[Witek_M]

So this vehicle has occasional habit of killing battery dead overnight. It happened twice before. Battery replaced in mid December. It was 5 years old and failed capacity test @350 CCA.
On January 15th it died overnight. It arrived on tow truck with battery voltage at 2.72V. It took 12 hours to get it back to normal at 10A charging rate.
Checked quiescent current. It starts at ~25A, after 30 minutes or so goes down to 0.007A. That is very low. I am used to 30-50mA on most modern MB cars.
I cannot find in WIS normal quiescent consumption or prerequisites to testing.
Maybe @konigstiger can help?
VIN 4JGDF7DE5HA745391
Attached scan after battery recharge. Obviously low voltage faults galore.

 

[HughV]

Hi W_M,
I am more than a little familiar with this phenomenon of intermittent dead battery. I installed a 200A current shunt and together with the 6 digit HP DVM could read down to milliamps (as it appears you are able to do as well).

My conclusion is that it is the headlights. I started looking at the air suspension pump behind the right front bumper area under the headlight. On my car (2017 gls63), any key or door-relates action causes the internal headlight motors to start running. Usually they time out after a few minutes, but sometimes they keep running. The bit of info I have been able to collect says that the pointing motor(s) are "homing" themselves. Both lights do it together, put your ear to them. Draw is 4A or thereabouts. Is there some gear or clutch slipping and the home sensor (optical, magnetic, current draw at end of travel? i dont know.) isn’t tripping?
I ran into something about a slipping gear somewhere iirc. At 2-3K/headlight it is hard to think about replacing them on a lark. Next time I am at the dealer (almost never) I will ask them to check them for a software update.
I thought I had a fuse ID-ed that I could pull, but nothing under the hood seemed to stop the motors. I have pulled the neg bat terminal once or twice to stop them from running on. I have the electrical diagrams so I should check there for a fuse to pull.
I have also purchased a RF headlight assy with a cracked case on ebay to a) practice pulling the lens cover off without destroying it for access, and b) to take a look at the motor & mechanisms and see what is up in there.
I also purchased a 12V/50A power supply and outfitted it with 2 jumper cable clips and a long extension cord. This lives in the tire well. (for emergency charging) (its voltage adjustment pegs out at 14.5 or so, so I might get the 15V version next time)

This is an extremely annoying problem, bordering on dangerous. I can't believe there isn’t more chatter about this issue. Maybe people are changing headlights out? Or indeed there is a software fix out there? Why doesn’t this show up in a service bulletin?

Anyway, take a listen to your headlamps afterstopping/locking the car. They may be tripped by key proximity, too, so I have a ritual to always lock the car and keep the key out of range when done driving. Along with jumping up and down on one foot while rubbing my belly and chewimg gum; something I learned while getting my BSEE....

looking forward to hearing about what you find!
-Hugh

So this vehicle has occasional habit of killing battery dead overnight. It happened twice before. Battery replaced in mid December. It was 5 years old and failed capacity test @350 CCA.
On January 15th it died overnight. It arrived on tow truck with battery voltage at 2.72V. It took 12 hours to get it back to normal at 10A charging rate.
Checked quiescent current. It starts at ~25A, after 30 minutes or so goes down to 0.007A. That is very low. I am used to 30-50mA on most modern MB cars.
I cannot find in WIS normal quiescent consumption or prerequisites to testing.
Maybe @konigstiger can help?
VIN 4JGDF7DE5HA745391
Attached scan after battery recharge. Obviously low voltage faults galore.

So this vehicle has occasional habit of killing battery dead overnight. It happened twice before. Battery replaced in mid December. It was 5 years old and failed capacity test @350 CCA.
On January 15th it died overnight. It arrived on tow truck with battery voltage at 2.72V. It took 12 hours to get it back to normal at 10A charging rate.
Checked quiescent current. It starts at ~25A, after 30 minutes or so goes down to 0.007A. That is very low. I am used to 30-50mA on most modern MB cars.
I cannot find in WIS normal quiescent consumption or prerequisites to testing.
Maybe @konigstiger can help?
VIN 4JGDF7DE5HA745391
Attached scan after battery recharge. Obviously low voltage faults galore.

(not sure my reply made it thru as such. copied here. -Hugh)

Hi W_M,
I am more than a little familiar with this phenomenon of intermittent dead battery. I installed a 200A current shunt and together with the 6 digit HP DVM could read down to milliamps (as it appears you are able to do as well).

My conclusion is that it is the headlights. I started looking at the air suspension pump behind the right front bumper area under the headlight. On my car (2017 gls63), any key or door-relates action causes the internal headlight motors to start running. Usually they time out after a few minutes, but sometimes they keep running. The bit of info I have been able to collect says that the pointing motor(s) are "homing" themselves. Both lights do it together, put your ear to them. Draw is 4A or thereabouts. Is there some gear or clutch slipping and the home sensor (optical, magnetic, current draw at end of travel? i dont know.) isn’t tripping?
I ran into something about a slipping gear somewhere iirc. At 2-3K/headlight it is hard to think about replacing them on a lark. Next time I am at the dealer (almost never) I will ask them to check them for a software update.
I thought I had a fuse ID-ed that I could pull, but nothing under the hood seemed to stop the motors. I have pulled the neg bat terminal once or twice to stop them from running on. I have the electrical diagrams so I should check there for a fuse to pull.
I have also purchased a RF headlight assy with a cracked case on ebay to a) practice pulling the lens cover off without destroying it for access, and b) to take a look at the motor & mechanisms and see what is up in there.
I also purchased a 12V/50A power supply and outfitted it with 2 jumper cable clips and a long extension cord. This lives in the tire well. (for emergency charging) (its voltage adjustment pegs out at 14.5 or so, so I might get the 15V version next time)

This is an extremely annoying problem, bordering on dangerous. I can't believe there isn’t more chatter about this issue. Maybe people are changing headlights out? Or indeed there is a software fix out there? Why doesn’t this show up in a service bulletin?

Anyway, take a listen to your headlamps afterstopping/locking the car. They may be tripped by key proximity, too, so I have a ritual to always lock the car and keep the key out of range when done driving. Along with jumping up and down on one foot while rubbing my belly and chewimg gum; something I learned while getting my BSEE....

looking forward to hearing about what you find!
-Hugh

 

[Witek_M]

Hi W_M,
I am more than a little familiar with this phenomenon of intermittent dead battery. I installed a 200A current shunt and together with the 6 digit HP DVM could read down to milliamps (as it appears you are able to do as well).

My conclusion is that it is the headlights. I started looking at the air suspension pump behind the right front bumper area under the headlight. On my car (2017 gls63), any key or door-relates action causes the internal headlight motors to start running. Usually they time out after a few minutes, but sometimes they keep running. The bit of info I have been able to collect says that the pointing motor(s) are "homing" themselves. Both lights do it together, put your ear to them. Draw is 4A or thereabouts. Is there some gear or clutch slipping and the home sensor (optical, magnetic, current draw at end of travel? i dont know.) isn’t tripping?
I ran into something about a slipping gear somewhere iirc. At 2-3K/headlight it is hard to think about replacing them on a lark. Next time I am at the dealer (almost never) I will ask them to check them for a software update.
I thought I had a fuse ID-ed that I could pull, but nothing under the hood seemed to stop the motors. I have pulled the neg bat terminal once or twice to stop them from running on. I have the electrical diagrams so I should check there for a fuse to pull.
I have also purchased a RF headlight assy with a cracked case on ebay to a) practice pulling the lens cover off without destroying it for access, and b) to take a look at the motor & mechanisms and see what is up in there.
I also purchased a 12V/50A power supply and outfitted it with 2 jumper cable clips and a long extension cord. This lives in the tire well. (for emergency charging) (its voltage adjustment pegs out at 14.5 or so, so I might get the 15V version next time)

This is an extremely annoying problem, bordering on dangerous. I can't believe there isn’t more chatter about this issue. Maybe people are changing headlights out? Or indeed there is a software fix out there? Why doesn’t this show up in a service bulletin?

Anyway, take a listen to your headlamps afterstopping/locking the car. They may be tripped by key proximity, too, so I have a ritual to always lock the car and keep the key out of range when done driving. Along with jumping up and down on one foot while rubbing my belly and chewimg gum; something I learned while getting my BSEE....

looking forward to hearing about what you find!
-Hugh



(not sure my reply made it thru as such. copied here. -Hugh)

Hi W_M,
I am more than a little familiar with this phenomenon of intermittent dead battery. I installed a 200A current shunt and together with the 6 digit HP DVM could read down to milliamps (as it appears you are able to do as well).

My conclusion is that it is the headlights. I started looking at the air suspension pump behind the right front bumper area under the headlight. On my car (2017 gls63), any key or door-relates action causes the internal headlight motors to start running. Usually they time out after a few minutes, but sometimes they keep running. The bit of info I have been able to collect says that the pointing motor(s) are "homing" themselves. Both lights do it together, put your ear to them. Draw is 4A or thereabouts. Is there some gear or clutch slipping and the home sensor (optical, magnetic, current draw at end of travel? i dont know.) isn’t tripping?
I ran into something about a slipping gear somewhere iirc. At 2-3K/headlight it is hard to think about replacing them on a lark. Next time I am at the dealer (almost never) I will ask them to check them for a software update.
I thought I had a fuse ID-ed that I could pull, but nothing under the hood seemed to stop the motors. I have pulled the neg bat terminal once or twice to stop them from running on. I have the electrical diagrams so I should check there for a fuse to pull.
I have also purchased a RF headlight assy with a cracked case on ebay to a) practice pulling the lens cover off without destroying it for access, and b) to take a look at the motor & mechanisms and see what is up in there.
I also purchased a 12V/50A power supply and outfitted it with 2 jumper cable clips and a long extension cord. This lives in the tire well. (for emergency charging) (its voltage adjustment pegs out at 14.5 or so, so I might get the 15V version next time)

This is an extremely annoying problem, bordering on dangerous. I can't believe there isn’t more chatter about this issue. Maybe people are changing headlights out? Or indeed there is a software fix out there? Why doesn’t this show up in a service bulletin?

Anyway, take a listen to your headlamps afterstopping/locking the car. They may be tripped by key proximity, too, so I have a ritual to always lock the car and keep the key out of range when done driving. Along with jumping up and down on one foot while rubbing my belly and chewimg gum; something I learned while getting my BSEE....

looking forward to hearing about what you find!
-Hugh

I like shunt idea. Do you mind sharing details.

I use meter connected in series between negative post and negative cable. During high current period like opening and closing door I jump around the meter. Blue wire in the picture. Meter can handle 20A only.

I have connected battery monitor and have been watching door open/close events. I have left C4 MUX on left front floor. I had to open door to get it and use on another vehicle.
Voltage drops low then recovers after few minutes.

But. Imagine busy family with few kids getting their stuff out. Doors get opened/slammed shut, liftgate gets opened/closed. By the time it is all done battery is low.

 

[HughV]

Hi Witek_M,

For the shunt, use a resistor like this:

I like this shunt resistor the best, easy to put lugs on the terminals, should handle starting the car OK (full load rated is 100A, but 200A for a few seconds shouldn’t be a problem).
https://www.digikey.com/en/products/detail/murata-power-solutions-inc/3020-01108-0/3438619
This will be .001V/amp or .000001V/mA.

Or anything on this page:
https://www.digikey.com/en/products...fAFo40EEkhiArnkIkQHTnIUgVpAHTVqYAAQEAFgFsucoA
(10 watt minimum, 0.001 ohms, separate terminals for the meter)

The next problem is the meter, one capable of reading to 1microvolt (1ma) or 10microvolts (10ma resolution, will be fine). I have a couple of older HP 5 or 6 digit meters that do this. There are others out there by other test equipment manufacturers, Fluke, Keithley, etc. This will be an ebay excursion for you, or and will be in the 100’s of $. There are some 5-1/2 digit ones out there for <$300. Possible meter pictures attached, one with another shunt plugged in.
For a new one, this should work OK, reads to 10uV, but probably a little jumpy down there.

https://www.newark.com/tenma/72-1020/multimeter-digital-bench-4-1-2/dp/74M7699?st=dvm

Picture attached also of the shunt setup. Used a battery post clamp to go to the battery and the body of a big lug to clamp the car’s neg battery clamp. With this setup you can start the car, etc. and watch what is happening for background current draw. You should be able to find some universal battery cables with lugs already crimped to get what you want. Screw-in posts for side-terminal battery, etc.

"But. Imagine busy family with few kids getting their stuff out. Doors get opened/slammed shut, liftgate gets opened/closed. By the time it is all done battery is low. "

Reading this part of your last post, however, it sounds a lot like something is wrong with your battery. It should have plenty of capacity to go in and out of the car, tailgate, etc. many times before the ability to start is compromised. The original post seemed to indicate that there is an intermittent overnight problem (which is what I have with the headlights). Please clarify this.

-Hugh

 

[Witek_M]

@HughV
Let me clarify. This is customers vehicle and I have limited time to offer fix or punt the job. Right now I cannot reproduce the problem. Every time I open/close door to wake vehicle up it goes to sleep with 7mA draw after some time. I cannot simulate normal operation of start, drive, shot down because meter will not handle it.

Reason I have mentioned unloading family vehicle is that I am floored by the fact that merely opening drivers door creates ~25A draw. I go camping in deep back country often. Not doable in GLS.

 

[HughV]

Hi,
Many things are powered up when the door is opened. I have seen similar results. Dash is powered, who knows what else. Many systems are initializing getting ready for full MB experience! I think this is completely normal. Mine settles to 4-6-8A until the headlights turn off, couple of minutes, then goes to <100ma. Except when the lights internal motor runs forever until the battery dies.

Battery is rated for a couple hundred amp-hours so 40A for even 10 minutes should not bother it. This is why I was confused by your comment. I think your meter setup with 40A short draw is fine to find this, but it happens rarely. Running the car does not matter. to my knowledge.

In the end, I do not know if your problem is the same as mine, but same year car, presumably with same smart LED headlamps (242?) and same symptom. So I am very suspicious that it is the same issue.

Did you put your ear to the headlights after opening/closing the door (or just lock or unlock)?
I tried to find my test sheet to compare headlamp software revisions. Have not found it yet (and scanner not running yet, either!). Maybe the frequency of occurance is affected by the condition of the mechanism in the lights. Listen to yours and see how long they run, if at all, after lock-unlock or door open/close. I will look for my software rev to compare.

One solution to back country problem (when we must guarantee no dead battery) is to add a battery disconnect switch. But tough to lock the car without power! This is a bad solution for the customer, of course, but it will work. Last time I heard the motor, I moved the seat up and pulled the neg wire. Next day reconnected and I was off, all ok since then (montha ago). I am always worried about it, that is why I carry the power supply.

(can we talk about my random P0299 turbo lag/underboost codes? this my wallet is very frightened of)

-Hugh

 

[Witek_M]

@HughV
You were spot on. Opening/closing doors would not trigger the problem. Each time current went down to 7mA.
Driving around for about 15 minutes. Key off, noise from headlights. Those are actually cooling fans for LED drivers. After two hours I have disconnected battery. Ambient temperature 30F. No reason to cool anything.
Vehicle is going to MB dealer.

 

[HughV]

Good to hear that I am not the only one with this problem!!! The MB memo is faulty as it does not address the problem, particularly when it leads to dead batteries.

I will now bet (a dozen doughnuts) that there is some internal fault that requires replacement of the headlamps and they are not interested in doing this to thousands of cars. The memo does not address the drained battery problem. Well, it is hard to imagine that they can't come up with a software solution.

I guess that without a big shunt that you cannot measure the current draw with your setup because that would require disconnecting the battery after starting and driving..... (was just curious what your drain looks like)

This should be a safety recall issue.

In the meantime, I will look for the cooling fans inside the destined-for-disassembly headlight assembly I got from fleabay. MAYBE, their wiring is accessible from the external ports, and MAYBE I can figure out how to wire them to a switched source or something like that. I am busy for the next month or so and probably won't get to it until March sometime.

Please post what you learn from the dealer, with a focus on what the heck we are supposed to do with our randomly occurring dead batteries!!

Thanks.
Hugh

 

[Witek_M]

@HughV

After road test and fans stuck on I have left scanner connected, key off, doors closed. Looked at actual values from battery sensor. Current consumption -14.8A. Again this is not a very good test because scanner keeps CAN bus awake. Too bad I cannot access that sensor directly. It is LIN device.

I have ordered shunt you recommended last Saturday. We got blessed with three day ice storm so nothing got delivered. Schools were closed from Monday noon through Friday. Today's forecast sunny and 67F.

There are two control modules in each headlight. No description as what each one does. One drives LED's other controls movement is my best guess.

 

[HughV]

Hi Witek_M,
How did things work out with taking the car to the dealer for the headlight/battery drain? Mind is doing it again today. I pulled fuses 207 and 211 and it seems to have stopped, battery on charger.

Anyway, interested how things turned out. Hopefully a software update fixes this and not some $5k worth of headlight assemblies!

Thanks
Hugh

 

[Witek_M]

Hi Witek_M,
How did things work out with taking the car to the dealer for the headlight/battery drain? Mind is doing it again today. I pulled fuses 207 and 211 and it seems to have stopped, battery on charger.

Anyway, interested how things turned out. Hopefully a software update fixes this and not some $5k worth of headlight assemblies!

Thanks
Hugh

I need to follow up on this with customer. Thanks for reminder.