Hacking [IMPORTANT] PSA: AutoRCM battery warning

[Deleted User]

It works because you have a charger plugged in and thus power, when a charger is plugged in the Switch will boot without pressing power, so that should be no problem as the "charging code" is loaded immediately.

When we are talking RCM the PC can't supply enough power to reliably keep the Switch alive if the battery is truly dead, so getting to a chargeable state can be impossible.

Then if that's the problem, can't someone just create a lightweight payload that gives the right code for the switch to charge and queue it. But even then, it takes a moment for the switch to turn on from a dead battery, and that kind of implies that it stocks a charge first before loading the os or any code for that matter.

 

[Taffy]

It works because you have a charger plugged in and thus power, when a charger is plugged in the Switch will boot without pressing power, so that should be no problem as the "charging code" is loaded immediately.

When we are talking RCM the PC can't supply enough power to reliably keep the Switch alive if the battery is truly dead, so getting to a chargeable state can be impossible.

Are you implying a switch in RCM is always connected to a PC?

 

[Lacius]

Then if that's the problem, can't someone just create a lightweight payload that gives the right code for the switch to charge and queue it. But even then, it takes a moment for the switch to turn on from a dead battery, and that kind of implies that it stocks a charge first before loading the os or any code for that matter.

Creating a payload that charges the Switch isn't the problem, since one can just boot Hekate while in RCM and charge it that way. The problem is, apparently, that a completely dead Switch with AutoRCM installed won't have any way to charge or enter RCM.

 

[Deleted User]

Creating a payload that charges the Switch isn't the problem, since one can just boot Hekate while in RCM and charge it that way. The problem is, apparently, that a completely dead Switch with AutoRCM installed won't have any way to charge or enter RCM.

then, that should apply to the horizon as well, because at that point it cannot launch it's code to have USB charging.

 

[Lacius]

Are you implying a switch in RCM is always connected to a PC?

If you're saying that a dead Switch with AutoRCM could be plugged into an outlet and receive more power, that apparently won't work since it won't charge while in RCM.

then, that should apply to the horizon as well, because at that point it cannot launch it's code to have USB charging.

A dead Switch without AutoRCM installed can enter a charging state when plugged in, since it's not going (or trying to go) into RCM.

 

[Taffy]

Creating a payload that charges the Switch isn't the problem, since one can just boot Hekate while in RCM and charge it that way. The problem is, apparently, that a completely dead Switch with AutoRCM installed won't have any way to charge or enter RCM.

oooooohhhhhhhh

damn software-controlled charging

then, that should apply to the horizon as well, because at that point it cannot launch it's code to have USB charging.

Does a standard USB port provide enough current/voltage to charge a switch?

 

[Deleted-442439]

then, that should apply to the horizon as well, because at that point it cannot launch it's code to have USB charging.

No, because if you plug in a charger, it will attempt to power on, one of the first thing that is loaded by the bootloader is the charging code, so it takes close to no power at all to get it running.

RCM however does not get passed the bootloader, and does not have the same power supplied + requires more power to get into HOS.

 

[Lacius]

Does a standard USB port provide enough current/voltage to charge a switch?

Usually, yes. For a Switch in RCM, apparently not.

 

[SanktyRG]

Ok so the Switch's battery only charges while the battery code is present, RCM has nothing like this, so in theory, even if sleep mode is fixed for the payloads, if the battery dries completely, only way to charge back is by placing the battery on another Switch, is this correct?

 

[Deleted-442439]

Usually, yes. For a Switch in RCM, apparently not.

Exactly. If you boot into RCM the usb charging will not do anything, it will only send data if autorcm. If not autorcm the initial "spark" will be enough to get stuff running.

--------------------- MERGED ---------------------------

Ok so the Switch's battery only charges while the battery code is present, RCM has nothing like this, so in theory, even if sleep mode is fixed for the payloads, if the battery dries completely, only way to charge back is by placing the battery on another Switch, is this correct?

Yep.

 

[Deleted User]

I think I'm going to test this. But I'm not going to install autorcm, I'm going to test it in a different way. So the switch boots into rcm mode when it gets powered on and vol up and a jig is in. So, in theory, If I where to hold the vol up button while plugging in the power cord for a completely drained switch with a jig, then my switch should enter rcm mode. If rcm mode lasts longer than 4 seconds, or long enough to enter horizon by payload, then this is flawed because it implies that the switch get's enough power to be in autorcm for a short amount of time. I'm going to test this. I'll see you in roughly 2 hours

 

[Deleted-442439]

OK.

So @ktemkin has just clarified this on Discord, turns out Tux was wrong.

If the SoC is completely dead (absolutely no power) the PD controller has a mode for saving the battery where it will "trickle charge" until the battery has a certain charge, so this is a false alarm.... however it can still fail requiring the battery to be removed and then reinserted to fix.

 

[Deleted User]

OK.

So @ktemkin has just clarified this on Discord, turns out Tux was wrong.

If the SoC is completely dead (absolutely no power) the PD controller has a mode for saving the battery where it will "trickle charge" until the battery has a certain charge, so this is a false alarm....

CALLED IT. Well glad this is over.

 

[rajkosto]

except the PD controller can still get stuck somehow and not let the console to power up, so you will have to disconnect and reconnect the battery in that case

 

[KazoWAR]

wait... if a switch can't charge unless its on, then is a switch with a completely dead battery not able to charge at all? even if its not hacked?

 

[James310]

I almost never fully turn off the Switch aince it’s always on sleep mode so I hope this won’t cause a bad day for me in the future...luckily sx os supports sleep

 

[Lacius]

except the PD controller can still get stuck somehow and not let the console to power up, so you will have to disconnect and reconnect the battery in that case

What are the odds of that happening?

 

[Deleted User]

I almost never fully turn off the Switch aince it’s always on sleep mode so I hope this won’t cause a bad day for me in the future...luckily sx os supports sleep

and atmosphere will support sleep when it gets it's official release

 

[GBA rocks]

So it’s all good.

Make sense since RCM is also a normal occurrence during Switch assembling and testing, why would they make it in a way that kills the emergency trickle charging.

 

[sj33]

Don't forget that even if this is a false alarm, the battery desync that will occur by leaving your Switch in RCM mode for too long is real. So don't stay in RCM mode for too long. TX people shouldn't leave their Switch on the TX menu screen because it won't sync the battery properly.