laptop battery charging stategy (Plasma)
I've heard and read many a time over recent years that the way batteries should be charged has changed. I've always been someone who lets devices run right down as low as I can before charging them back up to 100%. With one or two exceptions, this strategy still seems to work remarkably well for me, regardless of what others say. My previous Nokia phone would run for days under each charge, my current Android one goes five days or more between charges (I'm a light phone user), but conscious of this continual hounding from random people who may or may not be tech-inclined, and having both a laptop and a phone that offer flexi-charging options, I decided maybe I should change that strategy. So following a similar setting I just made on the Android phone, I set the Flexi-charging option in my Clevo laptop BIOS. It only allows adjustments in increments of 10%, so I set the lower Start point to 50% and the upper Stop point to 90%. Though I have a laptop, I almost always use it sat at my desk with it plugged in and hooked up to a bigger monitor and other peripherals. I like the versatility of taking my main PC with me on the odd occasion I go and stay with friends or family, but I don't frequent cafés with wifi because I'm not a hipster, and I'm terribly uncomfortable having the laptop on my lap or anywhere else. So for most of its life, it is plugged in being constantly topped up to 100%. Once or month or so I try to unplug it and use it on battery to let the battery drain and refill. When I do that, the laptop invariably throws a fit and suddenly warns about critical battery level even though it's still at about 45%, and switches off within a few seconds. I get these weird messages in my notifications: Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166486][ C0] Uhhuh. NMI received for unknown reason 20 on CPU 0. Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166498][ C0] Do you have a strange power saving mode enabled? Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166500][ C0] Dazed and confused, but trying to continue I don't think there's anything strange about my power-saving settings, it seems more likely the continual charging fools something in the subsystem as regards the actual charge level. Using flexi-charging is supposed to make battery reporting LESS accurate, but it's clearly already problematic. And usually, once I plug the laptop back in, the charge level shows back at 45%. So now having changed the BIOS setting, I see in Plasma (5.24) that the battery icon shows permanently in the system tray, because it's being constantly topped up... to 90%. I thought maybe what would happen is that it would decharge to 50% and then the charging would kick in and bring it back to 90%. I don't know if that would be better or worse, I can't see as hundreds of charge cycles will help, and in any case there's no options to set it to behave any differently, that would require additional software. Some of you must have a similar setup. What have your results been like with whatever charging strategy you use? I've had this laptop three years now and according to the Plasma Info Centre the battery health is at 94%, so it hasn't weathered badly. Should I abandon this experiment and disable the flexi-charging? The battery is replaceable anyway so if I really needed maximum battery life one day on this machine I could get another one. gumb
On 11/07/2022 22.16, gumb wrote:
I've heard and read many a time over recent years that the way batteries should be charged has changed. I've always been someone who lets devices run right down as low as I can before charging them back up to 100%.
Well, battery chemistry has changed. That strategy was recommended for Ni-Cd ones, to avoid the "memory" effect they had.
With one or two exceptions, this strategy still seems to work remarkably well for me, regardless of what others say. My previous Nokia phone would run for days under each charge, my current Android one goes five days or more between charges (I'm a light phone user), but conscious of this continual hounding from random people who may or may not be tech-inclined, and having both a laptop and a phone that offer flexi-charging options, I decided maybe I should change that strategy.
I charged my last 3 phones by leaving them connected the whole night on a slow charger: works perfect, no adverse effects, they all live long lives and still do. Only the oldest, maybe 8 year old, is suffering of old age. My later phone controls charging, and uses slow charging itself. It adjusts the charge to finish by the time of the wake up alarm. So, in modern laptops, continuously connected but limiting the charge to, say, 90%, should be perfect. ...
to unplug it and use it on battery to let the battery drain and refill. When I do that, the laptop invariably throws a fit and suddenly warns about critical battery level even though it's still at about 45%, and switches off within a few seconds. I get these weird messages in my notifications:
Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166486][ C0] Uhhuh. NMI received for unknown reason 20 on CPU 0. Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166498][ C0] Do you have a strange power saving mode enabled? Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166500][ C0] Dazed and confused, but trying to continue
Unrelated.
I don't think there's anything strange about my power-saving settings, it seems more likely the continual charging fools something in the subsystem as regards the actual charge level. Using flexi-charging is supposed to make battery reporting LESS accurate, but it's clearly already problematic. And usually, once I plug the laptop back in, the charge level shows back at 45%.
So now having changed the BIOS setting, I see in Plasma (5.24) that the battery icon shows permanently in the system tray, because it's being constantly topped up... to 90%. I thought maybe what would happen is that it would decharge to 50% and then the charging would kick in and bring it back to 90%.
If this happens, you would see it in the display. Better then to keep at 100. -- Cheers / Saludos, Carlos E. R. (from openSUSE 15.3 (Legolas))
On 12/07/2022 07:39, Carlos E. R. wrote:
My later phone controls charging, and uses slow charging itself. It adjusts the charge to finish by the time of the wake up alarm. So, in modern laptops, continuously connected but limiting the charge to, say, 90%, should be perfect.
I think it makes more sense on my phone, because it's only plugged in for charging, otherwise it's running on battery. Whereas with my laptop, it's plugged in 99% of the time. So I don't really know what advantage or disadvantage there is in the system continually topping up the battery to 100% as opposed to 90%. Does the latter prevent excess energy usage or overheating? If I were running it like my phone and doing repeated charging/discharging cycles I can understand the concept of letting it modulate between a restricted charging range, but if I've charged it once to 100% and then it just keeps it there, does that really do any harm longer term? By restricting it to 90%, will the battery start to 'lose its memory' of that extra 10% and render itself only nine-tenths as powerful? I never see these specifics addressed anywhere because all articles and conversations about the topic assume laptop users run primarily on battery like they would with a telephone.
I get these weird messages in my notifications:
Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166486][ C0] Uhhuh. NMI received for unknown reason 20 on CPU 0. Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166498][ C0] Do you have a strange power saving mode enabled? Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166500][ C0] Dazed and confused, but trying to continue
Unrelated.
Well, it is related. I should get typically around 3 hours' battery life, but rarely when I unplug do I not encounter this situation where with the battery level at anything between 99% and >20% (i.e. non-critical) and everything seeming normal, the laptop abrubtly goes into a critical battery level panic and attempts to hibernate. I can stop it if I scramble for the power lead and plug it in within x seconds, where x always equals half a second before I manage to plug it in. When I power it back on it sometimes recovers from hibernation, sometimes just does a fresh boot suggesting data loss or corruption could have occurred (though to date I've not noticed any). And I get these messages implying I must have some weird power saving setup and hence it's all my fault. Yet I have pretty much the same power settings as I've always used on my two previous laptops without encountering this warning. Usually, after this has occurred once, if I plug it back in and then unplug, it doesn't happen again, as if the charge level statistics have been reset and corrected. I opened the laptop the other day to do some cleaning since it's started increasingly making fan noise having been entirely silent for two and a half years prior. Fan and interior was all perfectly clean. Applied some fresh thermal grease to the CPU/heatsink. Still makes increasing fan noise. In putting it back together it requires a bit of squeezing tight to put the screws in, because it seems to me that the battery has puffed up to some degree. I didn't try removing it or any of the surrounding insulating material to inspect it in detail. I don't recall it being like that the only other time I opened it two or three years ago but I didn't pay too much attention back then. Makes me wonder if that's a common thing or whether my continual charging could be causing it to enlarge, which would be worrying. Unfortunately, unlike my two previous laptops which had detachable batteries, this one's removable but only by taking the device apart. Because if I could remove it easily I would, and just run on AC power. gumb
On 14/07/2022 10.00, gumb wrote:
On 12/07/2022 07:39, Carlos E. R. wrote:
My later phone controls charging, and uses slow charging itself. It adjusts the charge to finish by the time of the wake up alarm. So, in modern laptops, continuously connected but limiting the charge to, say, 90%, should be perfect.
I think it makes more sense on my phone, because it's only plugged in for charging, otherwise it's running on battery. Whereas with my laptop, it's plugged in 99% of the time. So I don't really know what advantage or disadvantage there is in the system continually topping up the battery to 100% as opposed to 90%. Does the latter prevent excess energy usage or overheating?
AFAIK, yes.
If I were running it like my phone and doing repeated charging/discharging cycles I can understand the concept of letting it modulate between a restricted charging range, but if I've charged it once to 100% and then it just keeps it there, does that really do any harm longer term? By restricting it to 90%, will the battery start to 'lose its memory' of that extra 10% and render itself only nine-tenths as powerful? I never see these specifics addressed anywhere because all articles and conversations about the topic assume laptop users run primarily on battery like they would with a telephone.
There is no memory effect on the current batteries. However, it is easy to overcharge, and keeping at 90% avoids that risk. Mind, as long as it is a constant 90%, not cycling.
I get these weird messages in my notifications:
Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166486][ C0] Uhhuh. NMI received for unknown reason 20 on CPU 0. Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166498][ C0] Do you have a strange power saving mode enabled? Message from syslogd@linux at Jul 3 15:34:04 ... kernel:[ 87.166500][ C0] Dazed and confused, but trying to continue
Unrelated.
Well, it is related. I should get typically around 3 hours' battery life, but rarely when I unplug do I not encounter this situation where with the battery level at anything between 99% and >20% (i.e.
Not related. The message is about some of the machine CPU sleep states, not battery modes.
non-critical) and everything seeming normal, the laptop abrubtly goes into a critical battery level panic and attempts to hibernate. I can stop it if I scramble for the power lead and plug it in within x seconds, where x always equals half a second before I manage to plug it in. When I power it back on it sometimes recovers from hibernation, sometimes just does a fresh boot suggesting data loss or corruption could have occurred (though to date I've not noticed any). And I get these messages implying I must have some weird power saving setup and hence it's all my fault. Yet I have pretty much the same power settings as I've always used on my two previous laptops without encountering this warning. Usually, after this has occurred once, if I plug it back in and then unplug, it doesn't happen again, as if the charge level statistics have been reset and corrected.
Battery problem just happens at the same time, then machine tries to sleep to conserve battery, and then the kernel is confused. But the message is related to CPU estates.
I opened the laptop the other day to do some cleaning since it's started increasingly making fan noise having been entirely silent for two and a half years prior. Fan and interior was all perfectly clean. Applied some fresh thermal grease to the CPU/heatsink. Still makes increasing fan noise. In putting it back together it requires a bit of squeezing tight to put the screws in, because it seems to me that the battery has puffed up to some degree. I didn't try removing it or any of the surrounding insulating material to inspect it in detail. I don't recall it being like that the only other time I opened it two or three years ago but I didn't pay too much attention back then. Makes me wonder if that's a common thing or whether my continual charging could be causing it to enlarge, which would be worrying. Unfortunately, unlike my two previous laptops which had detachable batteries, this one's removable but only by taking the device apart. Because if I could remove it easily I would, and just run on AC power.
gumb
Dunno. 2yrs is too soon for degradation in modern laptop. -- Cheers / Saludos, Carlos E. R. (from openSUSE 15.3 (Legolas))
On 2022-07-14 03:16:26 Carlos E. R. wrote:
|> I opened the laptop the other day to do some cleaning since it's started |> increasingly making fan noise having been entirely silent for two and a |> half years prior. Fan and interior was all perfectly clean. Applied some |> fresh thermal grease to the CPU/heatsink. Still makes increasing fan |> noise. In putting it back together it requires a bit of squeezing tight |> to put the screws in, because it seems to me that the battery has puffed |> up to some degree. I didn't try removing it or any of the surrounding |> insulating material to inspect it in detail. I don't recall it being |> like that the only other time I opened it two or three years ago but I |> didn't pay too much attention back then. Makes me wonder if that's a |> common thing or whether my continual charging could be causing it to |> enlarge, which would be worrying. Unfortunately, unlike my two previous |> laptops which had detachable batteries, this one's removable but only by |> taking the device apart. Because if I could remove it easily I would, |> and just run on AC power. |> |> gumb | |Dunno. 2yrs is too soon for degradation in modern laptop. | |-- |Cheers / Saludos, | | Carlos E. R. | | (from openSUSE 15.3 (Legolas))
All of your arguments seem to assume a good battery, but this sounds to me like a defective battery. The swelling of the battery is significant, IMO. Leslie -- Operating System: Linux Distribution: openSUSE Leap 15.4 x86_64 Desktop Environment: Trinity
participants (4)
-
Carlos E. R. -
Carlos E. R. -
gumb -
J Leslie Turriff