Kev, were the readings any better or still the same?

Almost the same, I may have to grease a friend of mine for a sticker if I can't get to the bottom of this

This is a copy of a post I made several months ago on exactly this subject...



My car failed 3 times, passed on the 4th, very recently. I bought it from a used car dealer who had 'acquired it' and was only looking to recoup what he had put into it.

The reason for my fails were high NOx levels. Basically the opposite of your problem, but that would suggest doing the opposite of what I did would have a positive effect.

The first 3 times the dealer was trying to sort things out by checking and cleaning normal culprits...fuel injectors, vacuum system, spark plugs, etc.

After I did a lot of research around the net I went down to work with his mechanic to fix the problems.

We adjusted the bypass on the MAF unit to bring the O2 sensor reading back into line with what they should be. I adjusted the DME to retard the ignition by about 3 degrees. The next day the dealer also changed out the catalytic converter for a new one.

The day after that the car was tested and the emissions had dropped to about 1/4 of what they had been. I've listed the results below:

Item HC (PPM) CO (%) NOx (PPM)

Limit 200 0.64 1400
Test 1 133 0.70 3337
Test 2 124 0.68 3804
Test 3 101 0.35 3865
Test 4 34 0.05 779

As noted by earlier posters your car seems to be running rich (too much fuel, or too little air), whereas the NOx I was dealing with is running lean.

Given that the results of my tests showed improvements in the HC and CO levels over the first 3 tests I would say cleaning injectors, spark plugs, checking vacuum lines, etc will help, but don't make a huge difference.

The 3 things that made the world of difference for me likely would for you.

Adjust the DME (adjust it back after the test)
Adjust the air bypass on the MAF
Replace the cat

Unfortunately all 3 things were done at once, so I don't know which one makes the biggest difference, but I'll give you my thoughts on them all.

Adjust the DME

Actually, it's the FQS (Fuel Quality Switch) in the DME that you're adjusting.

http://frwilk.com/944dme/fqswitch.htm - this tells you about the FQS adjustments. I put mine in position 4, you would likely be looking to position 2.

http://www.geocities.com/motorcity/1760/dme.html - this tells you how to access it.

http://www.promaxmotorsport.com/fqs.htm - everything on one page.


Adjusting the Air Bypass on the MAF

With the car running you need to back probe the O2 Sensor to find out the voltage. Anything over 0.5v means you're running rich. Below 0.45v is running lean (mine was at 0.32v when we started).

On the 944 the by-pass is adjusted using a 4mm allen key to a bolt on the MAF.

http://frwilk.com/944dme/afm.htm - the bolt head can be seen on the first picture of this page, lower right corner. It is covered by default. The cover is removed by screwing into the top of the cover with a sheet metal screw (just a couple of threads into it), then removing gently with a set of pliers or vice grips.

http://frwilk.com/944dme/lambda.htm - This gives you the voltage it should be.

Turning the 4mm allen bolt counterclockwise (I recall) will lower the voltage you read at the MAF. Take it slow and let the reading stablize before continuing to adjust. The reading will also jump around a bit so you need to average it.


Catalytic Converter Replacement

I would pick up a universal cat, then take it to a muffler shop and have them install it. You should be able to get one and hae it put in for close to $100. Might be a little more, but shouldn't be much.


Long winded post, but I hope this helps.

Cheers

Pete

Pete, GREAT POST, thank you very much for copy/pasting that, i will start looking into the details today!

EDIT 1

reading into this it seems like this is a three way test...
i need to read the voltage off the 02 sensor, not sure how im going to do that, aside from maybe snaking small wires into the barrel plug itself to connect to the voltage meter...

adjust the DME (i never knew it was adjustable!)
and adjust the AFM

a note about the AFM, that adjustment screw thats supposed to be covered, i know ive tampered with it about 2 years or so ago, mentally i thought i 'put it back' but theres really no telling...

Pete, my readings are very high NOx, if you have more description of systematically, physically doing this, im all ears. i assume these adjustments will need to be done on a warm engine at idle.

i suppose i should try position 3 +6% or position 1 +3% pete, why did you retard the timing?
as far as adjusting the AFM im guessing it should be set at REFERENCE voltage while idle? not high, not low, so im looking for VReference = +0.475V ... correct?
how do i get a voltage reading from the 02 sensor?


Edit 2
aside from just brushing this issue off and paying for a sticker, i feel like its probably in the engines best interest to resolve it and not run lean...
I do know that the PO put a chip into this car, im not sure if that changes matters!

You can back probe the O2 sensor, pull back the rubber insulation for the barrel connector and stick the test lead in to test the voltage.

The adjustment screw on the AFM is for CO adjustment, you can richen or lean out the mixture using that, but you need to monitor the voltage at the O2 sensor (to get a rough idea of the AFR). Ideally you want a completley warm engine. The voltage at the O2 sensor should be roughly .5 volts at idle, which indicates you are in the 'right' mixture range. The ideal way to do this is to tune the car using a 3 or 5 gas analyzer. The mechanic doing the inspection should know how to do this using the emissions equipment.

I would definetley put in a stock chip for emissions. Some of the aftermarket chips ignore the O2 sensor signal and rely only on preset maps.

BTW, another trick to passing the dyno is tell them to run it in the lowest possible gear. It may say run it in 3rd, tell them to run it in 2nd. The gearing in these cars is pretty long and not ideal for the dyno tests.

thanks guys, I have lots of good data in my pm box from techno duck as well as the data from pete. I'll post the info from my pm box to the thread for reference. I have some days off coming up I'll be able to give the car some time. I think this is exactly the info I needed. I will also attempt to install the stock chip. I will increase the mixture 3%, then make the afm adjustment per the spec. I will report back.

This community never stops amazing me, it's half the fun of the car, thanks guys

Ps. Front suspension bushings from pelican should also be in today.

I would try to get the AFM back to spec, put in the stock chip then retest with the FQS set at zero. I think it will pass fine.

Hi Kevin,

Sorry, got caught up with some things and didn't get back to this thread until now!

If you go through my post you'll see the tests and results that were taken. The original NOx figures were a bunch higher than yours, so it's the same general issue.

I apologize for part of the post (saying you should set to pos. 2) perhaps being a little misleading...I had posted it for someone having HC issues I think, which is basically the opposite problem to NOx.

Here are some further notes:

DME RELATED

The DME FQS adjustment is done with the engine off, battery disconnected, to protect it.

NOx is produced when the temperature in the cylinders at iginition gets above 2500 degrees. Retarding the ignition so that it doesn't fire until the cylinder is in the down stroke reduces the temperature as there is no further compression to happen (compressing anything raises the temperature).

Retarding the timing could either have minimal effect, or a huge one...depending on how close to 2500 degrees things are.

Accessing and adjusting the DME is detailed on the sites I posted. It’s under the floor in the passenger foot well. I adjusted it using a piece of a ball point pen lid that I shape using a box cutter!!


AFM RELATED

[ This was done with the car at a warm idle, as you noted.]

To adjust the AFM you need to remove the cover on the adjustment screw. I did this by screwing a self tapping screw (only just into it enough to get some bite) into the cover and pulling the cover out with a pair of pliers.

This exposes the allen screw which needs to be adjusted.

Back probe the O2 sensor by sticking one lead of the volt meter in, under the cover where your red arrow is pointing on an earlier post, and ground the other lead on the body.

The readings you’ll get cycle a little bit, so you need to make an estimate about where the average is. Mine started out at 0.32v which is WAY below spec, as you can see from the chart.

As you adjust the AFM screw take it slowly and let the voltage reading stabilize before keeping going. It wasn’t a fast process, but I think it was a big help.


Does this make sense?

Pete

need some help..
Got the isv jumper in,o2 sensor is disconnected, stock chip is in, have a good ground, and the + lead of DVM is plugged into the dme side of the harness for 02 sensor. I'm getting a steady reading on .28volts and I can't get it to change.

W the sensor plugged in I can get it vary adjustable from around .55-.7 v. On the afm adjust that's from
full stop clockwise to about three full turns CCW. Once I adjust the afm past 3 full ccw turns back the dvm starts reading negative numbers

Ok I think I'm getting it. (typing this from garage). Right around the point where the meter starts going crazy, if I microscopically adjust the afm and make very small changes to idle it seems to be jumping around now in low to mid fives. Guess I just need to keep at it. I wasn't having any luck trying this w the sensor unplugged. I'll report back shortly

You can put a capacitor inline between the volt meter and O2 sensor connection to steady the #'s a little.

well i got it as close as i could. the DMV did not like 'bumping' the throttle, if i did, it would take a while for the readings to come back. also, i would get it almost exactly 5.0 with the jumper, take the jumper off, the engine would re adjust the idle higher than where i had it, and i had to start over. it was a matter of getting the jumper on/off idles to match then getting the number as close to .5 . once i took the jumper off, and i got the idle dead on the numbers would move around A LOT, like into the .25-.65 sometimes steady around .5 for a moment or two then back around again.

i must have been idling with the car for about 2 hours.

im not sure what the big deal is about these DME chips. to be honest, i think the car runs smoother with the stock chip, and the more i think about it, i really didnt even feel any difference shifting around through the gears, even bringing the revs up?! the chip i took out said Authority on it.

well i'll give this a go at the testing center and see how were doing :)

i tried to back plug the lead on the DMV to the DME side of the harness and to the 02 sensor side of the harness while the 02 sensor was unplugged and i could NOT get the meter to read >.50 it always seemed 'stuck'. but with the 02 sensor plugged in i was able to take readings in line with the help here.

failed again, but there was some improvement.

NOx needs to be below 1323
Test 1 2991
Test 2 3086

Test 3 (today) 2241

huge improvement but im still 900 over.

can you guys clarify taking the .6 volt reading with the 02 sensor unplugged? what side am i taking it from? the sensor side or the DME side?

i guess we're on the right track here. do you guys think continuing to tweek AFM will get this down 900 units? or is there something else i should be looking at as well?

side note. HC and CO readings went WAY down as well. CO2 seemed to stay the same.

it really sounds like you have a vacuum leak at this point. or maybe a leak at the exhaust header.

all vacuum lines have been replaced, as well as intake gaskets and ISV as of last year.

which side is the reading taken from when the 02 sensor is unplugged with the DVM?

You want to test on the O2 sensor side of the connector. You ideally want the volt reading to be around .5v with the O2 sensor disconnected and the ISV disabled.

Try it again. The first thing i would do is reset the idle. Let the car warm up to normal temp, then jump the pins on the diagnostic port to disable the ISV. Set idle to ~900 rpm.

Now, disconnect the O2 sensor and connect one lead of the DVM to the respective O2 sensor pin and the other lead to a good ground on the block or chassis. You will get a bouncing around voltage in the sub 1v range. So you will have to estimate the average and set the CO adjustment on the AFM. Turning the screw counter clockwise will lean out (lower voltage) the mixture and clockwise will richen (higher voltage) the mixture.

When you get a rough average of .5 volts, reconnect the O2 sensor and remove the jumpers from the diagnostic port. The RPM will go up a little, then should return to normal.

Let us know what voltage you are getting from the O2 sensor before making any adjustments to the AFM. I would bet its <.5volts.

Just take note, this method gets you pretty close but is not exact. If you want exact (not really neccesary in my opinion), you need to have the guy doing the emissions testing tune the car on the dyno using the 3 or 5 gas analyzer on the emissions equipment. The #'s he wants to tune it to are in the factory manual.

mine, too. i had a vacuum leak this year. a big one.

You can always buy a cheap vacuum gauge from the local auto place and hook it up to the car. Warm idle you should get around 18-20'' hg for a 'healthy' system.

Here are my findings...

with the ISV jumper on, the 02 sensor disconnected and the DVM lead on the sensor side; it was at .3 when i started. the numbers are pretty steady. I was doing the AFM adjustments by 1/4 turn.

the FULL range i can change the DVM between AFM adjustment full out and full in is between .08-.30 volts. those values are while maintaining a steady 900 RPM.

if i plug in the 02 sensor the readings become more logical i can get more range while adjusting the AFM, and i can get it to between .48-.52 while it 'seeks'.

once i remove the ISV jumper the idle stays DEAD ON 900, but the DVM freaks out, it'll go from (-) to 0 to the .2's .4's .6s and back usually in the .2s - .3s and (-0) but it does it so fast its hard to tell.

i can only think of two things at this point.
1) try a new DVM
2) could my AFM be screwed up? <---- im ruling this one out, i just checked all aspects of the AFM per clarks, its working and within specs

You are reading volts DC right?

Im really not sure why your getting such low numbers. But that does account for the lean emissions #'s.

Todays progress....
Vacuum on the engine reads 18-19" Hg checked at the throttle body.

the new DVM gave similar results.
the only way for me to get a .55 volt reading on the 02 sensor is bring the idle up to about 1800 RPM! ive tried so many combinations of 'starting points' between the AFM and Idle adjustment im running out of ideas.

the absolute highest i can get the 02 sensor to read with it disconnected, at a 900 RPM idle is about .3 volts.

once the o2 sensor is plugged in, and the ISV is jumpered i can get a steady reading to very slowly drift across .50 volts. .48-.52v

once i remove the jumper i get lots of different readings, mostly LOW, on the DVM with the 02 sensor plugged in, but i can see it hitting on or around .5volts every few seconds.

i tried backing the AFM adjustment all the way out, raising the idle way up with the bypass, then lowering it a little at a time. i tried AFM adjustment full in, then backing it off, different combinations between the two.

i guess, since i got it close with the jumper, i should be in the ballpark? it just bugs me that i can get >.5v with it unplugged. doest make any sense.

im going to take it back tomorrow AM for the inspection (4th time now). Im on the fence if i should adjust the FQS to retard 3degrees and add 3% or just retest after making these small adjustments.

id say over the past 3 attempts of trying to adjust the AFM the car must have spent at least 6 hours idling in my garage