Flash back in history to this post: https://mbworld.org/forums/e-class-w...elt-codes.html

So back then, when dealing with the SRS codes (all fixed) I ran into a problem with the drivers seat. It got stuck. Deep in that post I discovered the seat would not go forward/back, banged the motor a couple of times with a mallet and got lucky and the seat forward/back motion was fixed. Yeah.

Forward to current, son tells me the drivers seat has been acting up lately - sort of sometimes wont go forward, won't go backwards. Sometimes will. He says it has been slowly getting worse, starting really showing up last winter and again this past fall of 2023 when it start getting cold.

I put it on the star and test the seat switches, they are all working correctly. But I can't reliably test the forward/backwards, get encoding error. Looks to be an issue with the encoder in the motor. Ok, very similar to what happen in that past post. So back then I avoided replacing that motor - but now I ordered a new one. Its a very simple replacement, just two screws hold it to the seat track bracket, it slides off, you slide the new one in and it comes with a new wire harness.



Put the seat back, 4 bolts torqued to 45NM, seatbelt bolt to 32NM. and your done.

Using the star go through the retraining on the forward/back range. And I have the same issue! Sometimes can move the seat forward/back, sometimes can't . Get the same encoder error on the star when it fails.

I rechecked the grounds, power - no issues. Its not the switches - they pass fine. That test also tests all the wiring to the seat controller for the switches. Its not the motor nor its wiring - it is brand new. Must be the seat controller!

Start searching - these are expensive new. But I see some sellers on ebay selling what are clearly gray-market versions. New units for less than ebay junk yard used units. For the first time, I decide to try one out. $85 free shipping from China, less then the $130+ for junkyard units, beats the $400+ for new units. So we order one. Maybe we will get lucky and it will work - if not. well - time to get a junkyard MB one.

We received the new controller - it is still in its box since I did not make the time to replace it yet.

Flash forward until early Jan - we are having an extended sub 0F week. Son tries to start his car (he is still driving it, just can't move the seat) and the battery is dead. As in really dead. 1V - yep - multimeter read it as 1V. What the heck?

Ok, Walmart has that size in stock (FYI, Walmart has the best prices on these batteries and they are Interstate). Get a new one and we have to replace the front one - wish Walmart carried the small front one - those are expensive- it is reading low from that week of 0F temps. Two new batteries now. The car is now inside the heated garage and on the charger to make sure they are fully topped off.

Two days later the the car won't start - battery too low. What the heck is draining that battery?

Charge them back up - put an amp meter inline on the neg terminal - and we have a 2A parasitic draw that drops to 1.5A when everything should be asleep. ???

For the next two weeks, my son treats his car like an EV. I let him park in the heated garage, and he keep it on the charger when not driving.

Finally I clear a weekend to try and figure this out. First use the star and other than the seat still giving issues - no new codes - EXCEPT that the can buss is failing. The tests all pass but the overall quick test fails with no error code. Humm. Since I don't have an oscilloscope, EDIT: I can't look at the can buss but my hypothesis is : a defective module is not going to sleep and broadcasting on the can buss keeping most of the modules from going to sleep mode. That would be around 1.5A. (member below corrected that if a module does not correctly respond to a CAN poll, the F-SAM will stay awake which could cause the excessive current draw). And know I have a defective seat controller.

I first thing I unbolt the seat, and remove that drivers seat controller and replace it with the Chinese clone one. Run all the star tests on that module - everything passes - seat is working perfectly. Ok, hook back up the amp meter inline with the battery and monitor what happens as the car goes to sleep. And sure enough - the 4A quickly drops to 2A in a few minutes, then drops to 1.5A in a few mins, then 1A, then 0.8A, then 0.5A, then 0.2A. Hangs there for about 45 mins, and then drops to 0.04A 40mA, the expected parasitic draw from a sleeping car.

So it was the failed seat controller. It was sort of working, enough for star to talk to it but it was somehow causing a high parasitic current draw - EDIT: quite possibly keeping the F-SAM awake., but it must have been failing to go to sleep among its other issues of unable to read the motor encoders for forward/back. So it was most likely chatting on the can buss - keeping all the other modules awake.

So replacing the seat motor was a waste of money. Oh well.

Summary - if you run into a 1.5A'ish parasitic draw - be thinking that something is keeping some module(s) awake. Oh and that super low cost Chinese seat controller clone - seems to work perfectly.

 

Two humbs up for not giving up

I can really imagine the nerve it takes to tinker with this chaos below 0°F.


seat controller
I have fixed mine so I have a fairly good idea how it should look.... Can you post pictures of the internal so I can point at what the problems are!

Your next step will be to go inspect passenger seat controller.

BTW no need to remove seat... the plastic module case is just snapped in.


conformal lacquer 👍
Here is what you need to easily prevent internal short circuits from developing.

As far as sleeplessness: it is caused by any CAN-B member that is disrupted... it can be more than one and more than one way too.
The F-SAM keeps polling members at regular interval. When that takes minutes instead of seconds, battery pays!

Try to use a DC clamp meter instead of serial 10Amp Max connection (prevents from disturbing system).

Personally I went to town and fixed all these known issues at once:

1. the painted GND screws
2. the bare unprotected circuit boards
3. the marginal solderless connections!
4. the dry mechanical assemblies
5. the leaky connector seals
6. ....

no more non-sense chaos drain!
You really need to get to all 4x door modules (DCU) + trunk modules

 

CAN hypothesis corrected and references to F-SAM added - thank you Plutoe and CaliBenzDriver.

I do have a clamp style meter (60A) but running in DC mode I find it gives very inaccurate results on those thick negative battery cables - numbers were all over the place. Maybe a better unit would not do that (its not a Fluke). Putting a multimeter inline in 10A gave me what appears to be accurate current measurements but naturally disconnecting the battery and reconnecting the meter takes awhile for the car to go back into sleep mode - and you better not do anything that would exceed that 10A draw.

After closing the car door and the interior lights not yet timed out so still on (mostly LED) initial draw was around 4A (trunk was left open for battery access and those lights are still bulbs) and dropped in steps as stuff turned off and ended up right at 40ma after I checked it about 45 minutes later.

CaliBenzDriver - I still have the controller. I'll open it up and look at it and post pics.