V12 ECU's Reporting 'Not Ready' - PeachParts Mercedes-Benz Forum

MB-Dude · #1 11-02-2007, 02:33 PM

Although the vehicle in question is a ’96 SL600, this situation may be applicable to other V12 models. About 6 weeks ago, the CEL came on when the engine began running quite rough, very poor acceleration, etc. The problem – whatever it was – cleared itself for the very next time I started the car (to begin troubleshooting) the engine ran incredibly smooth. A test drive revealed the engine was again running as good (quite strong) as it had before the CEL light. What a mystery.

I use an OBDII Scan Tool made by ProScan (www.myscantool.com) and it showed the following Trouble Codes (listed in order as retrieved) from the SL’s memory:
P0307 – Powertrain: Cyl 7 Misfire Detection
P0309 – Powertrain: Cyl 9 Misfire Detection
P0308 – Powertrain: Cyl 8 Misfire Detection
P0300 – Powertrain: Engine Misfire Detection
No other codes were listed and the CEL was, of course, registered as ‘On’.

Because the engine was running smoothly and I had documented the codes, I reset all codes and the CEL. And because of CEL activation, I expected the ECU’s to be in a ‘Not Ready’ state, and in fact, they were ‘Not Ready’. So I figured the ECU’s would settle down after an appropriate length of a Drive Cycle routine.

Here’s the problem… Even after 6 weeks and over 120 miles of mixed residential, city and highway driving, the ECU’s still register ‘Not Ready’. The attached file is a screen capture showing the reported state of the components. In addition, the misfire problem has *not* come back (this is good) and excellent vehicle performance has been there.

My question/concern revolves around two comments on the left-hand side of the screen shot: “CL-Fault: Closed Loop, but fault with at least one oxygen sensor. May be using single oxygen sensor for fuel control.” There are two entries, one for each bank of cylinders. Does this entry make sense? Are the ECU’s sophisticated enough (1996) to ‘fall-back’ to single O2 sensor operation, if a fault is detected in the other one? For information, there are a total of 4, Oxygen Sensors in the exhaust. Two per side/bank of cylinders; one each of these before the cats and one after the cats.

To my knowledge, the O2 Sensors have never been changed and the vehicle has 105,000 miles on it. I have always heard mixed thought on whether Oxygen Sensors should be changed periodically, though I have never had to change one. I know they are easy to change; just never had an occasion to change one.

Any and all thoughts/comments on the above problem would be greatly appreciated.

Cheers,
Jeff

mbdoc · #2 11-02-2007, 04:46 PM

In short, yes 1 "weak" sensor can cause the car to not pass the readiness tests.

We see that all of the time especially on 1996 models.

MB-Dude · #3 11-02-2007, 04:51 PM

Quote:

Originally Posted by M.B.DOC

In short, yes 1 "weak" sensor can cause the car to not pass the readiness tests.

We see that all of the time especially on 1996 models.

Thanks for the information. Not knowing which of the 4 are 'marginal', I'll change them all.

Follow-up question: Any particular reason why the '96 model year seems more prone to this issue, in your opinion? More curiosity, I guess, on my part. Regardless, thanks again for the feedback.

Cheers,
Jeff

mbdoc · #4 11-02-2007, 04:55 PM

Front ones are the culprits...rear ones ONLY monitor the catalysts & that takes several weeks AFTER the front ones do their jobs!

MB-Dude · #5 11-02-2007, 06:14 PM

Quote:

Originally Posted by M.B.DOC

Front ones are the culprits...rear ones ONLY monitor the catalysts & that takes several weeks AFTER the front ones do their jobs!

AWESOME! Sounds like you're cutting my parts bill in half as these suckers are expensive - MB list is ~$205 each. AutoZone sells a Bosch unit (#13638) for $139 each, but the lead time is ~1 week. All the other major parts houses in the area are quoting similar price/delivery. Apparently, this Bosch part is used only mid-90's Mercedes, so unlike spark plugs, etc., it's sort of a special order kind of thing. Guess I have to wait until next weekend to see how this saga plays out. Thanks!
Cheers,
Jeff

MB-Dude · #6 11-10-2007, 11:45 AM

Well… replaced the front O2 Sensors with new Bosch (p/n 13638) units from AutoZone. Unfortunately, there is no change in the ECU status. At over $300 for sensors and tax, this was a very expensive experiment. Guess that’s what I get for wanting to install OE parts as opposed to cheaper 'universal' components. Anyway, for reference I have attached a pic of the old O2 Sensor (top) sitting beside the new sensor (bottom). As you can see, the old sensor looks pretty clean, with very little soot or residue on the probe head.

OK, back to the problem… In my original post, I attached a screen short of the Engine Monitor. That screen is still identical, even after the O2 Sensor replacement and a 20 minute drive. While this is not long enough to complete an infamous ‘Drive Cycle’, it should be long enough to clear the errors noted - both Fuel Systems are still reporting “CL-Fault: Closed Loop, but fault with at least one oxygen sensor. May be using single oxygen sensor for fuel control.”

So I set up the OBDII reader to capture real-time O2 Sensor output. The attached pic is a screen shot of that process. Each graph represents 30 seconds of time duration. You will note that Bank1, Sensor 1 (upper left graph) seems to be oscillating somewhat normally, while Bank 2, Sensor 1 (lower left) is flat, at zero volts. The other 2 sensors are post-cat and are steady as a rock at ~0.5 volt - at least the cats are doing their job.
If I accelerate the engine up to ~2000 RPM, I do see a slow, steady change in both post-cat sensors, and by the same amount. Thus, they seem to be reporting and responding correctly.

But why is Bank 2, Sensor 1 reading zero volts? Going back through previous notes, I see the exact same output BEFORE the O2 Sensor change out. Thus, the O2 Sensors are not the root source of my problem.

If I watch the O2 Sensor readings long enough, Bank 2, Sensor 1 will jump inexplicitly to the normal range, then drop back to zero volts. I am interpreting this as an intermittent signal. Based upon this information, I need to trace signals.

QUESTION: Can anyone help me identify the signal path from the pre-cat O2 Sensors to the computer? Cable connections, wiring routes, etc.

However, I believe the problem is more electronic than wiring. Remember – my CEL is *not* lit and the engine runs incredibly strong. Also, the sensor that is post-cat for Bank 2 is operating normally. This suggests Bank 2, Sensor 1 is actually operating correctly, that the Bank 2 ECU is receiving the data and adjusting fuel accordingly. For some reason however, the computer(s) are not ‘reporting’ the correct O2 Sensor signal. This, I believe, is what is causing the Fuel Systems to report “CL-Fault: Closed Loop, but fault with at least one oxygen sensor. May be using single oxygen sensor for fuel control.” This in turn, does not allow the emission stuff to satisfy; causing the ECU’s to remain ‘Not Ready’.

At this point, I am open to any and all ideas, suggestions and recommendations. Unfortunately, my R129 Repair DVD is in transit to New Jersey for upgrade from MB, so I’m really blind here.

Cheers,
Jeff

Brett San Diego · #7 11-11-2007, 01:15 AM

Jeff,

I'm having what I think are identical issues with Bank 1 on my 97 S600 sedan. I have an O2 sensor, which I think is the pre-cat sensor, that is flat-lined at low voltage. I'll have to hook up my scanner again and re-confirm which sensor it is. New O2 sensor did nothing to the bad signal on the OBD2 scanner. Swapping MAF's and O2 sensors between banks did nothing. I'm close to beginning the hunt for the root cause of this problem since I'm now having a high, surging idle issue that began with removing and cleaning the throttle bodies. grrrr. That was supposed to be a good thing to do, but it seems to have caused some problems...

Brett

MB-Dude · #8 11-11-2007, 12:33 PM

Quote:

Originally Posted by Brett San Diego

Jeff,

I'm having what I think are identical issues with Bank 1 on my 97 S600 sedan. I have an O2 sensor, which I think is the pre-cat sensor, that is flat-lined at low voltage. I'll have to hook up my scanner again and re-confirm which sensor it is. New O2 sensor did nothing to the bad signal on the OBD2 scanner. Swapping MAF's and O2 sensors between banks did nothing. I'm close to beginning the hunt for the root cause of this problem since I'm now having a high, surging idle issue that began with removing and cleaning the throttle bodies. grrrr. That was supposed to be a good thing to do, but it seems to have caused some problems...

Brett

Thanks Brett for chiming in. Yes, it sounds like we have similar O2 Sensor behaviors, though in my case, the engine performs great. My low O2 Sensor is Bank 2, Sensor 1; should be Driver's side, pre-cat.

Through a series of PM's with forum member myarmar, I am following a theory that maybe the O2 Sensor is operating correctly and attempting to compensate for a very lean condition – i.e.: a vacuum leak. Armed with a vacuum schematic and some wiring diagrams, I am plodding through connection-by-connection. I’ll be reporting progress in a day or so, as some of the vacuum lines have become quite brittle; breaking as I move them. As we know, the V12 compartment gets very hot through normal operation. As a preemptive measure, I am going to replace all the small diameter vacuum lines tomorrow (Monday); assuming I can get the appropriate sized tubing. I also want to swap the MAF’s to see if the problem moves to the other bank, but have to make vacuum repairs first.

Cheers,
Jeff

Brett San Diego · #9 11-12-2007, 12:42 AM

We may have more similar problems than I first thought.

I hooked up my OBDII reader again today. I was able to confirm that it's the bank 2 pre-cat sensor that is flat at 0.0 volts not bank 1. The bad signal from my pre-cat bank 2 sensor is identical to what you show in your attached pic above, flat-lined with an occasional spike to around 0.5 volts. Bank 1 is fine. The O2 sensor signal may be indicating a lean running condition, but the problem I'm having with that diagnosis is that the long term fuel trim is reasonable on bank 2 (actually slightly negative, between 0 to -4%). As far as I know, if the ECU were compensating for a lean mixture, the fuel trims should be running positive. Nevertheless, I'm going vacuum leak hunting myself. I'm concerned about the PCV connection at the back of the engine. When I removed the throttle bodies, I noted that the right side metal PCV tube that wraps around the engine was kind of floppy, which worried me. I tried having a look with a mirror, but I really couldn't get a good view to confirm that the connection in back was solid. The PCV tubes connect directly to the throttle body.

I should clarify that my car wasn't having any running issues before I went and removed the throttle bodies. My condition was just like yours, seemed to run perfectly fine. I think I buggered something up in the process of doing the throttle bodies, and it may have been the PCV hoses.

And in case you haven't checked, yet, the vacuum hose tubing is available by the foot from the dealership parts counter. I bought some a while ago.

As I said, I have switched MAF sensors with no satisfaction. I have also switched O2 sensors (In fact, I just did it again today.) with no satisfaction. Oh, and my OBDII diagnostic checks all come up completed, and usually, the fuel systems read closed loop with fault, but today as I was checking things, I was seeing "open loop driving" for a long time. I'm not sure what that means versus simply "open loop."

Please keep us informed of your progress, and I'll do likewise.

Brett

MB-Dude · #10 11-12-2007, 04:39 PM

Local MB dealership does not sell vacuum tubing by the foot; I had to order it (!) pre-cut in about 5 foot lengths. It’s MB p/n 000.158.14.35 Price is fairly reasonable at US$3.60 per 5 foot section. Oh well, they only had one piece in stock. Ordered 10 pieces since I will be changing all the vacuum tubing on both my SL’s. The first attached pic shows one of the 5 (yes, five!) vacuum repairs I found on the 96 SL600… after taking ALL the plastic engine covers off! It’s amazing what one finds buried under all that fancy covering. Thus, my 94 SL500 is probably just as bad.

While waiting for the vacuum tubing, I went ahead and poked around the vacuum fittings. Lo and behold, I found a loose one – the vacuum connection to the master brake assembly. See second attached pic. This fitting was very loose and does connect to Bank 2. I was encouraged. But after tightening the nut and taking a quick warm-the-engine jaunt, I once again was staring at 0 volt output of the Bank 2 Sensor 1 O2 Sensor. Swapping the MAF’s between banks also made no change.

However, I did make an interesting observation. If I created a vacuum leak (at the vacuum repair in pic #1), I witnessed an immediate decrease in output voltage of Bank 1 Sensor 1 O2 Sensor. Remember, this is the bank that is operating correctly. Within 10 seconds of initiating the leak, the computer compensated for the leak and rebalanced fuel (presumably) so that normal O2 Sensor operation resulted. Reconnecting the vacuum line caused the O2 Sensor to output high, but only momentarily; the return to normal operation was a much faster response. In both cases, I witnessed the post-cat sensor following, though not as radically, the pre-cat sensor.

What this means is that small vacuum leaks – say, the size of small vacuum tubing – will not adversely affect this engine; the computer can compensate. In order for the computer to not compensate, the leak would have to be massive and, thus audible.

As a test, I revisited a previous post of mine…
Failing EGR Valve?
…and disconnected the vacuum line. Even with this fairly large, albeit metered, leak, the computer compensated fairly well. You could watch the O2 Sensor as the valve open/closed, but the computer maintained control.

Conclusion: While I definitely need to replace some vacuum tubing, I do not believe a vacuum leak is causing the problem.

So I’m back to square one. I’ll continue to poke around for vacuum leaks and will replace the tubing. However, I believe this problem is, unfortunately, electronic. Finally, myarmar asked if I could print out the ME-SFI data. Attached is a data snapshot. ProScan (the OBDII scanner I use) captures the data over and over, so the file is a series of successive snapshots. This is good for comparing oscillating or varying data. The file is really a .csv file, but to upload it, I had to rename it a .txt. Simply change the .txt to .csv and MS-Excel will read it correctly formatted. If further shots are needed, but only of a few variables, I can easily set that up.

QUESTION: Is there any harm in swapping the ECU’s between Bank 1 & Bank 2?

Not sure if there is any special information or bank-specific stuff that could cause a problem. Again guys, any and all ideas are welcome.

Cheers,
Jeff

Brett San Diego · #11 11-12-2007, 06:38 PM

Jeff,

We should compare notes more often. I, too, had the mysterious thumping noise that you describe and had no idea what it was. But, I also had the flashing gas gauge and P0455 fault code (gross leak, evaporative emissions system). I found a similarly disconnected hose, but it was in the center of the wiring harness tray where the hoses from the two purge valves meet with the main hose going back to the charcoal canister. I'm surprised you didn't have the fault indications.

http://www.benzworld.org/forums/w140-s-class/1346673-p0455-code-problem-solved-i-think.html

I never read your post because it was titled EGR valve. To my knowledge, the V12 doesn't have an EGR valve (at least the OBDII 1996 and later models). I've never seen it mentioned anywhere in alldatadiy.com nor have I seen anything that might be an egr valve in the engine compartment. The only thing attached to the exhaust side of the cylinder head is the air injection valve.

At any rate, I reconnected the evap system hosing, and the fault code has stayed away, and now that I read your post, I realize that the thumping sound is gone. I never connected the two issues before now.

Brett

lesrrt · #12 11-12-2007, 06:40 PM

When I removed the intake, I noted all the small ports were all clogged with carbon. Both egr valves were also carbon contanimated too; allthe way up to were it bolts to the bottom of the intake. The flimsy metal hose that is around the back Brett, is also part of the emissions recirc sysem...I think; removed them while doing the work; they were also clogged. One thing you can ck, is the air-port plug right next to the oil fill cap. Remove it and see if you can pass a stiff wire. They are probably clogged as were mine. All these tie into the emission/adaptation curves and how the computer compensates. The long runner design will have secondary back pressure gases and will cause carbon build-up. This was a documented problem with GM's TPI engine as on one of my older Vettes. Removing these and servicing/cleaning are a test in your abilty..believe me. Now I know why someone removed the cel before I bought the car. Good luck

PS don't forget to ck the wire harnesses!!

Roncallo · #13 11-12-2007, 06:50 PM

96 SL600. Sounds exactly like one bad ECU and an Engine wire harnes. Visit this guy, he will take your old ones off your hands

Cash for Dead 6 or 12 cyl ECU's and Connectors

John Roncallo

Brett San Diego · #14 11-12-2007, 07:01 PM

Quote:

Originally Posted by MB-Dude

While waiting for the vacuum tubing, I went ahead and poked around the vacuum fittings. Lo and behold, I found a loose one – the vacuum connection to the master brake assembly. See second attached pic. This fitting was very loose and does connect to Bank 2. I was encouraged. But after tightening the nut and taking a quick warm-the-engine jaunt, I once again was staring at 0 volt output of the Bank 2 Sensor 1 O2 Sensor. Swapping the MAF’s between banks also made no change.

Conclusion: While I definitely need to replace some vacuum tubing, I do not believe a vacuum leak is causing the problem.

Does the booster vacuum line connect directly to the intake manifold? I'm not sure mine does. I think it comes off the vacuum reservoir so that your brakes have a check valve protected source of vacuum. I was perusing the vacuum line diagram for my 97 model today, and there is no mention of a vacuum line from the intake manifold to brake booster, just to the reservoir. But, I'll have to look.

Quote:

So I’m back to square one. I’ll continue to poke around for vacuum leaks and will replace the tubing. However, I believe this problem is, unfortunately, electronic.

I think you may be right for both of us, unfortunately. I'm not looking forward to digging up and tracing the O2 sensor wires through the passenger compartment.

Quote:

Finally, myarmar asked if I could print out the ME-SFI data. Attached is a data snapshot. ProScan (the OBDII scanner I use) captures the data over and over, so the file is a series of successive snapshots. This is good for comparing oscillating or varying data. The file is really a .csv file, but to upload it, I had to rename it a .txt. Simply change the .txt to .csv and MS-Excel will read it correctly formatted. If further shots are needed, but only of a few variables, I can easily set that up.

QUESTION: Is there any harm in swapping the ECU’s between Bank 1 & Bank 2?

Not sure if there is any special information or bank-specific stuff that could cause a problem. Again guys, any and all ideas are welcome.

Cheers,
Jeff

From what I've read on alldata over the last couple of days, I don't expect them to be interchangeable. As far as throttle body control goes, the bank 1 ecu is sort of a master ecu while bank 2 is a slave. The throttle position sensor information feeds only to the bank 1 ecu, and then this information is relayed by bank 1 to bank 2.

I'll be checking my brake booster vacuum connections this evening. Thanks for that one.

The first pic that you show above is of the vacuum lines to the manifold absolute pressure sensors. Funny you mention those, too, because I broke those very lines a while ago while I was doing my water pump and coolant pipes at the front of the engine. That's why I had to go to the dealer to get some vacuum tubing.

Brett

myarmar · #15 11-12-2007, 07:46 PM

I think all differences in ME ECUs are external and they are interchangeable. EPC gives same p/n for left and right ME ECUs. I know for a fact the old LH ECUs are the same for my 600SEL.

Looking at you csv file, it may have an answer for your problem. What if one sensor was declared bad and no longer used for fuel control. I don't know, but maybe this ECU has a capability to control using O2 sensor from the other bank? Look at the xls file attached. I re-formated your data and made a chart. I does look like fuel trim for bank 2 follows bank 1. We don't have enough samples to be sure.
At this point I would reset adaptations and all faults in the ECU memory (if possible) and see what happened as ECUs re-adapt. Swapping ECUs may be a good step too.
Mike