Mercedes-Benz Forum banner

1 - 20 of 48 Posts

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #1
All,

I've read many posts on the forum but none seems to address my issue, which is that ever since I replaced the vacuum lines on my 1997 SL600, the engine is rough at idle and during acceleration, although it is as smooth as ever at steady speed and during deceleration. Idle is rock steady at 650rpm. In the last couple of months I have replaced the fuel filter, fuel pump, fuel pressure regulator, and both camshaft timing solenoids, but the problem is still there.

My Chinese C3 SDS reports only the misfiring fault codes P0300 and P0305 for both A and B. What is the difference between A and B? Are these the left and right cylinder banks? Why are there two ECU menus called ME1 RE and ME1 LI? I thought these would correspond to the two cylinder banks since in German REchts = right and LInks = left, but both report data for all 12 cylinders! I ask because I'm looking at lots of actual values but having trouble deciphering the data. Many data fields have a highlighted checkmark, none of them have a highlighted "-F-" for fail, but several of them have nothing highlighted. What does it mean when nothing is highlighted?

Based on what I've read in other threads, my current prime suspect is a timing issue, perhaps the camshaft position sensors. But how can I determine if these are good or bad before replacing them? Maybe this is a clue: unhighlighted actual values include:

1519 Right camshaft timing
1522 Left camshaft timing
1525 Right camshaft timing actuator
1533 Left camshaft timing actuator

Plus, ME1 LI tells me that Camshaft actuator is OFF.

Any help at all would be greatly appreciated, as this is driving me nuts!

Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
No idea about your problem, but the RE and LI are indeed the right and left bank ECUs. Maybe they show values for all cylinders just to make realtime diagnostics a bit easier? Codes for both banks register on the right bank ECU. Actual sensor values etc are only shown for the particular bank though, I think.
 

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #3
Some sample data

Thanks juuso, that helps a bit.

To stimulate the conversation, here are some examples of the actual values I'm getting:

1. The left and right upstream O2 sensors both have erratic voltages, while both downstream ones are very steady. Is this normal, or an indication of failed or failing upstream O2 sensors?
2. Both the left and right MAP sensors react very little to throttle inputs. They're both at around 200 mbar at hot idle, but I can't get them to go much higher than 300 mbar, even when accelerating quickly to over 3000rpm. Is this normal, or an indication of failed or failing MAP sensors?
3. There is a Smooth Running display that I don't understand at all. It has a number in units of 1/s2 for each cylinder that is normally zero, but I can get the number to go non-zero on cylinder 5 by blipping the throttle. It also goes nonzero on some cylinders in the other bank if I blip the throttle harder. There is a second display that shows faults for each cylinder, which shows zero for all cylinders but #5, which shows a number that increases with time (it was up to 22 by the end of my testing). I'm sure this is related to my P0305 code, but I don't understand how.
4. Both Camshaft timing sensor signals switch between 55 and AA, which I don't understand the meaning of, but I assume this is normal because the overall test results summary tells me the Camshaft Hall sensors are okay (checkmark highlighted).

I'm sure HHTWIN is trying to tell me what is wrong with my car. Can anyone help me interpret these results, or point me to a reference that will help me understand them?

Thanks
Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I guess you haven't found the test and adjustment data PDF yet? That should point you in the right direction at least with some of the values.

I'm actually trying to find an issue with an occasional slightly rough idle, and I suspect some power is missing, so I too am curious about real life experiences on what each value should be and what to look for. I checked the values at idle the other day and took photos of the relevant HHTWIN views. I'll try to post the values for reference at some point when I have time.
 

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #5
Thanks again juuso, that pdf helps a whole lot! It tells me my camshaft position sensors are behaving normally (as I thought). It tells me that my MAF sensors may need cleaning as they both registered around 8 kg/hr at idle when the expected range is 12-18 kg/hr.

Unfortunately it doesn't tell me what the expected range of pressures is for the MAP sensors.

It also says the upstream O2 sensor voltages should oscillate around 300 mV after 2 minutes idling. But it doesn't say if this oscillation should be visible as large and seemingly random changes in voltage levels as I am seeing, or an essentially steady voltage as I see with my two downstream O2 sensors. The pdf doesn't mention the downstream O2 sensors.

We're making progress, but we need more data!

Cheers
Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I checked the idle values this morning after driving about 20 minutes, mostly slow stop&go in the city. These are naturally just snapshots and I didn't check all of them many times, just recorded what I saw. My symptoms are different to yours, as said, but maybe these help a bit. And maybe my issue gets diagnosed by accident here - not trying to hijack your thread though :)

The downstream O2s were pretty stable for me too and the upstream ones were oscillating, but I didn't really check by how much (I was in a bit of a hurry).

My air flow was about 8kg/h too, though it sometimes went up to 12 at least on the right bank. Intake manifold pressure was around 330mbar.

The "smooth running" values were oscillating a bit. For the most part they stayed at 0 but sometimes some went up to around 0.5 (one cylinder went just above 1 once). When I gave even a little bit of gas, all went to 0.

The values that struck me as strange were the purge ratio which was lower than specified on the right bank, and "Throttle valve stop programmed" which was NO for both sides. Apparently the system should relearn the value when the ignition is turned on for 60 seconds and then turned off for 10 seconds. I did that but I don't know yet if it made any difference. Maybe some cleaning is needed?

Code:
                                            Specified values (idle)     ME1RE       ME1LI
RPM                                         600-750                     650 +-20    650 +- 20
Coolant temperature (C)                                                 92          93
Air mass (kg/h)                             12-18                       8 - 12      8
HFM-SFI voltage (V)                         0.6-0.9                     0.64        0.64
Intake manifold pressure (mbar)                                         334         324
Ingnition angle (CA)                        5-20                        7           8
Injection time (ms)                         3-5                         2.8         2.6
Camshaft actuator                           OFF                         OFF         OFF
Camshaft Hall sensor                        55-AA                       55          AA
Intake air temperature (C)                  >20                         34          41
Purge valve on/off ratio (%)                10 +-5                      2           6
Battery voltage (V)                         10                          13.7        13.8
Potentiometer actuator
r1 (V)                                      4.0-4.6                     4.47        4.49
r2 (V)                                      0.3-0.9                     0.49        0.51
Accelerator pedal position sensor           0.1-0.4                     0.34        0.38
Closed throttle recognition                 ON                          ON          ON
Wide open throttle recognition              OFF                         OFF         OFF
Torque (Nm)                                                             48          45
Throttle valve angle                                                    0.6         0.9
Throttle valve stop programmed              YES (learned)               NO          NO
Lambda control upstream of TWC (%)          0 +-25                      1           -3
O2S voltage upstream of TWC (mV)            -200 to +1000               805         746
Lambda control downstream of TWC (ms)                                   0           0
O2S voltage downstream of TWC (mV)                                      453         443
Self-adaptation Part load factor            0.77-1.28                   1.04        1.03
Self-adaptation Idle speed (ms)             -1 to +1                    0           0
AC compressor                               OFF                         OFF
Purging                                     ON/OFF                      ON          ON
Safety fuel shut-off                        OFF                         OFF         OFF
 

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #7
I checked the idle values this morning after driving about 20 minutes, mostly slow stop&go in the city. These are naturally just snapshots and I didn't check all of them many times, just recorded what I saw. My symptoms are different to yours, as said, but maybe these help a bit. And maybe my issue gets diagnosed by accident here - not trying to hijack your thread though :)

The downstream O2s were pretty stable for me too and the upstream ones were oscillating, but I didn't really check by how much (I was in a bit of a hurry).

My air flow was about 8kg/h too, though it sometimes went up to 12 at least on the right bank. Intake manifold pressure was around 330mbar.

The "smooth running" values were oscillating a bit. For the most part they stayed at 0 but sometimes some went up to around 0.5 (one cylinder went just above 1 once). When I gave even a little bit of gas, all went to 0.

The values that struck me as strange were the purge ratio which was lower than specified on the right bank, and "Throttle valve stop programmed" which was NO for both sides. Apparently the system should relearn the value when the ignition is turned on for 60 seconds and then turned off for 10 seconds. I did that but I don't know yet if it made any difference. Maybe some cleaning is needed?
Juuso, nice spreadsheet. And I'd be quite happy if we're able to sort your problem in this thread too. Your data look very similar to mine, which is reassuring. Here are some observation based on things I learned since my last post.

A mechanic friend of mine told me the downstream O2 sensor outputs are usually used to validate the upstream O2 sensor values by comparing them. In both our cases this comparison would suggest faulty upstream O2 sensors. However, apparently O2 sensor output depends on a healthy engine. An engine with a problem such as a misfire or vacuum leak can make an O2 sensor behave as if it's faulty. So maybe that explains the discrepancy between our upstream and downstream O2 sensor outputs. If this is incorrect, I would love to be set straight.

Based on my recent experience, I'm beginning to think that large (above 1.0) Smooth Running values indicate a spark problem on that cylinder, most likely a failing coil on the later cars with individual cylinder coils. I was able to diagnose my bad coil by moving the coil from Cyl 5 to Cyl 4, and the high Smooth Running value, along with the fault count on the next page, moved with it. Small fluctuating values like yours may indicate coils or connectors or spark plugs that are on their way out. Again, I would love to be corrected if this is wrong.

I'm sorry but I didn't note the purge ratio or "Throttle valve stop programmed" status when I did my tests, and I won't be doing them again for some days now. I will try to let you know the next time I run a test. Other than these, nothing jumps out at me in your results as peculiar.

EDIT: There was actually one thing I noticed in your data, the discrepancy between the intake air temp values on each bank. In my data both banks reported exactly the same value of 27C when the car had been idling for a few minutes in my garage and the external air temp was 18C. Not sure if this discrepancy in your data is significant, but I thought I would mention it.

Cheers
Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I think the idea of the downstream O2 sensors is to validate the performance of the cats by comparing values to the upstream ones. So some kind of difference is to be expected, but I don't know how much fluctuation there should be on either end.

My fault counts are zero and there are no misfires which makes me think there are probably no issues with ignition. A slight vacuum leak somewhere could be an option. Also, I removed the intake boots and cleaned some gunk out of the throttle bodies today during my lunch break (huh, some things are surprisingly easy and accessible on the M120!), let's see if that makes any difference.

Thanks for posting your findings too, Tom. It's really useful to be able to compare the values and observations.
 

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #9 (Edited)
I think the idea of the downstream O2 sensors is to validate the performance of the cats by comparing values to the upstream ones. So some kind of difference is to be expected, but I don't know how much fluctuation there should be on either end.
I don't know enough about it, but would the cat change the oxygen content in the exhaust gases that the O2 sensors are measuring? My understanding - from my mechanic friend - is that the main difference is that the upstream O2 sensors don't benefit from the cats heating the exhaust gases, so the downstream O2 sensors reach their operating temp range first. But after some time running, the system should reach thermal equilibrium and all O2 sensors should be at nearly the same temp, and give nearly the same readout if all is well.

But as always, I am willing to be educated!

Cheers
Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I was browsing through the STAR TekInfo site again and found that there is some info on what the smooth running values mean:

The cylinder balance test must be performed at idle. If the misfire recognition shutoff threshold is exceeded, a fault will be stored and at the same time the fault count will increases. After repairs, clear the diagnostic memory.
So unless the values go above the misfire recognition threshold, there shouldn't at least be anything serious. Would still be great to know what the values should be a for a 21-year-old engine :)

I noticed your comment about the intake air temperatures only now. I think the reason is simply that I took the right bank value soon after driving into the garage and the car was idling for at least several minutes before I took the left bank one. The left bank intake tube area (which is on the right side of the engine) is also more crowded with all the reservoirs full of hot fluids so I'm not surprised it gets hotter faster when the car is stationary.
 

·
Registered
1996 S500 & 1997 SL600 & 2013 C300 4Matic
Joined
·
67 Posts
Discussion Starter #12
I agree, that is useful information. But based on my experience, you can still use the smooth running values, and especially the fault count, when blipping the throttle from idle to diagnose ignition problems. I say this because it helped me solve my problem, which is now (mostly) sorted - I have replaced the ignition coil on cyl 4, where the nonzero values and rising fault count had moved when I moved my suspect coil from cyl 5 to cyl 4, and this completely cured my rough idle and rough acceleration. I still think the car hesitates upon initial acceleration, which may or may not be due to a failed MAP sensor, but the car is perfectly driveable now, so for the moment I am happy.

I wish you much luck in solving your problem, Juuso!

Cheers
Tom
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
Yeah absolutely, if the fault count goes up there's something wrong. I just meant that it seems to be normal if the figures aren't constant at zero as long as they stay low. Anyway, great that you got your issue sorted out!
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
Does anyone know how the throttle valve is supposed to behave when the engine is not running but ignition is on and the accelerator is pushed? I took the intake boot off and checked while a friend floored the pedal, and the valve opened probably something like 20%, if that. It's not mechanically stuck or anything, as it opens easily by hand on both banks.

Also, is the position learning process (ignition on for 60s, not touching the accelerator) supposed to automatically learn the wide open position too? The servo was clearly doing something during it, but it looked like it was just closing the valve, and didn't open it at all.
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
Well this is interesting. Today I decided to just try and reset the adaptations for the ECUs to see if that makes a difference to my lack of power or idle. According to Star TekInfo, on ME-SFI engines, after this you have to accelerate to 3500rpm in 3rd gear and then coast down to 1200rpm at least 3 times. That's a bit difficult to do over here on public roads but I tried anyway. I probably should have been a bit gentler with the throttle while doing this but what ended up happening was that after a few runs (some of which I had to abort due to traffic etc), the engine lost power. First it made a weird noise (not loud) somewhere above 5k rpm and lost power there.

I stopped the car and engine, hooked up the Star and there were no fault codes. I then continued driving with the Star connected and noticed that nothing happened above 4k rpm and even below that it felt I had 250hp missing. Otherwise the ride was smooth so I drove home. The Star laptop ran out of juice so I couldn't look at the live data for very long and didn't get much out of it. When I got home and charged the laptop enough, I checked the codes again and there was P0300 (random misfire) accompanied by misfire codes for cylinders 3, 4 and 5. Funnily enough the car idles rather ok and the smooth running values show almost constant zero at idle for the misfiring bank, but for the other bank there's slight fluctuation, with values going above 2 sometimes. All zeroed out when I blipped the throttle, as before.

I reset the adaptations again, and this time I also reset the crankshaft position sensor adaptations (there's a separate reset for it, not sure if it's included in the "bootstrapping" task in HHT). Later I'll try to do the 3500 -> 1200 rpm exercise again, this time without revving higher than 4k until it's done. Let's see how it goes.
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I'll continue my monologue :)

The problem didn't go anywhere. I did a long drive yesterday evening, including a few (I think) successful 3500 -> 1200rpm exercises, and afterwards the misfiring started again at 4000rpm. This morning there were codes for cylinders 3, 4 and 9, so also on the other bank. After clearing the codes I drove my short commute and did one hard acceleration in first gear, and it pulled rather OK until 5k when it started misfiring. No codes were stored this time though.

I have a theory. Ever since I got the car in June, I've suspected some power is missing, and in the absence of any clear culprits, I've been thinking the cats are clogged. Could it be that this is indeed the case, and before I reset the ECU adaptations, the ECUs had gradually adjusted themselves to take the deteriorating air flow into account, and now that I performed the reset, the factory defaults just don't work?

I read somewhere that the ME-SFI takes 10 trips of at least 20 minutes to re-adapt. In case my theory is true, I suppose it's possible that the misfires go away once the adaptation is finished. In any case, I'll start looking into replacing the cats. I know one of them was replaced already in 2002, which sounds suspicious to me.
 

·
Outstanding Contributor
1997 SL600 Brilliant Silver Sport with grey/dark metal interior, 234K miles
Joined
·
1,977 Posts
Move the coils from 3, 4 and 9 to other cylinders (say 10, 11 and 12). Verify the cylinder order before moving the coils. You may just have some sketchy coils. I've replaced all my coils (and a bunch of other things) by now.

Concerning your 20% throttle opening. I was working on my car (getting ready for the track) and noticed (as it always does) when I start the car cold, the throttle only opens about that far. The car only revs up to maybe 2500 until the third rev, and then it can rev further. It's a cold start thing.
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
Move the coils from 3, 4 and 9 to other cylinders (say 10, 11 and 12). Verify the cylinder order before moving the coils. You may just have some sketchy coils. I've replaced all my coils (and a bunch of other things) by now.

Concerning your 20% throttle opening. I was working on my car (getting ready for the track) and noticed (as it always does) when I start the car cold, the throttle only opens about that far. The car only revs up to maybe 2500 until the third rev, and then it can rev further. It's a cold start thing.
Yeah, I actually already switched the coils 3<->1 and 4<->2. I haven't read the codes again yet. I just wonder how likely it is that (at least) 4 coils go more or less simultaneously? Also, would you get the "random misfire" P0300 code then? Admittedly most of my coils are original. One of them has been changed though based on the stamps, and I think it was either 3 or 4.

As for the throttle opening, I was checking the angle with the Star while driving and it went at least up to 80 degrees so that's probably not an issue.

I'll check the codes at some point, got some misfiring at 5000rpm again today, but I'm leaning towards getting new cats in any case. Magnaflow seems to sell direct fit cats for the SL600 at about $600 a piece. Anything involving welding is crazy expensive over here, so that doesn't actually sound too bad. I've been quoted 1800 EUR for a custom exhaust (including muffler and resonator) with stainless steel piping which I don't really want, and 1500 EUR by another shop for fitting universal race cats with new pipes from the exhaust manifolds. But that might be a story for another thread...
 

·
Registered
97 SL600 Pano, ex: 96 (late 95) SL320
Joined
·
247 Posts
I was looking at the smooth running values today while driving and also when revving in P. Values for most if not all cylinders jumped up and down, going to pretty high values at times, some higher than others. Out of curiosity, I took one coil apart from the spark plug boot, and chips of metal fell out. I think it's time for the 21-year-old coils and boots to go before I touch the exhaust.
 

·
Outstanding Contributor
1997 SL600 Brilliant Silver Sport with grey/dark metal interior, 234K miles
Joined
·
1,977 Posts
The metal is part of the connection from the coil to the start of the plug "wire".
 
1 - 20 of 48 Posts
Top