Sounds goo to me .

I hope you're using a good quality analog meter when testing ! .

__________________
-Nate
1982 240D 408,XXX miles
Ignorance is the mother of suspicion and fear is the father

I did then what I knew how to do ~ now that I know better I do better

The difference must be someplace else in your car Shern. We’ve just hacked it so it works. Hope you’re okay with that and it doesn’t give you sleepless nights that the car is no longer as shipped.

Yes, there must be an offset or drift in the circuitry in the controller. One interesting thing through this entire process I discovered, or I think I understand is that the entire circuit that handles the thermostat control is contained in the little white board inside the knob assembly. It looks like just two components.



Out of curiosity I’d love to see a clear photo of the other side of the board. It must be a very clever design to do basically a difference amplifier with hysteresis with what looks like two components. Somebody at VDO was extremely clever and designed a minimalist circuit. It’s like “Name That Tune” for electrical engineers and this guy is a rockstar. “Difference amp with hysteresis…I can design that circuit with…Two components.”

I cannot. But I think it can be done especially when I’m looking at what looks like a transistor and resistor behind that board. Well let me think, naively I would form a voltage divider with the thermistor and the potentiometer and use that to drive a transistor that would put out just enough current to activate the relay when the thermistor resistance goes higher than the resistance dialed into the potentiometer. The point at which the transistor turns on is highly dependent on the transistor gain which will be subject to drift. I’m still not sure how the designer put hysteresis into the design - that is you don’t want it switching on and off rapidly when you hit that magic temperature. The on threshold has to be higher than the off threshold.

Maybe the transistor junction naturally does that or the relay magnetic coil lets go at a lower current than when it catches. I don’t know just a guess.

It’s pretty amazing because the ACC2 on my 300D has dozens of parts. Okay, it has proportional control rather than on off control but still, the difference in part count and circuit complexity is dramatic. Bottom line the beloved manual ACC on the 240D is brilliant.

I don’t think there is a separate brain box hiding somewhere else along the circuit. From your descriptions it looks like it is made up of the thermistor, thumbwheel dial and an ice cube relay under the hood that activates the compressor clutch. Pretty cool.

Next time you’re fiddling with that dial at the yard snap a clear photo of the top of the board with the components on it. You got me curious as to what it inside there.

Anyhow makes me think a bit. Time for breakfast.

ETA - found this. This guy did it in 3 components. Not sure how to get away without the 10uF capacitor. That probably gives it the hysteresis. …no, it probably doesn’t.

https://circuitdigest.com/electronic-circuits/thermistor-based-thermostat-circuit

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles

It possibly is. I just can’t account for where that might be… kinda run out of places to look
In the meantime, the thumb wheel/sensor combo I’m running works fine.
Though at the moment, I also have my dash wood out. Makes pulling switches and swapping cables pretty easy. Naturally I’m curious to see if any of these subtle differences account for anything.

Im out of town, but when I return, I’ll post a few images of the dissected thumb wheel.

__________________
1981 240D 4sp manual. Ivory White.

To close the book on this, I tried all combinations of new/used sensor, new/used thumbwheel. The final arbiter is indeed the resistance of the evaporator sensor. The worst performer was the "brand new" evaporator sensor. That thing left my AC cycling at around 58F, although the resistance value was the highest of the three sensors. The thumbwheels were nearly identical... Actually no discernible difference. And Carl, as you requested:









Now I have two sensor cables: my trimmed original, and the factory trimmed I pulled from the yard. I currently have the yard cable installed. It measures a resistance somewhere between my trimmed cable and the new cable. The reason I've installed the yard cable is this: it may not bring the compressor quite as low as my trimmed cable(trimmed was theoretically below 32, yard was 41F before cycling) but it will never freeze my evaporator in the autumn/winter should I need to defog the windscreen. The other reason is that in practice, the compressor will run around 44-45 on the hottest days without cycling. (The cutoff point seems to be around 41).
I'll use this one for a while and see how things go.

__________________
1981 240D 4sp manual. Ivory White.

Hey thanks for posting that. Looks like the circuit is more complex that I’d originally thought. The TAA865 is an operational amplifier. This is actually a textbook difference amplifier with hysteresis. There are a good dozen components on the little outrigger board.

Sounds like a wrap. BTW when you say

That actually makes total sense. High resistance at the same temperature means it is reporting back that it’s colder than it really is. If it tells the little circuit it is cold it’ll turn off your compressor and leave you with hotter air. Just saying it is doing what it was wired to do.

A second note is that little disk with the slotted head in it next to the thumbwheel pot is indeed a small trim potentiometer. It’ll change the overall resistance of the knob assembly a touch in either direction to adjust the temperature switch points. It’s basically a trim for the knob side of the circuit. I believe MB tweaks the temperature at the factory there. The trim resistor on the sensor is a coarse trim that gets you close and the fine adjustment is done on that screw pot. Had we known what that was we might have been able to achieve the same results by turning that screw. But we barely understood how the whole thing worked back then. This info will benefit the next guy who reads this. (Don’t touch this anymore you’re done).

Such is life. If we just had a time machine to go back and leave little notes to our past selves life would be awfully different.

The good news is it works just in time for the heat. This thread is going to be an informative one for future readers debugging manual climate control. I think you’ve added thousands in value to that car. It’s befitting for your clean example of a car to have perfect climate control. Great job.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles

That’s interesting re: the trim pot on the wheel. I have duplicates of everything now, so I may continue to experiment -on the bench!

Thank you for all the guidance. On a sweltering day like today, the tinkering sure pays off…

__________________
1981 240D 4sp manual. Ivory White.

I was not impressed with the state of the wiper traces in general. Just saying.

Other than that any needed component changes look easy. If you can find pretty much the same integrated component today.

I serviced a lot of very early transistorized items many years ago. In those days transistors and very early Integrated circuit devices just naturally deteriorated with age to some extent.

Typically was component drifting in value as well with time. Having to re align them almost always strongly indicated that after a repair. To restore their original performance. Technology advances in electronics have outpaced any other developments I can think of. In my lifetime.

This is the best thread on PP in a long time. Thanks to Shern and Carl for seeing this through to the end. I’m about to embark on a full AC rebuild in my CDI swapped 83 wagon and this process has been immensely informative.

__________________
1983 300D Midnight Blue
4 speed, W115 intake mani, non-EGR exhaust mani, KKK K26, 3 pc Euro bumpers, Lo-spec Euro headlights, AL129X, hubcaps, custom A/C

1987 300TD Smoke Silver
Euro headlights, thermostat relocation, coolant bypass mod, rebuilt Becker + Jehnert + Helix audio, OEM Oris roof boxes and surfboard racks, Euro towbar

1983 300TD Reed (Moss) Green / Dark Olive MB Tex Euro spec

1983 300TD Silver Blue / Blue cloth Euro spec, OM648 and 722.6, 15" hubcaps, W126 S2 brakes

When I upgraded to a parallel flow condenser, I saw a huge benefit. Much colder. I use R134a. The most important upgrade is the condenser.