Repair-It

Car demonstrates intermittent surging during lower RPM operations. The problem is, there are so many variables that are making this hard to isolate.

[X]- I thought unplugging the oil pump solenoid cured this. The likelihood that it does the surge increases with oil temp, generally. In the ~120F range it might behave the best? Without plugging the solenoid back in to do A/B testing it's hard to say exactly what the effect is. I unplugged last Nov, and have changed the oil since. Plus some other stuff.
[ ]- So then I thought, maybe I need thicker oil so my cam adjusters can function properly. Not positive about this diagnosis. I am considering Motul 8100 5W-50, but if I can source Amsoil for a reasonable price maybe I'll try that - some are using their 5W-50 with good results.
[X]- I replaced the main vacuum pump check valve (that goes to the boost control solenoid) AND the vacuum line from the pump to the boost solenoid AND the lines from the solenoid to the wastegates. I don't think there was anything wrong with my old lines.
[X]- Now I am learning there is another vacuum type pipe that contains two more check valves. Mine look OK but I am going to replace the line anyway because it's cheap and easy. By cheap and easy, I mean the hose assembly (three total snap fit connections - shown below) is $32 and takes 1 minute to remove/install.
[ ]- Going to throw a boost control solenoid at it, as well. The solenoid is under the central engine cover and may require a little more fiddling but is otherwise on top and looks easy. That was $82.
[X]- Did two MCT adaptations and that really hasn't solved this particular symptom and frankly I think the MCT is trying to adapt to whatever weird stuff the engine is doing. I still think doing the wet clutch adaptation was beneficial but at this point I think any perceived weird MCT behavior is just a symptom of something else causing it to get confused.
[X]- I threw plugs and coils at it thinking MAYBE that would improve this. It has not. It runs the same. My coils were original and probably still good.
- May have gotten worse after the BenzNinja updated my ECM to the latest SW. If I wanted to, I think I could have him flash it back to see if this theory holds any water.

This might be the boost pressure control solenoid: A0081535428

Ordered a boost (vacuum) control solenoid and this other tube that contains two check valves that can fail.

https://mbworld.org/forums/w212-amg/888 ... oints.html

There are more things I can do with Xentry as well.

Two key things:
- Graph/display cam timing commanded vs actual and see if anything undesired is occurring. I haven't done this. Hopefully I don't need HP Tuners to do this in a manner that I need. And even if I do, I am not sure I know what "good" vs "bad" is.
- Graph/display boost pressure control duty cycle - apparently there are ranges that it expects although if mine were falling out of range I'd probably have a CEL for it.

What I will do is throw the vacuum pipe at it and see what changes. Then I will try a new boost control solenoid. Then I will graph some things and go in the direction of the cam timing and see if maybe after all this I still need thicker oil. I just want to narrow this down.

It seems to drive most consistently when oil temps are in the 120F range.

I do need to narrow this down to either cam timing control issues or boost control issues. Cam timing issues are not to be trifled with. Left long enough this can actually trash the entire cam and possibly the head. The fix would be to replace the adjusters with new if it is having difficulty controlling cam timing in all conditions. Or try thicker oil if I have some evidence that this is actually occurring.

Boost control issues of this type aren't really a problem at all - just annoying. On the other hand, if it is slamming the waste gates around that's no good for anything. I want it to be right because when it drives right it's very enjoyable. Smooth, effortless torque. And pretty good throttle response. I think if I can get to that across all conditions I will finally be satisfied with this powertrain.

Did some logging. Will elaborate later. Here are some Xentry log files for cam adjusters in different conditions.

kevm14 wrote: ↑Mon Jun 17, 2024 7:56 am Car demonstrates intermittent surging during lower RPM operations. The problem is, there are so many variables that are making this hard to isolate.

[X]- Now I am learning there is another vacuum type pipe that contains two more check valves. Mine look OK but I am going to replace the line anyway because it's cheap and easy. By cheap and easy, I mean the hose assembly (three total snap fit connections - shown below) is $32 and takes 1 minute to remove/install.

Installed this. It's comically easy. Step 1: open hood. Step 2: snap onto all three nipples. That's it.

I also poked around the Launch tool for a while and decided I wanted the full power of Xentry. I examined:

- Boost control solenoid. Shows duty cycle. It was within the correct range at idle but I don't know what this means. There is also a bidirectional test that activates the solenoid, passing vacuum to the wastegates. You are supposed to watch the wastegate rods and activate the solenoid. Both activate fine as expected.

- Looked at cam adjuster timing commanded vs actual. I have some CSVs that I need to graph. They contain 2 minutes of data each.

- I also did a cam adjuster test that is done at 3000-3500 rpm. It passed that, too.

Road test reveals that it drives great. It's way too early to conclude this $32 PCV pipe fixed my issues but it had no issues during my fairly extensive test drive.

Lots of analysis on the cam timing actual vs commanded.

Attaching source but also the plots I made. Within each of the two Excel files are four plots. Each one is 120 seconds of:
1) Left intake and exhaust commanded vs actual, plus rpm
2) Right intake and exhaust commanded vs actual, plus rpm
3) Left cam position deltas, plus rpm (my calc with the data)
4) Right cam position deltas, plus rpm (my calc with the data)

Captures 1 and 2 were done within 10-15 minutes of each other, both fully warmed up. I just wanted to get two data points so if one of the plots seemed to "show something" there is another 120 seconds under somewhat different driving conditions.

And capture 2.
Things to look at, not because I really know anything but because of logic:
1) left to right variation in terms of response. Generally the wiggly plots are the actuals and the smoother ones are commanded.
2) Left to left variation between captures 1 and 2
3) Right to right variation between captures 1 and 2
4) Left and right deltas, vs RPM - between captures 1 and 2. Any of the big deltas are transient in nature which is kind of expected.

Overall I don't see anything wrong and the car was driving well, even hot, so this checks out. It has never set a code for cam timing. It passed every cam timing test I did with Xentry (again, as expected). So I would assume this is more or less known good. I don't know enough about these to say if anything is getting a little sluggish with miles. But it's still data and we like data.

And speaking of data, gathering similar data at, say, 120F oil temp, might be interesting. Do the adjusters respond better when the oil is thicker? It would be better if I had a capture at 200F oil temp with that jerky drivability that was the original complaint. But it drove well for every test yesterday. Meaning that perhaps the issue is more a boost control side issue rather than a cam timing issue. At the very least it's intermittent.

Needs more driving in other conditions, more heavy throttle. So I'll happily do that and see if I can get that jerky drivability back that I hate, and go from there. If not, then I submit that this $32 PCV pipe actually helped. IF it helped, the explanation would be fresh check valves didn't cause interruptions to the vacuum pump signal which would then disturb the boost control.

https://mbworld.org/forums/w212-amg/888 ... ost8990476

It does appear that if the central check valve in this PCV pipe was failing, boost could back feed the vacuum pump and disturb boost control. I could further postulate that this presented more at low RPM because the vacuum pump is spinning slower. As RPM builds, the vacuum pump is perhaps able to overcome the leaking check valves and keep the wastegates where they need to be. I did not have any low boost codes so it didn't trip whatever threshold that is.

I think I want to do the centrifuge cover even though it looks more invasive here than it was on my S550. I think you just remove the false firewall to gain access and it's not a big deal at that point.

I did change one throttle coding parameter documented in another thread. That shouldn't really have anything to do with this issue.

Drove it in this morning. Still seems to drive well. It will do a light amount of MCT engagement hunting at light throttle but the heavier jerking seems to still be gone. Driving home today will be upper 80s and most likely over 90 on the hot road, so that will be a good test.

Somehow this is actually fixed. That's pretty nuts. As I said in the thread here, this $32 hose has done more for drivability than doing all 8 plugs and coils. Go figure.

https://mbworld.org/forums/w212-amg/888 ... ost8991760

I need to do the centrifuge breather cover to round out the remaining things that are related (and this one is a good PM to ward off oil consumption, as well).

It definitely seems fixed. Adam also noted that jerkiness when you take off from a stop seems to be gone.

Also, he did a WOT 2-3 and it snapped off the smoothest, cleanest shift I have ever felt from this car. Usually you hit it and sometimes the boost doesn't feel all there, and after the shift it sometimes comes back stronger. That feeling also seems gone. This is pretty crazy.

As I said, I will put the centrifuge cover on the list. And the forum is trying to convince me that the evap line with check valve could also result in boost loss but that boost may end up in the fuel tank which would surely piss off the evap monitoring schemes. So I'm less concerned about that.