O2 sensor electronics testing
 

Hi guys,
I have 1992 500sel.m119
I understand that oxygen sensors put out 0.1 to 1 volt of DC depending on the exhaust fumes. Now if I disconnect O2 sensor and connect 1volt battery to terminals that go to computer will you notice the difference in engine sound? I suspect my computer to not working properly as I connect the 1volt battery to O2 terminals I hear no difference in engine sound, it would run the same as O2 sensor connected or not or 1 volt battery connected or not. Does that make you think that computer went south?
Mechanic pulled codes. Idle switch on gas pedal and starter lock out/back up switch on the transmission came up. I have both checked with ohmmeter and they both work fine.
Any ideas what is happening with my car?

Thanks

I have tested my O2 sensor with better ohmmeter and readings came up to be normal. I also have my duty cycle checked and it was oscillating up and down a bit so I assume lambda system is working. There is one problem: engine runs rich. I get 72 to 82 percent duty cycle reading on either idle or 2500rpms. O2 sensor readings are from .2 to .8 volts and it gives out high HC reading at inspection on idle.
I used to have 2 other MBs where I could adjust mixture with long octagon key through air/fuel meter in the middle of the engine. This car has m119 and air/fuel mixture is adjusted by computer. Does anyone know how to bring my duty cycle to 50% again? What causes the computer to run rich mixture?

Check Reg for fuel evidence at vac port.

http://catalog.worldpac.com/mercedesshop/sophio/wizard.jsp?partner=mercedesshop&clientid=catalog.mercedesshop&baseurl=http://catalog.peachparts.com/&cookieid=21H0MN3YU27C0R8G5H&year=1992&make=MB&model=500-SEL-002&category=C&part=Fuel+Pressure+Regulator

* You've probably checked the fuel pressure, the coolant temp sensor, and the intake air temp sensor values.
* What do the adaptation values tell you? Is the fuel management really trying to correct an over-rich condition? Would corroborate the HC values.
* Beckmann Technologies will test your controllers for a small fee if you suspect, but can't prove, a problem there.
* On this site, have you checked this out. Go to DIY Links (see the tab at top of page). From there go to DIY Links by Parts Catagory. Posts are organized by relevant "parts area". You might find something useful in "fuel injection". Good luck.

Never mind the adaptation values. My mistake. Somehow had the idea your MB was later model.

It does have adaptation values. You need a scanner with live data capabilities to get them.
Mike

Thanks for replies guys.
I will check my fuel pressure along with intake temp. sensor next. I replaced coolant temp. sensor already.
I would love to check adaptation values. Where can I find a scanner to get live data capability? Is it expensive?

Yes Victor, it is expensive. I have a Snap On MTG 2500 and Mercedes cartridge. Ebay is a best source now. Here is one for example.
MB cartridge:
http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&item=330087276384
And the scanner from the same guy:
http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&item=330174183204
I remember paying about the same for the cartridge from Snap On and much more for the Scanner, but it had cartridge for Domestic cars.
Mike

* If you have a good working relationship with an independent shop, and can talk with their MB tech, describe the values/interpretation you're looking for and pay them for some diagnostic test time. Of course it's not easy to find a shop that will take an honest interest in your project, and even harder to find a shop that will let you past the service writer to talk with a tech.
* A tool that will do live data, some coding, as well as scanning and other functions will likely cost you at least $3K. And that's a lot to spend for just the occational test on your own car. My PC Retriever (Assenmacher Specialty Tools) was around $2500, but I got a discounted price and I already owned a multiplexer which would've added $1K to the cost. Baum Tools has a nice tool also, but it's expensive too. Be wary of cheap equipment. Comparison shop. You might be able to find a reliable used high quality tool on eBay, or from a member of this site. Good luck.

When I read that you have a 119 and it is running rich, it reminded me of something I read on Jim Forgione's site about DTC #19. http://pages.prodigy.net/jforgione/MB_S500.html

Maybe your car has not reached the adaption limit to set off the CEL, but it may be worth scrolling down towards the middle of the above link for some good information about DTC #19. Good luck.

I went to Jim's website and built home-brew code reader to pull codes. It works great! Thanks for your time and effort to create and maintain your GREAT website Jim:) I have found lots of info I've been searching for on it.

There were many codes found. Mostly because I never cleared them after engine harness was replaced. All of them are cleared now and I will drive the car to see which codes will come back.

I was checking my lambda and it is still in 75-85% range. It gets to the point when you hold 2000rpms you can hear that the engine is slowing down a bit when readings reach 89% and when readings drop to 75% engine picks up rpms again. Can adaptations be a cause of it? I never got code#19 of on pin 19 though. Maybe I will order new eprom just in case. I don't think that my mechanic has that scanner. He has similar looking code reader he used to pull codes with but home-made blinking code reader pulled way more codes.

I have found my purge valve inoperative so I will replace it. Lambda was checked with purge hose plugged.

Fuel pressure regulator check did not reveal anything. I had it removed and cleaned. It looked good and held vacuum.

Air intake sensor was changing value when I blow into it. It reads around 1.7kohms at 70 degrees. Coolant temperature sensor is also in correct range.

Can someone tell me what signal get to pin 40 of Base module? I suspect Base module to be either rebuilt or tampered into. I opened it and found scratches on one diode and pencil marks on the other side.

Thank you for help, guys:)

Update: I just drove the car few miles and have duty cycle checked. On idle it stays at steady 68.9% and when you keep 2500rmps it jumps from 42 to 55%.
From what I understand D/C always has to be around 50% on either idle or 2500rpms. Does it prove that my adaptation values are out of limit even though no code19 is not present?

Victor300E, did you clean the MAF and IAT as shown in MENU#4a

Hi Jim,

If checked the price of the on/off lambda ratio tester (M0039) and it was kind of expensive ($355). Any cheaper alternatives to this tool--like at Radio Shack? Nothing's wrong (that I know of), but I was going to check my ratios out of curiosity. Thanks again for all your time and information.

I've not done this but a "meter" that does duty-cycle should work ok.

My comment in the previous post was meant for Victor300E to clean the MAF and IAT. Then check the lambda . . . . he may be pleasantly surprised! :pleased:

Understood. I was thinking of doing the same with my MAF and IAT sensors, but I wanted to see what the lambda reading was beforehand. Forgive my ignorance, but do they have meters at Radio Shack that measure "duty cycles"?

Yes, I think so . . . spent 20mins on their site and all I can say . . . :grim:

You can't find summary specs for their VOMs except in the "PDF" files which you'll have to wade through. Personally, I don't like RS for anything like this.

A better choice is "Fluke". . . be prepared to do some 'googling' and if you find something that looks interesting, I'd be glad to check it out.

Update:
 

Thanks for replies, guys.
I have checked everything again and here are the results. I had couple of shorts in my self made engine wiring harness going to ECT sensor. O2 sensor plug was broken apart and wires were swapped. Purge valve is leaking. I have fixed everything except purge valve. Just blocked vacuum tube to it for now and have new one on order. Erased all codes with home made reader.

No new codes are stored:) Car gained more power and revs quicker but there are 2 problems. 1st is that when the engine is started and its warm duty cycle stays at 50% for 30seconds and then starts to drop down to 10% and stays there +/- 5%. At 2500rpms duty cycle is jumping around 27-35% and when I let go off the gas pedal it jumps to 50% and stays there and then 1st scenario repeats. 2nd problem is that when the car idles and duty cycle is at 50% the idle is very smooth. When cycle drops to 7-10% idle gets rough, engine shakes to the point that one can feel it inside the car.

emerydc8: I have used Craftsman multimeter I bought at Sears 2 years ago for $40 and it works great. I used black lead on ground and red on pin3 in connector x11 on driver side wheel well. Did I do it right?

JimF: Thanks for lots of info you have collected on your website. I went to menu 4a and followed instruction on cleaning MAF sensor. It's drying as I type. I will check it in few minutes and update this thread.

Thanks

update:
 

I have MAF and IAT sensors cleaned as per Jim's instructions. Duty cycle on idle is 2-9% and at 2500rpms 23-35%. Cycle jumps to 49% if you give a little gas and let go. After it jumps to 49% it slowly drops to 2-9%. Engine starts to run a little rough. Exhaust smell bad. Does it mean that I have mixture too lean? No codes are stored.
Any ideas what can cause lean mixture?
Please help:confused:

Have you tried this . . . .

Disconnect the MAF sensor connector by turning it a quarter turn. This should force the car to run on the precoded fuel management mode and go to the 50% area. Supposedly, airflow info is generated by using rpm and throttle position instead of reading the MAF sensor.

If the lean drivability symptoms go away, the MAF sensor is probably bad. If they don't, the MAF sensor is probably OK and it might be time to replace the O2 sensor esp since it's been misconnected and sitting in the exhaust stream with no heater.

I disconnected maf sensor and duty cycle went to steady 59.7%. When maf is connected cycle goes to 50% and going down to 3-12%. I have O2 sensor on order. It will not hurt the car to have new O2 sensor since it's old anyway.
I did have a stalling problem before and I bought a used maf from good running car with 125k. I know the guy and it was his personal car he parted out. After maf replacement my car never stalled once. Maybe his maf was not good?

Jim, by the way, does duty cycle has to go to 40% when maf is disconnected? Do I measure cycle correctly? Red positive lead from multimeter to pin3 and black ground lead to ground? If I reverse leads then I get 40% maf code.

I think the 'test' showed that the MAF is bad since it goes to the 10% area unless it's contaminated and that contamination can't be removed.

Is it a "hot-film" type vs "hot-wire"?? In trying to clean them, the H/W MAF clean up better than the H/F MAF. My tech has tried a number of them with no success.

Red on pin 3; Black on Pin 2.

Thanks, Jim. This is a good string of information. Any way to tell by looking if my 1993 400E is a "hot film" vs. a "hot wire" type MAF? Thank you too, Victor, and good luck.

. . . . look down the 'spout' . . . . it's somewhat obvious; L/H has a thin wire, H/F has black plastic assy that sticks into the venturi.

I was looking on electrical diagrams and it seems that maf sends signal from pin5 to ECU pin17. What kind of signal is it? I have read how maf works but it seems to have a solid state device inside that converts current needed to heat hot wire to some other signal that is sent ECU pin17. What kind of signal can it possibly send to ecu? It has to be some voltage that varies with amount of air passing through maf. I want to get to basics of my lean conditions. Also I have new universal O2 sensor from Bosch. Hopefully it will improve something.

Another question is can one adapt MAF from m104 engine to work with m119 engine? m104 mafs are way cheaper than m119 engines. I have no problem with resoldering the plug to fit m119 plug, but I have doubts about m104 maf ability to calculate extra air that m119 use. Also does m104 maf has same dimensions as m119 maf?

Menu#4a has a link to general info for a MAF (MAF Troubleshooting and Testing) along with typical voltage output levels. As you surmised, the amount of air flow is turned into a voltage; the more voltage = more air.

As far as interchanging MAFs from 119 and 104, it is "theoretically" possible if you could design a ckt to translate the output of one (119) to be equal to the other. Practically, it's not possible.

Well when you amass all that info you can teach us . . . . :sultan:

PS: the other day, I checked my Lambda to see what's what; oscillating between 48 to 56 at idle and the same at 2K. Seems like the 'cleaning' is still 'effective'. I sold the K&N filters and now using paper!

Fyi
 

Inside LH MAF

I just bought an autoranging multimeter at Sears (a step above Radio Shack) for $49.95 that measures "duty cycles." The Flukes were right next to them, but they were all well over $100. Jim, I know you have to plug the purge valve (Y58/1) line to check Lambda. Is that the valve that's mounted to the left fender well and makes a clicking noise most of the time?

Thanks, Myarmar. Is that a picture of a hot-film type MAF or hot-wire?.

Hot wire

I had one that I almost took apart . . . but I didn't. Tnx for the pic of the guts!

emerydc8, yes that's the device; the rubber hose slides onto a rigid metal tube. Good Luck w/ the new meter.

Thanks to all. I hope I have the hot-wire type, but I guess I'll find out soon. The rubber hose on the purge valve has hardened (just like every other hose in my engine compartment that was once flexible), so I'm expecting to break it when I take it off for testing the Lambda.

Jim, I was reading your Menu 4(a) on MAF Contamination and Lambda under the headnote Fixing The Problem. Am I reading this correctly--that if Lambda is high (above 65%), you can disconnect the MAF at the canon plug and if that causes Lambda to drop to 50% then there is likely contamination on the MAF?

emerydc8, please re-read post#19 and #20 . . . .

In Victor's post, it appears as though his is running lean but I think yours is running rich. So I "hope" that it should drop to around 50% but may not. I guess that remains to be seen. If it does, then the DI pre-coding is working as it should.

I thought you cleaned the MAF?? If so, maybe it's really 'broke'??? Then what to do next . . . ??

Thanks, Jim. It's always passed through emissions okay. I just thought I would take the opportunity to learn something about duty cycles and Lambda, so I jumped into this string. (I hope Victor doesn't think I hijacked it.) I'll probably end up breaking something that didn't need fixed to begin with, but at least I'll learn about MAFs and duty cycles. I've never cleaned my MAF. I talked to a friend that said he cleaned his (hot-wire) on a Chevrolet. He said he actually wiped down the wire, but I'm not sure I want to go that far. Maybe regulated air to blow it off would be the safer way.

Ok, since it's never been cleaned, suggest that you do that; remove it, get some CRC MAF cleaner, spray it thoroughly, and if you have shop-air use it, backing way off from the MAF unit to blow dry. If you don't, just let it dry . . .won't take long.

Do NOT try to touch-it or use a brush for the H/W (LH) version, you could break it. It's just a very thin wire suspended inside the center.

I'll bet it will change its response for the better . . . :)

I hate to tell you this Victor but a o2 sensor is nothing but a low level battery...that activates when heated......test my theory by taking it out of the car ...hook it up to your VOM meter and use a propane torch on the tip and see what happens to the voltage. interesting.

and it the only way to tell if its working or not

Thanks again, Jim. I have the rubber connector for the MAF on order. I figure that will break when I take off the MAF. I'll post the results.

And you can get a feel for whether it's good or bad by testing it following the procedure in MENU#4 under "O2 Sensor Testing".

While I am waiting for the rubber boot to arrive for my MAF, I hooked the autometer up to my car to check the Lambda. It was consistently between 48%-52%--both at idle and 2500 RPM. Does anyone know what this plug is with the numbers 052 stamped on it? It's on the right side of the engine bay.

That's perfect! Where did you get the connection points?? You may be able to find the connections in the 38 pin Diag Connector on pin 14 for late '94 and and on.


That looks like the connection from the ETA via its flying cable.

At what I think is X/11 next to the purge valve on the left side of the engine. I think my car was made in November 1992. I put the red on pin 3 and the black on pin 2, selected frequency and it came up as a %. Is that right?

Do I have an ETA? I thought my car just had the mechanical linkage. Maybe I'm confusing the two. In any case, I unscrewed the plug with the engine running and the idle increased.

Looks good to me . .

I thought that "ALL" MBs had one of those PITA . . . check the connector . . .

You're right. It's the same connector. Why are they a PITA? Has yours given you trouble?

Yes, I think they continually are a PITA. . .

Checkout this thread started almost 4 years ago . . .
http://www.peachparts.com/shopforum/showthread.php?t=74272

Check post #32 . . . and that will bring you uptodate.

I don't think my car had the ASR option. It looks like the ETA has other functions besides just anti slip.

I knew I'd Break Something!
 

Well, I think I found one of the functions of the ETA besides ASR. Since I took the canon plug off with my engine running last week, my cruise control is completely inoperative. Also, I'm getting a DTC 3 on Socket 7 of my 38-pin connector. The code cannot be cleared. I noticed that my idle speed has dropped from about 500 RPM to 250 RPM too. I checked the fuses in the regular box. I don't want to jump to the conclusion that it's a bad cruise control amplifier, since I think the problem originated from my own monkeying with the ETA wiring. Any ideas?

Simply put . . . you can't do that! EA/CC/ISC and ETA are married together. The "CC" is 'cuise control'.

Don't think you found anything new but you certainly created a DTC! :grim:

Now, if I could just get my cruise control back I promised myself I would never screw with something that wasn't broken.

Some have reported losing the CC when their ETA went bad; for me, I never did lose my CC when mine went bad.

Have you read all of the DTCs from your car??? It would help to show what's what.

When I checked all the codes yesterday, I got a DTC 4, DTC 6 and (dreaded) DTC 19 on socket 19, along with a DTC 3, 6 and 11 on socket 7. I was able to clear all but the DTC 19 on socket 19 (as expected) and the DTC 3 on socket 7. I know what's causing the DTC 19 and it's probably been there for an half decade. The only remaining code that's a problem is DTC 3 on socket 7, which (I think) is the cruise control. I'm wondering if I spiked something when I took off the ETA with the engine running. I talked to the tech at the dealer today and he said that maybe my home-brew scan tool will not allow a reset. He offered to hook up to his equipment and attempt a reset, but I still think my cruise control will be inop and, besides, taking it down there is like admitting defeat.

I read that there is a fuse in the diagnostic module. Could that be the problem (or maybe another fuse in one of those modules)?