My 2006 CLS 320 has been sitting on my drive for nearly 5 weeks without running. I went out this morning and there was a quite loud buzzing noise coming from the engine near the passenger side. I opened the car and started it but the noise continued. I had to shut the car down and disconnect the battery to make it stop. Any ideas?

 

My idea is that as this is a GLA-dedicated forum you may be better served at a Benz forum that discusses a number of different/older models, such as Mercedes-Benz Forum

 

My 2020 GLA 250 was parked in the garage for 2 hrs. I heard something under the hood. Opened the door... to pull the hood latch... and the noise quit. So, I can't detect where the noise is coming from. WHY would opening the door have anything to do with it?

 

To add some of my own observations regarding electrical activity during sleep mode for a 2019 GLA 45:

-The blower motor sometimes does turn on after an hour or so after shut off. I've timed it, and it goes for just about 30 minutes. You can indeed feel a very gentle flow of air coming out of the vents during operation. At one point, using an amp clamp on the battery negative cable, I noticed a current of around 500 mA. This was after the car had been parked for a period of time, and the immediate module activity directly after parking was already long gone. When I pulled FUSE 202 in the engine bay fuse box ("Stationary Heater Control Unit"), the amperage dropped to the typical 50 mA quiscent draw that is always present when the car is sleeping. I looked at the WIS electrical schematics, and noticed that the blower motor is in some way connected with this fuse (it appears to also be related to a fuse in the passenger's footwell, and I have confirmed this by measuring voltage drop across the fuse while the blower was operating).
-When the car is in sleep mode, and even after it has been sitting over night, I have heard a hardly audible noise that has lasted for several hours that sounds like the flow of liquid through a pipe or tubes. It sounds exactly like a toilet tank refilling after a flush. To my ears, and using a sound detector app on my phone, it appears that the sound is coming from what appears to be a small black, cylindrical pump that is behind the driver's side of the front bumper. This may very well be the pump referenced in Post #25 above. I pinched most of the tubes coming out of it, but they simply felt as if they had air in them; I did not detect any change in sound, feel any vibration or evidence of flow, etc.
-On a few rare occasions, I have heard a very brief whistling noise from the engine bay after the car was in sleep mode for a while, exactly like the noise made when you let out a short burst of air from a balloon by pinching and stretching outward on the balloon's opening (didn't we all use to make that noise to annoy everyone when we were kids--or was I just a butthead? [Rhetorical question here]). It definitely sounded like a gas being released. I remember looking into it, and I concluded it likely had to do with the AC/evaporation system.
-Indeed, when there is an electrical draw such as the blower motor running, when the car is locked or unlocked, the windows are remotely lowered/raised, etc., the eletrical draw will cease.
-The SOS system goes off with some regularity as the car sleeps (maybe once per hour or even less). I am not sure if this requires that someone is seated in the driver's seat such that the seat weight sensors are activated. I have observed this SOS check many times while sitting in my car. I believe that I have witnessed it while not sitting in my car as well, since I have seen a few flashes of light coming from that location while walking past the car at times; but I am not certain.

Here are the results of a 2.5-day timelapse I made with a Go Pro camera set up to take pictures every 5 seconds while a voltmeter and amp clamp monitored the OEM MBZ main battery during sleep mode on my 2019 GLA 45. The X-axis shows time in seconds while the two Y-axes show amperage in A, and voltage in V. I opened and closed the door once every 12 hours in order to wake the system up and let it go back to sleep again, as well as to correct any drift that might occur with the voltmeter/amp clamp. The final spike is me starting the car (I had to truncate the A and V peaks in order to make the overall scaling appropriate).

Here are some conclusions from the graph:
-It's hard to tell due to the scaling, but quiscent amp draw for my GLA 45 is definitely just about 40-50mA (I have tested/confirmed this many times beyond the above data, including using a higher resolution amperage setting on the amp clamp than that used above). I had no peripheral devices that draw from the battery with the vehicle shut off.
-There are uknown voltage drops that occur every once in a while (I suspect that at least some of these are due to the aforementioned SOS testing, as some of the timelapse images show a light flash near the top of the windshield--perhaps all are due to SOS, but I would not know, as an image is only taken once every 5 seconds, and the flashes do not perist for quite that long).
-It can be seen that voltage and amperage are inversely related, as would be expected. If you look at the raw data, it's amazing how closely they are correlated. I used two completely separate devices to measure the voltage and current, and even tiny fluctuations in current consistently correlate with tiny flucutations in voltage--the sensitivity and consistancy of this amazes me and bolsters my faith in the fidelity of the measurements.
-The overall battery voltage recuperates after each drop, but it definitely slowly decreases over time. In this case, my battery went from ~12.55V to ~12.35V over the course of 2.5 days in ~50F weather (I should note that my battery's health during this testing was not the best; in fact, these tests were taken in order to gather information about the health of my battery during this time--I won't go into more details, but I believe the above data are applicable to the average OEM GLA battery (A001-982-8108, 12V 80AH 800 CCA, AGM) and shed light on some of the electrical activity going on while in sleep mode.

Anyway, the above are some observations I have made regarding electrical activity in the GLA 45 while in sleep mode.

 

Great attending to detail. I've seen the same. Also, sometimes the battery discharges significantly faster than other times. I haven't figured that one out.

I feel I'm getting repetitive here, but being retired with two cars, one car or the other can go a couple weeks without being driven. Short of hooking up a battery tender every time we get home, I've found a monitor that automatically disconnects the battery has saved us a bunch of grief. So I've put one of these on each car ...

LI LEAD Auto Starter 12V Car... https://www.amazon.com/dp/B07LGCY3RR?ref=ppx_pop_mob_ap_share

With the reconnect button mounted on the lower corner of the windshield we no longer worry about going out to the car to find a dead battery. Pam was ready to dump the 45. She's now happy again.

No, this isn't a fix to the parasitic drain. I consider it insurance to avoid getting burned by that drain.

 

@ Wayne:

I would love to see a picture of your setup under the hood. I've thought about going the disconnect route so many times, but I've been uncertain about how everything would fit. I was also concerned that disconnecting the main battery for more than a few days might cause the small auxiliary battery to get drained, thereby leading to reset and other problems.

I was thinking of experimenting with removal of the battery negative that's above the driver's side fender. Maybe replacing that brass nut with a high quality wing nut? I don't know if removal of just that one cable would serve as an appropriate disconnect, whatwith the sensors and other added complexity of the OEM battery connections at the battery posts.

 

What I have right now isn't super pretty but it seems to work great. There have been times that the battery has been disconnected for several days and upon reconnection the car has started immediately with no codes or problems. Even when the main battery dropped to 12 and disconnected the aux battery stayed at 12.6, leaving me to realize that I don't understand the isolation!!!

On my Porsche if the disconnect occurs I do need to reset my auto window limits (manually drive them down and then up) and drive a short distance to clear the PSM (Porsche Stability Management) warning light. But that's easy.

I installed this about six months ago on the 45. Making it permanent would have required cutting the battery cover, so I opted for a wait and see approach. The knee pad you see in the picture covers the + terminal in case of a hood crushing accident (yes, unlikely). The pictures were taken in a dark driveway. But they should convey the idea.

The wait and see period is over. I'm happy with the solution. So ... I found this alternative with longer cables that might have worked really nicely ... but the item I received was the older model and didn't work properly. So back it went.

RIEIEK Smart Car Battery... Amazon.com

This would be neater, but required carrying a remote for reconnection. But it is supposed to auto connect any time your phone is near. I'm not sure if I like that or not. It could defeat the whole purpose of the unit.

I hadn't thought of attaching to the fender cable. I don't see any reason why this shouldn't work. I'll have to look at that tomorrow when I pull the car into the garage to replace the brakes. Thanks for the idea.

 

That's great. Thank you for sharing. Really not a bad setup, and I'm surprised you managed to fit it so well, given the bulk of the unit. I think the issue with the negative post above the fender is that the hood comes down pretty close to it, and doesn't leave much room in that area for anything. I'll probably stew over some possible solutions over the next week or so, and I'll post again if I come up with any other worthy ideas.

 

@ Wayne:

Quick question: Assuming you have Keyless Go, does that feature go away once you've disconnected the main battery? I assume it does, but I'm not sure if perhaps the aux battery might be in control of that.

 

If the doors don't unlock then the battery is disconnected. We press the button on the windshield which connects the battery, and touch the inside of the door handle to unlock the car. Everything returns to normal, no hassles.

 

@ Wayne:

Quick question: Have you ever tried to start the GLA without engaging the battery disconnect first? I was researching the pros and cons of having a battery disconnect in line with the negative vs the positive battery cables, and I was told that if such a device is in line with the negative, and one tries to start the vehicle, the alternator will try to pull current from wherever it can, and if there is a ground wire still connected somewhere in the system--say, to the radio--that wire will get toasted. A guy who has sold these devices for over a decade said he's seen it happen twice now.

In other news, I took some measurements under the hood to see how easily a disconnect device like yours can be put on the fender end of the battery negative cable. I now think there's easily enough room above the battery cover for such a device. For example, above the battery cover, there's at least 2.25" of height (more like 2.5-2.75" even) and there'd be no problem in the forward-backward or side-to-side directions. I put a stack of 3x5 cards 2.25 inches tall all around on the battery cover, and there was no problem with fitment. I think that the design of your particular disconnect switch would require some specific cables/adapters for connection, though, since the device has a built-in "post" connector vs terminal screws that take eyelet connections.

 

In order to get into the car (unlock it) I need to connect the battery. I have not waited a minute before starting to allow the battery monitor to disconnect again.

It would be easy enough to ohm the disconnected terminal cable to frame. But if it did have an alternate ground path that wouldn't matter as long as the ground terminal at the battery is disconnected ... there's no path for the positive side of the battery to get to the negative post.

Now ... if you use the fender bolt as your disconnect that might be a different story, since there are multiple wires connecting at the battery terminal. You've made a good point as to why this disconnect should only attach directly at the battery.

 

In order to get into the car (unlock it) I need to connect the battery. I have not waited a minute before starting to allow the battery monitor to disconnect again.

It would be easy enough to ohm the disconnected terminal cable to frame. But if it did have an alternate ground path that wouldn't matter as long as the ground terminal at the battery is disconnected ... there's no path for the positive side of the battery to get to the negative post.

Now ... if you use the fender bolt as your disconnect that might be a different story, since there are multiple wires connecting at the battery terminal. You've made a good point as to why this disconnect should only attach directly at the battery.

I see. By "multiple wires connecting at the battery terminal," I assume you're talking about the electronic battery sensor at the negative battery terminal (the rectangular unit with a wire going to it that is used to measure battery current, voltage, temp, ...). Other than that sensor, it appears that the negative cable simply goes straight from the battery post to the fender connection point. Unless there's more to it than I understand--and there probably is--that would be the only wire that would be vulnerable? I'm imagining that the battery is like a canister of compressed/pressurized electrons, and if the alternator is initiated, they rush out of the negative terminal of the battery as long as there is a downstream "valve" open. If the disconnect switch is at the fender location, then the only path the electrons can take is not through the large battery negative cable toward the fender ground, but rather through the small sensor wire that connects at the negative post. Since the demanded starting current is so large, the friction from the flow heats up and melts that sensor wire, effectively turning it into a fuse. Is that basically the idea here?

 

When you remove the cable terminal with all of its connections from the negative post of the battery there can be no current flow even if you connect the positive side of the battery directly to the frame.

Then you put the disconnect switch on the negative post and the cable terminal that used to be there on the other side of the switch.

Now if the switch opens its the same as pulling the entire cable. No connection. No current anywhere. No damage possible.

 

When you remove the cable terminal with all of its connections from the negative post of the battery there can be no current flow even if you connect the positive side of the battery directly to the frame.

Then you put the disconnect switch on the negative post and the cable terminal that used to be there on the other side of the switch.

Now if the switch opens its the same as pulling the entire cable. No connection. No current anywhere. No damage possible.

Yes, that makes sense. I'm guessing that if the disconnect is put on the fender side, while leaving the original connection to the battery negative, then the only path for current would be down that tiny set of wires that go to the battery sensor still connected to the negative of the battery. I'm just trying to pinpoint where the vulnerability would lie.

 

I have a 2017 GLE350 and there is a headlight fan that continues to run all the time. the car has 2 new battery's and every night they are drained and I have to jump the car to get it started every morning. Other tan that the vehicle runs perfectly. I have takem it in to 2 different mechanics shops and it hasn't been fixed. it is put on the computer and they are ablt to shut the fan on and off and it will stop running for a while, but within a week or 2 it starts running again. do you think it could be a computer module that gets stuck or does anyone have an idea of what it can be?