Large difference between %CO in P/N & 4/R

[Berggreen]

Hello everyone,

I have been trying to adjust the engine on my 280SL from 02/71, as it was running much too rich. It is an automatic, and I have adjusted the fuel/air mixture on the injection pump and with the air-screw, while maintaining idle at around 750 RPM in both P, N, 4 and R by adjusting the idle solenoid magnet, and at the same time monitoring the CO % in the exhaust.

But one thing really puzzled me: At idle (750 RPM) I can push the CO down to around and even below 2.0% with the gear lever in 4 and R, and still have the engine running smoothly and pulling strong both at idle and while driving. But by moving the gear lever to P or N and also at idle (750 RPM), the CO increases rapidly and stabilises around 8.0%, with the engine running less smoothly.

Why is there such a big difference in CO between P/N and 4/R??? 😮

I would have expected the opposite in idle (750 RPM), as the engine have to overcome more resistance in the automatic gearbox in 4 and R. Ok you give the engine more air, when the idle solenoid magnet is pushing on the linkage to maintain 750 RPM, but from 2% to 8 % CO is a huge difference.

I know that the CO should lie between 3.0-4.5% CO at idle, but if I adjust the car to this in P/N, then CO in 4/R would be much too low. Opposite, if I tune to 3.0-4.5% in 4/R, the CO would be much too high in P/N.

Do anyone have a good clue or explanation? 🙂

I myself have a thought about the venturi switch on the throttle house and its connection to the automatic gearbox modulation solenoid, because the gearbox is shifting (especially down) pretty hard, especially just before stopping the car and I have large jerks when shifting into R. But I am not much into how this system works. Could the reason for the large difference in CO lie here?

Looking forward to hear what you say and think.

Thanks! 😀

Cheers,
Christian

1971 280SL aut. (dark olive 291H)
1970 280SL man. (silver 180G)

[teahead]

Your CSS hooked up?

Your dashpot is what's maintaining your idle speed when dropped into 4/R?

[Berggreen]

Thanks, as my car is a very late 280SL (02/1971), it does not have the CSS. The injection pump (model # ending R24Y) on the last generation 280SL's did not have a CSS anymore. The function of the CSS was instead built into the pump.

Yes, I adjusted the idle to 750 RPM in 4 and R using the idle solenoid magnet, but adjusting the length of the piston in the solenoid.

I should also say that my 280SL does not have aircon, thus the drop in RPM going from P/N to 4/R is solely due to different resistance in the gearbox.

[Tyler S]

It sounds like the throttle linkage or throttle plate are out of adjustment or any wear on the balls, sockets, or bushings and what ends up happening is when the idle solenoid actuates it is influencing the IP either first or more than the throttle plate. They should both move simultaneously off of their respective stops. The whole idea of the solenoid is to bump up idle speed (air and fuel) when the engine is loaded.

in order....Get the throttle linkage correctly adjusted. Set your base Idle speed and CO in park. Then set your loaded-in gear engine speed via the solenoid actuator while in gear.

[teahead]

Ya, definitely adjust your shift linkage as well as check pressure on the trans.

Then once that's dialed in, it's a matter of dialing in the IP mixture, air mixture screw, and that dashpot adjustment.


I noticed my car would run a bit rough if the dashpot would be adjusted too high.  That will open the throttle flap more, but will not move the IP rod.  However, I got most everything adjusted right and it seems ok now.

[Berggreen]

Thanx for your input....all makes sense.

Sorry, I mixed up the terms here. I thought you meant CSS is the Cold Start Solenoid on the side of the injection pump. But I guess CSS can also be Constant Speed Solenoid, thus the solenoid magnet pushing the linkage, when in 4 or R. In German it is called Gleisstrom-magnet. But in any case, I have it on my car, and it is hooked up and reacts when in 4/R.

The linkage is activating the injection pump and throttle at the exact same time, and I also checked that the length of the linkage to the injection pump is exactly 233 mm as it should be. I could check if the throttle is closing fully at idle, I think it is, but will double check.

There is no wear in the linkage system, as it was recently renovated. However, I do not have the dashpot on the throttle house.

How do I check the pressure in the gearbox? Any chance to check the modulation solenoid connected to the venturi switch on the throttle house?

Thanx again! :-)

[Tyler S]

Don't not rely on the 233mm rod length spec if your car has the hole for the adjustment dowel in the linkage bracket on the passenger side. The throttle plate rod and socket should be removed from the cross linkage. Then the cross linkage ball should be lined up in this hole. You can make a dowel to aid/lock off the cross linkage, (a deep 1/4 socket with the back cut off works well for this). Then adjust/set your IP rod length so the IP is on its stop. Lastly adjust/install the throttle plate rod accordingly.
You wouldn't think it would make a difference but even slight misadjustments in rod lengths changes the geometry of the system.

[teahead]

https://www.sl113.org/wiki/Automatic/Start#SmoothShifting

Ya, the venturi switch could be adjusted.  I can't recall how I did that now.  WIth a vacuum gauge?

[Berggreen]

Thanx Tyler.....but unfortunately, my car does not have the hole for the adjustment. So I guess I have only to rely on the 233 mm measure, or? 

I also have the linkage tour document and followed it carefully, when I adjusted the linkage.

Cheers,

Christian

[Berggreen]

Thanx for the link on the venturi....but damn, it looks quite complicated to mess around with.

Any short cuts I can take? ;-)

[jeffc280sl]

my thought is the FIP reacts to the under load condition when the SL is in gear.  the FIP reads a different location on the space cam and enriches or leans the co/afr.  i contributed with others to this part of the tech manual

https://www.sl113.org/wiki/Fuel/Injection#Pump

[Berggreen]

Thanx, I will study this in detail. :-)

[jeffc280sl]

im not sure how the fip measures engine load.  i tuned my pump in accordance with the on road dyno test specs called out in the bbb which includes a load specification.  my suggestions here are more theory than anything else.  let us know if you have any comments

[Pawel66]

im not sure how the fip measures engine load. 

I am curious about that - how it is done...

[jeffc280sl]

Interesting post related to Bosch FIP

https://www.pelicanparts.com/techarticles/101_Projects_Porsche_911/30-Mfi_Info/30-Mfi_Info.htm

[jeffc280sl]

After reading the post above you learn that the "internal mechanisms of the MFI pumps are coordinated precisely with the timing of the engine".  Timing of the engine is vacuum related which may explain in some way how the pump relates to engine load

[Pawel66]

Acccelerator pedal posiiton and engine speed...

This is a very good article - thank you for sharing!

[jeffc280sl]

more reading for those interested.

https://www.pelicanparts.com/techarticles/911_MFI/TipMFI.htm

[jeffc280sl]

i have another theory about how the FIP responds to engine load.  i think it may involve a relationship between rpm and throttle position as read by the stylus on the space cam.    Linkages from the throttle give an indication of load, while the speed of the pump varies with the rpm of the engine itself.  when on the road the car reaches a hill and while under load throttle position is increased by the driver with a lag in rpm.  similarly at idle throttle position is constant but engine rpms fall as in gear load is applied to the engine.  a new "fuel program" is provided to the fip via the space cam.  engine load causes a vacuum change in the intake manifold causing the ignition system to advance or retard.  in this case the fip and ignition operate as two partially related systems to respond to engine load.

i think we know from the co tuning results provided in this post the co varies at idle while the SL is in or out of gear.  i have no idea what the correct reading should be.  if someone has a perfectly tuned engine and newly calibrated fip they could take some readings for us to use as a gold standard spec.



'

[Tyler S]

I went back and re-read the OP. What is the one thing that differs from park/neutral to 4/Reverse. Is someone holding the brake pedal during the in gear test? Plug off the brake booster and retest. Something is introducing additional air while in gear.

[stickandrudderman]

After reading the post above you learn that the "internal mechanisms of the MFI pumps are coordinated precisely with the timing of the engine".  Timing of the engine is vacuum related which may explain in some way how the pump relates to engine load

This refers to crank timing, not ignition.

[jeffc280sl]

thanks for the information.  makes more sense

[Berggreen]

Tyler S.:
I went back and re-read the OP. What is the one thing that differs from park/neutral to 4/Reverse. Is someone holding the brake pedal during the in gear test? Plug off the brake booster and retest. Something is introducing additional air while in gear.

Yes, I have to hold the break during the in gear test. Have not tried to unplug the brake booster. Do you think that will make a difference on the CO?? :-O

[Tyler S]

Yes it could make a big difference if the brake booster is leaking. Definitely worth checking if not cross it off the list.

[jeffc280sl]

You should hear a sucking sound at the booster near the connection with the master cylinder if there is a leak when depressing the brake pedal.   Instead of disconnecting the booster line I suggest you use the emergency brake and chock the wheels.  Be safe.