Hacking Battery Problem After

[lemmings123]

Just to confirm. Also had the battery issue after backing up nand with kobos linux img. Tried running down with brightness and sleep down/off on home menu. Switch would shutdown around 15-30% and wouldnt turn back on until docked. Disconnected battery today and instantly fixed the issue. Thanks Shadow LG for the tip

 

[Bitsoft]

Does this only happen if you run Linux? Or does it happen if you just booted RCM as well?

 

[Puppydogpals]

The battery pull imo will always work in all cases. Not that I've had this issue yet.

My reasoning for this is if you have a faulty battery and change it with a replacement the switch will will auto recalibrate to the new battery.

If it did not auto calibrate at this point, and thought the new battery was at a low level when in fact it was 100% it would never boot on battery.

As there is no software reset available from the the os or recovery this is the logical way Nintendo have decided to reset the battery memory (custom boot loaders for android have added this functionality)

As for should I press the power. It's good practice to do it, does it matter in this instance? Maybe not but if you took the effort and time to take your switch apart to disconnect the battery pressing the power for 15 seconds is not a huge hindrance

Other ways mentioned may work but they put extra stress on your battery by cycling the battery with charges and discharges

Sent from my SM-G935F using Tapatalk

 

[subcon959]

The battery pull imo will always work in all cases. Not that I've had this issue yet.

My reasoning for this is if you have a faulty battery and change it with a replacement the switch will will auto recalibrate to the new battery.

If it did not auto calibrate at this point, and thought the new battery was at a low level when in fact it was 100% it would never boot on battery.

As there is no software reset available from the the os or recovery this is the logical way Nintendo have decided to reset the battery memory (custom boot loaders for android have added this functionality)

As for should I press the power. It's good practice to do it, does it matter in this instance? Maybe not but if you took the effort and time to take your switch apart to disconnect the battery pressing the power for 15 seconds is not a huge hindrance

Other ways mentioned may work but they put extra stress on your battery by cycling the battery with charges and discharges

Do you envisage someone coming up with a software fix for this or is it likely to be an ongoing issue?

The reason I ask is the tinkerer in me was considering putting an external switch to disconnect the battery.

 

[Puppydogpals]

Do you envisage someone coming up with a software fix for this or is it likely to be an ongoing issue?

The reason I ask is the tinkerer in me was considering putting an external switch to disconnect the battery.

That I don't know. It could be (possibly is) possible but when....


If your having this issue now on multiple occasions. Every time you open up and remove the battery this increases the chance of damage (plus the inconvenience)

So a button mod might be a viable option to lets say capable people.

I won't say go for it as a software fix may turn up either through Linux or cfw then you will have a redundant switch on your switch!

If that's not an issue going forward (redundant switch) and you are capable enough to switch mod the battery and it would be of benefit to you then go for it!!

Sent from my SM-G935F using Tapatalk

 

[aut0mat3d]

Probably someone skilled is willed to dig into the Theme a little more deeper?
I think there must be a register in the PMIC that can be set/reset to have a forced recalibration which probably can be done via accessing it from linux or with a FC payload..
Think about how the Switch OS can mention the "full power loss" or Brownout. - this can be a register from the SOC or (hopefully) a register read out of the PMIC
Unfortuanally i can not find a detailed datasheet with register description for the m92t36 which should be in the console.
I believe it is a circuit from rohm from the bm92txx range.
Thats what i found so far:
http://www.rohm.com/web/global/datasheet/BM92T30MWV/bm92t30mwv-e
Unfortuanally mostly hardware side implementation, only references to appnotes which i did not found on their homepage for free access...

 

[Shadow LAG]

Just to confirm. Also had the battery issue after backing up nand with kobos linux img. Tried running down with brightness and sleep down/off on home menu. Switch would shutdown around 15-30% and wouldnt turn back on until docked. Disconnected battery today and instantly fixed the issue. Thanks Shadow LG for the tip

It's Shadow LAG , and happy to help.

Also looked into attempting to reset the power state / clear nvram from Linux; unfortunately I don't think this is the way forward. A payload may be better suited for a software fix.

https://unix.stackexchange.com/questions/317864/how-to-clear-reset-smart-battery-memory

 

[Syren1]

Does this only happen if you run Linux? Or does it happen if you just booted RCM as well?

Same question here !

 

[x0x0]

I would like to share with you the battery experience. The first time I launched linux 2 days ago only for 5 minutes. I went to test the battery switch turned off at 40%. Yesterday I played linux a few hours without starting Horizon. Today I decided to do another battery test, this time the switch turned off at 30%. The result is that it does not matter how much you sit in linux the battery will be decalibrated from the first start. I'm sure we'll get a payload reset battery or something like that, so do not panic and enjoy Linux.

 

[DerProGamer2000]

when my switch turns off at 20% or so is the battery completly dead and it is just displayed wrong or is it the firmware which turns the console off even tho the battery is 20% charged?

 

[alkar]

when my switch turns off at 20% or so is the battery completly dead and it is just displayed wrong or is it the firmware which turns the console off even tho the battery is 20% charged?

" is it the firmware which turns the console off even tho the battery is 20% charged" this part

 

[DerProGamer2000]

" is it the firmware which turns the console off even tho the battery is 20% charged" this part

well the first part would be more understandable but when it is the second one thats really weird because it displays 20% so it knows that its just but at the same time it thinks its dead? wtf

 

[alkar]

well the first part would be more understandable but when it is the second one thats really weird because it displays 20% so it knows that its just but at the same time it thinks its dead? wtf

that's right

 

[Vicfr21]

Anyone solved this issue only trying to drain battery?? I tried to do with fusee gelee payload and it turn off with 7%
(I don't have the screwdrivers...)

 

[naitsa]

Anyone solved this issue only trying to drain battery?? I tried to do with fusee gelee payload and it turn off with 7%
(I don't have the screwdrivers...)

Can anyone please state if this kind of battery problems are "just" a linux related thing, or are they caused by everything RCM related. Im thinking abut messing around a bit with RCM payloads and hekate (using linux is not intended). Now im affraid of risking battery issues. Or is this just linux related?

 

[Vicfr21]

Can anyone please state if this kind of battery problems are "just" a linux related thing, or are they caused by everything RCM related. Im thinking abut messing around a bit with RCM payloads and hekate (using linux is not intended). Now im affraid of risking battery issues. Or is this just linux related?

It's just linux related because I used hekate for weeks and it was doing well until then... but if you want more security I recomend to wait atomosphere cfw which is way more stable

 

[killplaystation]

Can anyone please state if this kind of battery problems are "just" a linux related thing, or are they caused by everything RCM related. Im thinking abut messing around a bit with RCM payloads and hekate (using linux is not intended). Now im affraid of risking battery issues. Or is this just linux related?

The switch OS doesn't measure the actual remaining battery capacity. It keeps track of a percentage and adjust the percentage based on how much you use the switch or charge the switch.
So when you boot into Linux it uses that battery power but the switch OS isn't decreasing that percentage because you're not running that OS. So if you're in the switch OS and you have 50% then you boot into Linux and use 25% more you would think you'd be at 25% remaining.
However the actual switch OS will still think it is at 50% because it's keeping track not measuring remaining capacity. So it's going to power down with what appears to be 25% remaining but actually there's nothing remaining.
It's also not going to charge to an actual 100% but rather to 75%. Because it has lost track of that 25%.
It's not hurting the battery and it's not caused by RCM. It's just the way the switch keeps track of battery percentage

Edit: based on this logic the easiest way to prevent or fix this without disconnecting the battery is the same. Boot to Horizon aka the actual switch OS and charge to, what it thinks, is 100%. After disconnecting from the charger immediately boot into Linux and charge it to an actual 100% in Linux. When you disconnect the power immediately boot back to Horizon.
Now the actual battery percentage and the Switch OS "keeping track of" percentage are synced again

 

[jakkal]

Lol at you people messing your switches up over incomplete Linux ports.

"Hackers"

 

[x0x0]

Ok, everyone who has a problem with the battery. Let me read. I did not use the switch for a good week. He was in sleep mode in the dock all the time. Today I started Zelda to test the battery, I have 5% battery and the Switch still did not turn off. It looks like it has been repaired.

 

[Bartesco]

https://twitter.com/SciresM/status/999232271843442690