Originally Posted by barry12345

I would make an effort to sub another one in as a test. Depending on your background in Electrical/ electronics.

I do not have a schematic of the internals.

Know anybody there with another car that uses it? Or with a lot of luck where one is in a local salvage yard. Failure prone part makes it highly suspicious. Wiring issues in these older Mercedes not usually a big factor. Until they got newer.

I may be wrong but it is an over voltage protection unit. To test off the car you would probably require a dc variable voltage supply. If you have that your testing may be valid. Again I do not have a schematic. It is also fused on top I believe. Since it has a relay. I suspect it is an on off device. Rather than a variable ouput.One might test it in place by carefully winding a wire on the output pin by winding it around it.Keep it very high on the pin. Then seenig if there is voltage output with the car running. I assume it is too hard to get a reading from the bottom of the socket.

I might also check the output voltage of the alternator at the battery. Sitting at more than 14 1/2 volts with the engine running the OVP may just be doing it's job.

Bad voltage regulator might be causing an issue with it. You did test it off the car. Did you establish the voltage at the point the relay drops out? Or only that the relay was still closed at 12V? Caught a lot of vlotage regulators that were not regulating properly. Those cars were okay still as they did not have any over voltage protection incorporated. Plus many aftermarket replacements for thos regulators had so many issues I only use good origal types. Unfortunatly they are a lot more expensive.

Now I am also cheap.If I could not find a substitute ovp unit. If the voltage on the battery was in the normal range. I would locate the voltage input to the device plus the voltage output. Solder two wires to scrap pins and insert them into the socket. Without the engine running and the battery voltage at 12.75 or less. Connect the two wires. but since I do not know if there is something critically sensitive to the voltage. I might just put one pin in the output of the socket using an external voltage supply of 12 volts. or close. If tack and everything starts to work when you start the car. I asume then the device is defective. Of sourse I would have probed to make sure the device was being voltage supplied before doing that. Ignition key has to be on for that. Now my warning is I am capable of doing all kinds of tests because of my electronic background. Of over fifty years of extensive constant troubleshooting.

So i am just mentioning things I might do depending. So what you do is up to you and I cannot recommend my approaches to everybody. Any electrical or electronic issue that is not intermittent is easy compared to those that are. There are a lot of methods to deal with intermittent conditions as well. I am glad to be retired from that field for quite some time now. What they call the over voltage protection. Sounds very simple.

Yet without seeing either one internally or a schematic.I am not positive.I am curious about that ten amp fuse on top especially. I suspect they are just using a zenner diode inside but this is European electronics. Perhaps built by some horrific outfit like Phillips in Europe in the day. They sometimes engaged some really strange approaches in engineering design. There should only be three active pins on that device if the design is rational. One will be a ground.

Why I went into this is the device can be working as it should be, Youi have to check the output of the alternator voltage at the battery terminals with the car running..First before looking at the device. If the voltage is okay. Then the ten amp fuse is suspect on top of the device. They do seem to age out. you clear those two things first. Before actually dealing with the device, You mentioned the tach works sometimes. If that is only with the headlights on. Or the car is under a heavy electrical load situation is interesting. I still would have to see a schematic. Or examine an actual unit, Cutoff threshold is perhaps too low. Or even just dirty contacts in the relay.

Description was very long winded. Just never saw anyone dive into troubleshooting these before. So just thought it might be worth expanding as the part is expensive for what it probably only is. Plus failures have occurred often enough.