Electrical Issue 98 E300 - PeachParts Mercedes-Benz Forum

jake12tech · #1 12-31-2013, 06:47 PM

Odd electrical issue.. Working on a 98 E300, and I've yet to see problem like this in the days of w210s. The the front windows don't work. Rear windows are fine. Fuses are all good, regulators are good. Window switches are good. No coffee spilled in the center console to gum up the switches. Nothing. I'm lost after this. Pointers would be excellent.

ROLLGUY · #2 12-31-2013, 07:45 PM

I had a similar problem on my '98 as well. If I remember correctly, the computer needed to be reset somehow (don't remember exactly how). I remember holding the switch down for a period of time. It had something to do with the automatic window opening feature (all windows open/close when a certain button/s is pressed on the fob). I am sure someone will chime in soon......Rich

jay_bob · #3 12-31-2013, 07:55 PM

The W210 power windows work like this:

Console switch -> electronics on the pc board under the console wood -> message placed on the body CAN bus -> message received and decoded by the door control module -> door control module switches power to the motor -> motor spins -> actuates really crappy regulator mechanism that breaks easily -> glass moves (hopefully) -> door controller monitors motor current, looking for spike in current -> door controller stops moving glass if it determines, due to prior training, that the spike in current coincides with the glass at the stored position for the end of travel. If the spike does not coincide with the previously learned end of travel, it is assumed that the glass hit something so the motor reverses a little bit.

What I am trying to convey with that little thesis is that it is nowhere near as simple as a 12x body window, where the power went straight from the fuse to the switch to the motor (or through the convenience module on gen 2 124/126, but even that is still a dumb switch as far as the motor is concerned).

It is not possible to determine the health of the switch without analyzing the CAN bus data from the control board under the wood.

It is not possible to just jump 12 V dc to the motor and expect it to move. The door controller and motor are integrated into one assembly. The window motor connector has about 10 pins on it. It receives constant 12 V dc from the fuse, and is a node on the CAN bus. The puddle light on the door also ties into this module as well.

Unless you have access to STAR DAS, or have a CAN bus analyzer, you are probably going to have to swap parts to see what failed.

If it was me I would try swapping the window switch panel first, since that part is in common between both doors.

I don't know how they manage the power window relay on a W210. I suspect it's arranged differently than a 12x because the 210 has the ability to roll the windows up or down with the remote control.

Speaking of which, what happens if you aim the remote at the driver's door handle and hold down the "unlock" button? That is supposed to roll down all 4 windows and open the sunroof. Holding the "lock" button closes everything up. This function is IR based so you have to point the key fob end (the end you put into the dash to start the car) at the door handle. If you look real close at the door handle you will see a little IR receiver window.

If you can successfully operate all 4 windows in this method, that tells me the switch board on the center console is bad.

jay_bob · #4 12-31-2013, 09:40 PM

Here is the "training" process:

Think of the window control switch as having 5 states:
Up, past the click = express up
Up, before the click = momentary up
at rest
Down, before the click = momentary down
Down, past the click = express down

What you want to do is run the window down about 3/4 way using the "down, before the click" position. Then press the switch to the "up, before the click" position, and hold it there while the window runs up to the stop. The motor will continue to push the glass up. Continue holding the switch until you hear the glass relax and the relay click in the door. Then let go. Now the window should go all the way up and down automatically.

What you did was taught the window controller the relative position of the mechanism at the end of travel. The next time you express roll the window up, the controller will watch the relative position of the motor (via a position encoder on the shaft) and the current draw. If the stored end of travel coincides with the spike in current, then the controller knows the window is at the top. (Once it knows where the top is, it can figure out where the bottom is, because the number of motor revolutions it takes to go from top to bottom is a fixed amount, and this number is hard coded into the controller.)

The second gen 124/126 had the convenience control module, but it was not as sophisticated as the W210. It just ran the window up until it detected the spike in current and assumed it was at the top (or bottom) and stopped. But if it was someone's fingers caught in the window it would not know it should reverse instead. The W210 window knows that if it is not at the top when the current spike occurs then it should reverse due to obstruction.

The sunroof has to be trained in a similar method, except that you close the sunroof and continue to hold the switch until the motor relaxes.

ROLLGUY · #5 01-01-2014, 02:04 PM

Quote:

Originally Posted by jay_bob

Here is the "training" process:

Think of the window control switch as having 5 states:
Up, past the click = express up
Up, before the click = momentary up
at rest
Down, before the click = momentary down
Down, past the click = express down

What you want to do is run the window down about 3/4 way using the "down, before the click" position. Then press the switch to the "up, before the click" position, and hold it there while the window runs up to the stop. The motor will continue to push the glass up. Continue holding the switch until you hear the glass relax and the relay click in the door. Then let go. Now the window should go all the way up and down automatically.

What you did was taught the window controller the relative position of the mechanism at the end of travel. The next time you express roll the window up, the controller will watch the relative position of the motor (via a position encoder on the shaft) and the current draw. If the stored end of travel coincides with the spike in current, then the controller knows the window is at the top. (Once it knows where the top is, it can figure out where the bottom is, because the number of motor revolutions it takes to go from top to bottom is a fixed amount, and this number is hard coded into the controller.)

The second gen 124/126 had the convenience control module, but it was not as sophisticated as the W210. It just ran the window up until it detected the spike in current and assumed it was at the top (or bottom) and stopped. But if it was someone's fingers caught in the window it would not know it should reverse instead. The W210 window knows that if it is not at the top when the current spike occurs then it should reverse due to obstruction.

The sunroof has to be trained in a similar method, except that you close the sunroof and continue to hold the switch until the motor relaxes.

That is the exact procedure I could not remember:

Quote:

I had a similar problem on my '98 as well. If I remember correctly, the computer needed to be reset somehow (don't remember exactly how). I remember holding the switch down for a period of time. It had something to do with the automatic window opening feature (all windows open/close when a certain button/s is pressed on the fob). I am sure someone will chime in soon......Rich