brettob

The ECU on the car was bad (defective coil driver found in it) and sent it off for repair. ECU driver was repaired, but then other problems with the ECU became apparent (it does not read the RH Intake and Exhaust Cam Sensors in the Star Tool - they are fixed values and not changing when running). This then leads to injectors 4 and 6 failing to work as soon as the Check Engine Light is turned on due to the cam sensor errors (my assumption?). So I picked up a used ECU and had the repair place "Clone" the original. The cloned ECU now has the same errors. But with my oscilloscope connected to the ECU inputs for these sensors, the proper signal is clearly entering into the ECU.

My conclusions are the ECU has either 1) hardware or 2) software/firmware/programming problems. Hardware issues seem unlikely to have 2 ECU's with the same defect. This now leads me to searching this forum to figure out what I need to really be asking to get my ECU repaired/update to the latest and greatest.

a) can the software/firmware/programming actually become corrupt and then when cloned, the corruption follows into the donor ECU?

a) if I want someone to update the ECU to the latest software/firmware/program for this vehicle, what terms and Mercedes speak do I need to ask? What tools/service do they need?

b) can this be done outside of a Mercedes Dealership using an independent?

c) if this is to be done at a Mercedes Dealership, what would one expect to have to pay for (ie, maybe 1 hour of labor plus the cost of the software/firmware/programming update)?

d) I have a Star C4 tool and can someone actually send me what I need to do an offline transfer (meaning not connected to Mercedes Corporation) of this update myself to the car using something like the developer mode? If so, any recommendations on who I could contact to get this sort of update?

e) can anyone point me in a better direction if my questions and thoughts are heading off in the wrong direction?

Thanks for any clarifications anyone can provide. New to this stuff, but fascinated to both learn and get this car and others running.

CaliBenzDriver

I think there may be issue with the car itself (nasty GND, or connectors, or power supply,...) rather than multiple ECU having the same problem after cloning.

As you've found out software + hardware is a fragile combination with countless ways something won't work right.

At this stage, you want to update firmware to refresh your used cloned ECU. Your best bet is to contact @BenzNinja to get this working again.
🤞

JettaRed

iTrader: (1)

Ditto! Reach out to @BenzNinja .
www.benzninja.com

brettob

Thanks for your thoughts CaliBenzDriver. I have checked the grounds, connectors, power and they all appear to look and read great. The main reason I believe it is something in the ECU is that when you start the car and the RPM's are held above 1200 or so, the check engine light doesn't come on at all. During this time the Star tool clearly shows the Left Bank Cam Sensors Values are fluctuating, but the Right Bank Cam Sensors are always frozen to the default values. But if I use my oscilloscope and measure the signals for all 4 of these sensors at the ECU connector, all of the sensors work great and have similar wave forms. Then when I let the RPM's drop to idle, exactly as soon as the check engine light starts to blink, only injectors 4 and 6 stop receiving the signal to inject fuel and the car starts running really rough and makes the car non drivable.

With this super strange phenomenon, I can't imagine how a ground or connector could be causing this to happen always at the exact same time as the check engine light coming on. Plus it seems super weird for a lack of RH Intake and Exhaust Cam Sensor inputs to normally, by design, force only 2 LH injectors (#4&6) to suddenly stop functioning (which are also on the opposite side of the engine). Just crazy and too weird to imagine it could be anything but a problem coming out of the ECU. Seems to me to sound like corrupt software/programming would be the most likely problem (and it was copied over to the "new" used ECU).

Now I suppose, in theory, my used ECU could have the same hardware failure, but the probability of this would seem rather small unless this is a common failure of the ME9.7 ECU's.

Thanks for your great suggestion to contact BenzNinja. I will be reaching out to him to see if he can help. Much appreciated the direction and suggestions!

JettaRed

iTrader: (1)

Or, it may not be the ECU at all. Is the affected engine your M113? The good thing is that your car doesn't use direct injection and new injectors are relatively cheap. You can even get some remanufactured injectors, balanced and matched, on eBay for cheap. It might be something going on with those two injectors.

1. You can try to swap injectors on cylinders 4 and 6 around and see if the problem remains on those two cylinders.
2. You also have two spark plugs per cylinder. Have you replaced your plugs?
3. How did you know the original ECU was corrupted?
4. Did you replace the cam sensors?

brettob

Affected engine is a M272 (6 cyl). Here are answers to your questions or comments:
1. Injectors were swapped to other locations and the problem stays with 4&6 as the signal to fire the injectors drops out as soon as the check engine light comes on. Also each injector tested off line for proper function. All are good.
2. The later M272's were moved back to a single spark plug. Plugs have been changed. Problem is not spark, but injectors 4&6 stop injecting when the CEL comes on.
3. Original ECU corrupted??? This is the true unknown, but the most likely cause at this stage. But the ECU works fine and fires off all injectors at high idle, and this should also indicate that the hardware external to the ECU is fine. Proper waveform and voltages are exhibited at the injectors and the car runs smooth. But once the CEL comes on at the low idle stage, injectors 4&6 suddenly stop providing fuel to the engine. Verified by oscilloscope. Of course there is a theoretical possibility that the hardware in the ECU is failing somehow, but this would seem to be to be a low probability with 2 different ECU's off of 2 different cars. And from a programming stand point, it is hard to imagine any engine development engineer would design the software to kill only LH bank injectors 4&6 when the RH Cam Sensors set the CEL. In fact, I am not sure what the case would be for a engine development engineer to on purpose kill only injectors 4&6 under any condition (or maybe only in a case of deactivating cylinders for better efficiencies- but this probably would be a rotating pattern between all the cylinders)
4. Cam Sensors were also swapped from RH to LH. Problem does not move with the injectors, but stays with the original 4&6 injectors stopping and always the RH side Cam Sensors not being read by the ECU (when by the oscilloscope, the RH Cam Sensors are actually providing the correct signal into the ECU when measured at the ECU connector.

Engine is a 6 cyl, so the RH bank is 123 and the LH Bank is 456.

All great questions, but so far, still leads me to believe I have an ECU issue. Either corruption causing the strange affect of shutting off cyl 4&6 when the CEL comes on, or an ECU hardware issue that is causing the CEL to come on at idle while not reading the RH Cam Sensors at anytime.

Anyone else have any thoughts on what else to try? or any answers to any of my original questions in the first post indicated by a) thru e)?

JettaRed

iTrader: (1)

Sorry, but I missed where you mentioned which engine you were talking about. So, here's a crappy, but readable picture of the firing order of the spark plugs. We can assume the fuel injectors are firing in the same order. What are the DTCs when this happens?

JettaRed

iTrader: (1)

Now that we know which car and engine you are talking about, does your car work with XENTRY or DAS?

a) Unlikely unless someone was futzing around with Vediamo or DTS Monaco, etc. Has anyone been trying to reprogram the ECU?
a-2) Check with @BenzNinja. In addition to XENTRY/DAS (or whatever software is used), the person updating your firmware will need an account with MB to access the latest versions.
b) @BenzNinja can update firmware, I believe. You need to ask him directly.
c) No idea what the dealership would charge. AND not all dealerships charge the same. Best to ask the dealership directly.
d) See answers above.
e) See answers above.

brettob

The car is a 2007 SLK 280. When I use the Clone Star Tool, I go in through XENTRY menu to select the car, then I think when I click on the scope icon, it enters DAS. So I think I am running DAS during my diagnosis.

Codes that come up when the car is allowed to idle are as follows:

P0016, P0017 Which are RH Bank Cam Sensor Errors
P0300 Engine Misfire
P0304, P0306 Which are Cylinder 4&6 Misfires

Thanks for all your thoughts and guidance. Trying to figure out the Mercedes programming stuff on line is a little bit confusing to me. Some things I read say only MB can reprogram the ECU, some say you need SDN, some call it reprogram, some call it reflashing, some call it firmware, etc. I have not found anything super clear to say step 1 do this, step 2 do that etc.

brettob

Oh, also, I do not believe that anyone was messing with reprogramming the ECU. Great looking car was purchased not running for dirt cheap, but the engine fundamentals were good (ie, turned over fine and great compression, etc.). A friend needed a car and this seemed to be a great one if I can get it working properly.

For me, it is an education and a personal challenge (I always want to WIN over the machine!!!). For them, it would give them a nice car at an affordable price.

Car was pick up from a shade tree mechanic trying to fix it for a friend. Either this was maybe a victim of a lightning storm or the shade tree mechanic fried some circuits? 3 or 4 coil packs bad, Crank Sensor bad, and MAF sensor bad plus the wonky things with the ECU.

Would love to learn how to use my Star Tool to reprogram or reflash (whatever the right terminology is) this ECU with the latest and greatest to verify that my diagnosis is correct... an ECU problem. Maybe I just need a little guidance on how to do this along with the files needed for this ECU/car. VIN number WDBWK54F07F154049 (2007 SLK 280 with a M272 engine).

JettaRed

iTrader: (1)

I don’t think you’re going to find a step-by-step procedure for fixing the ECU. Such knowledge and skill are valuable and not freely given.

You will need to contact an expert like BenzNinja who not only has the knowledge, but also the authorization to access Mercedes servers, etc. and actually perform whatever programming is necessary.

Yes, the car is old enough to use DAS, and you may be able to make changes because you have Developer Mode, but can easily brick your car if you don’t know what you’re doing.

brettob

I have reached out to BenzNinja and awaiting what his advice is. Hopefully this is something I can do with the Star tool if I can get someone to send me the needed files so I can do this myself. Reality is it could go to the dealership for reprogramming and probably much more cost, but more than half of me want to learn what is possible and also let me WIN over the machine.

Also if anyone else has access to these files and some basic directions on how to do this, please reach out to me. Thanks so much. I am determined to WIN over this thing!

brettob

oops, made a posting mistake on this one. not sure how to delete it

CaliBenzDriver

You are doing great as far as troubleshooting goes... I think your ECU is genuinely ok.

Something is causing your ECU to get in this weird LIMP MODE with shutdown injectors.
Google that search about 4 & 6 injectors - I have to go work on my own car: AAC passenger actuator this afternoon.

Double check the crankshaft position signals, I think CKP is a dual-path sensor. See that your CPS has no oil-in-harness" or bent-open female side connector pins to create a poor signal connection.

Read your fuel pressure data.
Compare your LTFT: 1 good + 1 bad ??

Eye ball all live sensors data for any odd missing input.

Pls let's not be hell bent on this being two bad ECU.

+++ bad camshaft position...
The code you are getting may be related to the camshaft sensor giving positions tge ECU flag as bad... I am talking about a correlation issue. I know the code does not call for poor "timing correlation"... this sensor code points "camshaft issue" -
Try to swap Bank1 CPS with bank2, you know.
Limp chain tensioner? Go look .


You have all it takes to nail this one fixed

brettob

Hi CaliBenzDriver! I really like your thinking!!! Now you are giving me a deeper investigation challenge, and let's see where this goes. It would be great if it was not the ECU. I also don't think it is likely for 2 ECU's to be bad.

We know the first one was bad as the chip that controls the coil packs was fried. This was replaced and then the car would start. But it now had the wonky and strange situation where it works fine above idle (although the MB Star Scan Tool does not see varying RH Cam Sensor values like the LH Side always exhibits --- and also I don't see a LH O2 sensor varying, but so far, this is not setting any codes). But when you go to idle and the check engine light gets set, it immediately sets the following codes:

P0016, P0017 Which are RH Bank Cam Sensor Errors
P0300 Engine Misfire
P0304, P0306 Which are Cylinder 4&6 Misfires

Maybe I am stuck on the fact that the RH Cam Sensors never vary in the Star tool view. This would indicate to me that the ECU is not seeing any signals coming into it from these sensors. But when I put my oscilloscope on these 2 sensor inputs at the back of the connector going into the ECU, they are sending a great signal just like the LH working Cam sensors do. Using this a the only facts that I know about these signals, it would point to something in the ECU that is causing this problem since it see the LH properly but the RH Cam sensors are not varying. This is where I was thinking hardware or corrupt software problems. Hardware seems like a long shot with 2 different ECU's having the same problem (unless this was a common problem with the ME9.7 ECU's.... but I am not finding this as a common problem with this according to my Google Searches). So this is why I am asking about the possibility to reload or reflash the ECU with the latest and greatest to see if this eliminates a possible corruption issue in the original ECU being cloned into the newer used ECU.

I tried searching many ways about cyl 4&6 injectors and limp modes for Mercedes. Couldn't find anything with what I was trying to search with. But maybe you are on to something. It could be triggering a Limp mode, but I have not found what that could be other than the RH Cam Sensors that send a signal to the ECU, but the ECU can not see the values in the MB Star Tool.

It is freezing temps here now so I will do more investigation and report back.

As for your other questions: Crank Sensor is new and old one was dry - wiring is clean. Will check the pins once I can get to the car and it is a little warmer. Checked fuel pressure with guages and all is good. Will check actual values when I check the pins above. Will also check the Long Term Fuel Trim and compare sides once I get out there to run it some. Because the fault codes set quickly upon idle, I am not sure what my LTFT values will look like or represent since the car has only been started and troubleshot. Runs super bad with #4&6 injectors not providing fuel. Probably drive-able, but really rough. I am guessing I would need to drive it for some time to see valid LTFT values? Will study other values when running it next. Have already swapped LH and RH Cam sensors. Problem doesn't move with sensors, but always stays on RH side of engine. Chain checked with marks by service manual to make sure the chain has not stretched too much or that the earlier motor recall for worn gears. All looks good here, but I have not confirmed the tensioner for tension. Perhaps this might not be strong enough to return the chain to a good position when the RPM's are dropped to idle? Will check it out. Good thoughts!!!

The thoughts on the chain tensioner could make some sense if the RH Cam Sensors were actually being read by the ECU. But since I can't ever see varying RH values while it is running (like I can on the LH side), I don't know how the ECU would then see a chain elongation or slack issue. Sort of leads me back to the ECU corruption theory... maybe corruption was cloned to the 2nd ECU?

Thanks for the kudos on me being able to win over this machine. I am determined to learn and win in any case. These things were designed and made by humans so we must be able to figure out how to fix it (but just might need a little great advice such as yours and others on these wonderful forums)! We will WIN

JettaRed

iTrader: (1)

What are the "static" cam values being reported. (Interestingly, mine are always "static" on both banks when reading live data with my scan tool.). These are mine and they don't change with the engine is running. Is this what you are measuring?

CaliBenzDriver

okay, go make yourself a tea. Today we have bad news and we have GOOD news with possible SHORTCUT fix.

You said hardware and you were right with that. Software survived as good as new.


> GOOD / BAD / PLAN-B...
A1-- The good news is computer is A-1 okay

A2-- The bad news is the engine repair calls for its retirement.

B! -- You said you wanted to win this machine back to life. We are going to give it a second life.

-- Thought process... you said:
- it runs great above idle! Where code sets and shuts off injectors.
Once you know what happens outside idle, that reads like good hope. Poor oil pressure loosens tensioner chain slack, camshaft timings get shifted

- That's where software takes over to dislike the loose position of camshafts.


> Long story short...
- Your chain when loose has to much slack: duh!
That's what tensioners work to keep consistent.

- Complete job calls for a new chain (forget the balance shaft 30Hr) with tensioner.

Everybody likes to blame that as being a cheap oil issue. At this stage this translates as being engine wear.

Your engine balance shaft has reportedly expired. Mercedes repair calls for a 30 hours job with engine out...


> RIP.. Maybe NOT ??
Well that's the standard story but since you report that engine runs perfect above idle, I am going to recommend replacing your chain + tensioner.
Double check that solution if it has helped your engine line.

This may be good money against bad money... at least it will drive you around for a while perhaps 25kMi.

Weak tensioners are starting to appear like a favorite. Here we see tensioner forcing ECU in limp-mode.

Check out this
Good chain timings Tensioners are important.
Hope that makes sense
🤞


++++ Repair plans...
We are are trying to find a shortcut to deal with bad correlation.

gear driven exhaust... VVT pin bad??

Exhaust camshafts are interconnected by gear not by chain. This leave very little room for freeplay besides VVT Lock pin, right?

So not just long chain + tensioner + Bank1 VVT Gear.

Test:
rotate crankshaft CW
does VVT skips forward on bad lock pin?

brettob

Yes, this is what I am measuring. Your software doesn't seem to be the Star DAS stuff. My just shows raw numbers without any symbols.

The Cam sensors vary when running as follows:
LH Exhaust Cam between -19.66 to -19.88
RH Exhaust Cam fixed at -20.00
LH Intake Cam between 36.21 to 36.41
RH Intake Cam fixed at 36.21

I guess since my Star tool shows the left to vary, this is why I expect the right to vary. I will have to try another scan tool to see what it says. In the Star system, it allows one to do a Sensor check. It has you run the engine and verify with a yes or no if the values are varying. If not, it sends you troubleshooting the sensor, the wiring or the ECM if I remember it correctly.

This is the main reason I am sort of hung up on my oscilloscope shows the proper signal going into the ECU, but I can't see it in the software... so I have been assuming a hardware issue within the ECU or some corrupt software that got cloned into another ECU.

brettob

CaliBenzDriver, once again, I really like your thoughts... very similar to what I have been thinking. A couple of things on the good side of what I am running into: 1) this engine serial number is above the recalled balance shaft sprocket (so I think I am good there), and 2) I have seen the alignment video of the marks to check for timing chain issues and I believe this car is spot on (but very difficult to get your head in there to see it squarely... so I used mirrors... and if there is anything off, it is only off a hair). I will also take your advice and figure out how to check to see if I have any freeplay in the VVT lock pin. WIll also report back on this when I can get to the car again to work on it.

It makes some sense to me that when the engine is running at higher RPM's, the chain tension is different than when it is decelerating down to idle. If the tensioner is weak, then perhaps the slack moves to another place and then the Cam Sensors detect a problem with correlation, then hence setting the Check Engine Light. So easy enough, I will get a tensioner on order as this is a super low cost item to replace and good for a preventitive. Have not yest checked to see how easy it is to change, but if there is clearance, I imagine it is only a few minute job. WIll report back once I get it installed.

This theory is a good one. Hope it works.

But what keeps going through my head is why doesn't the RH Cam sensor values change like the LH ones? And the next thing is why would a development engineer drop out #4&6 injectors if a Check Engine Light goes on? Surely they could have the engine run in reduced power by changing the mapping conditions or something else so the low power mode runs much smoother? It would be great to know if anyone out there knows that Mercedes actually uses dropping of 2 injectors as a means to go into a lower power mode or limp home on a car like this. If this is correct, then things are making more sense. I have search the net and come up empty as to what technique Mercedes used to get lower power out of their engines when they go into limp home mode. Anybody aware of how they create the lower power for Limp mode?

CaliBenzDriver

I think we can learn from this other member currently dealing with the same troubles here: P0016 after all these...

• timing chain
• chain tensioner
• chain guides
• ok-Balancer shaft/sprocket

Earlier today, I concluded above pointing at VVT Gear lock pin needs to be tested.

Replacing the long list of worned parts did not alter the timing issue between each camshafts.
The computer complains that the intake and exhaust camshafts positions are wrong in relation to each other. THEY ARE LOCKED BY GEAR BESIDES the intake camshaft is connected by a VVT phaser.

These phasers have an internal lock pin that wears out making the position uncontrollable without proper oil pressure at idle.

There is test where the crankshaft is rotated by 2 Turns while observing the VVT ... if at any point it shifts forward you know it need replacement.

Hope this helps point you towards a solution.

brettob

CaliBenzDriver, great link to something very similar. Enthused to see the link and what he did to find and resolve out of correlation Cam Sensor values. Was writing a reply last night and while I was in the middle of it our power went out and I lost the reply. But armed with new information and a possible cause to my problem, this morning I rotated the crank many times slowly by hand and no strange noises. Then I took off the covers on the RH bank to watch what is happening when I rotated the engine by hand. Absolutely no clicking, ticking or jumping of anything. Watch both gears, and the variable teeth on the Exhaust Cam. I didn't tear any of it down to see if I could find the suspect pin, but I am assuming the clicking or jumping would have shown up in the VVT gear movement compared to the Exhaust Cam Gear. Doe this sound like I am looking for the correct thing? Bottom line is I could not find anything strange. All looks pretty clean in there and almost no wear.

So this leads me back to a couple of thoughts. Tensioner is weak or leaking at idle causing the chain slack to show up elsewhere and then causing the P0016 and P0017 Cam Correlation Faults. New tensioner on order and should show up next week. Another possibility would be to have oil that is too thin in nature which is causing the tensioner to be weaker than normal. And finally, maybe I am getting back to some of my original thoughts as the ECU is corrupt.

If someone has the knowledge or can point me in the direction so I can find out how Mercedes M272 Engines actually go into reduced power output during a Limp Mode it would greatly be appreciated (meaning do they drop injectors from providing fuel on a couple of cylinders like 4&6 in a V6 case). If this is actually true, then it still might be possible for this to be a car side problem and not a corrupt ECU.