Rocksolid Light

Welcome to novaBBS (click a section below)

mail  files  register  nodelist  faq  login

You have the capacity to learn from mistakes. You'll learn a lot today.

tech / / Re: Battery charge tests - running a battery to 0 frequently - checking re-charge times

Subject: Re: Battery charge tests - running a battery to 0 frequently - checking re-charge times
From: Andy Burnelli
Organization: NNTP Server
Date: Sat, 7 May 2022 04:31 UTC
References: 1 2 3 4
From: (Andy Burnelli)
Subject: Re: Battery charge tests - running a battery to 0 frequently - checking re-charge times
Date: Sat, 7 May 2022 05:31:29 +0100
Organization: NNTP Server
Message-ID: <t54smd$17mm$>
References: <t510ud$2t9$> <> <t53fmp$l1h$> <>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Info:; logging-data="40662"; posting-host=""; mail-complaints-to="";
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.6.1
X-Notice: Filtered by postfilter v. 0.9.2
Content-Language: en-GB
View all headers
Jeff Liebermann wrote:

No, it's not.  I guess I should be more specific.  I would like to
know why you find it necessary to test a LiIon cell in a charge range
of zero to 20%, where literally every recommendation by the
manufacturers declare that to be an RBI (really bad idea)? 

I _know_ it's a really bad idea, which is _why_ I want to see what happens. I got a handful of these phones for free, so I have _plenty_ to work with.
It's just a phone. Phones are a dime a dozen. They're a commodity.

This is a Samsung Galaxy A32-5G, which, even if I paid for it, is cheap.

People do drop tests all the time, don't they?
That's a really bad idea too. :)

Looks that
specs for any BMS (battery management system) found inside most LiIon
battery packs.  There is a feature that literally disconnects the cell
if the terminal voltage goes below some value which usually works out
to about 20% charge.  Maybe this will help you understand the problem
you're creating for yourself:
"Lithium Ion Cell Operating Window"
Notice that the "operating area" is between 20% and 90% SOC.

I designed microcontroller-based batter chargers as experimental tools back
in the 80's in graduate school but it has been a long time since then.

I've probably forgotten more than I learned by then, but my main point is
that I consider all smartphones merely a tool to play with to learn about.

If you don't do destructive testing, you won't learn as much as if you do.
(As an aside, you should see the destructive testing I did on the iPad!)

I take risks with the mobile devices all the time in order to learn more.

For example, I recently figured out a way to change my supposedly permanent
supposedly unique Google-supplied GSF ID, which may be the reason that
the Android 11 to Android 12 update on my Samsung deleted hundreds of apps.
 *Have you ever tried to CHANGE your unique GSF ID on your Android device?*

So it's not just me who is curious what happens in the real world.

Yes, but you are not the entire real world.  Your currently
undisclosed operating criteria is not the same as every user and
certainly not the same as the cell phone manufacturer.  The
manufacturer wants big numbers because big number sell phones.
Whatever it takes to produce big numbers balanced by cost and safety
issues.  Big numbers are rather useless if the phone catches fire in
the owners pocket.  So, the game of battery specmanship degenerates
into squeezing as many watt-hours out of the battery as possible by
any means deemed economical (and maybe reliable).  Do it wrong, and
you have a situation like Apple, where the phone had to be slowed down
to produce a reasonable runtime as the battery aged.  At that point,
the user gets involved and tries to squeeze out as much power as
possible.  However, they can't because the manufacturer has already
done that with a complexicated BMS algorithm.  So the user looks to
see what can be gained by breaking the safety rules.  Good luck.  If
you are actually able to run the phone at extremely low SOC, then the
manufacturer has screwed up and is selling an unsafe phone, battery,
or both.  What phone and battery are you using and I'll be sure to
blacklist it.

It's the Samsung Galaxy A32-5G and it's the T-Mobile model SM A326U.

You'll see some of my experiments on the XDA Developers' forum:

I've always been curious about the best way to do almost any thing.
And destructive testing is a fantastic way to figure out what really
happens in the real world under real world conditions, even as you can't
hope to run a "consumers report" style full-fledged scientific
investigation with basic home equipment.

Please note my domain name,  The intent is
slightly different from yours.  It's my contention that one does not
understand how something works without first breaking it, and
subsequently fixing it.  Destructive testing, without subsequent
understanding (and enlightenment) is useless.

Like you appear to do, I try to learn why things fail when they fail.

For my BMW, a lot of parts break, where I take them apart to find what
caused the failure, whether that's the FSU or the expansion tank.

You can't put that stuff back together, although I did try to reseat the
myriad "angel hair" wires inside the ABS control module, which are just
impossible to do without specialized machines.

Having said that, my cars are decades old where I fix everything I can,
even to the point of mounting and balancing my own tires at home, so if I
_can_ fix it after taking it apart, I will fix it after taking it apart. Analyze the wear Keep close track of wear Watch wear over time Inspect root causes Run experiments Gather more data Test your assumptions ID engineering principles Fix root cause Buy a new tire Break the bead Replace the valve Mount the tire Balance the wheel Recycle the carcass

Bear in mind it's difficult to find out why some wear patterns happen
simply because they are on perfectly well aligned vehicles sometimes.

Still... I try to learn... and destructive testing is part of learning.

It's a tiny part but admittedly the fun part.  It's lots of fun to
blow things up.  It's less fun, but more educational to understand how
the device you just destroyed functions.  When you destroy something
(like your phone battery), do you take or record measurements?  Do you
record a video for an instant replay?  Have you worked out in advance
what you expect to happen?  Do you look for anomalies?  Do you own a
data logger?  How would blowing up a cell phone battery demonstrate
anything if you don't know at what voltage (or SOC) and temperature it
blew up?  Did you put a plastic bag over the phone to capture any
gasses (and flying glass) produced?  Do you have a new battery or
phone available for comparisons?  Without these, all you've "learned"
is how to blow up a battery or phone.

I can't disagree with anything you've said as I run experiments all day
every day, some of which are detailed (like those tire experiments) and
others aren't as detailed (as my phone battery experiments are).

When I was a kid, my dad kept a box of old "stuff" for me to take apart.

Hint:  I still act like I'm kid.  I even take things apart BEFORE I
try operating them.

The one thing really convenient about destructive autopsies is you don't
have to put anything back together. You just put it in the recycling bin.

Why does anyone run any experiment?

Usually because they are suspicious of the established theories of
operation and have reason to suspect that parts of the theories are
wrong or badly understood.

Notice the tire experiments above where there is precious little
information on the Internet why all the cars that are perfectly aligned in
any given twisty road area wear the tires the exact same abnormal way.

Take a look at the Scotty Kilmer video below, and note particularly that
the photos I've been posting (which I've posted for years) are the _same_
as in Scotty Kilmer's videos (just look and you'll recognize my pictures!).

This is a shot he took of my BMW tires, for example, with wear lined up: He stole my images! :)
I find it a testament to my thoroughness that Scotty Kilmer stole my photos
(which clearly are the same photos as I've been posting here all along).

o Battery charge tests - running a battery to 0 frequently - checking re

By: Andy Burnelli on Thu, 5 May 2022

132Andy Burnelli
rocksolid light 0.7.2