymore.

I put 20psi in it to see how long it takes to drain. It took 40 seconds to go from 20-15psi and that's with my boost leak tester leaking a bit of air. Then like 3 minutes to get it to 5psi and I got bored and let the rest of the air out . I will probably do another one tomorrow to double check. After I fixed the boost leak pre MAF, it seemed to have gotten better AFR. I did not play with the global deadtime yet.

You may notice that my ISC position is 67. This is the BISS screw screwed all the way in. With it all the way in, the rpm doesn't stick anymore. It also doesn't lower the idle anymore than like 850 but it will raise it if I back it out. BISS screw is new by the way. I was originally trying to get it to the target idle of 750 and it didn't work.

Another strange thing, is my boost gauge reads like 14-15 vacuum before my boost leak test and tightening the coupler. After it was still 14-15. I read they are supposed to be around 18 or so at idle? I assume that indicates some sort of boost leak that I have failed to find. Hopefully these new logs will help.

 

[Dryppical]

Well I discovered I was incorrectly adjusting the BISS screw, I will do it properly by grounding the pins tomorrow and report back!

 

[SixBolt_16G]

Pins need to be grounded on a 2g? I thought that was only 1Gs, no?

 

[We're on Boost]

I tried to do some research on the IntTemp and the gm maf doesnt log that? I don't have an iat sensor either.

This is true. GM MAF doesn't have an output for intake air temp and you don't need an IAT sensor to use a GM MAF. There's no BARO coming from a GM MAF either. So in your logs you'll see dummy numbers for those 2 things.

In your 1st log in post #9, your MAF numbers look pretty realistic for 2g MAF Hz numbers. It looks like you must be using a MAF Translator, rather than the GM MAF CABLE that ECMtuning was selling for a while?

 

[jed344]

So it has been year's but as I recall when using GM MAF and if the cable is being used you have to be looking at MAF raw. I do not see that as a option in your list to look at. The normal MAF will show the same ish results but you then bring up MAF RAW and its wildly different and that is the hz you use not the normal MAF hz.

Same reason on a SD car you can watch MAF and it even moves yet there is not one. There must be a option to log it as I grabbed a log from a SD car I finished tonight and MAF RAW came up as a option to view.

Next thing. What wideband is in the car as I have never seen a preset for AEM log accurate. Being you are simulating this is very important as if its not logging accurate it will mess up the fuel trims too. Your battery voltage looks quite low. Air flow per rev is more or less correct, TPS volts close enough, idle SW is turning on. The fuel trims I won't trust until we know the wideband log is accurate.

ISC position is high. This normally means you need to open the BISS. This also I have seen when a ISC motor is sticky too and no amount of BISS adjustment will bring it correct.

 

[We're on Boost]

I checked in the ? Help in the MAF Comp tab in the application to read about how to choose Base MAF Type when you have a GM MAF.
It says this:

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

The OP has GM 3" selected for Base MAF type on his MAF Comp page. So that would be correct if he is using the GM MAF cable. But it would be wrong if he is using a MAF Translator. (according to the Help)

That's mostly why I'm asking my question up in post #12.
When I was using a GM MAF I didn't have ECMlink at all, so I don't know for real how this is supposed to go.

 

[Dryppical]

SixBolt_16G - I don’t know if they do or not, I was just reading and some people said to do that. It didn’t fix my issue anyways but was worth a shot.

We’re on Boost - I am not using a MAF Translator. I bought a GM MAF to stock MAF harness and it’s plugged in.

jed344 - I have a AEM wideband, I’ll look into not using a preset and set up a linear wideband function and calibrate it myself. I believe the alternator is dying. I’ll look into replacing it. The TPS reads .63v with the car off and when I turn it on it goes to .65v. I thought that was strange. With the BISS all the way in, the rpm’s will fall like they are supposed to. I ohmed the ISC and got 40 ohms all across. I’m going to remove it and turn the key and check the function.

I’m also going to remove the throttle body and clean it, look at the FIAV and check throttle plate position as I did remove it last year with replacing the throttle body shaft seals. I’ll report back after I go through my checklist. Side note: My car will not idle any less than 800 with BISS screw all the way in and target set at 750. I believe that means air is getting past the throttle plate someway or another.

Thank you guys for the help and suggestions, I really appreciate it

 

[Dryppical]

Ok, I did check out the throttle plate and I managed to put it in backwards when I replaced the seals and I had a ring of light around the throttle plate. I switched it around so the 50 is upside down and on the ISC side. This is the best I can get it to seal. Just a little bit of light on both sides by the shaft. Is this ok? I tested the ISC on the car and it moves in and out like it should. I don’t know what I’m supposed to be looking for on the FIAV though. I’m a little confused about that

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

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

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

 

[We're on Boost]

Ok, I did check out the throttle plate and I managed to put it in backwards when I replaced the seals and I had a ring of light around the throttle plate. I switched it around so the 50 is upside down and on the isc side.

Good find with the throttle plate, hope it makes a difference. Your 3 pics were all shot after switching it around, right?

I bought a GM MAF to stock MAF harness and it’s plugged in.

You did not say the words "ECMTuning GM MAF Cable".
A GM MAF to stock MAF harness could be just a GM MAF to stock MAF harness, with no electonics in it at all. I know they exist because in another thread there was a guy who showed one.

The best way to clarify what cable you have is to put a link to the web page you bought it from. Another way, if the web page doesn't exist anymore, is to shoot a screen shot of your order confirmation email that you probably would have gotten from the seller. That way we'd know who the seller was, and we'd know what they called the part. With the web page we'd probably also have a picture of the part.

 

[Dryppical]

We’re on Boost - Yes, they are all shot after I fixed it, I forgot to take before pictures.

This is a screenshot of the one I bought. I bought it from Rix Racing as well.

I did put the throttle body back on and I have a leak at the throttle body so I’m going to fix that and boost leak test when I’m done again and report back

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

 

[We're on Boost]

We’re on Boost - Yes, they are all shot after I fixed it, I forgot to take before pictures.

This is a screenshot of the one I bought. I bought it from Rix Racing as well.

I did put the throttle body back on and I have a leak at the throttle body so I’m going to fix that and boost leak test when I’m done again and report back

Oh good, thanks for that screenshot. That is for sure a genuine ECMTuning GM MAF Cable. It's a version 2 (the latest one) because it has the ground wire coming out separately with a little ring terminal on the end, which was supposed to be bolted down to the firewall or some chassis ground point. The version 1 cable went to sensor ground and that caused some trouble. I think it messed with other sensors that use sensor ground.

Well if you have a minute, take a look at posts # 14, 15, and 16 in this thread:
https://www.dsmtuners.com/threads/g...to-ecu-running-ecmlink.547218/#post-153923375

Guess what, I'm still looking for someone logging a car who is using a GM MAF with the GM MAF Cable rather than a MAF Translator. And that's you. You are the only one I know of!

So if you can find something in your Captured Values list that looks like MAFRAW or RAWMAF, turn it on in your future logs so we can find it in Available Values.

Here's how it looks in Available Values on a 1g with a normal 1g MAF. MAF, and right under that is MAFRAW. But it has to be getting logged in the first place before it will show up in Available Values.

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

In that other thread, jed344 said the raw MAF values on cars using the GM MAF Cable were about 10 times higher than the stock MAF values. But he apparently couldn't find his old logs of cars done that way.

 

[Dryppical]

Yeah, for sure! I’ll do a log tomorrow before I take the throttle body back off to fix the vacuum leak just so you can see the rawmaf value!

 

[Dryppical]

Here's the log with the raw maf value. Also, I noticed when I drove my car down the road that the idle issue with it sticking is magically gone and I haven't touched it since last time so I guess that's some good news

 

[We're on Boost]

Here's the log with the raw maf value. Also, I noticed when I drove my car down the road that the idle issue with it sticking is magically gone and I haven't touched it since last time so I guess that's some good news

Awesome, thanks!
Yeah the MAFRaw is about 12 times the MAF in this log. That's about what ECMtuning said to expect.

Looks like your idle speed will probably come down from what we see in this log, as the car warms up and the ISCPosition comes down.

 

[Dryppical]

Yeah, that log was from a cold start. I’ve been a little busy but thought I could get a quick log for the rawmaf values. Hopefully the throttle sticking issue is gone. I should be back to working on it Monday. I get to put in rear wheel bearing hubs as mine are not good, then I get to actually drive it without fear of a wheel falling off . After I fix my throttle body leak and check if there’s any others of course. Oh, and make my aem wideband a linear and calibrate it. I’m very excited!

 

[We're on Boost]

I get to put in rear wheel bearing hubs as mine are not good, then I get to actually drive it without fear of a wheel falling off .

Important!

Oh, and make my AEM wideband a linear and calibrate it.

I looked just now at your wideband scaling using the tables AEM has in the install instructions, and I see the scaling you have now is correct for the old AEM wideband but would be way off for the X-series (the newer model that first became available in 2016).

Is yours the X-series? Or is it the older one (they are still available new).

I displayed the raw value for AEMWBGauge which is volts on that input, then looked it up on the AEM charts for older and newer widebands.

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

AEM part # 30-4110 is the old one which they call the "Classic" now.
AEM part # 30-0300 is the newer one called the "X-Series".

 

[Dryppical]

I have the normal AEM wideband. Not the x series.

Anyways, I replaced my wheel bearings and took my car out for a cruise. The idle issue is completely gone. It fixed itself and I didn’t touch it. It was still there after taking off the throttle body and fixing the throttle plate, turned it off and next day went to mess with it and it fixed itself. I’ll take it .

I also fixed my fan always being on, I had to wire in the a-32 front harness plug and took the male end from my old neutral safety switch from my auto trans. Wire colors were not the same and I managed to wire the plug in backwards because I was holding it upside down . All good now

 

[We're on Boost]

I have the normal AEM wideband. Not the x series.

Anyways, I replaced my wheel bearings and took my car out for a cruise. The idle issue is completely gone. It fixed itself and I didn’t touch it. It was still there after taking off the throttle body and fixing the throttle plate, turned it off and next day went to mess with it and it fixed itself. I’ll take it .

I also fixed my fan always being on, I had to wire in the a-32 front harness plug and took the male end from my old neutral safety switch from my auto trans. Wire colors were not the same and I managed to wire the plug in backwards because I was holding it upside down . All good now

Sounds good! Glad the idle issue went away and hope it doesn't come back .

I looked at your last 3 logs again just to refresh memory about your wideband readings.
Actually they are in the 14's when the O2 voltage is cycling up and down, and 14's would be correct AFR there.

Also, the scaling you have for the wideband is correct for the AEM you have, which is not the X-series, it's the older model. If you do the Linear Wideband scaling yourself, and you do it according to the chart AEM gives you, you'll have the same scaling.

All I see wrong is the same thing I said in post #8 -- sometimes at 800-900 rpm idle the O2 quits cycling and then the AFR goes a little too lean. So I still think the easiest thing to try to make your idle AFR's better is this:

One way to bump up your idle richness is by adding some Global Deadtime on the Fuel tab. You can use that kind of like the idle mixture screw on an old carburetor small motor (think outboard motors and so forth). I like it because it is so darn easy to do, and then to undo if you don't like it. It does make a difference.

We can tell more about your AFR reasonableness when you get into the throttle more, getting the load factor up there a little more. But right now, in the last 3 logs I don't see any load factor higher than 0.85 and at that load you are still in closed loop, the O2 is cycling, and the AFRs look reasonable, near stoic like they should be.

So I think just try the Global Deadtime idea for now. Give it maybe 800 microseconds.