Mercedes-Benz Forum banner

Key Fob Diagnosis and Repair

37K views 38 replies 10 participants last post by  93SoCal400E  
#1 ·
I unlocked my 2000 E320 with the remote, as usual, but when I put the key in the ignition, nothing - no sounds, no steering unlock, cannot turn key to start. So I tried my spare, and it worked perfectly to lock/unlock and start.

Tried the first key again, and it no longer can lock/unlock, or start. I put new batteries in, although they should not be needed to start. I get an LED blink when pressing a button, and see the IR blink with my phone camera.

My feeling is that the key induction coil is broken or disconnected, and that somehow the sync for unlocking has been lost. Your comments and suggestions, please.

I disassembled the key, and some photos are attached. I can not see any obvious problem. I am reluctant to attack it with a soldering iron, as my fine soldering skills are not great. On the other hand, if I have nothing to lose...

Google tells me that a shop in the UK will repair my key for about $50. Seems reasonable. Are there any in the US?
 

Attachments

#2 ·
I don't know much about soldering but $50 sounds too inexpensive to be true. Just reprogramming alone is more than $50, so I'll disregard that quote. I am thinking more like $200+

First, try YouTube if you haven't already - I found A VID showing where to solder for a ML.

If YouTube is no good, find a local locksmith capable of reprogramming the key - I just Googled Mercedes key fix in SC and found a bunch. I have a ML500 and just went through the same problem you are having. Found a local locksmith in VA. They ordered a shell and the internals, put the thing together and reprogrammed. You can get the shell for cheap on eBay, if you want.
 
#3 ·
#6 ·
This is a wild shot in the dark, but I used to have a late friend - master mechanic - who prided himself in being able to fix a lot of (at least 80s) Mercedes electronic stuff. Many times there was just a crack in the solder somewhere. He used a big magnifying glass to look over all of the solder and if he saw a crack just put the soldering iron on to seal the crack.

Fortunately unlike our R129s you can still get electronic keys for the W210. I got one for mine when I took possession of the car - with my dealer discount about $200. And they put it in a modern housing now!
 
#8 ·
If you are good with a multimeter, you can check the resistance of the coil

1) at the tip of the coil where the coil wire ends are soldered to the the two posts. Just measure the resistance. It should be less than 100 ohms. If not you have a break in the coil wire typically at where it is soldered to the posts.

2) The next point to measure is at both sides of the white cylinder where the posts are soldered to the tracks. These points are the weak points that have cracks that you may not see easily. You should receive 100 ohms or less.

3) Follow the tracks from both sides. One goes to a plated through hole and the other goes somewhere under the 68 microfarad 10V black capacitor. Once you determine where they terminate, you can take a measurement there too.

Your key is a generation 1 key. The video shown is for a gen 2 key.

In my case, one of the soldering strips of the white cylinder at the tip was cracked. I soldered it and it was good after that. Also the one of the posts at the tip of the coil (next to the IR receiver / sender LEDs) had intermittent connection. I ended up removing the coil wire, and wind it back on the spool, and solder the ends on the posts. It works with no problem.
 
#11 ·
#12 ·
Hi, got your PM.

I would be happy to try to help, but I need to state that Gen 1 keys use a different EEPROM and the inductive coil and IR parts all look different too.

So, my experience and extra parts are not directly applicable. In fact I have never handled one of those early keys, though I have been curious to check one out. They were only used a couple years and are less prevalent.

You are in a unique position to find the issue. You have a known working key as well as the dead one. So if you have a meter with small enough probes you can A/B it. Check resistance at the points Mr Boca mentioned. If a wire or solder joint is broken you should find an open circuit on bad key where good one shows a value.

With keys I work on, usually if the red light flashes, the key is working.

Since I don't have a single one of those Gen 1s I don't have anything to compare with. I would be more than happy to solder anything, I have a hot air station and can work on the various chips. I have been doing all the key work for free. I just charge for the keys themselves and return shipping. (just sent another off today !)

The Gen 1 EEPROM is Motorola while the Gen 2 (black plastic) and Gen 3 (first Chrome) use a NEC EEPROM. I doubt they are pin compatible so I don't think I can move the EEPROM to one of those. I will probably go on Ebay and buy a Gen 1 just to check it out. (If someone has a known working one in a drawer somewhere I would accept it as a donation)

To sum up, I would be happy to help but can't guarantee success. If you can isolate the issue I can probably do any needed soldering.
 
#16 ·
The Thick Plottens

Hi, got your PM.

I would be happy to try to help...
My thanks to you, mrboca, Deplore and MAVA.

My cheapo multimeter tells me there is no resistance when I measure over the solder strips on either side of the coil. Same for the terminals on the front by the IR LED. I can still see the IR blink with my camera, and also the side LED.

Here is where it gets stranger. We recently returned from a 10-day trip. This morning, while I was once again at the car outside, trying the key, the neighbor came over and said that the alarm had gone off briefly a few times while we were away. Each time it had silenced itself after a short time. The car is parked in the driveway.

So something strange is going on. Meanwhile I still have one working key, and one dud.

I am happy to send you the key to fool with, if we decide that it is in fact the problem. Right now I have nothing to lose.
 
#13 · (Edited)
Gen 1 with Motorola is incompatible with anything. It exists in its own time bubble.

Gen 2 and 3, as you noted, are interchangeable. No real issues.

Gen 4 (that'd be the square buttons) are BGA and have no real "eeprom" so to speak of. Chinese copies have eeprom, original MB do not. It's easy to reuse the chinese copies, but original ones are next to impossible....well, impossible by tools as of this writing. Maybe in future....


It's possible to make a new key from Gen 1, but you'd need expensive tools to do it. I have a VVDI, and I read the EIS info, then removed the motorola chip from the gen 1 key that I have (My car came with 3 Gen 1 keys, all original... an extreme rarity given how easily keys are lost these days), got the key password, then used the EIS info + password to generate a new key.

So basically a locksmith can do that.
 
#19 ·
Make sure none of the keys on the printed circuit board is stuck "down".

1) Take the printed circuit board out of the fob shell.

2) Check for stuck down keys (one of the keys may have been stuck down due to improper installation into the fob shell)

3) Hold the printed circuit board from the rear edges, and gently push it into the ignition slot.

4) Do you hear the steering wheel unlocking noise ?


The reason why the alarm is turning on and off maybe due to the panic button at the back of the shell (mounted on the battery tray), being stuck too. I never used the panic button, but I think it just disconnects the battery momentarily.

If one button is stuck down, pressing the other button (lock/unlock) will blink the LED on the key, as well as the IR LED, as simultaneous pressing of lock and unlock will have a meaning in terms of global locking / unlocking.
 
#21 ·
I assume you inserted the coil end fully to activate the switch that activates the transponder coil around the ignition switch hole. When activated the EIS applies around 115 KHertz power on the transponder coil which is supposed to be inductively picked up by the inductor on the tip of the key. The picked up power is then rectified by a diode, and the resulting power is supplied to a regulator to generate 5 Volts for the Moto microcontroller (not EEPROM). Also it is used to signal the controller that the key is inserted so the signalling interchange can take place.

It is possible that the coil has an internal short that does not develop sufficient voltage over the regulator to start the controller. One can read the voltage across the black capacitor marked as 10V but it is of short duration which can be traced with a scope. The EIS cuts down the transponder signal if it does not receive IR signal from the key within few seconds. Then it re-tries.

If the signal is received by the EIS, and the authentication fails (for example you inserted someone else's key), the transponder power is cut off and no re-tries are done. This is a clue that further diagnoses the problem.
 
#24 ·
UPDATE: I got 2 used keys off Ebay and Nugent's dead key.

I could find no obvious faults with Nugent's key.

I picked the best Ebay key and moved the EEPROM chip over.

The red light would flash in response to the buttons. With Gen2/Gen 3 keys this has always meant the key was functional.

Sadly Nugent reports that the symptoms have not changed. The key behaves as if from different car. I will add pix shortly.
 
#26 ·
Have you checked the IR function of the remote with a cam / smartphone?

Try the following:

Instructions on how to sync Mercedes remote:

Insert the key into the ignition.
Quickly remove key from ignition and within 10 seconds hold down the lock button.
Continue holding the lock button, and at the same time press, the unlock button five times.
Next release all buttons and press the gate/trunk button one time.
Your Mercedes remote should be synchronized. Test to see if lock / unlock functions are working.
 
#28 ·
Well, well, well... Fixed?

After the dealer quoted $300 for a key, my motivation was renewed. I have been trying everything I, or google, can think of to make the key work.

Today I found a slight variation on mrboca's procedure: "Reset your Mercedes-Benz key by placing the key in the ignition of your vehicle while the vehicle is turned off. Do not turn on the vehicle. Press a finger on the "lock" button of your key fob while the key is in the Mercedes' ignition.

Pull the key out of the ignition while continuing to hold the "lock" button down. Continue to hold the "lock" button for the duration of the key resetting process.

Use another finger to push down on the "unlock" button five consecutive times. Keep holding down the "lock" button.

Release the "lock" button. This is the end of the resetting process for your Mercedes-Benz. Use your key fob to ensure it is properly working."

The second time I tried it, I inserted the good key in the ignition, which unlocked. I removed to good key and inserted the non-working one, which unlocked and was able to turn! Removed the newly working key and confirmed that it would unlock the driver door only. I pressed the lock and unlock together to make it unlock all doors - aaand, it stopped working again.

After some cursing, I managed to get it going again with the procedure, and also managed to program it to unlock all doors. So it seems I have a working key again, although I have to say I am leery after my experience. I am still mystified why it stopped working, and also mystified why it started again.

Thanks again to everyone who stopped by to help, especially 93SoCal400E who transplanted the chip.
 
#29 ·
It appears that your key's EEPROM was not updated for rolling code sequence. This may be a glitch in writing into the EEPROM portion of the key microcontroller when the EIS writes the rolling code value, after a successful car start.

When you press a key on the remote, it sends a code to the remote control receiver in the car, and the code is authenticated. If valid, the car is locked or unlocked, just like your garage door opener. After the car is started, a new rolling code is written into the key, so that the key will start using this code on with the next key operation. This is also key specific, so each key has its own sequence to follow.

If the key is never updated fro some reason, the key will not be recognized, as the EIS expects the "updated" code.

The sequence you followed is a re-synching sequence that allowed you synch the key to the EIS again. However, for it to work, you needed a good key that would authenticate and validate your key, so that the follow up resynching can take place (you cannot just put any key and program it, the EIS needs to know that you have a valid key to start the car first).

My LS430 has a similar method. I bought a key with a blank chip, and got the blade cut at a locksmith for $20, then by using the valid master key an a set of sequences, I programmed the key to start the car.

It is a good thing you had a second (good) key. Else programming would have not been possible.
 
#30 ·
The only thing I found wrong with first key was that the first or 2nd time you hit a button the LED was bright red. But by 3rd or 4th press it got dim as if batteries were dying.

But the same batteries in the 2 Ebay keys ($20/each) could be used over and over and LED was bright. Only the original key from Nugent did the dimming thing. So I decided that indicated some part of circuit was acting up. When I moved the EEPROM to one of the Ebay keys the problem didn't follow the EEPROM.

Hopefully this key is fixed.
 
#31 ·
Here we go again

I had not driven my car for more than a week. This morning I went to pick up SWMBO at the airport. I used the refurbished key to unlock and start, and off I went. After parking at the airport, the key would not lock the car. In the ignition - nothing. So I had to lock the car with the spare working key, and use it to unlock again and drive home.

I have not yet tried to resync the key as I thought I'd post here first. Something is funky.
 
#35 · (Edited)
Key was bad.
He sent it to me and I took EEPROM and placed on an (unknown) Ebay key I got for $20.

Nugent is reporting similar issues to before the surgery.

The only constants are the EEPROM and the EIS. If main key always perfect, than issue is likely key.

Or maybe I got a bad donor key that was similarly broken to previous key.

It is also possible that the chip is not down perfectly and a leg has either lifted off it's trace or two legs are shorted.

The EEPROM was much larger than the Gen 2 one, I could have put a larger nozzle on the air gun.

Anyhow, the "do or die" moment for Gen 2 has been the red light flashing appropriately. I have probably done 20-30 keys in last few years. If the red light flashed I was done. The key worked. I think only 1 person ever had issues, and they too had odd issues before.

I believe he acknowledged up front that key may be killed by the procedure. This was my first Gen 1 Smartkey. I'm happy to move to the EEPROM to the 2nd donor key.

These keys differ from Gen 2 in one GIANT way. They have the delicate coil that turns magnetic field into DC current to run starting circuit swinging off the end of the PCB . Wire is crazy thin. And Gen 1 has an ENORMOUS coil that dwarfs rest of the PCB at the end.

It isn't necessary to pull the PCB to change batteries, but for every inquisitive owner who does so they have big risk of damaging the silly thing.

For Gen 2 this became a much coil that lives on just one side of PCB, no taller than nearby parts.

On Gen 1 it is the end of the key. It towers over PCB on both sides of key board. And at it's edges are that wire that is almost too fine to see.

The key Nugent sent and the other one I bought appeared to have damage from some over-eager-Einsten jamming the PCB in to the key housing with force.

Imagine having "delicate" (gender neutral) body parts wider than torso, you'd take your time going through door frames with close dimensions. Most delicate part of these keys is in most defenseless area. Easy to see what this changed.

Anyhow, Nugent is welcome to send key back and I'll either fix bad joint or move EEPROM to donor #2. At that point it would HAVE to be dying EEPROM not erasing/writing correctly.

Or he could take key to local who does component level solder work. (Not a "board replacing warrior" at a mall kiosk)

Have him reflow the EEPROM. It's the giant black chip with "Motorola" on it. (and sloppy hand solder work)

I'd love to see this resolved. These keys are getting rarer but I'd still like to know them better.
 
#36 ·
Key was bad.
I believe he acknowledged up front that key may be killed by the procedure. This was my first Gen 1 Smartkey. I'm happy to move to the EEPROM to the 2nd donor key.

The key Nugent sent and the other one I bought appeared to have damage from some over-eager-Einsten jamming the PCB in to the key housing with force.

Anyhow, Nugent is welcome to send key back and I'll either fix bad joint or move EEPROM to donor #2. At that point it would HAVE to be dying EEPROM not erasing/writing correctly.

I'd love to see this resolved. These keys are getting rarer but I'd still like to know them better.
The good news is that I got the key going again by following the re-sync procedure. I don't drive the car often, so wait and see if it lasts.

I had zero expectations about the key ever working again, so I regard 93SoCal400E's efforts as a success. None if this is his fault, and he won't respond to my offers to pay.

Thanks for the offer to reflow the solder, but as it's working I'll hang on to it for now. That may change if your offer remains open!