What other modules are reporting faults when your Keyless becomes unstable ??

 

During this incident, the only other info that grabbed my attention was a constant fluctuation of 12.3 to 12.5v reported by EZS Electronic Ign Lock(N73). I had the service menu active and when iCarsoft via EIL was showing 12.3 - 12.5v swings, the dash was giving more consistent 12.6 with occasional 12.5 drops at anywhere from -1A to -6A.

On another keyless incident in which the car was either hesitant to unlock via door handle or hesitant to start via start button (forget which specifically as I've had both somewhat recently), I had a Hall Sensor fault for the up/down actuation for steering column. "Comfort access". The column did function, but with audible mechanical clunks & crunches that day. I do believe I've seen you mention this in another thread regarding lubrication of part of the mechanism, but it hasn't been a regular occurrence. The code was historic, cleared, & never returned. During this incident (and other times when doing routine scans without symptoms), there was also a 3rd Central Gateway fault along with the other two that Jetta first mentioned in the 1st post here.

I'm still going to make that alternator/voltage thread because I have a bit more to the story, but this is everything related to Keyless. Really hoping I can narrow it down to this system, although I haven't really started digging too deep yet.


From EIL Module (N73)

From ESA-DR Electronic Seat Adjustment Driver (N32/1)

Central Gateway 1

CGW 2

CGW 3

 

That's pretty comprehensive detective work

You can ignore the "Diagnostic" faults of CAN-D - Its caused by OBDII disconnects... ok.

The most significant is that low voltage you captured. It's sort of "on the fence", I mean not super low in 11.xV but it's messing up your CAN-B.

If you want to test "normal voltage" with stable 14.1V and chassis reboot then consider ALT-LIN for relief.
The whole CAN-B benefit soldering door modules loose pins including KEY-Go module itself.

 

Thanks Cali. I think I did also come across similar info regarding the CAN-D stuff.

If I let things go long enough without a Ctek overnight meeting, that EIL module voltage might get into the high 11's. Not a theory I'm willing to test!

ALT - LIN disconnect & reboot is starting to look appealing. Oddly, some months ago I cleaned all of the chassis grounds I know of (minus interior somewhere under driver's carpet) because I was doing a washer pump & saw the sad state of the front driver's wheel well ground. Upon cleaning all (11 I believe?), giving a dab of Deoxit, & reconnecting battery, I was getting a steady 14.1v with no fluctuations! Some time after that, I decided to swap out the voltage regulator on my 15-16 yr old factory alternator, & after VR swap, it went back to the fluctuations.

Recently, I see prolonged 12.5 episodes under various states of driving with a few 12.4 - 12.2v glances while going from R to D or D to P. I wonder if ALT -LIN disconnect will reveal trouble with the alternator itself? I have heard noisy bearings enough for concern, but am unsure I'm capable of diagnosis on these "smart" dumb charging systems.

 

12.6V to 14.9V is working stock swings
12.2V... that is "drained by driving" yoyo
14.1V is ALT-LIN fixed self regulated voltage.

 

Well, it's been near a month now of disconnected ALT-LIN plus a new key fob battery & no keyless faults have returned. All modules are of course getting stable voltage as well, which is a nice change!

If 12.6-14.9v is stock swings, perhaps you've answered my alternator dilemma. Before ALT-LIN disconnect, I was not staying in that range consistently. I wasn't 100% sure on the proper numbers for the factory "swings". I am mostly getting steady 14.1v now, but when opening & closing doors (coupe having the windows slightly lower), or when rolling windows up/down, I get a drop down to 13.7-13v. The RPM's also drop & raise in perfect harmony w/ the voltage drop/raise. Normal driving will sometimes show 13.8 - 14.3 swings. I was hoping an ALT -LIN disconnect would reveal a weakness in my alternator if such swings were present compared to a hoped for steady 14.1v.

Perhaps this along w/ my noisy alt bearings & nearly 100% worn down og voltage regulator are enough to warrant swapping the 16 yr old alternator? This was a large part of my original suspicion with my keyless go intermittent issues.

 

These voltage variations are not supposed to be while ALT is regulated. This is significant ie. it means something: "drop-voltage"?

Self-regulated voltage should be: 14.15V +/-.05 st battery posts.
You're reporting 13.8 to 14.3V (from where??)

At least this range is guaranteed not below battery resting voltage. At least this effectively cancels "drained by driving".


Perhaps you have an oxidized GND strap?

Did you do the "battery float" step before disconnecting?

What sort of battery current does your IC-Display shows?

 

Observed 13.8 - 14.3v (as low as 12.9 while rolling windows up today at a red light) via the instrument cluster. The only times it goes below 14.0 are when windows are used or when shifting in/out of Reverse. And yes, "drained by driving" has been eliminated as well as the need to have a Ctek session. The battery has needed no topping off since ALT - LIN unplugged. It used to be a 1-2 wk necessity.

All grounds have been recently inspected, cleaned & given a small amount of Deoxit. Only the front driver's wheel well ground had significant corrosion. All others, including the one underneath on the side of the transmission were near spotless, but cleaned all anyway. Just after doing this ground cleaning & while ALT - LIN was still connected, I WAS actually achieving a steady 14.1v with zero fluctuations whatsoever. However, once I changed my voltage regulator (on a whim), the yo -yo, drained by driving, etc ensued. Old VR was significantly worn, & upon doing so, it was obvious my bearings were showing their age.

I did not float, but in retrospect curious what that result would've been. ALT - LIN was disconnected the day after a full Ctek charge & of course the battery was fully disconnected during the unplugging of the ALT - LIN. The IC current sits around 30A on cold start, goes down to around 9- 10A just after idle goes down, once driving away it sits at 3A, and once near op temp & thereon it stays around 1A.

What's best is just today on a warm start, I got the "key not detected" message again while initiating the IC voltage display. No issue unlocking the car or anything else. Turned ignition back off, & car started right away. Fresh FOB battery. No codes other than 1337 "Communication w/ Starter or Alt" fault & now have a transmission fault 0705 selection range sensor(Y3/8s1) defective as shifts have also been a bit more harsh & erratic since unplugging ALT - LIN, although not sure of the correlation of that last fault. Fun!!

 

This voltage instability you describe started after you replaced VR, right?

Before that ALT was doing a better job at self-regulating 14.1V, right?

I would fix that as a priority as it may be causing the other faults your experiencing.

ALT-LIN is only about getting sel-regulated stable voltage.

Where did you buy your replacement VR??

The IC current you report are well inline with the improved ALT-LIN expectations.

Why is your chassis voltage unstable under load... needs a little detective work: drop-voltage, where?

 

Yes, after VR replacement voltage instability ensued. ALT only did a good job at self regulating after all grounds were cleaned. Shortly after cleaning grounds, the new VR was installed. Before touching GND's or VR it was just starting to enter "drained by driving" territory & if I didn't top off the battery at least every 2 wks, I'd get slow cranking on cold starts. This went on for maybe 2 months, as I was just starting to try diagnosis.

An alternator is certainly in order asap. The replacement VR is a Bosch made in Taiwan unit from FCP Euro.

Self regulated voltage certainly is a beautiful thing....to go out to your car the next morning and your battery is actually still charged like every other Earth-dwelling automobile with a normal, sane, and rational charging system is a good feeling indeed! Glad to hear the current is within the expected/accepted range.

Detective work certainly needed! I've also just noticed turning on A/C while stationary cause the same drop voltage the windows do. I certainly hope this is all ALT related. The only other potential smoking gun I have is a fault in the old SG-Fond Rear Control Unit (aka rear window control module). It's a B code that states "Internal Electrical Fault", although all windows function fine as well as seat belt presenters. An ebay "known good" unit was bought which had no codes & car was taken to my indy to confirm software status as I don't have Xentry/Star, & all was good for months. I've just recently seen the code reappear, & it is current & doesn't clear. All corresponding fuses are good. More work to do.

As always, your responses are greatly appreciated, Cali.

 

You can load test your alternator, with or without current clamp at alternator positive output.....while using the car built-in voltage and amperage information.
Something like below :



full post here :
https://mbworld.org/forums/e-class-w...soak-test.html

At idle in D, brake hold or 540RPM..........and engine already hot, your alternator probably can not keep up if load is more than 60 amps to alternator* ( *current clamp needed ).
Assuming LIN is disconnected, the moment you see instrument cluster voltage information drop from 13.8 or 14.1 to below 13.5V, and amperage in minus (deficit ), that meant alternator can't keep up and'
the minus amperage is what your main battery is supplying.

MB amperage data is NET to or from the battery, it does not read true gross alternator output.

 

.... it's almost like your new VR is a mismatched spare part! Previous wirked better than new FCP trusted unit.

You want to measure or track what voltage ALT puts out under various loads such as rear defrost or windows motor. Voltage should not budge (+/- 0.1v)

> DECISION TREE TEST:
Find out if ALT output deviates under load
or if voltage is being dropped by chassis wiring, prefuse and the likes.

It's weird that your genuine Bosch VR does not self regulate the way our Valeo does.
Bosch invented all that jazz licensed to cheaper Valeo partners.

Was your previous reg. also a Bosch from original mount ?

The target is flat stable voltage regardless of normal load current.

 

The timing of your post is absolutely perfect. I happened to witness this exact scenario completely unplanned. I had made a short trip (but long enough to reach op temp) about 30 mins before leaving for work. This was early afternoon, and ambient temperature was about 93°F & very high humidity with a 73% dew point. Upon getting back into the car, I did something I rarely or perhaps never do.....turned on the A/C while in R with foot on brake before backing out of the garage, near or just at 500 RPM. The results were exactly as in the video you shared. I'm normally in motion at an expected 1500-2000 RPM (give or take) when turning on A/C. Being stationary with heat soaked engine bay also had the fan coming on right away & the draw was unlike anything I've witnessed in my 5 & 1/2 yr ownership.

My ALT - LIN is currently disconnected. This RPM correlation also makes sense with my window situation. This being a C207, every time my doors are opened/closed, the pillar-less coupes do their partial lowering & raising of windows. I get the same voltage drop & negative current draw. I also rolled up all 4 windows at a red light 2 days ago & had the same result. I've seen enough & am more than happy to replace this alternator.

Thank you as always for your extensive experience and knowledge.

 

I had also been thinking about this potential mismatched VR! I confirmed the P/N with Mercedes, but they had a voice of concern for me wanting to replace it rather than the alternator itself given the age, etc. They insisted they rarely sell just the VR's & almost suggested against it. I could assume they just want to sell more alternators &/or get more cores, but this particular dealership/wholesaler (MB of Covington, LA) has been nothing but extremely transparent & helpful over the years. They are not local to me, but when I need something straight from MB, they are my go-to for how easy they are to work with & being anywhere from 40-100%, yes 100% cheaper than my local dealership in NJ!

My original VR is also a Bosch unit as is my original 150A alternator. I had a suspicion of my alternator & my plan was to disconnect ALT - LIN & see if the alternator could hold its own, which clearly it cannot. These issues all started most notably just after unplugging ALT - LIN, & have only been witnessed & faithfully recreated under load, so as of now I have no suspicion of wiring issues. I pray ot stays that way!

The HVAC & windows have been the only noticeable draws. The "key not detected" & TCM selection range sensor fault have been the only electrical issues, & the strange shift behavior directly correlates with ALT - LIN unplugging, but has not been an every day occurrence.

I have a refund coming in the next few days via FCP from a lifetime warranty replacement on some suspension work. If the bearings weren't screaming, I'd put the old VR back on but that seems foolish given the same labor can equal a new unit for no money upfront.

Any thoughts on a new SEG unit? Apparently they are a Bosch subsidiary now handling their alternator & starter production. I believe I've read of significant troubles here from reman Bosch units. I believe the reman units are also more expensive & I don't remember even seeing a core charge either.

 

I dont know what is the best brand replacement ALT. New sound better than used.

You need a unit that can plug-in to operate with your car computers. Matching selection is limited.

Some modern ALT's types are controlled by serial ECU data, hence the ALT-LIN disconnect for self-regulation....


> NO CAN DO...
Perhaps your chassis uses a conventional style ALT that can not self-regulate ??

To find out.... :
scan your chassis for faults.
Look at ECU faults
Is a LIN fault registered: Y/N ?

> STOCK NOT LIN?
HOW about simply plugged in stock?

If your specific ALT is not LIN controlled...
Lucky no yo-yo "drained by driving"

Stay plugged in then.


> NO CHANGE: ALL GOOD
-- Overall this means that if your old ALT unit is not LIN controlled, a replacement will not be any different.

-- Keep existing!

 

Well, thankfully the aforementioned AEG (Bosch) unit shares the same P/N & Country of origin (Spain) as my original. So at least this is easy enough. The only other option is reman, including from Mercedes.

I double checked the P/N's on og VR & replacement for sanity's sake. The numbers do cross reference each other, my og VR being superseded plenty of times (along w/ alternator P/N's.) I still can't wrap my head around the "out of order" order of events.

I do have the IBS on my negative post as well as a fault in the ECU/ME from unplugging the ALT - LIN. A communication error with starter or alternator. Also, when 1st getting the car before I touched anything mechanical or electrical, I tried out the IC voltage/current display before & during a trip probably at my 1st weekend or so of ownership. I observed the (then) steady 12.6v readout under normal driving & the "regeneration" of 14.1v give or take during decel, coasting, & braking. Also, the voltage/current increase during R or when A/C set to max.

As for the "out of order" order of things;
-Slow cranking one day. Beginning of IC display monitoring. "Normal" smart charging behavior observed. 1-2 wks in would see low voltage (11.9v) before 1st day's cold start. Beginning of Ctek sessions. Charging would remain at 14.1v longer as the battery would get drained just before it was time for another Ctek session.

-Inspection & cleaning of all chassis GND's. Battery always tested "Good" or "Good Charging". Starter crank time and draw was always fine after Ctek entered the picture. After GND cleaning, an almost constant 14.1v observed, only changing slightly when in Reverse.

-VR changed for experimentation & observation. True beginning of constant yo-yo & consistent drained by driving. It started slow, then escalated quickly. Slow cranking & "key not detected" intermittently appeared. Bearings noticeably bad & then my concern with all other alt components having equal degradation.

-ALT - LIN unplugged, no more drained by driving, no more Ctek. Proper voltage every morning. However, intermittent (negative) difference in shifting patterns, but overall as it was before (which was good). Introduction of voltage drain & RPM drop during window actuation or A/C on. Only significant while stopped & in D, as Mr Prihadi showcased.
A constant ECU fault (naturally) from ALT - LIN unplugged.
All strange. All interesting.

 

Old vs New

Factory Alt

IBS. Sister's late model W203 has conventional Alt, so she shouldn't have this?

The only GND out of the 11 or so inspected/cleaned that showed obvious corrosion. Driver's wheel well, observed while doing windshield washer pump.

ECU fault from ALT - LIN unplugged. Thankfully TCM fault 0705 "selection range sensor defective" only came twice & hasn't returned. A noticeable "slip" in N was felt/shown before TCM fault. No dash lights, but was current in TCM. Cleared, stayed cleared.

 

Here is what I understand :

your ALT has a VR that's LIN Bus controlled
we don't know if this Bosch VR can self-regulate.
This all works stock to "drain by driving" flat battery below 12.0V

Your unit ALT-LIN'ed out has not been shown to hold self-regulated voltage under basic load.

You best need stable voltage to help prevent nightmares with module faults and low battery slow-cranks.

You have not tested for voltage drop, I assume your particular BOSCH VR does not self-regulate automatically like VALEO does.
I would say: as-is your unit is not ALT-LIN capable for lack of self-reg.

Can this ALT core use a Valeo VR or a more modern type of Bosch VR that is compatible with self-reg ??

As long as you know about missing self-regulation, you want to reconnect ALT-LIN to operate with stock voltage control.

 

When LIN disconnected , this soft DTC will happen, no worry.



Soft DTC is no engine check light and the icon for it in Xentry is i ( as in Infomation ) only.




.

 

Subject: Corrosion & poor connection in grounding circuits:
Some thoughs on fighting corrosion and poor connection in grounding circuits, as I've seen mentioned previously. This is a bit of a long babble, but hopefully of use to some. Highly experienced techs will know many of these things (I'm not a car tech, but know a bit about connections and grounding), I just want to ensure that everyone here has more info to help them in their troubleshooting.


Background:
If corrosion gets between connecting parts of a grounding circuit, or any circuit, (e.g. ground post to retaining nut, or retaining nut and lock washer to wire connector, or connections between circuit modules and wires), this can cause trouble in the circuit (as we know). If you encounter a loose connection, you can also have trouble.


What to do:
If you loosen the retaining nut and wiggle the wire and retighten or disconnect a connection and reconnect, all might seem well....for a while. It is important to clean or tighten the connection. Cleaning the wire's connector is great, but you also should clean any nuts and washers found in that connection. Also of importance is to clean the threads of the mounting (ground) post. I even go as far as cleaning the threads of any nuts involved. This way, you have clean and pristine mating surfaces. And yes, use an anti oxidation goo that also enhances conductivity between parts. Reconnect and/or tighten accordingly. If it is a connection between wire harnesses, the affected male and female pins on both sides of the harness connection must be cleaned. If it is a loose connection, it must be tightened.


Where to clean or fix:
Remember that it is usually the threads of both the mounting post and any retaining nuts involved that actually perform the connection to the mounting (ground) stud, and thereby to the car body. Sometimes there is a toothed washer used to connect between wire and car body as a compression connection. Think of how the connection physically goes from the car body to the ground wire (piece by piece) - but at a microscopic level.
Remember also that it is the mating surfaces of connector pins that perform connection between wires, along with how the pins are connected to the wires. Pins must be connected tightly to the wire and must mate tightly together when connected. I once found one single poor crimp that caused the complete isolation of a small town Central Office from the public telephone network. Can you imagine living in a small distant prairie town with occasionally no phone access to outside help? Simply due to a poor wire crimping job? Yup, there were many raised eyebrows, notably in the Emergency Services field. It was a simple 5 minute fix. Similar to an expensive car having an intermittent fault that made it occasionally not start.


Testing for poor connection:
Here's a great way to see if there is a poor connection between a wire and what it is connected to. Simply measure the voltage between the wire you are looking at and where it is connecting to. Try to measure at the bare wire itself.


Test method:
For example, let's consider testing the connection of a ground wire to a ground stud...

1) set your handheld meter to a low voltage range (if it doesn't auto-range)
2) press one meter probe onto the copper wire if possible
3) press the other meter probe firmly onto the car body beside the ground stud - yes, poke through the paint and touch up later (it's just a pin hole anyway).
4) ensure both probes are pressed firmly in place for good connection
5) ensure good current flow through the connection (for example - with car running and affected circuit operating) and measure to see if the meter displays any voltage across where your meter probes are contacting.
6) if the meter shows any significant voltage, there is an impedance between the ground wire you are testing and the car body. There should be no impedance and thus no voltage drop across a grounding connection (or any connection, for that matter).


Fault Isolation:

You can find where the impedance is by measuring across various points - wire to connector (sometimes connectors are not tightly connected to a wire, thus soldering helps out), connector to ground stud, connector to nut, and finally - connector to car body.
If a significant voltage across the connection is displayed, thoroughly clean the connector, (or re-crimp the connector to the wire, or solder the connector to the wire) involved in the connection. The voltage seen across a connection is called a voltage drop . It shouldn't be there as it robs voltage from the circuit it is connected to, and restricts current flow in the circuit you are attempting to fix - possibly causing it to fail occasionally due to not having sufficient voltage on, or current through the thing you're fixing. These are horrible gremlins to find!

Another thing to look at is where a group of wires go into a plastic connector that snaps into another plastic connector (for example where a wire harness connects onto a module). There are a few things that can go wrong there.

Explanation:

1) the wire may not be tightly crimped to the connector pin (on either side of the connection), or a pin poorly connected onto a circuit board.
2) connector pins might not fit together snugly
3) there might be oxidation on connector pin mating surfaces
4) there might be corrosion on connector pins (don't forget that corrosion bridging between adjacent pins can also be a problem).


Resolution and discussion:

Both oxidation and corrosion need to be completely removed from connections, and they must connect tightly. A crimped connection is meant to duplicate a seamless connection between wire and connector - as if it was welded together. If you wish to use an anti corrosion gel, ensure none of it gets between adjacent pins, possibly causing a short circuit, or cause signal bleeding onto another wire (a nightmare to find - easy to resolve by cleaning).


As I mentioned, that was a long exhausting read. I hope it helps someone find one of those miserable gremlins that cause occasional faults from vibration or poor electrical connection.

All the best!

 

Very nice write up!

You're going to find these issues implemented to produce unequal voltage differences in many circuits.

One of my favorite is solderless in CAN-B Modules to produce battery issues. Visit KeylessGo and door Modules for loose pins ready to solder.

When CAN-B becomes unstable it swarms the Gateway module that impacts CAN-C: engine + tranny.

when cleaning GND or fixing modules.... disconnect all batt power prior.

 

Thanks CaliBenzDriver, and you're right - soldering is an excellent method to make a good connection.

The most important thing in any grounding system is to ensure all operating circuits have low impedance paths to a common ground. In our case, the common ground is the vehicle body (which also needs a low impedance connection back to the battery and charging system. As you mentioned, MB seems to have done a brilliant job in creating unequal impedances in these ground paths (solderless connections, crimp connections, compression connections, etc.). The net result being inconstant current flows and varying voltages in circuits as current draws change, making rather interesting faults (I.e. change a part, things work great...for a while!).


Finally, the name of the game is to clean, solder or re-crimp to make solid low impedance connections. If you're looking for a fun project, find and test all grounding connections to ensure low impedance paths to common ground. Don't forget the return path to the battery and charging system. Test all connection points in the circuit you're working on. This is how I found why my old NA car's battery kept discharging - a poor connection from the battery to the engine block.


Most importantly as you mention, CaliBenzDriver, disconnect all power sources prior to repairs - main battery and any backup batteries or capacitors. Note any settings that will be lost due to disconnection that will be needed to be re entered after reconnection.

All the best!

 

All good
Lucky us the W212 does not loose any settings during battery out. In fact the unstable modules greatly benefit from getting REBOOTED regularly.

Have you cleaned your main chassis GND strap aka the "WT-F strap" ? Located all the way under cariage near the muffler.

You understand the reality between resistance and live impedance.

Beyond power supply being offset by randomly unstable voltage drops, MB engineers implemented that for CAN-Bus. Not all the nodes of ONE GIVEN BUS operate with the same impedance, in particular the body network CAN-B.

The HI/LO thresholds working on a different reference help make the network less stable with slower though put.

 

Holy smokes, you're right! tThat's a really different twist to CAN-bus! I thought it was only 120 ohm impedance. You're also right, that would complicate things. So, we treat them all as separate busses, and ensure they all have good solid voltage due to their signaling method. Poor voltage would definitely bugger up the signaling protocol on a bus and cause bus overwhelm from repeat transmissions from all modules on the bus. There is no broadcast control other than priority and the controller on each bus has to filter based on priority. Modules have to rebroadcast at a known rate until their message is accepted. Works ok, but yeesh can this cause problems.

Did a little searching, found a great CAN-bus tutorial for everyone to see...

https://www.csselectronics.com/pages...ntro-tutorial#

A good, fairly reliable system for automotive use.

Like the great comedian Benny Hill once said in one of his sketches "Learning all the time", lol.

 

See.... when MB officials repeatedly swear we drive the only reported cases, the reality is hand-crafted to create memorable ownership.

Once upon a time, long ago CAN termination resistors were found at the two most most extreme ends of networks.
Now on W212... these have been replaced by one single resistor built-into the central bus-bars.

That would be okay if all equal Star branches were not also built unequally like packs of tree branches.
Now that architecture guaranties unequal branch impedances. Its the wrong receipe for good networking hence CGW has a taste for reboots.