Stumbling when revving

Hi there

I am expreriencing a problem on my ‘72 450 SL (US spec). It has begun to run slightly rough, primarily in idle both when cold and warm. Also when I give the throttle a quick “jab” the engine seems to stumple and almost stalls. When I depress the throttle more slowly, the engine runs fine and revs without any problem.

I have verified dwell (32 degree), timing (5 degree ATDC) and fuel pressure (2 bar). I have also verified that all injectors are firing.

I have turned the the ECU knob all the way counter-clock wise and that has made the rough idle go away, but the stumbling when quick revving the engine is still there, so turning the ECU knob may only be a quasi treatment.

Any ideas?

Best, Peter

Hi Peter,

either there is an issue with ignition timing or more likely you have an issue with false air. Rheostat helps to correct in idle only. That is what does not let the engine rev up if mixture is basically wrong. Other options are wiring harness and injectors.

Hi Peter,
maybe the TPS enrichment-signals are missing.

Regards
Norbert

Hi Norbert,

not likely as it would not create such a bad mixture that car would not want to rev up. Remeber that we often recommend to drive without TPS connected to verify that stumbling does not come from TPS.

I have re-checked the fuel pressure and it is right on 2 bar. All injectors are clicking without misses.

When idling there is no problem, but as soon as I put it in gear (reverse or drive) it begins to stumble and almost dies. When in gear it practically has no power and barely accelerates.

When disconnecting the air temperature sensor the idle is practically the same, but now the engine acts normal when I put it in gear and accelerates.

Does it sound like I need to chase a vacuum leak/false air? 

/Peter

....I may be on to something. I just measured resistance on the engine temperature sensor. The area around the sensor was 53 degrees celcius and it measured 480 ohms. Even with a +/- 15% tolerance that appears to be too small a resistance. I measured both with/without connector plugged in and that made no difference.

I will measure the resistance again tomorrow when the engine is cold and see what resistance value I get there.

/Peter

This morning I measured the engine temperature sensor and got the following values (in Celcius and Ohms):

26 degrees ==> 1980 ohm
53 degrees ==> 480 ohm
75 degrees ==> 256 ohm

This suggests perhaps that when the engine (and the sensor) is cold the values are within spec, but as the engine heats up the sensor begins to deviate from specs and show to little resistance.

Upon start-up and while the engine was still cold I put it in gear and the engine acted normally (revs coming down slightly). During warm-up up and upon reaching normal operating temperature, the revs came down and the engine began to shake and miss when I put it in gear.

Could I be having a faulty engine temperatur sensor?

Hi Peter,
I don't think measuring the surface-temperature near the sensor gives reliable values. 

I would trust the values only after having measured the sensor in a fluid...

It wasn't reported the sensors tend to drift away.

- Did you check the MPS for manipulation?

Regards
Norbert
 

Hi Norbert,

I know for at a fact that the MPS is untampered as the black cap is still mounted and has not been removed (and re-installed). It holds vacuum.

I pulled out the engine temperature sensor and measured the resistance in water at different temperatures and got the following values:

80 degrees ==> 269 ohm
70 degrees ==> 320 ohm
60 degrees ==> 505 ohm
50 degrees ==> 703 ohm
40 degrees ==> 981 ohm

I still believe these values are too much out of spec which could explain that the mixture is too lean. At least it fits the story that the engine runs significantly better when I disconnect the air temperature sensor.

I've ordered a new engine temperature sensor (NOS) and will obviously check once it is installed. It did not come cheap, but the existing one also has a crack in the plastic connector housing so I figured that it was good to replace it anyways.  

Yes, Peter -the readings are out of the limits.

Maybe this makes the difference... you will check it out.

I never had a sensor with such a low deviation.

Please tell us the result after having changed the sensor.
It would be very helpful you could test the new sensor before install.

Regards
Norbert

Here's an update.

I've installed a new engine temperature sensor and honestly it has not changed much.

The engine is idling fine. When I am a red light and it turns green, the car is almost cutting out when I step on the throttle, but if I go VERY easy on the throttle I can barely manage to get it up to speed. Once I get to around 60 km/h it has no problem whatsoever in pulling from there. Very linear acceleration and no hesitation. Same behaviour regardless if the engine is warm or cold.

I can find no obvious vacuum leaks. Have sprayed break cleaner on injectors, and air hoses and there are no changes in idle speed.

I am totally at a loss here. Anybody has a good idea? 

Hi Peter,
I wrote more than two weeks ago the acceleration signals of the TPS could be missing.
Did you check this?
Have a look here: jetronic.org/en/d-jetronic/tps

In most cases opening the TPS and cleaning it is sufficient.

Regards
Norbert

Hi Norbert,

Definitely worth a try, so I pulled off the throttle body late afternoon and carefully cleaned the throttle switch with cotton pins and contact cleaner. It was not particularly dusty or greasy and after putting things back together again it made no difference.

Back to square one, sadly.

Hi Peter,
did you verify the signal does reach the ECU? Mightbe a broken wire or so.
Here is a download for you: jetronic.org/de/forum/attachment/346
Read the values at the plug after having pulled it out of the ECU.
You have to read the pins 9 and 20 against pin 12 to test the function of the acceleration switch.

Regards
Norbert