The Central Hub for DSM Community and Information

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. This is where the DSM platform history is documented and archived. Log in to help us in our mission, and to remove most ads from the browsing experience.

failed MDP sensor / gm 3 bar help

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

2point3Tsi

15+ Year Contributor
225
5
Jan 12, 2007
Maple Valley, Washington
was on a a short drive after setting it up yesterday ( just upgraded ) it was doing had a few degrees of knock I was working out by pulling timing then bam CEL came on and knock started to hit 4 degrees.

At that time I wasn't currently logging MDP, after checking the dtc's which were p1400 and p0300 I cleared it and basically putted the car home at less then 2000rpms.

I currently possess a GM 3 bar pn 12223861 that I've wanted to wire in to log but keep getting confused whether I'll need a buffer circuit like they talk about or if I can just wire it into the existing MDP sensor wiring.

this is my daily driver so any chance of fixing it tonight will keep the gf off my back LOL

-Chris
 
just to be clear I logged MDP on link and it was flatlined at -0.1hg

when I first started logging it showed -13.1hg even though my vacuum gauge showed I was idling at -21hg

would this be responsible for my 4+ degrees of knock
 

Attachments

You must be registered for see attachments list
Do a boost leak test. Make sure that you don't have any leaks. IIRC right click on gm3 bar, go to properties and make sure its set for your engine size and altitude.
 
80 370 590
GM sensor is installed and running proper

still random misfire code P0300

car throws CEL, wideband hits lean or ----- and I lose all power until I pull over turn it off for a couple seconds, clear dtc's then start it again.

also extremely erratic knock patterns between 1-6 degrees :ohdamn:

check seconds 80, 370 and 590

I know about my dead front O2 just got V3 about to simulate it with my aem w/b but got side tracked by mdp and misfire codes

any help would be greatly appreciated
 

Attachments

You must be registered for see attachments list
Sorry, I'm at work right now so I don't can't view your log, but since you have a 6bolt, you should have DSMlink setup so the ECU doesn't look for the random misfire. I assume you are running a 6bolt head with the green or black lid CAS?

As I have mine setup right now(6bolt long block), I have an OmniBar 3bar map sensor wired directly to where my MDP sensor used to be, so I have DSMLink useing the MDP input logging my MAP sensor. I also have dsmlink ignore cylinder misfires.

this is my daily driver so any chance of fixing it tonight will keep the gf off my back LOL


FWIW-I had a gf like that once upon a time... One of the reasons I broke up with her.
 
7 bolt head 6 bolt block green top cas

why would you ignore cylinder misfires? and does that ignore the random misfires too not just the cylinder specific?

after setting up my mdp to use the gm 3 bar everything logs fine, although my mdp does still vary its reading even though there's no wires hooked up, do I just set it to ignore factory mdp signal and it will still log the gm sensor right?
 
Last edited:
7 bolt head 6 bolt block green top cas

Green lid CAS is the worst CAS for throwing random misfire codes. I had one and I got the code ALL THE TIME. I changed it to a black lid CAS and I got the code maybe 2 times a day. However, with DSMLink, you should have the software ignore the code all together, so you shouldn't be getting the code at all.

From ECMLink Tuning. The "cylinder misfire" box should be unchecked.

You ignore the misfire all together because the 1g style CAS is not quite as accurate as the 96+ crank and cam signal, so you will randomly get misfire codes.

EDIT: I'm not quite sure what you mean about the MDP sensor? Now that you have a GM 3bar, you should have cut the wires to the MDP sensor and used those wires to wire in the GM 3bar. In Link you use the MDP sensor input to log the GM3bar.

See in this screen shot They have the AEM 3.5-bar utilizing the MDP sensor input. That is exactly how you should have yours with the exception that you would use the GM 3-bar sensor selected instead of the AEM one. I hope I understood your question.

EDIT2: Yes it ignores all misfires.
 
then why when it throws the code does my wideband hit lean?

it happened every time but 1 out of 9 between 1 hour worth of driving (took me a lil longer since I pulled over every time it hit lean, then I restarted)

I still have both the mdp and gm 3bar logged.

the gm is ran through the factory mdp wiring and reads almost dead on to my gauge. I don't know why the factory mdp still is logging values that fluctuate but I vaguely recall a check box somewhere that tells it to ignore it

also I do have another cas but its the 1990 one with the pigtail harness so a few minutes would be required to adapt it =)
 
Last edited:
then why when it throws the code does my wideband hit lean?

it happened every time but 1 out of 9 between 1 hour worth of driving (took me a lil longer since I pulled over every time it hit lean, then I restarted)

I still have both the mdp and gm 3bar logged.

the gm is ran through the factory mdp wiring and reads almost dead on to my gauge. I don't know why the factory mdp still is logging values that fluctuate but I vaguely recall a check box somewhere that tells it to ignore it

also I do have another cas but its the 1990 one with the pigtail harness so a few minutes would be required to adapt it =)

I'm not really sure why it goes lean. My narrow band gauge did that back before I had a wideband. I think it goes into a limp like mode and kills the injectors, hence the lean condition. Sort of like when you decelerate while still in gear, your WB goes lean.

EDIT: Yes, there is a check box for the EVAP/EGR/MDP. But that just ignores the DTC code, it doesn't ignore the input. You should be utilizing the MDP input for the GM3bar. Log the 3bar and never look at the MDP

EDIT: There is a check box to ignore a CEL for teh EVAP/EGR/MDP. But that doesn't ignore the signal. You should be using the MDP input to log the GM3bar.
I never had a wideband when I had the problem, just a narrow band, but the narrow band gauge did the same thing. I believe it shuts off the injectors, hence the lean reading. It's sort of like a limp mode.

What input on the ECU are you using to log your GM3bar?

Your Greenlid will be just fine, just uncheck that box and your problems will go away. The 90 style uses the same signal as the 91,92, & i think 93 greenlid sensors. I believe the 94 is when the black lid came out, which uses a signal much more similar to the 2g style CAS. But either way, unless you have that misfire unchecked, you will still get a misfire code, regardless of what 1g style sensor you have.

Sorry if some of this doesn't make sence, my everytime I edit my posts, they get all scrambled and f-ed up and deletes some of my content...
 
Last edited:
No.
In the "DTCs" tab on the left, you should uncheck the "cylinder misfire" box
In the "Misc" tab on the left, you want to keep the "Use Non-95/98 style cam angle sensor" box checked

You should be ready to rock then.:thumb:
 
random misses gone, I'm pretty sure I forgot to check the box that saves to ecu as you make changes.

I still have some knock when I actually get on it.

gonna post the log later I guess Mcdonalds wifi won't let me attach the log :bang head:

some inclement weather and bald tires delayed a new log til today, things are running much better still have more knock registered when not in boost mainly just trying to figure oiut if its real or not.

will have wideband logged by wednesday hopefully.

mainly cruising with a short WOT test near the end.
 

Attachments

You must be registered for see attachments list
I don't think the knock is real. It may be picking up some noisy lifters, or an exhaust leak. My knock sensor will occasionally pick up some noises from my exhaust system as it rattles at certain engine frequencies.

Your short pull shows virtually no knock, except for the .4* but that's nothing really to concern yourself about.

Whats the condition of your knock sensor? If it's oozing out the back, it's time for a new one. It also might be loose, or even over tightened. I believe the proper torque is 16 lb ft.

EDIT: I found the torque spec and 16 lb ft is correct (23nm)
 

Attachments

You must be registered for see attachments list
Last edited:
haven't torqued it since the motor was built, no black stuff, currently fighting ecmlink connection issues after wiring in my wideband :banghead:

Lol, I've been in your shoes. Make sure you plugged everything back in. I couldn't get Link to connect to save my life. I went around my entire car checking grounds, and voltage, and in the end, I had a plug disconnected. Lol, I felt like a jack ass.
 
Add Value - Be Respectful - No Trolling - No Misinformation - Participate Often!
Support Vendors who Support the DSM Community

Latest Classifieds

  • For sale Garage clean out
    Changing setups on the car and getting rid of some stuff as well that's been laying around. Will...
    • 92GSXtacy
    • Updated:
    • Expires
  • For sale 4G63 Griffin intercooler cores
    Griffin intercooler cores. Top to bottom flow. High cfm and heat transfer. 24x8x2.75 and...
    • Galant665
    • Updated:
  • Wanted wtb black 2g dashboard
    Looking to buy a 2g black dashboard. Located in southern california but willing to travel.
    • randizzle420
    • Updated:
    • Expires
  • Wanted 1g Wtb: uncut 6 bolt crank
    6 bolt uncut crank
    • erikoberdorfer
    • Updated:
    • Expires
  • For sale 1g Used 1991 Turbo Engine Control Harness, Auto, with ABS
    Used engine control harness for a 1G turbo, automatic with ABS vehicle
    • 19Eclipse90
    • Updated:
    • Expires
Back
Top