What would make this 73 450SL run fine on start up but then after 10 min start missing and running rough? I have changed the coolent temp and aux air, checked out the MPS and the plugs are not fouled, it is like the amount of gas or air is changing. What would be a logical troubleshooting procedure for this problem, all the vacuums have been checked and verified with a vacuum guage
Thanks In Advance
George in Appalachia

I would suspect lean running after warm-up. Best course of action is to get the hot mixture checked with a CO meter, but failing access, you could try the following:

Make SURE you don't have a vac leak somewhere (or that you have not recently fixed one!). If you find any, fix them.

Make sure the ignition is up to snuff -- that means good cap, rotor, wires, plugs, and points. If the ignition is weak, it will run poorly at the correct mixture because the spark is weak.

Check the tailpipe -- if it is black and sooty, expect an ignition prolbem.

lastly, and only if you cannot find anything else, pull the cover on the ECU and turn the adjustment wheel a couple clicks clockwise. This will cause it run run a bit richer. If there is no change, turn back to the orignial setting and llok for the vac leak. Remember, the intake manifold is in two parts with seals (8 of them) between, and there are two seals on each injector, too! Bad door locks or climate controls can also give you a vac leak, so can the large hose to the MAP.

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!

Thanks for the reply Peter,

I was fighting a too rich condition on this so I plugged off the vacuum feed for the climate control and the door locks right at the intake manifold, I upgraded from a points system to a Crane electronic ignition with a photo sensor and then back to points and I get the same symptom, I think you are right and it is now running too lean but the ECU has been moved during the course of working on this so I do not have a good starting point for the ECU , it has about 22 clicks so I set it about midway. So that is about where I am on this right now....
Thanks again
George

If you have "lost" the setting for the mixture, the best way I've found to get close is to adjust for maximum rpm, then ccw two or three clicks. This will get you close. You can then monitor fuel consumption and tailpipe condition -- if you are too rich still, the pipe will stay black and sooty, and your milage will be poor. Adjust one click lean and see what happens.

The range is very small, and the bad part is that milage will decrease too lean as well as too rich, since the fuel won't burn properly. If another click lean doesn't help, go a couple clicks rich instead.

However, unless you can get access to a CO meter, this is the best I've managed. It will run best at the rich end, though, so max rpm will be just a shade rich (about 3% CO, I think).

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!

The adjustment on the computer only affects idle. The way to set this system up is to get your timing high (at least 5 degrees above spec don't get fooled by the vacuum retard - the spec intends you to be retarded). I would remove the vac retard line and set it to 10 degrees BTDC (maybe more - go for 17in manifold vacuum after the fuel is optimum).

Now to adjust the mixture: disconnect the throttle switch. This removes the computers idle comp circuit (so one can work on just the main system). To adjust one must open the black cap on the manifold pressure sensor and use a 4mm allen wrench. I do it on a dyno so that I can check it at various loading but it will work OK to set the CO to 1% at 2000rpms. After setting reconnect the throttle switch and adjust the control unit to best idle : usually 2-3% CO. If you wish to adjust this system you must do it with an exhaust gas analyzer. I have probably done it a thousand times maybe ten and I would only do it by ear in a pinch but since you are going to do it that way anyway, use the above technique.

Without a dyno one will only be sure of proper adjustment under all loading if the manifold pressure sensor is still at full range. Since its action is based on an evacuated bellows it looses its range over time. A functional test can be done with a scope by measuring the injection pulsewidth. Hook scope to injector and turn key on engine off. By hand, move the throttle. There will be 20 firings one at a time as one moves the throttle to full. Only two injectors will pulse (the ones that correspond to the trigger point closed at the moment). Hook to one of these injectors and view the pulse on the scope. At the moment there is zero manifold vacuum (not running) so the pulsewidth is maximum - probably 7-10ms. Pull a vacuum on the manifold sensor and the pulsewidth should go down. At 17in it will be about 2.5-3ms.

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

Steve:

I suppose this means I need to adjust some more -- I only set the idle mix.

I get black smoke a full throttle, and I'm only getting 15 mpg, seems a bit low to me. Tailpipe is still sooty, too, although not so much as before.

Hmmm.

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!