Car was idling a bit rough (also missed sometimes) after warm-up and it was suggested to me that adjusting the lambda control might eliminate it.

I measured the lambda statically with ignition on and engine off and got the following readings. I used the sears dvm set on %.

30% ignition on, engine not running
80% with throttle opened completely
90% with air flow sensor deflected


After delecting the air flow sensor, the duty cycle changes from 30% to 60% and returns to 30% if I turn the ignition off and then on. Almost like something is being reset.

When I start the car and warm it up, the duty cycle reads at idle in the 38%-43% range (rich) and 41%-49% when I race the engine. The plug on the adjustment screw has been removed so someone has messed with this before. When I push down on the screw with a 3mm allen engine speed changes (runs better), but returns to rough idle when I remove the pressure on the screw.

How do I interpret these readings?

Thanks,

Steve

Hi Steve,

I am no expert, but as nobody has chimned in yet, I give it a try.

You have to give more info on the car (year) for the experts, as the injection system might differ.

In general, when the car is warmed-up, the CO2 sensor is measuring how much air is left unburnt in the exhaust. The system tries to match fuel and air as best as possible in order to avoid "dirty" exhaust gases. It does that by fluctuating around too rich and too lean.

The fact that you have it fluctuating tells me that the system is working.

I would suspect that spark plugs or wires or distributor rotor are more likely to be responsible for a rough idle. How old are they?

There are also many previous threads about reading duty cycle, so a search might get you faster results.

Hope this helps.

Reinhard Kreutzer

Its a 1987 560sl. I've already replaced the following:

plugs (bosch platinum)
wires
cap & rotor
fuel filter

I know the wires were bad (could see arching), but the rest made no change. Good practice anyway on an old car.

Steve

PS. I'm starting to think that the coolant temp sensor (b11/2) may be bad. How do I check this?

I think the Bosch Platinums are not recommended for this car, the old copper plugs set with a somewhat increased gap are better to smooth out the idle.

Reinhard

It sounds to me like the lambda system is working. I would think that in that range the idle would be OK. You could have a failing idle air valve or vacuum leaks. Agree - change plugs back to copper.

You need a wiring diagram to figure out what the coolant sensor does. On the 380, the oil temp kicks the lambda on and the coolant sensor controls the idle.

__________________
Chuck Taylor
Falls Church VA
'66 200, '66 230SL, '96 SL500. Sold: '81 380SL, '86 300E, '72 250C, '95 C220, 3 '84 280SL's '90 420SEL, '72 280SE, '73 280C, '78 280SE, '70 280SL, '77 450SL, '85 380SL, '87 560SL, '85 380SL, '72 350SL, '96 S500 Coupe

Why can't you use platinum plugs on this vehicle? Bosch website even shows platinum as an option for this vehicle as well as the oem's copper.

I disconnected the sensor leads for the water temp and the car went into a high idle mode. No missing or roughness just high idle.

Steve

You can use the platinum plugs. Bosch would love to sell them to you at their inflated prices. That's why they list them as an option. They won't do any harm. The body of experience from many members however indicates that they don't run as well. Seems the best bet is just the good old cheap copper core Bosch Supers.
Back in the '70s, when Porsche came out with the Turbo 911, the factory put some fancy $8 ea. (in 1970s $$$$!) plug in the engine. The teams that raced the cars however just used the regular plugs at $1 ea. They didn't find enough benefit to justify the slight extra cost.

For any KE equipped engine go to this site for an excellent write-up on how to check duty cycle, interpret the results, and adjust the basic mixture, if required.

http://www.landiss.com/mixture.htm

Duke

Replaced the platinum plugs with the standard copper. Started it up and of course there was no difference in the idle. Still rough! Now on to the next idea.

Yup, it's frustrating to chase a problem that is affected by many things...Just look how many responses you get on the search "rough Idle".
I recall a posting maybe 2 months ago where the idle was finally fixed by adjusting the valves. One valve was out of spec...
Here's another easy one, cleaning the idle control valve. When pulling the connector to it, idle should increase to over 2000 rpms.
It and the connecting tubes can be cleaned for smoother operation. It is easier once the air filter housing is off.
Reinhard

Attached Thumbnails

 

When in closed loop mixture is controled by the system. The common idle problems are caused by lean misfires either due to vacuum leaks or injector flow restriction.

Because the system is controlled based on O2 sensor judgments of exhaust mixture and because the O2 sensor reads the richest cylinder, the mixture is too lean for any cylinder even slightly different in flow or air intake.

After fully warm try pressing down on the air flap. If the car instantly smooths out and then gets rough again it is running good as the lean cylinders are richened and then the correction mechanism leans it back out and it returns to its prior condition.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

Reinhard,

I had previously removed the air control valve and cleaned it, but there was no change in the idle problem. Tonight I removed the electrical connection to the valve and like you said the rpm surged to approximately 2000.

Steve,

The history of this car is that it was driven infrequently in the last few years since the owner lost interest in it. She already had the use of 3 other cars (2 company cars & a lexus rx300). So it sat alot which I guess could cause the injectors to gum up. I've added techron several times without any improvement. Also, I've replaced the fuel filter. Car runs fine on acceleration and cruise, but at a light you can feel the engine twitch. I've also hunted for vacuum leaks but can't find any.

If the result from your test confirms a lean mixture that is causing the miss, what's my next step to remedy. Do I replace all injectors or can I narrow down to just the bad ones?

Thanks much,

Steve

Steve -

Could you expand on this? Wouldn't the O2 sensor be picking up a "blend" of all the cylinders?

Thanks.

__________________
Chuck Taylor
Falls Church VA
'66 200, '66 230SL, '96 SL500. Sold: '81 380SL, '86 300E, '72 250C, '95 C220, 3 '84 280SL's '90 420SEL, '72 280SE, '73 280C, '78 280SE, '70 280SL, '77 450SL, '85 380SL, '87 560SL, '85 380SL, '72 350SL, '96 S500 Coupe

Yes, it picks a blend, but the blend will look like the richest cylinder. The exception might be a misfire where gobbs of O2 aren't burned. This can make the whole mixture look lean. (To the O2 sensor... not an exhaust gas analyzer)

If the car is running at .5 to .8% CO as a Lambda system will try to achieve. The CO levels are read from the rich cylinders as lean ones have none (virtually). A cylinder or two running at .1% CO makes very little difference to the average in this type of testing.

The controlled mixture doesn't compensate for single cylinders and with mixtures as lean as they are a 10% difference in injector flow will be felt at idle. These situations can be determined by the test I described. The test proves that mixture is a problem. It doesn't prove whether it is a difference cyl to cyl or whether the whole motor won't run at lambda control mixture.

It is hard to test individual cylinders without a differential flow test, which will cost as much as a few injectors and probably can't be done in most cases for lack of equipment.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

Steve,

Is the air flap the same thing as the air flow sensor plate?

Thanks,

Steve