FYI Hesitation / surging at WOT - Solved!

[gsxr]

A few weeks ago, one of my cars started acting up. It was very intermittent though, and few if any codes were stored. The main symptom was a slight surging at high RPM at full throttle, like you were moving the gas pedal ever so slightly back & forth (i.e., it felt like going from 95% throttle to 100%, on/off). Just enough to feel that something wasn't right. Also, one time when hitting the gas full throttle from a stop (dragstrip launch), nothing happened for a full half-second, then it went to full throttle. Huh... THAT ain't right. Never had that happen on any M119 before. Also, when the car was parked for 1-2 weeks, I'd get a misfire under heavy load after driving about 10 mins... never misfired at idle, or at light load / cruising, only under hard acceleration / high load.

These were the ONLY two codes present, which would eventually return after a number of full throttle runs to redline:

EZL code 002 - Maximum retard setting on at least one cylinder has been reached​

EZL code 018 - Magnets for CKP sensor (L5) not recognized​

​

Code 018 will sometimes appear even though the CKP is good, I've seen this on other cars that are running fine. There's a TSB about adding a shim to the sensor, click here. I plan to add that shim if code 018 keeps appearing, but it's a PITA to R&R the sensor so I'm avoiding that for now. Code 002 was weird - never seen that before. This had me checking live data via HHT/HHT-Win to see what was going on. I was able to reproduce the misfire at high load after 10 mins of driving, the EZL voltages were all abnormally high, and would spike when the misfire occurred. After 20-30 mins of driving this would clear up. Most ignition components were ok/decent, but the coils were 30 years / 194kmi old, so I figured I'd start by replacing the coils. That lowered the ignition voltages a bit and I didn't have the misfire afterwards, but I also didn't have time to wait for 1-2 weeks to re-test. (The new coils may not have had any effect - but, new ones don't hurt either.)

Anyway, the WOT surging was still present intermittently. And, live data showed ignition timing retard was way more than normal. !!! It seemed like the surging was ignition related. Pulled the plugs to inspect, they all looked clean - almost too clean. The insulator was closer to white instead of tan. The plugs had been in service for ~10kmi and the gap had opened to about 1.1mm, so I re-gapped to 1.0mm and re-installed. I next tried replacing the rotors only (new OEM Doduco) in place of the old/worn Bosch. No change. New Beru caps were next... no change. The CKP code 018 was popping up occasionally, so what the heck, install a new OEM CKP. A failing CKP can do some seriously bizarre stuff with fuel mixture, so I figured that was a reasonably gamble. Still no change. WTH!!! Tried a different MAF, no change. Tried different LH module, different EZL, different E-GAS... still no change. At this point I'm running out of ideas on what can affect fuel mixture / ignition, I'm not sure what else to swap for testing. To recap:

• New coils
• New rotors
• New caps
• Plugs inspected / regapped
• Wires/insulators OK, cam seals bone dry
• Different EZL
• Different LH module
• Different MAF
• Different E-GAS
• New throttle cable (not related, but the old one was tired, and while I was in there...)

And with all that stuff swapped... it still had the SAME WOT SURGING.

More testing: To verify this wasn't a fuel delivery issue (new fuel pumps, old FPR) I connected a gauge visible while driving. Nope, fuel pressure is right on spec, and pressure was dead steady when the WOT surge occurred. It wasn't fuel. Per @emerydc8's catalyst failure testing, I also connected a vacuum gauge for grins. At high RPM the vac did get near zero, but then dropped to 2-3" vacuum near redline... seemed like it wasn't bad cats either. And, when the throttle wasn't surging, there was normal / full power to 6000rpm. Cats don't fail intermittently, so it had to be something else.

With the excessive ignition retard, I was thinking the surging was related to timing being pulled. I did not want to replace the knock sensor as that job is nightmare fuel, not to mention many hours of labor. The ETA was a 2000 datecode and had been working fine. But, since I was low on ideas I figured the ETA was worth a closer look before chasing a knock sensor. Keep in mind, there were ZERO fault codes on the E-GAS module (i.e., no ETA faults), and no knock sensor faults either. Sure enough, live data showed the throttle angle was not at maximum at WOT (this should peg at ~81° and remain there at WOT). Instead, throttle valve angle was in the mid/high-70° range, and when the surge occurred, the throttle angle was also changing on the live data! Light. Bulb. Dug up a spare ETA from the parts bin, and a bit over 45 mins later was able to test drive with the spare ETA installed. NO MORE SURGING! Finally!!



Figured I'd share for all y'all who are chasing other maddening issues. The 002 code had me focusing more on the ignition retard and possible lean surging issues, since there were NO faults for the ETA.

• Lesson 1: Don't expect fault codes to always appear and tell you what's wrong.
• Lesson 2: Don't get blinders from red herring data that takes you down the wrong path (mixture/timing, in this case) when the root cause was exactly what it felt like - throttle control.
• Lesson 3: Just because an ETA has good wiring and is not a mid-1990's date code, does not mean it's "good". The internal stuff can and does wear out.

I still have to figure out what is causing the abnormal timing retard, but that's a separate issue, that may have been present for many years - I never looked at that specific live data while driving at WOT. Now that the throttle is staying open at WOT, there's no seat-of-the-pants symptom when the timing is getting pulled. Without live data, I'd have no clue this was happening. The engine isn't stock, which complicates things slightly. I have a feeling the throttle plate not fully opening may have been slightly leaning out the mixture and maybe causing some of the timing problems. (??) Now that the throttle is fixed, I'm going to keep a close eye on fault codes and see if either code 002 or 018 return. I'll add the CKP shim if code 018 continues to recur. And, I have some colder heat range spark plugs on the way (thanks, Jono!) as an experiment to see what that does to the timing retard at WOT. I may also experiment with 96-100 octane unleaded race fuel for grins, but winter is coming soon, so that may have to wait until spring.


EDIT: Forgot to mention earlier, but at part throttle, there was NO problem at all. The engine started, ran, and drove flawlessly when not at WOT. Cruise control was smooth, in normal daily driving you'd never know something was wrong. Also, the timing retard is normal (near zero) at part throttle.

 

[asultan]

A few weeks ago, one of my cars started acting up. It was very intermittent though, and few if any codes were stored. The main symptom was a slight surging at high RPM at full throttle, like you were moving the gas pedal ever so slightly back & forth (think, 95% throttle to 100%). Just enough to feel that something wasn't right. Also, one time when hitting the gas full throttle from a stop (dragstrip launch), nothing happened for a full half-second, then it went to full throttle. Huh... THAT ain't right. Never had that happen on any M119 before. Also, when the car was parked for 1-2 weeks, I'd get a misfire under heavy load after driving about 10 mins... never misfired at idle, or at light load / cruising, only under hard acceleration / high load.

These were the ONLY codes present, which would eventually return after a number of full throttle runs to redline:

EZL code 002 - Maximum retard setting on at least one cylinder has been reached​

EZL code 018 - Magnets for CKP sensor (L5) not recognized​

​

Code 018 will sometimes appear even though the CKP is good, I've seen this on other cars that are running fine. There's a TSB about adding a shim to the sensor, click here. I plan to add that shim if code 018 keeps appearing, but it's a PITA to R&R the sensor so I'm avoiding that for now. Code 002 was weird - never seen that before. This had me checking live data via HHT/HHT-Win to see what was going on. I was able to reproduce the misfire at high load after 10 mins of driving, the EZL voltages were all abnormally high, and would spike when the misfire occurred. After 20-30 mins of driving this would clear up. Most ignition components were ok/decent, but the coils were 30 years / 194kmi old, so I figured I'd start by replacing the coils. That lowered the ignition voltages a bit and I didn't have the misfire afterwards, but I also didn't have time to wait for 1-2 weeks to re-test.

Anyway, the WOT surging was still present intermittently. And, live data showed ignition timing retard was way more than normal. !!! It seemed like the surging was ignition related. Pulled the plugs to inspect, they all looked clean - almost too clean. The insulator was closer to white instead of tan. They had been in service for ~10kmi and the gap had opened to about 1.1mm, so I regapped to 1.0mm and re-installed. I next tried replacing the rotors only (new OEM Doduco) in place of the old/worn Bosch. No change. New Beru caps were next... no change. The CKP code 018 was popping up occasionally, so what the heck, install a new OEM CKP. A failing CKP can do some seriously bizarre stuff with fuel mixture, so I figured that was a reasonably gamble. Still no change. WTH!!! Tried a different MAF, no change. Tried different LH module, different EZL, different E-GAS... still no change. At this point I'm running out of ideas on what can affect fuel mixture / ignition, I'm not sure what else to swap for testing. To recap:

• New coils
• New rotors
• New caps
• Plugs inspected / regapped
• Wires/insulators OK, cam seals bone dry
• Different EZL
• Different LH module
• Different MAF
• Different E-GAS

And with all that stuff swapped... it still had the SAME WOT SURGING.

More testing: To verify this wasn't a fuel delivery issue (new fuel pumps, old FPR) I connected a gauge visible while driving. Nope, fuel pressure is right on spec, and pressure was dead steady when the WOT surge occurred. It wasn't fuel. Per @emerydc8's catalyst failure testing, I also connected a vacuum gauge for grins. At high RPM the vac did get near zero, but then dropped to 2-3" vacuum near redline... seemed like it wasn't bad cats. And, when the throttle wasn't surging, there was normal / full power to 6000rpm. Cats don't fail intermittently, so it had to be something else.

With the excessive ignition retard, I was thinking the surging was related to timing being pulled. I did not want to replace the knock sensor as that job is nightmare fuel, not to mention many hours of labor. The ETA was a 2000 datecode and had been working fine. But, since I was low on ideas I figured the ETA was worth a closer look before chasing a knock sensor. Keep in mind, there were ZERO fault codes on the E-GAS module (i.e., no ETA faults), and no knock sensor faults either. Sure enough, live data showed the throttle angle was not at maximum at WOT (should peg at ~81° and remain there). Instead, throttle valve angle was in the 70° range, and when the surge occurred, the throttle angle was also changing on the live data! Light. Bulb. Dug up a spare ETA from the parts bin, and a bit over 45 mins later was able to test drive with the spare ETA installed. NO MORE SURGING! Finally!!



Figured I'd share for all y'all who are chasing other maddening issues. The 002 code had me focusing more on the ignition retard and possible lean surging issues, since there were NO faults for the ETA.

• Lesson 1: Don't expect fault codes to always appear and tell you what's wrong.
• Lesson 2: Don't get blinders from red herring data that takes you down the wrong path (mixture/timing, in this case) when the root cause was exactly what it felt like - throttle control.
• Lesson 3: Just because an ETA has good wiring and is not a mid-1990's date code, does not mean it's "good". The internal stuff can and does wear out.

I still have to figure out what is causing the abnormal timing retard, but that's a separate issue, that may have been present for many years - I never looked at that specific live data while driving at WOT. Now that the throttle is staying open at WOT, there's no seat-of-the-pants symptom when the timing is getting pulled. Without live data, I'd have no clue this was happening. The engine isn't stock, which complicates things slightly. I have a feeling the throttle plate not fully opening may have been slightly leaning out the mixture and maybe causing some of the timing problems. (??) Now that the throttle is fixed, I'm going to keep a close eye on fault codes and see if either code 002 or 018 return. I'll add the CKP shim if code 018 continues to recur. And, I have some colder heat range spark plugs on the way (thanks, Jono!) as an experiment to see what that does to the timing retard at WOT. I may also experiment with 96-100 octane unleaded race fuel for grins, but winter is coming soon, so that may have to wait until spring.

would be interesting to see if there was a mechanical failure or play in the actual internal gearing/ plate actuator on the left hand side — or if it was as easy as a cold solder joint

 

[gerryvz]

would be interesting to see if there was a mechanical failure or play in the actual internal gearing/ plate actuator on the left hand side — or if it was as easy as a cold solder joint

My conversations with Don Roden, who recently retired from rebuilding ETAs, indicated that despite biodegradable wiring being a major factor in ETA problems, that many if not the majority of ETAs he took in, with mileage/usage on them, had internal issues with potentiometers and other mechanical problems.

Even my own ETA that I sent him from my E500, which only had ~140k miles on it at the time, showed several problems with internal mechanicals, all of which he remedied in addition to the internal and external rewiring.

I think it's a common misconception that wiring is the only problem with ETAs. It certainly is the best-known problem, but veterans here, including @jhodg5ck, can attest that the internals can be just as bad as the wiring.

 

[gsxr]

would be interesting to see if there was a mechanical failure or play in the actual internal gearing/ plate actuator on the left hand side — or if it was as easy as a cold solder joint

I think it was more than a cold solder joint. I removed the bad ETA when it was still warm, and noticed the throttle plate did not move easily. When pulled manually to the fully-open position, it did not "snap" back to idle/closed position quickly. It felt like the shaft was lubricated with molasses.

However - the next day when the bad ETA was at room temperature, movement was normal again (snapped closed rapidly). What I don't know is if a "good" ETA does the same thing. This also makes me wonder about repair shops that bench test their rebuilt ETA's only at room temperature...

 

[Alphasud40]

Since we have no rebuilder for the ETA units it would be interesting to take it apart and see what could be causing the excessive friction. I would think it would be a potentiometer issue but not bad enough to trigger a fault code. As far as the knock sensor goes a good way to see the knock sensor performance is to use a scope to look at the waveform. You use the crank sensor to provide reference and a pickup on number one. By doing this you can determine if it’s cylinder related or just sporadic noise. I am always amazed that knock sensors work as well as they do considering the harsh environment they live in.

 

[dionphaneuf]

This sounds not dissimilar from my issue. I had my ETA rebuilt by Don Rodden in 2021 and I'm 90% sure he didn't give me back mine, but actually a rebuilt spare, so I'm a bit surprised and ought to run through your list of issues to test.

Were you evaluating these readings with the car static and engine off but key On? Or while driving?

 

[gsxr]

I wish there was a way to test the knock sensor easily. A resistance test doesn't work, it's open/infinite across all 3 terminals. I don't know how to connect test equipment to the sensor without cutting into the wiring (not a good idea), or what to look for on a scope, if you could attach it somehow. And you'd have to have the scope visible while driving at full throttle. I rely on the EZL to throw a fault code if the knock sensor has failed.

Fortunately, the knock sensor appears to be a rare failure. I've never had one go bad, which is good because (1) replacement is a miserable job, and (b) they are - of course - NLA. (It appears a replacement can be fabricated using later M119 sensors and a Kurth custom harness. Edit - Kurth sells a complete replacement using this method, click here. Naturally, it costs double what the OE sensor was when available.)

 

[gsxr]

Were you evaluating these readings with the car static and engine off but key On? Or while driving?

All my testing was done driving the car at full throttle, while trying to read tiny jumping numbers on a screen... don't try this at home, kids!

 

[dionphaneuf]

It'd be awesome if SDS could log to an SD card as a CSV or whatever.

 

[gsxr]

It'd be awesome if SDS could log to an SD card as a CSV or whatever.

YES. I would LOVE to get data logging capability. This would be so so so so helpful!!!

Data logging is possible using HFM Scan to the LH module (details here), but it appears to be limited to variables available on either the LH module itself, or what is passed via CAN bus. I believe knock/retard values per cylinder, and ignition voltage per cylinder, would not be available (but I don't know for certain). In post #18 of that thread I was wishing for the same data mentioned here - knock angle and cylinder voltages. That thread was from 5 years ago, I wonder if any advancements have occurred since then.

 

[195910]

EZL code 002 - Maximum retard setting on at least one cylinder has been reached

I have seen this code from a bad distributor cap, but a bad ETA!? I guess its a worn Potentiometer, its the only part that cannot be remanufactured, even the OE manufacturer cannot produce them anymore and the tooling is gone.

 

[gsxr]

EZL code 002 - Maximum retard setting on at least one cylinder has been reached

I have seen this code from a bad distributor cap, but a bad ETA!? I guess its a worn Potentiometer, its the only part that cannot be remanufactured, even the OE manufacturer cannot produce them anymore and the tooling is gone.

I believe the 2 issues are not related. The 002 code for maximum retard was likely NOT caused by the ETA. Only the surging was caused by the ETA. It seems to be coincidence. However, if the throttle plate not opening fully was somehow causing a lean mixture, I guess that could cause abnormal ignition retard and indirectly cause the 002 code? I'm not sure.

I have a theory about the 002 code and am working on that, but it will be a few weeks before I have an update. Also, the 002 code is intermittent as well.

 

[gerryvz]

Since we have no rebuilder for the ETA units

We do have a rebuilder, Victor at "RestoreYourMercedes.com" I believe does this and folks have been happy with his work.

I believe also perhaps one or two other firms that do this work here in the US. Beckmann maybe?

 

[gsxr]

There's also a guy on eBay in FL but I'd probably go with Victor for now, unless someone finds another Don.