Have I stumbled into the Gremlin's lair? EHA valve??
 

O.K. I been walking by my 300 all week without even looking toward it cause it makes me sick. Well, just moments ago while straightening up my workbench, I have my Dmm in my hand and it occurs to me that I have what I need to test the EHA. I rig up a crude connector and set the DMM to 20MV. I start the car and walk around to see read the display. 0.00mv. Hummph, maybe something isn't connected right. Yup, one of the test lead alligators has jumped off.... but wait a minute... this thing is running pretty good!!! The EHA is disconnected completely. O.K. It's in open loop. I'll plug the EHA up and see what happens. Engine starts running ragged again. Disconnect and it smoothes out!!! Just for sh**s and giggles. I reach in the car and push the A/C button. The compressor engages and cool air comes out. Drive the car down the road, pep is back!! Check engine light alternates on and off but no effect on the way it runs. What am I to make of this. Have I finally found the Gremlin, or at least one of them??

As I have pointed out, these cars do just fine when warm without the electronic correction.

There are two possibilities to what you are seeing. Just last week we had a car come in complaining of poor cold starting. A quick look to check EHA current (ma not mv) showed that the connector was off. Easy cold start repair, but, plugged in the car ran worse and worse. While monitoring, the EHA current would slowly drop (once the connection was made) till the lambda correction was all used up. It went to -12ma. This means the controller thinks the car is running rich and it subtracts fuel (lean correction). So, next we disconnected the O2 sensor hoping that the sensor was giving a false reading (it would have had to see a voltage on the sensor as a ground would have sent the system the other way to +12ma). For those ready to test, the EHA ma will go to +12ma if one grounds the O2 lead to the controller. If one places one hand on 12v and the other on the sensor it will see enough voltage to go the other way, as we saw it, at -12ma. This test is a very good way to see if the system has full corrective capability. Depending on initial calibration the system usually goes far enough in both directions to run poorer with an especially good area somewhere in the middle.

It turned out that our car had a bad controller (car was a recent purchase) and someone had disconnected the EHA to solve the lean correction.

That possibility is very rare. Controllers seldom fail. It's possible for the heater circuit of the O2 sensor to bleed enough voltage into the O2 sensor signal for similar to happen. A short will make it rich and that usually makes it run better not worse.

The other possiblity is that the car doesn't like running at the lambda control mixture. This is common but usually is only an idle thing.

You need to get that meter back hooked up. Remember that you are checking current. For current to flow it will pass through the meter. When in the mv scale the current will not pass through the instrument so the circuit is not complete and the car will run as if disconnected. To verify your hook up turn the key on with engine off. When warm you should see about 18-20ma. During cranking it willl jump by 30-50ma. With the O2 sensor disconnected (under carpet passenger side), you should get 0ma when warm. If you drive with the meter hooked up the reading will jump +6-10ma with a sudden acceleration. While at constant speed the reading should fluctuate up and down about 2-3ma, centered about zero or a little negative (I usually set them to -2ma to zero ma).

Stephen,

I had an old 89 190E 2.6 with the same problem to a tee. I would drive down the street....and when accelerating, the check engine light would blink on and off, and when lit, the car ran like crap, no power. I assumed it was my O2 sensor, but my mechanic verified it was working fine. Just for fun, I had a spare computer that was out of a 86 300E. I popped it in, the car ran PERFECT ever since. the light never came back on, it idled right, and accelerated fine.

To this day I dont know if the old computer was bad or what, but I didnt complain because it fixed the problem. I know the computers rarely go bad.

Best of luck.

As usual I'll have to THINK about Steve's reply but:
 

What's the connection to the A/C compressor? It won't work as long as the EHA is plugged in. Unplug it and I've got A/c.

You got me on that one. Are you sure it wasn't a coincidence? Is it repeatable?

I can't think of a direct relationship.

Yes, With climate control on and EHA disconnected, I have compressor cut in. Plug in the EHA and the compressor is idle. I tried it several times and the results were the same.

Solved the AC part of the problem
 

The EHA connector and the temp sensor for the aux fan were plugged backwards (oops). The plugs are identical except for the pin size. I switched them and the AC clutch works correctly.
I hooked the meter back up to the EHA and get -0.00ma on all tests. I checked the voltage from the wiring harness plug and get 4.10v. Anyone have a clue?

I was thinking along those lines, but couldn't remember if the connectors were close enough. One part of MB engineering, I have always liked, is that by design they usually make such happenings hard to accomplish. Much more so than other manufacturers.

I had bragged on MB about that very thing
 

But the only two similar connectors on the entire wiring harness branch from the harness in the same spot, hence the ability to confuse. So much for foolprooffing:rolleyes: Now. What do those numbers mean? This engine has to be started about ten times before it will run!!:confused:

Bad OVP maybe.

I think it's time for a generic wiring harness and a FORD 302. Then it will be a 500F!