MichaelBrown

There have been a few discussions over the years on the subject of the 02 light on California cars but no definitive solutions or answers to the continuous ight issue. I have a 93 2.3 that has the issue where the light stays on despite resetting the computer or replacing the sensor. I understand that the light on pre 87 cars illuminated by mileage and the solution was to simply remove the bulb, but on this car the light means there is an issue. There is no code being thrown and if you unplug the sensor the car throws a 02 code and the check engine light comes on so I know the sensor is communicating with the engine control module.

When the 02 relay reset button is pressed, the 02 light goes off for a few seconds, flashes and comes back on. I have read where some people simply remove the relay but if there truly is an issue with the car, I want to resolve it. I suppose there is a possibility that the relay unit has failed but there are none available anywhere that I can find. If anyone has any input on this subject, I would love to hear it. I have also tested the heater voltage and voltage output from the sensor and all is good.

dolucasi

Share with us what your duty cycle measurements are telling you Michael.

- Cheers!

MichaelBrown

I’ll check them tomorrow morning and report back

dolucasi

Great!
Make sure and send us all the readings below.
Static - Ignition On, Engine off
Warm-up - Engine running cold
At operating temp idle, cycle variation
at 2500 rpm.

MichaelBrown

Video of duty cycle readings this morning

dolucasi

Thanks. Comments below:

Static Duty cycle - 81% - This is closer to 80% (an error code) than to 85% which is CA emissions code
Cold Duty cycle- Not in the video as it seems you started the car prior to taking the video. This would have been helpful because we do not know how accurate is your meter since you mentioned you can never get to 50%. With engine stone cold it should read 50%. If it does not I would question your meter first.
Open loop duty cycle - Starting at 38% or so - This should also read 50% till it switches to closed loop, this is why I am suspecting your meter's accuracy.
Closed loop 25% and cycling - Not sure how accurate this is either but at least it is cycling.
2500rpm - Missing. You need to rev it to 2500rpm in neutral and keep it there for 10 seconds min. Just giving a little boost will not help us. But hold off on this as we need to verify all the previous readings first.

So let's go down to basics and calibrate your meter by using the DC voltage method.

With ignition on, engine off (your meter says 81%) please switch to dc volts on your meter and tell us what you read. Then switch the lead to pin-3 on the X11 port to pin-6. Report on that voltage. It should be around 12V but we need the exact value.

- Cheers!

MichaelBrown

Dc voltage test

dolucasi

Hi Michael,

looks like you are having some measuring issues. I tend to believe the 1.84V static DC voltage pins 3-2. Assuming 12V battery voltage that is almost exactly 15% which in mercedes terms is 85% duty cycle. However your 6-2 measurement which should have been system voltage did not measure ~12V. Possible the system voltage is either pin4 of pin5 in your car.

However I noticed that you have dwell angle capability in your meter as well as duty cycle. So we have yet another way of verifying.

Let's repeat this same experiment with ignition on, engine off.

Measure your system voltage first. Clearly in your car it is not pin6. On my '89 it is pin 6. I think they may have moved that around thru production years. Try pin5 and pin4 till you get approximately 12V.

Then measure 3-2, DC voltage again. Then switch to duty cycle, then switch to dwell angle with 8, then 6, then 4 cyclinders. Only 4 cylinders will make sense here but I would like to see it.
I'm suspecting your duty cycle measurement was/is not that precise/accurate.

We have to establish baseline measurements before we can do real measurements.

MichaelBrown

I will go down after tomorrow mornings festivities. I really appreciate your time and helping me out with this.


There are a couple of things to note here. I believe it was pin five that I got 12 V on, but I will verify that tomorrow. Also, on my car and my particular meter, I have to invert pins two and three in order to read duty cycle. When I first started messing around with this car, I was using pins two and three, which is the standard and just got wild readings that made no sense. Then I ran across a blog post by an engineer, saying that in certain meters you needed to invert in order to read duty cycle whixh was the case with mine.

dolucasi

Aha, this explains things. Never invert pins, always keep your black/ground on pin2. That engineer might have even been myself that you read in another forum.
Indeed some meters read incorrect values for duty cycle when you invert them. Mine does, clearly yours do to.

So measure 5-2 and 4-2 for grins. Also measure the other suggestions I made so we have multiple confirmations.
For duty cycle, just subtract 100%- (your reading) for Mercedes defined duty cycle.

What you expect to see when you switch to 4cyl dwell is the Mercedes duty cycle. It will be close anyway but again this may be meter dependent as well.
I can explain this later why this is true later if you like.

MichaelBrown

I’ve completed the various tests in this video. Hopefully by doing this video, you can see how confusing all of the various posts are when you get these types of readings. Hopefully this thread will be of great value to a lot of people.

dolucasi

So your meter is funky. At times it is measuring true average DC voltage (as in when you are measuring static duty cycle).
Your pin4 is the timing signal from the EZL
pin5 is the system voltage
pin6 - I have no idea what this is.

I am not sure if I made the request clearly. Do not start the car and take the following static measurements. We do not need a video you can just tabulate if you like.
Static, ignition on, engine off.

DC voltage - 3-2= 1.84V, 4-2=?, 5-2=?, 6-2?
Duty Cycle: 3-2 = 14.7%, 4-2=?, 5-2=? 6-2?
Dwell Angle: 3-2 = ?? in 4cyl mode, 3-2 = ?? in 8cyl mode


Please do not turn the ignition until we go to the next step. So far we are just trying to figure out how to make the measurements with your meter.

However, so far we know.
(1) You meter is not reporting "Average" voltage like most meters do. It is measuring "peak" voltage for most conditions except when the average voltage is low.
So you really can not trust your DC measurements. I am not sure if it is on some weird mode but I could not read the display from your video, it was fuzzy.
(2) There are no error codes stored in the Duty cycle controller - This is good. You are measuring ~15% which is 85% - You have a CA car as you mentioned
(3) Since we can not trust your average DC voltage readings I would like to make sure the Duty cycle is reporting correctly in all conditions, thus I need to confirmation from above with the dwell

Send us the table from above when you can. All with ignition on, engine off.

Merry Xmas and Happy Holidays,

MichaelBrown

OK I took the measurements, and here are the results:

DC
2,3 1.78
2,4 11.62
2,5 11.62
2,6 1.51

Duty
2,3 14.5
2.4 0.6
2,5 0
2,6 0

Dwell 4
2,3 76.2
2,4 0
2,5 0
2.6 89.6

Dwell 8
2,3 38.2
2,4 0
2,5 0
2,6 44.9

190Efan

I own a 91 2.3 but my car isn't a California car. Does your car have any additional symptoms besides the O2 sensor light being on?

MichaelBrown

No, no additional symptoms other than the O2 light remaining on and my failure to get duty cycle to get to 50 but that might be my meter. We’re working that out now.

dolucasi

OK, so we are calibrated with your meter now. FY here it is, so you can do the same in the future every time you measure Duty cycle:
Ignition On, engine off
DC: 100% - 100% * (1.78 / 11.62) = 100% - 15.32% = 84.68% ~ 85%
Duty cycle = 100% - 14.5% = 85.5% ~ 85%
Dwell 4cyl = 76.2 degrees * 100% / 90 degrees = 84.67% ~ 85%
Dwelll 8cyl = 38.2 degrees * 100% / 45 degrees = 84.89% ~ 85%

You can probably guess what you would have measured Dwell in 6cyl mode ( x * 100% / 60 degrees)

Since your DC measurement is not dependable at lower duty cycles always use duty cycle (100 - duty cycle actually) and report that in this post. And if you have any doubt you can switch to Dwell and do the math there for confirmation. Remember to never switch the polarities on your meter.

Now step 2. With stone cold engine next morning, I would like you to video tape the duty cycle from first crank to warm up. And after warmed up throttle the engine to 2500 rpm and keep it there for 10 seconds so we see your duty cycle.

BTW, you battery is worn assuming you did not have headlights on during the above measurements. It is only registering 11.62 volts.

- Cheers

MichaelBrown

Will do and yes, battery is on its last leg. I will charge it tonight as well.

MichaelBrown

Actually, I have not run the car today so I can actually make this video now. For clarification, do you want me to acquire duty cycle with the duty cycle setting or the voltage for this video?

dolucasi

Keep it on Duty cycle with black lead on pin2. Your DC measurements for duty cycle are not very useful. (which is strange but true)
At the end of the video, keep the engine running and switch to 4cyl dwell just for confirmation.

Cheers!

MichaelBrown

video of results

dolucasi

Hi Michael,

Not sure what was going on at the end when you tried to confirm with dwell measurements. It kept bouncing around but let's ignore that.

So there are a few oddities.

(1) Your ECU immediately goes into showing some duty cycle which makes no sense because at this point the ECU is forcing severe enrichment, something in the EHA current of 15-25mA out in CA with today's weather. This is out of range for the duty cycle translation so for my '89 the ECU just spits out a solid 50%, telling the operator it is not in closed loop yet and operating in open loop. It will take 2-3 minutes to get into closed loop at which time it will switch from 50% to EHA current translation.
(2) Yours does not do this either because this is production year dependent or because there is a malfunction and the ECU is thinking the engine is warm already and goes straight into closed loop
(3) And during this warm up period there is quite a bit of idle hunt it seems, so something is not stable. Item one and items 2-3 may well be related to the same issue.
(4) Your 2500 rpm duty cycle drift is less than 10% from idle so you are good there in that you do not need any adjustments in the FD/EHA at the moment for that. We can check this item off.
(5) Your duty cycle is too high, in the 70% range meaning the system is trying to enrich so your FD plunger is adjusted too lean and the ECU is trying to enrich it constantly. We will have to adjust this but not yet, we have to fix the other oddities first.

(A) So what production year is your car?
(B) How well does the car run if you start it first thing in the morning and not wait for it to warm up and just drive away? Does it have any hesitations, etc?
(C) Have you ever pulled the codes from your diagnostic port (9-pin or 16-pin depending on production year). I am wondering if there is a code there.

- Cheers!

190Efan

I've noted that you have no additional symptoms other than the O2 sensor light being on. I agree, they're typically on for a reason. I've found with my car that any time there's any dash light on, there's a reason for it. I also noticed from your video that on initial start-up, your car is alot noisier than mine is. That might be explained by your car being a California car and mine isn't but maybe not. I also observed the hunting after the start-up that Dolucassi has pointed out. A low battery can cause strange things to happen. I'm also wondering what the status of your alternator is? Might it be possible that there's some other additional draw on your electrical system, ie, security alarm, etc that might be drawing enough electrical current to upset the ECU? It sounds to me like the ECU is getting faulty information but I'm not yet sure where that bad information is coming from. Keep doing the testing that Dolucassi is recommending and hopefully that'll help better pinpoint where the problem is.

MichaelBrown

MichaelBrown

I wonder if one of the temp sensors has gone south? Seems like I remember replacing those but honestly I’m not sure.

dolucasi

This is good info. So one of the sensors to the ECU must be off if you already swapped and nothing changed. You can start with checking the coolant temp sensor. Attach your multimeter and tell us what the resistance it at room temp and then at operating temp. Let us know if you do not know how to do this.

But even before that can you read your error codes in your 16 pin terminal by the battery that has the built in LED bulb?
Some error codes will also trip the CEL. So we will also verify why your CEL is on.

That should be the first thing to do. If you do not know how to do it, here it is:

This used to be free but Kent is now asking $3.99. Probably worth the money if do do not know how.

- Cheers!