1G 1992 N/T Knock question and log

[AskaFa]

Hi, I've been having problems with my DSM lately.

About a month ago my car started acting weirdly around idle. The idle ever since I bought the car has been over/or 1000 rpms and I decided to fix that. I first deleted the FIAV (the free way method), cleaned the ISC and tested it. Idle didn't kick all the way to 1700rpms on a cold start after this fix, but it still didn't idle correctly at 750rpms.

I adjusted the BISS screw using the manual (grounding pin 10 and the plug behind the battery). But even with BISS fully screwed in, my idle was still around 900-1000rpms.

I then disassembled the whole throttle body and intake, resealed everything with new seals, O-rings and verified that there are no vacuum leaks from the intake.
After this I was able to get my car to "idle" at the designed 750rpm, but it was horrible. The car was shaking heavily and the vibrations didn't stop until i adjusted the idle back to 900.

I've been driving with my idle set to 900 for past few days and the car feels alright during acceleration, but when decelerating or cruising in gear at lower speed the car "twitches" forward and back for a few moments (I don't know how to describe it better than if you repeatedly go WOT and close the throttle immediately).

I've also checked my logs and noticed there is certain amount of Knock (sometimes even 15) which surprises me, because the ignition timing is set correctly to 5 degrees BTDC (I set this by getting the engine to TDC and reseating the Crankshaft sensor the way it is described in the DSM manual).

I've also changed my spark plugs to Iridium ones in hopes to fix this. I've used BPR7E NGKs gapped to 1mm (they don't sell BPR6E here).

The injectors have been cleaned and resealed with new O-rings, fuel pump replaced, fuel tank also cleaned, fuel filter replaced, spark plug wires replaced (NGK 8741), ignition coils replaced, the ECU capacitors changed (old ones were not leaking at all), EGR sealed and unsealed (made no difference), I even ran the car with ISC and FIAV deleted completely from the throttle body using only the BISS screw to hold idle, yet the car still idles poorly at 750 (when deleting the ISC the car also loved to stall randomly when put into neutral).

The last thing I can think of is the O2 sensor or the knock sensor somehow doing something, otherwise I'm pretty much in a blind corner right now and don't know where to continue without pulling the whole engine out.

I think that before my ownership no one really did any things to this car other than changing the oil and filters. The car was still on it's old blown out shocks and the whole engine bay even with seals, bushings etc.. looked OEM.

If anyone has any ideas, please tell me. Log is here (I used TMO) https://drive.google.com/drive/folders/1itsq9KGZ2CFFYUlCfLEkWJ8QzOYy7sq6?usp=sharing

And sorry for bad grammar, not my native language, thanks.

 

[dwb]

The black ISC is PN MD628051. Mitsu, Hyundai, and Dodge used them in the 90s.
Before you attempt to track one down, you should open up your ECU and check for blown driver chips (https://www.ecmtuning.com/wiki/driver01) and capacitors for corrosion (https://www.ecmtuning.com/wiki/ecurepairhome). Both can cause issues with the idle circuit. Feel free to post a picture of the board here if you're unsure.
As far as repairing an ISC, maybe it's possible if you have a doner. I've cracked a few open, and while I'll attempt to repair almost anything, including my own ECU, I didn't feel confident I could repair an ISC.

 

[steve]

These pages have been around almost as long as I have. http://www.stealth316.com/2-isc-iac.htm The 3000GT ISC is the same as the DSM version. Don't use any of the links to repair companies.

 

[AskaFa]

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.

Hi
I’m sending the pics of the ECU
I’ve been here before doing the capacitor job and socketing the EEPROM (I’m looking to buy a DSMlink in future if possible).
I don’t see any problems on the PCB and nothing is shorted on the underside.
I’ve rechecked all the contacts of the EEPROM socket and the capacitors after soldering and they all have a good contact.
I’ve also ordered a new black-body style ISC motor, luckily Amazon ships here, which saves my ass big time.

(Also on the pictures it may seem that the previous capacitors leaked, but they didn’t. I’ve scrubbed away a layer of protective film with rubbing alcohol and microfiber cloth because it was hard to solder with it on the joints)

 

[AskaFa]

Oh and the number on the EEPROM is E935. I've taped it over just to be double sure when I had it apart.

 

[steve]

I don’t see any problems on the PCB and nothing is shorted on the underside.
I’ve rechecked all the contacts of the EPROM socket and the capacitors after soldering and they all have a good contact.
I’ve also ordered a new black-body style ISC motor, luckily Amazon ships here, which saves my ass big time.

(Also on the pictures it may seem that the previous capacitors leaked, but they didn’t. I’ve scrubbed away a layer of protective film with rubbing alcohol and microfiber cloth because it was hard to solder with it on the joints)

From what I see it's clear that the previous capacitors leaked. I don't see damage offhand that I would attribute your knock problems to but the corrosion damage around the SMD capacitor C19 is to the voltage reference for the ADC and that can cause all sorts of issues. The pins 45 and 44 are to the ISC driver IC's. Any place where the copper is brown or black is corroded, the SMD capacitor C14 is another good example

 

[dwb]

Your driver chips look OK to me but it looks like you have or had a leaking cap. I don't see any immediate damage but I question what's going on around/under c106, as that is a pretty common spot for a broken trace on the idle circuit. Here's an example of how many traces I had to fix on my ECU to fix the idle circuit after leaking caps: https://www.dsmtuners.com/threads/resolved-1g-ecu-no-idle-control-board-repair.542567/post-153883477 (some more pics in that thread too of how I fixed some of the traces).

Since you already ordered a new ISC, might as well see if that fixes your issue first.

 

[AskaFa]

Hi again.
I've cleaned up the ECU board a little more and went to check on the SMD caps.

The C19 measures at 12.9mF, C20 at 161.4μF, C14 at 12.92mF, and C44 at 9.45nF

I've also checked for continuity of their traces. They all have good contact.
However, I've also measured C40 which from looking at it doesn't look black on one side like the others and nothing showed up. Multimeter just switches between μF and mF without ever showing anything.

My question now is, should I buy new SMD caps, and which? And if someone has a datasheet for a 1G ECU for all the electric components on it.
Thanks

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.

 

[AskaFa]

Update: I found a bad trace between R46 and C44

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

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

 

[AskaFa]

After repairing the broken trace the ISC motor has started working again.
The car still stalls on neutral, but has since started acting more like an Idle Surge with RPMs dropping to low 500s and climbing to 1000 a few times before the engine finally stalls.

 

[AskaFa]

Update:
Maybe a little too early to judge, but it has stopped stalling, at least for now.
The Fix? I reverted the only thing I didn't; the cut up air filter I did while following the Vfaq. I took the cut off bits and just quickly fixed them back to the filter housing with a duct tape.

Took it for a spin to the city - lots of stops, made many circles back and forth, I was basically stopping on an intersection or a roundabout every 2 minutes. After about 45mins I finally parked at the mall. No stalls whatsoever or any Idle Surge.
I'm perhaps yet uneducated on how the airflow is designed for the MAS on an N/T car, but it seems that too much access to air might not be good for an untuned ECU with stock MAS.

I'll try and drive the car as most as I can now, to see if this really did fix the stalling or not. I'll report back in a few weeks.
Thanks everyone for helping guys, great community

 

[steve]

The Fix? I reverted the only thing I didn't; the cut up air filter I did while following the Vfaq. I took the cut off bits and just quickly fixed them back to the filter housing with a duct tape.

I'm perhaps yet uneducated on how the airflow is designed for the MAS on an N/T car, but it seems that too much access to air might not be good for an untuned ECU with stock MAS.

Which VFAQ, there's modifing the air can and then there is hacking the MAF, which is a really bad idea today.

 

[AskaFa]

Which VFAQ, there's modifing the air can and then there is hacking the MAF, which is a really bad idea today.

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

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

"Modding, NT" is what I've done and since there are no pictures I used "Modding" for 1G Turbo as reference. I've also removed the resonator and the tube with no exit under the aircan.
There were no mods I've done to the MAS or any electric components around. One odd thing is that the guide says you can remove a silencer which should be a part of the MAS if I understand right. I didn't have any silencer in the MAS or anything that would resemble it so I ignored that.
I'm also posting a picture of how the engine bay looked after the modification. I've since changed a few things (went for a Hyundai valve cover since the Mitsubishi one was cracked, and converted to an Oil catch tank)

 

[AskaFa]

Well, after more than a month I can surely say: The car was indeed stalling because of the open air intake. I've welded the cut up airbox and the car has not stalled a single time in over a month. I don't know if this is a sign of a dying MAF sensor or if too much air is going through the unmonitored holes in the MAF. I'm at least fully happy that the stalling has stopped completely.

A new ISC and a new BISS came through, and I fitted them ASAP.
For the ISC it made little to no difference, the old one was working, so is the new one. The BISS change however lower the idle RPMs a little bit, so I definitely had a vacuum leak around it. I hope the new BISS will last long, because for whatever reason (manufacturer cheaping out ig), it is made of plastic. I couldn't get the OEM screw where I live, so I had to order from China.

I've also bought a new knock sensor, however the ECU throws a code 31 as soon as I connect it. It looks the same as the old one and has the same connector. Maybe it just doesn't work (It was ordered from China too).

The car has been running fine, the idle is sometimes still a bit undecisive on what it chooses to be, sometimes it idles at around 900, sometimes 1000 and sometimes even closer to 750. It never stalled yet, so I'm thinking the problem may not be so critical at this moment.

I also wanted to ask if anyone has experience using a DSMLink on an NT car and if it's even compatible? I'd like to have more diagnostic power in my hands rather than the OBD1 port with only a few codes.

Thanks guys.