Hardware o3DS - Weird battery behaviour

[Kaiboo]

Hey all, I have a o3DS (~10 years old) which recently started to exhibit some weird behaviour.

For a bit of context (not entirely relevant to the issue I suppose), after ~2 months of not using it, I decided to finally mod my 3ds with Luma (no CFW prior) (start of November).
Now after (successfully) playing on it relatively infrequently ~1 or 2 weeks ago something weird happened. I fully charged my 3DS and started playing on it. Suddenly, after like 5 to 6 minutes, without warning the red LED started flashing and within 15 seconds shut down entirely. Well, after some research I looked at the battery as it fit the symptoms and indeed it was a little bloated so I ordered a replacement. Now, this replacement came in 2 days ago, so I put it in and started playing on it. Guess what, after some time of playing, same thing it starts blinking red and almost immediately shuts down. Since then I have been researching possible issues while charging/discharging it with the hope that maybe the charge indicator was 'smart' and not voltage based, but to no avail.

I already did some small experiments using a Multi-Meter (MM) where I tested the voltage when the battery was 'full' and when the battery was 'empty' and time measurements:

• Battery full: 4.12V according to MM, 4.07V according to Rosalina
• Battery empty: 3.28V according to MM and 3.27V according to Rosalina
• Charging takes ~50 minutes with device off
• Discharging (in homescreen) takes ~55 minutes
• Once battery hits 50% (Roselina menu), it almost instantly decays to 30%, then to 10%, then to 5% and shuts down
• Charging with device on to look at Roselina menu indicates that after reaching ~66% charge it suddenly jumps to 100% charge and charging ends

The replacement battery is a CTR-003 rechargeable battery 3.7V 2000mAh from ZedLabz. (I already saw that batteries with 1500+ mAh are falsely advertised as so, but even so I would expect it to live longer than an hour and behave properly)

Now my question is, is there a chance this replacement is actually also bad (but then why would it last 10 times longer despite exhibiting the same bad battery symptoms?) or is there potentially some other hardware issue going on that I'm not aware of? I would rather not ruin this new battery (if it is not already) by continuously having it deep discharge and recharge and/or ruin my 3DS/SD card by having it randomly shut down in the middle of playing all the time.

Thanks in advance,
- Kai

 

[impeeza]

before considerate a bad battery for the replacement, let make 5 or 6 full charge-discharge cycles, to the firmware get battery calibration values.

 

[Kaiboo]

I think I already did 5 or 6 cycles, as I initially hoped that it just had to recalibrate, but I will run a few more cycles just in case and let you know if anything changes in the recharge duration and/or discharge duration

 

[IC_]

It might simply be a very low "chinese" quality battery, falsely advertised capacities are just a red flag for that. I don't think the 3DS has any battery calibration systems at all.

 

[Kaiboo]

Ah oke, I ordered another replacement battery from Cellephone which I saw pop by in a few other Battery related threads and will let you know how it fares when it comes in .

Just for curiosity sake though, why would this battery last ~11 times longer than the OG battery if they are both bad and not just (near) insta-fail ? Does it have to do with how the battery is structured? Like does the battery contain smaller energy cells of which some appear good while they are bad and then fail when they are relied upon or something?

 

[Kaiboo]

Yep, as you suspected the first replacement battery was also faulty . The second replacement solves the issue , thanks!

 

[Sono]

Leaving this here for people in the future having the same problem:

Doing the voltage test both unloaded (out of the 3DS) and loaded (as returned by the MCU) is actually a pretty easy (although kind of crude) indicator of how worn the battery is. Due to battery chemistry shenanigans, an unloaded battery (even if very empty!) is prone to recover voltage, sometimes to a stupidly misleading degree. Because of this, measuring loaded voltage on a half-empty battery while the 3DS is doing a light task, then unloaded voltage should give a crude approximation of how worn the battery could be. If the voltage drop is small enough, then the battery is probably fine. However if the battery drops a few hundred milivolts then it could be of concern, as the 3DS shouldn't use *that* much power to cause such voltage drop.
The battery has to be around half-empty (preferably around 70%-60%), as a fully charged battery has different curve characteristics, which skew the results of the already crude test.

Secondly, battery chemistry has still advanced since the release of the old3DS. It *might* be possible to cram 1800mAh into such small area, but it would need to be slightly thicker. It's definitely slightly more dangerous though, as for that you'd have to make the outer plastic much thinner than on the original battery. 2000mAh is really pushing it though, that sounds pretty unlikely.

Third, I found some tables related to voltage monitoring code, which may be what is used to calculate the battery percentage (hard to tell, the math used is pretty garbage). If that is the case, then that means that no matter how many times you deep-discharge and deep-charge the battery (which is not even possible, because the charge controller is forcibly turned off when the voltage reaches around 4.2V), the battery percentage curve will be the same, as the MCU (which reports and co-ordinates all of this) has no permanent storage, and it only writes to itself on software update, to replace its own code. That means that if your battery is worn out, then you'll get incorrect battery percentages, and it will not re-callibrate itself.

Oh, and the sudden percentage drop applies to every device running a battery due to the first point: because the battery has less capacity, voltage drop is very proportional to the current use and the battery wear/remaining charge, the output collapse point will happen at a much higher voltage.

 

[Youthanaser]

Yep, as you suspected the first replacement battery was also faulty . The second replacement solves the issue , thanks!

Hi! I recently buyed a battery that has the very same behaviour that the first one you posted... So, just to be sure, you just buyed a new battery (a 2nd new one) and the problem was solved ?
I mean, if that's the solution then I will buy another one but I wanna be sure

Thanks in advance!

 

[KaySutcliff]

THIS IS THE POST I HAVE BEEN LOOKING FOR!

I have the same exact issue as the original poster (o3DS from 2011, battery dies at 50%). However, in my case the original battery lasted around 2 or 3 hours and I decided to change it, also for a ZedLabz battery because they were the only ones who would ship it to me. This ZedLabz replacement battery is the one that dies at 50%, the original one went through the battery lines really fast, but this one goes from two lines to red in a matter of seconds.

At this point I'm considering putting back the original battery and seeing what happens because at least I can take it to a different room without having to worry about it turning off before I can save my game. I did follow the ZedLabz instructions about how to use it for the first time and have gone through several charging cycles and it doesn't show signs of improving whether I'm playing or just have the menu open.

It seems Cellephone is out of 3ds batteries at the moment, but I'll try ordering one soon and will report back once I get it.

 

[KaySutcliff]

Short update, I went back to the original Nintendo battery and it lasts between 2 and 3 hours of playing Pokémon with the internet connection on. Not ideal but WAY BETTER than the ZedLabz battery.