2G Not passing smog because of monitors

[toojung2die]

1997 Eclipse Spyder GST a/t

OBD-II indicates I have three monitors stuck in a not ready condition. Nevada smog will allow up to 2 monitors not ready. Not ready are Catalyst, EGR system, Oxygen sensor heater.

There are no problem codes logged and no CEL. The car passes this part of the test. My friend runs the smog test station and he ran some diagnostics for me. Both O2 sensors are feeding back and the pre-cat sensor is in closed loop during idle. He reset the ECU and I drove almost a full tank of gas before bringing it back. The systems are still reading not ready and there is no CEL. The CEL turns on for about 5-10 seconds during start-up and goes out normally.

I replaced both O2 sensors and ran another half tank of gas, multiple trips. The sensors are still not ready with no errors logged and no CEL.

Mods: 3" catback exhaust, Injen style intake w/filter, "knock" LED on boost solenoid. All hoses, sensors and solenoids connected. Basically stock. Car runs great. What can I do before I have to take my baby to visit Satan.

 

[akdsmer]

So you diagnosed the failing component in this instance to be the starter and nothing else was required to eliminate the excessive voltage drop?

Yep, the starter was the only thing bad. I actually put my 1/2 rachet on the crank and it spun over easily enough...and not at TDC either.

 

[fwdeclipse]

I am now having this same problem. I have a 95 eprom with a dsm chip and all five of mine say not ready and it says the mil is on buts its not. I am going to try my stock 97 ecu first then I guess I have to pay the dealer to pass it.The other thing that pisses me off is to change the ecu I have to put my 450s back in and this just freakin sucks.

 

[Atuca]

hey guys, wanted to share some weird information about my experience smogging my car.

I recently unplugged my battary for a stereo installation, therefor resetting my ecu. I than needed to pass smog for registration, so i fugre drive around for 20-100 (ok so that is a big gap) but i needed to get the monitors reset for my smog test. Well, that didn't work after 150 miles. So i looked up resetting the monitors on my car.

I found some links to the driving prucedurs for resetting, did them for about 2-3 hours, and still, the monitors did not set.

Well, i was taking my car to this noname shop in fremont, part of a gas station, and for me, had to be a test only station. They had a policy of no pass, don't pay, so i figure before i give into Satan to have them rest my monitors manually with a computer, i would make sure i could pass the emissions test.

The guy there was pretty cool, and he ended up testing my whole car knowing that the monitors weren't set (he checked the for me)

Well my car did pass the emissions, but he went ahead and tested everything, gas cap, looked at my engine and stuff, was doing a throuogh job. Well, after the test, he clicked next next blah blah assuming i was going to fail. We were talking and in the back ground i see it printing out, and guess what?

I passed!

We were both at a loss of words, we both knew the monitors were not up, he didn't do anything illegal or anything. He said he wasn't going to question it, i paid and i was on my way.

I don't know if this means anything to you guys, or maybe his computer was broke, but with monitors not set, not even one, 5 minutes before the test, and even after the test they were not set, i still passed smog. He told me mitusibhi as well as nissians and some volvos would require going to the dealership to be reset, but he suggested maybe the goverment had by-passed the requirement for the monitors on Eclipses.

I don't know what happened, or how to explain it, but, now you guys know, and if there is anyone else out there who had something similar happen to them, please post here and aloow other people to know that this fluke may be the new way we smog our eclipses

 

[spyderturbo007]

Yep, I had the same problem. Apparently there is a bug in some of the 2g ECU causing the monitors to reset to the "Not ready" state every time the car is turned off. The dealership has no idea why it happens to one car off the line and not the next. They also refuse to do anything about it, so everyone with this problem has to go to the dealer so they can fudge the test.

They hook the ECU up to a MUT-II and force the monitors to set, then run the test. Technically in PA, it's illegal because you aren't supposed to start the car until part way through the test, but I guess they don't care. Normally they charge you about $150 and send you on your way. It's a bunch of crap if you ask me, but I don't have to go through it anymore since I don't drive my car more than 5000 miles every year.

 

[Atuca]

well the point of this post was mine didn't reset, and i still passed.

that doesn't make sense to me, that is what i was questioning.

in my experience, no non contiguous monitors, is a o k.

i was just seeing if anyone else had the same experience.

 

[wolvorine4424]

Sorry to bring this back...

Have a question about this, I just found this info and desided to ask a question...

I have a 97 Eclipse w/ 420a Non-Turbo FWD. Do I have this problem or not?
If I do, I want to BUY an ECU now before I get screwd... (build date: 6/97)

What ECU do you all recommend me to get. I want an Updated ECU that has no problems (If posible)

I Just got this car and I will keep it. So I better not have any problems (Or the deler will LOVE me, In a bad way. Me have way to many friends in high places)


If anyone needs more info about the car/ecu/etc., Please let me know.


Thank you in advance,
Roman

 

[Bigglesworth87]

From my understanding it was just the turbo models that had the problem.

 

[toojung2die]

You see that the voltage drops well under the 9.5V that can reset the PCM. For many vehicles this isn't a problem, but Mitsubishis have always had capacitor problems and I believe that this problem could be resolved by appropriate sized (or maybe just good) capacitors in the PCM to hold it over during this very short period.

I checked the cold start voltage using my Fluke 87's min/max hold for a few days. The voltage at the battery dropped to 9.88 volts consistently. This is above the 9.5 volt threshold, but as you mentioned, the Fluke may not be fast enough to capture a drop that happens for less than 100ms.

 

[toojung2die]

I started this thread 3 years ago when I first discovered my ECU wasn't setting the OBDII readiness monitors required to pass the then new OBDII smog test. Since then I paid the dealer insane money each year to set the monitors and give me a passing smog certificate. Today I got my first smog pass certificate the same way everyone else gets it; at a smog testing station.

My solution was to install DSMLink and a '95 ECU. First I had to turn on the DTC checks that DSMLink turns off by default. After a couple of short drives on the highway and local streets all the readiness monitors were completed. DSMLink also fixes the false "MIL is ON" bug inherent to '95 ECU's. I could've bought DSMLink three years ago with the money I was going to spend at the dealer.

Big thanks to akdsmer for his invaluable contributions to this discussion. If you have this OBDII smog problem and are stuck using the original ECU your car came with his suggestions will probably help solve your problem.

 

[akdsmer]

I'm glad this has helped out. These vehicles are a torn in my side I/M-wise. I wish EPA had the resources to actually go out and verify this concern and get the manufacturer to correct this weakness. I've given EPA what I know and the data from my state to prove it, but they have bigger things to deal with. Until then....

 

[toojung2die]

There are several possible solutions.

1. Take it to the dealer each year and pay $118 to $236 to have them set the monitors to ready using the MUT-II diagnostic tool to obtain a passing smog certificate. Trouble with this solution is when diagnostic mode is exited and the ignition is turned off the monitors reset to not-ready again.

2. Buy a 1999 ECU. Some 1997 DSM owners on this board have said their readiness monitor problem was solved by installing an ECU out of a '99 DSM.

3. Buy a 1995 EPROM ECU and add DSMLink. The '95 ECU will set the monitors as required to pass emission testing but has a "MIL Bug" that indicates a CEL condition even though the CEL is not lit. DSMLink fixes the "MIL Bug" and has lots of other nifty features.

4. Take your car and the failing smog test to the state's emission control lab and point out that your car's model year has a known problem with setting the readiness monitors. If they're real nice they will give you a waiver.

5. Buy a 1995 EPROM ECU and add a KEYDIVER Chip. They finally solved the "MIL Bug" problem. Quoted from the KEYDIVER website http://www.dsmchips.com/2g.html

"MIL Bug fix. The 1995 code, which you will now be using, has a bug which tells the OBDII scanner that the MIL light is ON, even when it is OFF. Thanks to Thomas Dorris this is FIXED!"

If DSMLink is beyond your budget then this is the next best solution.

 

[TunaTalon]

The undead thread is back again. My Talon failed the new OBD-2 emissions test in Missouri.

The wooden stake for the readiness flag problem of our DSM’s may be in the EPA standards. What better source?

The EPA’s guide for State agencies references document EPA420-R-01-015 Performing Onboard Diagnostic System Checks as Part of a Vehicle Inspection and Maintenance Program. (http://www.epa.gov/otaq/regs/im/obd/r01015.pdf)

Appendix D of that document recommends checking for MIL command only for 1996- 1998 Mitsubishi cars.

“APPENDIX D

Manufacturers Known to Have OBD Readiness Issues

1996 - 1998 Mitsubishi vehicles - These vehicles may have a high degree of “Not Ready” for catalyst monitor due to a “trip based” design. Mitsubishi has provided driving cycles in its service information to allow monitors to operate. These vehicles should be scanned for MIL illumination without regard to readiness status. “

Emphasis mine.

Individual states testing programs should be in compliance with the EPA document. For example the Missouri bid for emissions Acceptance Test Procedures RFP B2Z07028 reference http://www.oa.mo.gov/purch/bids/b2z07028.pdf required compliance with EPA regulations as follows:

“2.1.2 The ATP shall be a series of verification tests on all emissions inspection equipment hardware and software to show compliance with the contract between the State of Missouri and the GVIP contractor. The ATP shall also verify the entire Vehicle Inspection Database (VID). The ATP contractor must certify and demonstrate to the state agencies’ satisfaction that the entire system is fully functional and that it is operating in full compliance with all current state and Environmental Protection Agency (EPA) data collection and reporting/analysis requirements. All inspection hardware and software shall comply with the final ATP.”

Emphasis mine.


Now if only we can find some reasonable people in each State to recognize that their rules are deficient, not the DSM, then we can get back to tuning. (Hey it could happen.)

Also please see the TSB at http://autocenter.weber.edu/OBD-CH/documents/newdocuments/Mitsubishi/tsb0313005.pdf for details of the required drive cycle and requirements for temperature, tools, and type of road recommended to set the flags.

According to the TSB the flags can be set by two highly trained technicians and a dealer proprietary computer in only a few hours on a flat test track on a lovely spring day. Don't know how long it would take a normal driver in the winter in the hilly hinterlands. This is obviously not something that can be reasonably expected from the vehicle owner.

 

[akdsmer]

The dealer proprietary scan tool puts the computer in "enhanced sensitivity". It doesn't take two people and only takes a short while. I've done this on a few cars using their scan tool. The problem lies in the fact that once you remove the factory scan tool and cycle the key (a component of the OBDII test) the monitors all clear once again. Unless the dealer is doing the test it won't be set when they are done (or the key is cycled). The standard drive-cycle is best done with the dealer tool also as it will show you as each monitor is set (done), which cannot be done with other scan tools. The factory programming only shows incomplete until all monitors are set on ONE drivecycle. It then needs to be shut-down, restarted and the drivecycle ran again. On the second one, the monitors will all start to show complete as they are set. You need a perfect drivecycle to get these to run. My .02 is to make sure the idle switch is correctly set and that the battery is good. If I am questioning the battery (as I've stated before) I use a jump-box at each start-up to ensure that the PCM doesn't reset.

The 2007 updated Appendix D keeps these vehicles on the list of recommended to be scanned for MIL illumination only (EPA420-F-07-060). Alaska is in the process of using the EPA RECOMMENDED testing procedure, with an added full visual/functional inspection to verify no modifications, but it won't be in effect for a while. Another note: the Alaska I/M testing programs look to be being eliminated by Dec 09.

 

[BoostFrenzy]

My car had the same problem dealing with the smog check. So since they wouldn't set to ready, I swapped a '99 GSX ECU in and then drove it around for a bit. Everything passed fine and (for now) I have a '99 ECU in my car.

Anyone have any info on what they changed to make it work better form the '98 ECU to the '99 ECU?

 

[TunaTalon]

The dealer proprietary scan tool puts the computer in "enhanced sensitivity". It doesn't take two people and only takes a short while. I've done this on a few cars using their scan tool. The problem lies in the fact that once you remove the factory scan tool and cycle the key (a component of the OBDII test) the monitors all clear once again. Unless the dealer is doing the test it won't be set when they are done (or the key is cycled). The standard drive-cycle is best done with the dealer tool also as it will show you as each monitor is set (done), which cannot be done with other scan tools. The factory programming only shows incomplete until all monitors are set on ONE drivecycle. It then needs to be shut-down, restarted and the drivecycle ran again. On the second one, the monitors will all start to show complete as they are set. You need a perfect drivecycle to get these to run. My .02 is to make sure the idle switch is correctly set and that the battery is good. If I am questioning the battery (as I've stated before) I use a jump-box at each start-up to ensure that the PCM doesn't reset.

The 2007 updated Appendix D keeps these vehicles on the list of recommended to be scanned for MIL illumination only (EPA420-F-07-060). Alaska is in the process of using the EPA RECOMMENDED testing procedure, with an added full visual/functional inspection to verify no modifications, but it won't be in effect for a while. Another note: the Alaska I/M testing programs look to be being eliminated by Dec 09.

Thanks akdsmer. That document will help the cause. Google did not find the document number. Could you please include a link to it on this thread?

My comment about taking two people to get the flags set is related to the description of the drive cycle on the TSB. I’m sure that as you describe the procedure it’s possible for a DSM Wiseman to get and keep the flags set. That leaves the rest of us novices out in the cold paying the dealer to fake a fix.


BoostFrenzy,
Mitsubishi changed the code in the ECU for the 1999 model year to relax the drive cycle requirements to set the flags. See the TSB referenced above

 

[jrprich]

Sorry if this is incorrectly posted. I DID search for TSB 18-005-01

My 96 Talon Tsi AWD Turbo will not pass Oregon DEQ tests due to "Not Ready" messages in the PCM. DEQ tells me Talons are not on their exception list but they do know that they are really Mits Eclipse which is on the exception list

They tell me: "The 1996 Eagle Talon Is not included in the vehicles subject to the alternate gas idle test when unreadies are the reason for vehicle failure. The readiness issue with this vehicle is due to an improper design that usually follows a “key-off” event. Chrysler Recall Event #678, addressed by TSB 18-005-01 contains the information about this situation, and also stipulates that the service will be performed at no cost, regardless of warranty expiration."

Does anyone have any info on this TSB ? I called the local 5 Star Chrysler dealer and the Service Manager says there is no such TSB............

 

[STM tuned]

Found some information on a thread where others were having this issue on various 96 model year chryslers

Seems like what you said, when you turn the key to turn off the car, it erases some readiness monitors everytime, which is why your not passing.

This problem occured when the ignition key was turned off which was erasing the internal memory pertaining to Oxygen Status, EGR Status, Oxygen Heater Status. This did not erase Misfire Status, Fuel Status and Comprehensive Status!!!

There was a recall campaign on this problem #678. Describbed in TSB 18-005-01.

According to what was posted, they found out that the dealership SHOULD cover it supposedly

later in the thread

Well here it is. After reading the above information you provided, I looked into this with the Dodge dealership in town. They covered the PCM flash update (no charge) under a lifetime warranty for this model. I drove the car about 35 miles or so and performed some drive cycle requirements. The car passed the test! The PCM readiness monitors will now hold after turning the engine off.

You should call another dealer and see if you can find out any information.

Can view it here OBD II internal system check question in 1996 3.5 - Page 2 - DodgeIntrepid.Net Forums - Dodge Intrepid, Concorde, 300m and Eagle Vision chat


I'll let you know if I find anything else out.

 

[jrprich]

Thanks,

This is pretty much what I have found as well.......problem is the Dealers seem to claim this TS doesn't exist.
I was hoping someone had access to the actual TSB that I could show to the stupid Service Managers. WTF
Pretty sure thay have it in ALLDATAdiy.com but I don't have a membership.

Thanks again,

Jim

 

[STM tuned]

Check with another dealer and see if they can find out anymore information. Some dealers dick people around when it comes to recall work. I had one dealer tell me they weren't going to do my transfercase recall and that it wasn't a lifetime recall...

 

[TunaTalon]

Waivers are possible for OBD II Emission tests.

The state of Missouri has issued me a waiver for my Talon for readiness flags not set.

The research has been summarized in a Wiki OBDII Emissions Testing and the DSM - Dsmwiki which includes a sample e-mail requesting the waiver.

This applies to 32 states which are listed in the Wiki.

 

[jrprich]

THANKS SO MUCH, LOTS OF GREAT INFO HERE

BUT, there is a Catch 22 in my state, Oregon. The do recognise the Mits as needing the back up test BUT they DO NOT list the Eagle
All attempts to get them to have so far failed. I showed the DEQ Tech the Engine wich is clearly marked as a Mits.......no luch as their Software reads my VIN as a Eagle Talon and it has no exceptions for the Talons in the software so they will not test it.

I need to find the Chrysler version of the Mits TSB........so far lots of references but I cannot find the ACTUAL TSB to see if it mentions Talons

Chrysler Recall Event #678, addressed by TSB 18-005-01

I will keep trying,

Thanks again,

Jim

 

[TunaTalon]

Only a bureaucrat could think that the Eclipse and Talon have different ECU's.
My contact in Missouri was an engineer. He understood and acted professionally.

If you can't find an engineer in the Oregon chain of command, maybe show that the Eclipse and Talon have the same OEM part numbers for the ECU.

I did not find the TSB with the Eagle listed in my research, but only because after finding the Eclipse TSB I thought that was good enough. After this post I will search again.

 

[jrprich]

That about sums the situation up........the "tech" at the testing station did know they were both Mits, but since the DEQ software is based on my VIN and my VIN returns a Eagle, and the Eagle is NOT on the exeption list they can't get their software to report the car as tested and passed..........Catch 22. He did call the head office at DEQ, but as luck would have it seems all the senior managers were in a "training session". So I am to show up again Monday......the 4th time, to see if they can resolve this.

Thanks for all your time, would be nice if Mits AND Chrysler would do something to resolve this. I did send a long and detailed email to Chrysler. But since the car is 12 years old........I don't have much hope.

 

[PattyG]

I'm looking on motoralldata but I'm not finding a TSB matching those numbers.

I found it for the Eclipse of the same year but I can't find it for the Talon.

 

[TunaTalon]

California knows about the problem, see http://www.smogcheck.ca.gov/ftp/pdfdocs/APPENDIX_J.pdf and search for Talon.

NJ knows about it, see http://www.njrec.org/Downloads/OBD_Probs2.pdf.

I found many references to TSB 18-005-01 that describes the recall but ended up at a firewall at ALLDATAdiy.com :: Access Detailed Automotive Technical Service Bulletin and Recall Information that's asking $25.00 for the first car to get the full TSB. I guess I'm not that eager to help.

Ah, just checked back in. If PattyG is already in AllData and found nothing I feel even better about not breaking out my credit card.