Discussion:
Should I completely discharge my laptop battery?
(too old to reply)
a***@gmail.com
2014-12-30 19:53:15 UTC
Permalink
I have a Satellite A660 laptop with a Lithium-Ion battery pack. I've
read that you should completely discharge batteries every month so
that the charger can recalibrate on what is 0% versus 100%. Does this
apply to Lithium-Ions? This also means I turn off the safety feature
of hibernating when the charge dwindles to 8%. Doesn't the disk heads
crash once the charge dies?
~misfit~
2014-12-31 00:35:34 UTC
Permalink
Post by a***@gmail.com
I have a Satellite A660 laptop with a Lithium-Ion battery pack. I've
read that you should completely discharge batteries every month so
that the charger can recalibrate on what is 0% versus 100%. Does this
apply to Lithium-Ions?
Yes, especially to Li-Ion as they last so long that the control circuitry
needs new info on the state of the cells from time to time.
Post by a***@gmail.com
This also means I turn off the safety feature
of hibernating when the charge dwindles to 8%.
Yes.
Post by a***@gmail.com
Doesn't the disk heads crash once the charge dies?
No. Newer laptops are clever enough to stop in a controlled manner as the
battery pack is exhausted. This is another good reason to fully discharge
your battery now and then - so the control circuitry knows at what voltage
it's near-empty.
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
a***@gmail.com
2014-12-31 19:06:26 UTC
Permalink
Thank you, misfit. I googled both before posting and after your response. I didn't run across anything about saving hard drive heads from crashing when the battery runs out, but I did see that one could set Power Options' Advanced settings to turn off the hard drive at a certainly charge level.

About completely discharging the battery, Windows 7 safety features are frustrating my attempts to do so. It's not enough to create a new power plan (say, "Exhaust battery") where the computer never goes to sleep. There are other controls hidden in the advanced power settings for Low, Critical, and Reserve battery levels, and the last one seems to be the problem -- you can't specify "Do nothing" (at least not via the GUI). According to http://superuser.com/questions/622545/disable-critical-battery-event, you have to issue the following command:

powercfg -setdcvalueindex SCHEME_CURRENT SUB_BATTERY BATACTIONCRIT 0

*Sigh*. All I want to do is discharge the battery. Thank goodness for Google.
a***@gmail.com
2014-12-31 19:54:22 UTC
Permalink
Post by a***@gmail.com
Thank you, misfit. I googled both before posting and after your
response. I didn't run across anything about saving hard drive heads
from crashing when the battery runs out, but I did see that one
could set Power Options' Advanced settings to turn off the hard
drive at a certainly charge level.
<...snip...>

Oops, my bad. The setting doesn't turn off the HD at a certainly
charge level. It turns of the HD after a certain time.
a***@gmail.com
2015-01-01 23:13:11 UTC
Permalink
Post by a***@gmail.com
Thank you, misfit. I googled both before posting and after your
response. I didn't run across anything about saving hard drive heads
from crashing when the battery runs out, but I did see that one
could set Power Options' Advanced settings to turn off the hard
drive at a certainly charge level.
About completely discharging the battery, Windows 7 safety features
are frustrating my attempts to do so. It's not enough to create a
new power plan (say, "Exhaust battery") where the computer never
goes to sleep. There are other controls hidden in the advanced
power settings for Low, Critical, and Reserve battery levels, and
the last one seems to be the problem -- you can't specify "Do
nothing" (at least not via the GUI). According to
http://superuser.com/questions/622545/disable-critical-battery-event,
powercfg -setdcvalueindex SCHEME_CURRENT SUB_BATTERY BATACTIONCRIT 0
*Sigh*. All I want to do is discharge the battery. Thank goodness for Google.
Oops, my bad. [In the 1st paragraph] The setting doesn't turn off
the HD at a certainly charge level. It turns of the HD after a
certain time.
Darn, another error on my part. It isn't the Reserve battery level
for which I had to use the powercfg command to set to do nothing.
It was the Critical battery level.

From googling yesterday (sorry, I didn't capture where I surfed), the
Reserved level is something different. Windows & apps will save stuff
when Reserved level is reached. From what I could find, reaching the
Reserve level doesn't trigger anything that interferes with the
discharging, e.g., sleep, hibernate.
~misfit~
2014-12-31 23:29:12 UTC
Permalink
Post by a***@gmail.com
Thank you, misfit. I googled both before posting and after your
response. I didn't run across anything about saving hard drive heads
from crashing when the battery runs out, but I did see that one could
set Power Options' Advanced settings to turn off the hard drive at a
certainly charge level.
About completely discharging the battery, Windows 7 safety features
are frustrating my attempts to do so. It's not enough to create a
new power plan (say, "Exhaust battery") where the computer never goes
to sleep. There are other controls hidden in the advanced power
settings for Low, Critical, and Reserve battery levels, and the last
one seems to be the problem -- you can't specify "Do nothing" (at
least not via the GUI). According to
http://superuser.com/questions/622545/disable-critical-battery-event,
powercfg -setdcvalueindex SCHEME_CURRENT SUB_BATTERY BATACTIONCRIT 0
*Sigh*. All I want to do is discharge the battery. Thank goodness for Google.
Just another word or two here; You actually don't want to *completely*
discharge the battery pack - more like discharge it to the point where there
is no usable power left in it. (There needs to be some power left to run the
battery's internal monitoring electronics - if these die then the pack's a
brick.)

So just 'excercising' the battery by discharging to <5%, fully charging and
repeating the process (maybe twice) should do it.

Mike makes some good points. Doing this can brick your half-good battery
pack so deep discharging is usually something I only do when the pack has
already got to the stage where it's only borderline useful, in the hope of
regaining some 'capacity'.

Then again I use ThinkPads which allow me to set charging thresholds (I set
my machine to start charging below 50% and stop at 85% and maybe twice a
year perform a battery reset through ThinkVantage Power manager). Also I
store currently unused packs in the fridge at ~40% charge.

Best of luck - and happy new year.
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2014-12-31 20:05:26 UTC
Permalink
Post by ~misfit~
Post by a***@gmail.com
I have a Satellite A660 laptop with a Lithium-Ion battery pack. I've
read that you should completely discharge batteries every month so
that the charger can recalibrate on what is 0% versus 100%. Does this
apply to Lithium-Ions?
Yes, especially to Li-Ion as they last so long that the control circuitry
needs new info on the state of the cells from time to time.
Post by a***@gmail.com
This also means I turn off the safety feature
of hibernating when the charge dwindles to 8%.
Yes.
Post by a***@gmail.com
Doesn't the disk heads crash once the charge dies?
No. Newer laptops are clever enough to stop in a controlled manner as the
battery pack is exhausted. This is another good reason to fully discharge
your battery now and then - so the control circuitry knows at what voltage
it's near-empty.
I have a contrary opinion.
I don't have any NEW systems, so this is slightly out of date,
but the concepts haven't changed.
Surely someone will come up with a counterexample that proves me stupid,
so, you'll have to decide whether you believe me.

I've only done this on a dozen or so batteries, so can't guarantee all are
that way. All this assumes that a worn battery is just worn and does
not have shorted cell and not part of a stupid system design
that puts some of the cells right adjacent to the processor heat sink
and cooks them.

There are at least three levels of protection.
A coulomb counter basically counts the electrons as they go in/out.
Knowing the battery capacity, you can calculate the state of charge. I
believe the laptop polls the battery to determine
the state of charge and decides what to do. That's an orderly shutdown
or hibernation as you've instructed in the setup. There's likely a
voltage threshold too.

Battery and laptop vendors are very concerned that you don't sue them
over battery issues aka fires. They care not that you have to buy
a new battery. At retail, they probably make more profit off the
battery than they did on the laptop. I read it on the internet so it
must be true.

There are at least two other discharge protections in the battery pack.
There's a shutdown voltage below which the control FETs open up and
disconnect
the battery. This is NOT a controlled shutdown. The battery just
disconnects. This is likely what you get when you turn off the protection
and run it till it quits.

There's a end of life voltage below which the pack will not recover.
You can't normally (see below) get there by discharging.
But you can get there
by sticking a discharged battery in the drawer for a couple of years letting
the discharge protection circuit systematically discharge the cell.
Sometimes it's firmware in the microcontroller that cannot be
reset unless you know the secret method and can reprogram it.
I've seen some that have a thermal fuse adjacent to a resistor.
At the end of life
threshold, they light up the resistor and blow the thermal fuse.
In either case, that battery pack ain't coming back. The only pack
I've ever recovered had a PIC processor as controller and hitting
the reset did actually reset it.

Some of the old dell dimension packs would let you open them up,
charge the cells directly and they would run the computer again.
They'd continuously flash an error code and the battery gauge didn't
work, but you could run the computer.

I think it was a Toshiba that had an end of life mode that would
let you run the laptop off the battery. OR, you could charge the
battery with the computer off. If you wanted to run the laptop
off AC, you had to physically remove the battery. Strong motivation
to buy a new battery, or in my case, strong motivation to never buy
another Toshiba.

I don't think I ever got a Thinkpad battery back.

So, why do batteries fail?
Over time, the cells just can't store as many electrons and capacity
decreases some.
But my experience is that the primary failure is due to increased
internal series resistance.
The symptom is as follows...
You charge the pack and the battery level shows 100%.
You disable the battery warning modes.
You run the computer and the meter drops steadily like
you'd expect until it hits some level like 40% and drops
immediately to zero and the computer shuts down soon thereafter.

If you take the battery out and test it under controlled conditions,
you find that most of the electrons are in there and can be recovered
at low current. If you load with high current, the current x high
series resistance
subtracts from the voltage to the point that the protection circuit
shuts you down at shutdown voltage. The electrons are in there, but
the pack won't let you take them out.

Now, we get to the relevant part.
I've had a two systems that had about 50% usable battery capacity.
I decided I'd discharge them in hope of resetting the battery gauge.
I turned it on and let it sit till it quit. And it never came back.

Here's my thesis. I think the system sat there at low power with the
display backlight and hard drive off until it hit the point where the
laptop decided to take some action. Most of the capacity was gone.
When the disk started and the backlight started, the additional voltage
drop due to the increased current in the high battery internal series
resistance
dropped the battery voltage all the way to the end of life threshold.
Game over.

That's why I'm very wary of running systems 'till they quit. Seems to
work fine on slightly used batteries. When the resistance gets high,
you may just kill a perfectly half-good battery.

If it ain't broke, don't fix it.

YMMV
~misfit~
2014-12-31 23:55:54 UTC
Permalink
Once upon a time on usenet mike wrote:

[snip very useful if slightly disjointed info]
Post by mike
I don't think I ever got a Thinkpad battery back.
I have - the one in this 15" UXGA IPS 4:3 screen T60 - It's labeled 'Lazarus
Battery'. I got it in a second-hand machine that used factory presets for
charging (whenever charge drops below 95% top up to 100% - a usage pattern
designed to get you to buy a new battery every two years or so).

When I got it it reported only 10% of new capacity so I did a 'battery
reset' using ThinkVantage Power Manager (TVPM). It died, it wouldn't take a
charge, the battery LED in the ThinkPad just flashed yellow. When powered up
(with adapter) power manager said 'battery dead - replace'.

I threw it in a drawer for 6 months. Then I thought it's either rubbish or
it's not so put it in a zip-lock bag in the freezer for a week, then put it
in a ThinkPad, plugged in the adapter but didn't turn the machine on
(concerns about condensation as well as not wanting TVPM to make a hasty
decision and 'mark it' dead again). The battery LED flashed yellow for 15
minutes - then stayed on solid yellow (meaning charging at 5% capacity or
above)! I left it for a day or two (the LED had been on solid green for
quite a while) then started the laptop. It now said the pack held 25% of
original capacity (yay!).

That was two years ago and I've been using it since then in this machine
that is never taken out of it's dock, using it as a UPS, set to charge when
below 50% and stop at 85%. However I've since used TVPMs battery reset
feature a couple of times, about once a year, and the reported capacity has
risen to 40% of original. I have the laptop set to maximum power (as it's
essentially used as a low-power draw desktop) and when there's a power
outage it usually runs for at least 30 minutes before I decide to turn off.

That's a heck of a lot better than a 10% battery only good for 5 minutes
then dirty shut down - or a brick.

I have two new repack batteries (a 6 cell and a 9 cell) stored at 40% charge
in the fridge as I also have a 14" SXGA+ T60 that is used infrequently as a
portable machine (I'm an invalid and don't get out much). The 14" has a 90%
good battery in it ready to go - which reminds me, I must cycle that
battery.

Happy new year.
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
a***@gmail.com
2015-01-01 22:51:39 UTC
Permalink
Mike, Misfit,

That was an overwhelming amount of insight into how batteries degrade.
I'm not really an electronic technology (anymore), but I appreciate
your sharing. I don't anticipate going the lengths of
resuscitating a battery, though I'm sure that the information will
give the more technologically oriented readers some good ideas.

About Misfit's advice that one doesn't need to discharge much below
5%, this seems to imply (to me) that the charging circuitry isn't
losing track of what constitutes a low charge, it is losing track of
what constitutes a full charge. Is this correct?

And this risk of bricking the battery -- from reading your responses,
it seems that this is a very plausible risk even if I just run it to
the point where the computer shuts down (not putting it in a drawer
for half a year). It's a bit ambiguous, because one can interpret
this as the point at which there is "no usable power left", but then
again, there was that story about high internal resistance developing
a high voltage drop once the laptop draws current to do shutdown
activities. As I said, I've got the laptop set to do nothing when it
reaches the Reserve battery level, but I have zero trust that windows
actually does nothing in the same way as it does if you were to just
yank the battery.

Anyway, given this ambiguity, it looks like 5% is what you recommend
as a good level to discharge to if I just want to recalibrate.
Thanks.
~misfit~
2015-01-02 01:16:25 UTC
Permalink
Post by a***@gmail.com
Mike, Misfit,
That was an overwhelming amount of insight into how batteries degrade.
I'm not really an electronic technology (anymore), but I appreciate
your sharing. I don't anticipate going the lengths of
resuscitating a battery, though I'm sure that the information will
give the more technologically oriented readers some good ideas.
About Misfit's advice that one doesn't need to discharge much below
5%, this seems to imply (to me) that the charging circuitry isn't
losing track of what constitutes a low charge, it is losing track of
what constitutes a full charge. Is this correct?
Sort of. As Mike pointed out older cells can develop high internal series
resistance (ISR). So it all depends on what your particular battery pack
uses to trigger various states. Most use voltage and, if the ISR has risen,
at lower voltages the cells still have the energy stored but can't deliver
it at high load levels. So, with warnings turned off, when the computer asks
for quite a few amps (CPU usage goes up) it can die suddenly. The power is
still in the battery but it can only deliver it at low rates of current.

It's complicated stuff really and isn't helped by, as Mike says, laptop
manufacturers err on the side of caution so will often have their control
circuitry (in both battery pack and laptop) tell you a battery has less
capacity than it does - and so limit your access to lower voltage states as
the cycle count goes up. They make a lot of money out of selling replacement
batteries. This doesn't take into account however what conditions the
battery has been used under and whether it has been deep-cycled or
constantly topped off.

(The latter is the default for most laptops and is really bad for
batteries - especilly at raised temps, which a lot of batteries run at - and
increases their ISR more quickly than if the batteries were cooler and were
kept closer to the ideal state of charge for longevity - around 40% charge.)

My experience is with ThinkPads and the newer ones (post 2006 / T43 and
later) allow the user to manually set the charge parameters. If you do that
as I do - so the thing doesn't constantly charge to full every time the
charge drops below 95% - then the batteries last twice as long at least.
They also have software for 'battery reset' - where they fully charge /
discharge the pack and recalibrate the control circuit built into the pack -
basically what you're trying to do now.

So, not knowing how your system manages it's battery pack I can only give
generalised advice. Almost all laptop battery packs use solder-tab 18650
cells. I wish I could design and build a pack for my laptops that uses
replaceable 18650s (like my flashlight does) and contained circuitry to
manage them. Much like replacing AA cells in a remote controller. However as
Mike pointed out the manufacturers make big money selling replacement
batteries. :-/
Post by a***@gmail.com
And this risk of bricking the battery -- from reading your responses,
it seems that this is a very plausible risk even if I just run it to
the point where the computer shuts down (not putting it in a drawer
for half a year). It's a bit ambiguous, because one can interpret
this as the point at which there is "no usable power left", but then
again, there was that story about high internal resistance developing
a high voltage drop once the laptop draws current to do shutdown
activities. As I said, I've got the laptop set to do nothing when it
reaches the Reserve battery level, but I have zero trust that windows
actually does nothing in the same way as it does if you were to just
yank the battery.
Are you not running the OEMs battery mangement software?
Post by a***@gmail.com
Anyway, given this ambiguity, it looks like 5% is what you recommend
as a good level to discharge to if I just want to recalibrate.
Yeah, it might not reclaim a huge amount of charge 'marked bad' but it also
shouldn't brick your battery pack.
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
a***@gmail.com
2015-01-02 02:11:59 UTC
Permalink
Post by ~misfit~
Post by a***@gmail.com
About Misfit's advice that one doesn't need to discharge much below
5%, this seems to imply (to me) that the charging circuitry isn't
losing track of what constitutes a low charge, it is losing track
of what constitutes a full charge. Is this correct?
Sort of. As Mike pointed out older cells can develop high internal
series resistance (ISR). So it all depends on what your particular
battery pack uses to trigger various states. Most use voltage and,
if the ISR has risen, at lower voltages the cells still have the
energy stored but can't deliver it at high load levels. So, with
warnings turned off, when the computer asks for quite a few amps
(CPU usage goes up) it can die suddenly. The power is still in the
battery but it can only deliver it at low rates of current.
<...snip...>
Post by a***@gmail.com
And this risk of bricking the battery -- from reading your
responses, it seems that this is a very plausible risk even if I
just run it to the point where the computer shuts down (not putting
it in a drawer for half a year). It's a bit ambiguous, because one
can interpret this as the point at which there is "no usable power
left", but then again, there was that story about high internal
resistance developing a high voltage drop once the laptop draws
current to do shutdown activities. As I said, I've got the laptop
set to do nothing when it reaches the Reserve battery level, but I
have zero trust that windows actually does nothing in the same way
as it does if you were to just yank the battery.
Are you not running the OEMs battery mangement software?
It's a Toshiba Satellite A660, and the 64-bit Windows 7 OS is the one
from the OEM install discs. The battery is also Toshiba, for this
laptop (perhaps for others as well, I never really looked into that).
Li-ion Battery Packmodel PA3817U-1BRS. An OEM webpage I found is
http://www.toshiba.com/us/accessories/Power/Batteries/PA3817U-1BRS.
There just doesn't seem to be a way to set the upper percentage level
at which the charger stops charging, and the lower percentage level at
which it starts charging. It charges to 100%, and as it discharges,
it encounters the Windows thresholds (Low, Critical, and Reserve) and
responds with the notifications and/or actions which have been
assigned to those events, e.g., sleep, hibernate.
Post by ~misfit~
Post by a***@gmail.com
Anyway, given this ambiguity, it looks like 5% is what you
recommend as a good level to discharge to if I just want to
recalibrate.
Yeah, it might not reclaim a huge amount of charge 'marked bad' but
it also shouldn't brick your battery pack.
Thanks. 5% is good enough for me. I'm interested in the gross
practices that can achieve gross improvements to longevity. Thanks!
mike
2015-01-02 09:15:17 UTC
Permalink
M
Post by a***@gmail.com
Post by ~misfit~
Post by a***@gmail.com
About Misfit's advice that one doesn't need to discharge much below
5%, this seems to imply (to me) that the charging circuitry isn't
losing track of what constitutes a low charge, it is losing track
of what constitutes a full charge. Is this correct?
Sort of. As Mike pointed out older cells can develop high internal
series resistance (ISR). So it all depends on what your particular
battery pack uses to trigger various states. Most use voltage and,
if the ISR has risen, at lower voltages the cells still have the
energy stored but can't deliver it at high load levels. So, with
warnings turned off, when the computer asks for quite a few amps
(CPU usage goes up) it can die suddenly. The power is still in the
battery but it can only deliver it at low rates of current.
<...snip...>
Post by a***@gmail.com
And this risk of bricking the battery -- from reading your
responses, it seems that this is a very plausible risk even if I
just run it to the point where the computer shuts down (not putting
it in a drawer for half a year). It's a bit ambiguous, because one
can interpret this as the point at which there is "no usable power
left", but then again, there was that story about high internal
resistance developing a high voltage drop once the laptop draws
current to do shutdown activities. As I said, I've got the laptop
set to do nothing when it reaches the Reserve battery level, but I
have zero trust that windows actually does nothing in the same way
as it does if you were to just yank the battery.
Are you not running the OEMs battery mangement software?
It's a Toshiba Satellite A660, and the 64-bit Windows 7 OS is the one
from the OEM install discs. The battery is also Toshiba, for this
laptop (perhaps for others as well, I never really looked into that).
Li-ion Battery Packmodel PA3817U-1BRS. An OEM webpage I found is
http://www.toshiba.com/us/accessories/Power/Batteries/PA3817U-1BRS.
There just doesn't seem to be a way to set the upper percentage level
at which the charger stops charging, and the lower percentage level at
which it starts charging. It charges to 100%, and as it discharges,
it encounters the Windows thresholds (Low, Critical, and Reserve) and
responds with the notifications and/or actions which have been
assigned to those events, e.g., sleep, hibernate.
Post by ~misfit~
Post by a***@gmail.com
Anyway, given this ambiguity, it looks like 5% is what you
recommend as a good level to discharge to if I just want to
recalibrate.
Yeah, it might not reclaim a huge amount of charge 'marked bad' but
it also shouldn't brick your battery pack.
Thanks. 5% is good enough for me. I'm interested in the gross
practices that can achieve gross improvements to longevity. Thanks!
There are several sides to this coin.

As a user, you get the best life out of your battery if you learn
what it can do and manage YOUR activities to maximize that.
Maximum charge level is hard to work around because it's automagic.
Having a programmable setpoint is about the only viable alternative.
But when you use it, what are you gonna do?
You've decided that you want to stop at 25%. Doesn't matter whether it's
manual or automagic. If you're in the middle of something, do you shut
down? Or do you finish your task? For most of us, having a battery that
needs replacement less frequently is small consolation when you couldn't
finish the presentation and lost the order.
So, you charge as frequently as you can and use it as little as possible.
If it doesn't do what you want you bought the wrong laptop, or need a
second battery.

From the vendor's perspective, as head of sales, do you want the ad
to say, 8-hr battery life? Or 5-hr battery life and saves 20% on battery
replacement costs, cuz all our other laptops abuse the battery?
Or maybe you say 2X faster processor and hide the fact that it hardly
runs long enough to boot.
These marketing guys are pretty smart. They know what sells.
The tablet-inspired expectation of 11 hour battery life is
affecting the whole laptop ecosystem. May not be long until
virtually every laptop follows the "surface" model of tablet
with keyboard. Most of us have wireless everything and don't
really need external ports.

Other factors include safety and protecting the replacement battery revenue
stream. Nobody with the frugal gene would ever buy a new battery at
retail. You can get crap batteries on ebay for cheap. Sad news
is that if you buy a "OEM" battery, you may get a crap battery
anyway. The plot thickens because the NEW battery
has been sitting in a container in
Arizona since the turn of the century.

Vendors can't have any of that!
I've seen laptops that refuse to boot with a counterfeit battery installed.
Some will run but won't charge. (you see that same symptom with some
counterfeit power bricks)
Others, give you a warning that you have to dismiss.
Some run the computer, but the battery gauge doesn't work and any
test reports "battery not installed."
The vendors want you to buy new laptops frequently. And batteries
are priced/specified to encourage that. Everybody wins...unless
you're concerned that, one day, the entire surface of the planet
might be covered in busted laptops.

The technology is quickly evolving. Batteries, some, but management
technology has come a long way. Vendors find it more marketable
to manage the best they can and replace the battery at your expense
frequently.
There are a number of battery management technologies, but the trend
is to make is impossible for a mere mortal to fix one.

Back in the day, I'd match NiCd's and re-cell. Turn 3 bad ones into
one good one. You can't solder battery packs together.
Even if you're crazy enough to try to solder on them,
there's so little clearance that the blob won't fit the case.
If you get the tabbed ones and solder the tabs, there's often
insufficient room to fit the tabs in.
I have two generations of DIY tab welders and a real
CD spot welder. Most of that is useless today. You can change the
cells, but if the chip won't let you use it, you're done.
If you crack the case on a Lithium battery, you can sometimes charge
each cell individually to balance it and it may run the laptop, subject
to all the issues stated above. I buy laptops with busted screens.
The battery is worth more than the value of trying to find a screen.
Working laptops at garage sales always have bad batteries. That's
why they are for sale. Ones with busted screens sometimes have
usable batteries. It's worth a buck to find out.

So, who cares about any of this?
Cheap bastards like me!
Statistically, the percentage is near zero.
IF you use your computer to make money, the cost of a
replacement battery is noise level.

My life revolves around frugality and prospecting for deals.
I buy a lot of stuff at great prices.
I try not to think about the fact that I could have bought
one new computer for less than the prospecting costs of the
dozen crap laptop treasures I have hanging around.
That's why I call it a hobby.
THEY call it hoarding. What do they know? I can stop any time
I want... ;-)

There's a set of freeware diagnostic tools called PCWizard.
Last I checked, the sweet spot was the 2010 version, but it's
worth a try at the newer versions.
There's a battery tab that shows you what the pack thinks it
is, design capacity, current capacity, current state of charge,
voltage etc.
Whether it works for you is a crap shoot depending on whether it supports
the methods/protocols used by your laptop. Do a reality check on
the numbers presented by battery analysis software. Because of the
different protocols in the wild, sometimes the numbers are just WRONG.

Don't fix it if it ain't broke.
a***@gmail.com
2015-01-02 17:34:19 UTC
Permalink
Mike, your last post was entertaining. As I said, though, I'm not as passionate about it as to make it a hobby. I'm OK with the gross guidelines of discharging to 5% once a month. I appreciate the rationale behind those guidelines as much as I appreciate the guidelines (which are not trustworthy without rationale). Most of the time, I'm on AC, and the battery is only a UPS. If it goes, I'll buy another (accepting the risk that it might have been on the shelf since the turn of the century). I'll just complain to the vendor if it's crappy, and if they play hardball, I'll complain in public. I know it will not make it above noise level because so many people do that anyway, but I am riding the technology improvement trend that is driven by capitalism. Batteries today are heads and shoulders better than a decades ago...except to enthusiasts who intend to crack them open to extend their lives. More power to you if you do, though.
mike
2015-01-02 18:46:51 UTC
Permalink
Post by a***@gmail.com
Mike, your last post was entertaining. As I said, though, I'm not as passionate about it as to make it a hobby. I'm OK with the gross guidelines of discharging to 5% once a month. I appreciate the rationale behind those guidelines as much as I appreciate the guidelines (which are not trustworthy without rationale). Most of the time, I'm on AC, and the battery is only a UPS. If it goes, I'll buy another (accepting the risk that it might have been on the shelf since the turn of the century). I'll just complain to the vendor if it's crappy, and if they play hardball, I'll complain in public. I know it will not make it above noise level because so many people do that anyway, but I am riding the technology improvement trend that is driven by capitalism. Batteries today are heads and shoulders better than a decades ago...except to enthusiasts who intend to crack them open to extend their lives. More power to you if you do, though.
Consensus seems to be that you should take the battery out if using it
on AC.
Assuming the design of the charger stops charging when it gets to max,
there's still the issue of heat.
I use mine little, so just leave the battery in on AC all the time.
If I used it a lot, I'd take out the battery.
I had more than one laptop where the battery sits two plastic
thicknesses away from the processor heat sink.
When you take the case off the battery, the ones closest to the
heat are discolored...and much higher resistance than the others.
Some designer oughtabeshot.

It's been said that lithium cells degrade almost as fast on the shelf
as if you use 'em sensibly. A spare battery is great if you need to use
it. If it's sittin on the shelf justincase, it may not be a good
investment. Still might be better sittin' in your fridge than
having it sittin' in that container
in Arizona till you buy it. ;-)
a***@gmail.com
2015-01-02 22:25:44 UTC
Permalink
Post by mike
Consensus seems to be that you should take the battery out if using
it on AC. Assuming the design of the charger stops charging when it
gets to max, there's still the issue of heat. I use mine little, so
just leave the battery in on AC all the time. If I used it a lot,
I'd take out the battery. I had more than one laptop where the
battery sits two plastic thicknesses away from the processor heat
sink. When you take the case off the battery, the ones closest to
the heat are discolored...and much higher resistance than the
others. Some designer oughtabeshot.
It's been said that lithium cells degrade almost as fast on the
shelf as if you use 'em sensibly. A spare battery is great if you
need to use it. If it's sittin on the shelf justincase, it may not
be a good investment. Still might be better sittin' in your fridge
than having it sittin' in that container in Arizona till you buy it.
;-)
Understood, Mike. However, my usage for the battery is 95% as a UPS,
and 5% as a power source when away from AC. That's not necessarily
the proportion in which I value the two functions, but it is how the
battery spends its time (give or take a healthy margin of
uncertainty). I guess I could get a real UPS for slight more than a
battery:

http://www.amazon.com/gp/top-rated/electronics/764572
http://www.toshiba.ca/productdetailpage.aspx?id=4730

However, I need a battery anyway, so by keeping the battery in and
shortening the life by some amount, I get the convenience of a zero
footprint UPS and avoid having to buy a real UPS. Subjectively, it's
worth it to me at this point, just based on the fact that I haven't
spent the time thinking about a UPS til now. Within that scenario,
though, I still should try to maximize the battery life, which I did
by identify the gross best practice of discharing to 5% once a month.
~misfit~
2015-01-03 23:28:40 UTC
Permalink
Once upon a time on usenet ***@gmail.com wrote:
[snip]
Post by a***@gmail.com
However, I need a battery anyway, so by keeping the battery in and
shortening the life by some amount, I get the convenience of a zero
footprint UPS and avoid having to buy a real UPS. Subjectively, it's
worth it to me at this point, just based on the fact that I haven't
spent the time thinking about a UPS til now. Within that scenario,
though, I still should try to maximize the battery life, which I did
by identify the gross best practice of discharing to 5% once a month.
Andy - I don't know where you got this once a month figure from? IMO two or
three times a year is the most you'd need to do this considering your usage
patterns.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
a***@gmail.com
2015-01-04 00:48:45 UTC
Permalink
Post by ~misfit~
Andy - I don't know where you got this once a month figure from? IMO two or
three times a year is the most you'd need to do this considering your usage
patterns.
I saw it a number of times during my web wanderings. However, I just did another wander, and it seems that your advice above is more common. Thanks.
~misfit~
2015-01-07 01:47:48 UTC
Permalink
Post by a***@gmail.com
Post by ~misfit~
Andy - I don't know where you got this once a month figure from? IMO
two or three times a year is the most you'd need to do this
considering your usage patterns.
I saw it a number of times during my web wanderings. However, I just
did another wander, and it seems that your advice above is more
common. Thanks.
You're welcome. Once a month is simply wasting cycles IMO - there is a
finite number in each cell (the number of cycles depends on variables but is
still finite). This site is invaluable IMO; http://batteryuniversity.com/
especially this page
http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

I've just finished cracking open a Sony Viao battery pack for the cells to
use in a flashlight or other project. The (circa 2009) laptop was bought to
me to fix but was deemed to expansive to repair by its owner so left with
me. I salvaged various parts but for me the most valuable was the battery.
Sony make good quality cells and the owner said the battery was still good.

The pack had a very complicated control PCB - around 50 SMDs on it with one
IC having 48 legs, another with 30 legs, one with 20 leg and five ICs with
eight legs. However it treated the six cells in pairs so only ran three
'channels'. My Bosch 18v Li-Ion cordless power tools battery packs have five
channels, one for each cell. The tools all use the same battery pack so I
have two in use and two stored in sealed zip-lock bags in the fridge - which
reminds me - I should bring those up to room temp and cycle them soon before
chilling them again.

Anyway now I have another six 1st tier 18650 cells to play with. Learning
their (remaining) capacity is the hardest part for me. If I wasn't so broke
I'd buy a charger / analyser like the Maha Wizard One device that I bought
ages ago for managing my Eneloops (I have a strong dislike of primary
cells). However as it is I'm stuck with a 'dumb' charger that just charges
until voltage is ~4.2v. Then I to do a timed discharge in a flashlight. :-/
Very crude.
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-01-07 03:38:00 UTC
Permalink
Post by ~misfit~
Anyway now I have another six 1st tier 18650 cells to play with. Learning
their (remaining) capacity is the hardest part for me. If I wasn't so broke
I'd buy a charger / analyser like the Maha Wizard One device that I bought
ages ago for managing my Eneloops (I have a strong dislike of primary
cells). However as it is I'm stuck with a 'dumb' charger that just charges
until voltage is ~4.2v. Then I to do a timed discharge in a flashlight. :-/
Very crude.
But effective.

I built a setup with programmable power supply and programmable load.
Makes nice graphs,but not clear that they have any value beyond curiosity.

I do have a suggestion.
use a variable power supply and a diode.
set the power supply so the output of the diode is at the low
limit discharge voltage. Set the current limit sufficiently high
to run the discharge load.
Now, when the phone rings and you forget, you won't overdischarge
your 18650s.
If you use a transistor B-E junction as that diode, you can
use the collector to drive a circuit that stops a clock when
the
low-end voltage is reached, or starts a clock you can use to subtract
from the total time to get the discharge time and you don't have to sit
and watch it.
~misfit~
2015-01-09 00:36:27 UTC
Permalink
Post by mike
Post by ~misfit~
Anyway now I have another six 1st tier 18650 cells to play with.
Learning their (remaining) capacity is the hardest part for me. If I
wasn't so broke I'd buy a charger / analyser like the Maha Wizard
One device that I bought ages ago for managing my Eneloops (I have a
strong dislike of primary cells). However as it is I'm stuck with a
'dumb' charger that just charges until voltage is ~4.2v. Then I to
do a timed discharge in a flashlight. :-/ Very crude.
But effective.
I built a setup with programmable power supply and programmable load.
Makes nice graphs,but not clear that they have any value beyond curiosity.
I like to know cell capacity as I have some applications wher they run in
parallel.
Post by mike
I do have a suggestion.
use a variable power supply and a diode.
set the power supply so the output of the diode is at the low
limit discharge voltage. Set the current limit sufficiently high
to run the discharge load.
Now, when the phone rings and you forget, you won't overdischarge
your 18650s.
If you use a transistor B-E junction as that diode, you can
use the collector to drive a circuit that stops a clock when
the
low-end voltage is reached, or starts a clock you can use to subtract
from the total time to get the discharge time and you don't have to
sit and watch it.
That first thing sounds good but I'm a little challenged when it come to
electronics. I know basics (enough to make audio crossovers etc.) and have
some components around the place.

Would you be so kind as to explain in a bit more detail? I'd be obliged.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-01-09 01:34:29 UTC
Permalink
Post by ~misfit~
Post by mike
Post by ~misfit~
Anyway now I have another six 1st tier 18650 cells to play with.
Learning their (remaining) capacity is the hardest part for me. If I
wasn't so broke I'd buy a charger / analyser like the Maha Wizard
One device that I bought ages ago for managing my Eneloops (I have a
strong dislike of primary cells). However as it is I'm stuck with a
'dumb' charger that just charges until voltage is ~4.2v. Then I to
do a timed discharge in a flashlight. :-/ Very crude.
But effective.
I built a setup with programmable power supply and programmable load.
Makes nice graphs,but not clear that they have any value beyond curiosity.
I like to know cell capacity as I have some applications wher they run in
parallel.
Try to use cells from the same pack and don't sweat it. They're already
paralleled in pairs in many packs. Unless a cell is shorted, it's not a
huge risk.

The time you want 'em matched is when you run them in series. If they're
not matched, the pack gets imbalanced and can overdischarge or overcharge
one.

I avoided all that hassle with my bike light. I put 4 18650 cells
in a plastic pipe and brought out the wires of all cells to a DB-9
serial connector. The light runs them in series.
The charger charges them in parallel. I did put 4 1.5 ohm resistors
in the charger plug so that if something shorted, it would blow the
resistor instead of the battery.
Post by ~misfit~
Post by mike
I do have a suggestion.
use a variable power supply and a diode.
set the power supply so the output of the diode is at the low
limit discharge voltage. Set the current limit sufficiently high
to run the discharge load.
If you don't have an adjustable power supply, it doesn't matter.
The purpose of the diode and supply is to prevent overdischarge.
You set the power supply so that the diode is off and the supply
is completely disconnected during the whole test.
If you're gonna sit and watch a voltmeter during the whole test,
you don't need it.

When you do need it is when you aren't watching. When the phone rings.
Or the wife gets freaky.
Or when the kid falls and hurts himself.
If you get distracted, the diode eventually turns on, supplies
all the load current and saves your
cell from overdischarging below whatever voltage you set minus the diode
drop.
Post by ~misfit~
That first thing sounds good but I'm a little challenged when it come to
electronics. I know basics (enough to make audio crossovers etc.) and have
some components around the place.
Would you be so kind as to explain in a bit more detail? I'd be obliged.
Cheers,
Bottom line is that it takes a lot of equipment to get any meaningful data.
I had fun doing it, but it's not clear that the data made any real
difference in the
applications.
~misfit~
2015-01-10 01:59:25 UTC
Permalink
Post by mike
Post by ~misfit~
Post by mike
Post by ~misfit~
Anyway now I have another six 1st tier 18650 cells to play with.
Learning their (remaining) capacity is the hardest part for me. If
I wasn't so broke I'd buy a charger / analyser like the Maha Wizard
One device that I bought ages ago for managing my Eneloops (I have
a strong dislike of primary cells). However as it is I'm stuck
with a 'dumb' charger that just charges until voltage is ~4.2v.
Then I to do a timed discharge in a flashlight. :-/ Very crude.
But effective.
I built a setup with programmable power supply and programmable
load. Makes nice graphs,but not clear that they have any value
beyond curiosity.
I like to know cell capacity as I have some applications wher they
run in parallel.
Try to use cells from the same pack and don't sweat it. They're
already paralleled in pairs in many packs. Unless a cell is shorted,
it's not a huge risk.
When I charged the six cells from the Viao pack some took twica as long to
charge as others.
Post by mike
The time you want 'em matched is when you run them in series. If
they're not matched, the pack gets imbalanced and can overdischarge
or overcharge one.
Been there done that with NiMH. Same deal.
Post by mike
I avoided all that hassle with my bike light. I put 4 18650 cells
in a plastic pipe and brought out the wires of all cells to a DB-9
serial connector. The light runs them in series.
The charger charges them in parallel. I did put 4 1.5 ohm resistors
in the charger plug so that if something shorted, it would blow the
resistor instead of the battery.
Good solution. :)
Post by mike
Post by ~misfit~
Post by mike
I do have a suggestion.
use a variable power supply and a diode.
set the power supply so the output of the diode is at the low
limit discharge voltage. Set the current limit sufficiently high
to run the discharge load.
If you don't have an adjustable power supply, it doesn't matter.
Well, I have a few ~US$5 adjustable current / voltage LED drive / battery
charge modules. I thought I might be able to re-purpose one of those?
Post by mike
The purpose of the diode and supply is to prevent overdischarge.
Yep, that's what I want.
Post by mike
You set the power supply so that the diode is off and the supply
is completely disconnected during the whole test.
Can I do that with one of the modules mentioned above? I'm talking about
these things;
http://www.dx.com/p/buck-constant-voltage-constant-current-module-blue-dc-dc-5a-239099#.VLCG0dKUcTs
I use them mainly for driving LEDs off old laptop adapters.
Post by mike
If you're gonna sit and watch a voltmeter during the whole test,
you don't need it.
Yeah - however....
Post by mike
When you do need it is when you aren't watching. When the phone
rings. Or the wife gets freaky.
Or when the kid falls and hurts himself.
If you get distracted, the diode eventually turns on, supplies
all the load current and saves your
cell from overdischarging below whatever voltage you set minus the
diode drop.
.... and that's why I'd like to set something up to protect the cells during
discharge.
Post by mike
Post by ~misfit~
That first thing sounds good but I'm a little challenged when it
come to electronics. I know basics (enough to make audio crossovers
etc.) and have some components around the place.
Would you be so kind as to explain in a bit more detail? I'd be
obliged. Cheers,
Bottom line is that it takes a lot of equipment to get any meaningful
data. I had fun doing it, but it's not clear that the data made any
real difference in the
applications.
Understood. Still, if I can do something toward protecting the cells with
what I already have then it'd be a big help.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-01-10 04:22:44 UTC
Permalink
Post by ~misfit~
Post by mike
Post by ~misfit~
Post by mike
Post by ~misfit~
Anyway now I have another six 1st tier 18650 cells to play with.
Learning their (remaining) capacity is the hardest part for me. If
I wasn't so broke I'd buy a charger / analyser like the Maha Wizard
One device that I bought ages ago for managing my Eneloops (I have
a strong dislike of primary cells). However as it is I'm stuck
with a 'dumb' charger that just charges until voltage is ~4.2v.
Then I to do a timed discharge in a flashlight. :-/ Very crude.
But effective.
I built a setup with programmable power supply and programmable
load. Makes nice graphs,but not clear that they have any value
beyond curiosity.
I like to know cell capacity as I have some applications wher they
run in parallel.
Try to use cells from the same pack and don't sweat it. They're
already paralleled in pairs in many packs. Unless a cell is shorted,
it's not a huge risk.
When I charged the six cells from the Viao pack some took twica as long to
charge as others.
If the pack is decent, that shouldn't happen. If it's been sittin' in a
drawer for a decade, all bets are off. Also, there are some laptops
that put the pack right up against the heatsink and can cause serious
degradation
of those cells closest.

I'd charge 'em up to the design max, and measure it again.

Another cause I've seen is that you have two in parallel, but one of
them has popped it's internal breaker and you really only have one charging.
Post by ~misfit~
Post by mike
The time you want 'em matched is when you run them in series. If
they're not matched, the pack gets imbalanced and can overdischarge
or overcharge one.
Been there done that with NiMH. Same deal.
Post by mike
I avoided all that hassle with my bike light. I put 4 18650 cells
in a plastic pipe and brought out the wires of all cells to a DB-9
serial connector. The light runs them in series.
The charger charges them in parallel. I did put 4 1.5 ohm resistors
in the charger plug so that if something shorted, it would blow the
resistor instead of the battery.
Good solution. :)
Post by mike
Post by ~misfit~
Post by mike
I do have a suggestion.
use a variable power supply and a diode.
set the power supply so the output of the diode is at the low
limit discharge voltage. Set the current limit sufficiently high
to run the discharge load.
If you don't have an adjustable power supply, it doesn't matter.
Well, I have a few ~US$5 adjustable current / voltage LED drive / battery
charge modules. I thought I might be able to re-purpose one of those?
Post by mike
The purpose of the diode and supply is to prevent overdischarge.
Yep, that's what I want.
Post by mike
You set the power supply so that the diode is off and the supply
is completely disconnected during the whole test.
Can I do that with one of the modules mentioned above? I'm talking about
these things;
http://www.dx.com/p/buck-constant-voltage-constant-current-module-blue-dc-dc-5a-239099#.VLCG0dKUcTs
I use them mainly for driving LEDs off old laptop adapters.
Post by mike
If you're gonna sit and watch a voltmeter during the whole test,
you don't need it.
Yeah - however....
Post by mike
When you do need it is when you aren't watching. When the phone
rings. Or the wife gets freaky.
Or when the kid falls and hurts himself.
If you get distracted, the diode eventually turns on, supplies
all the load current and saves your
cell from overdischarging below whatever voltage you set minus the
diode drop.
.... and that's why I'd like to set something up to protect the cells during
discharge.
Post by mike
Post by ~misfit~
That first thing sounds good but I'm a little challenged when it
come to electronics. I know basics (enough to make audio crossovers
etc.) and have some components around the place.
Would you be so kind as to explain in a bit more detail? I'd be
obliged. Cheers,
Bottom line is that it takes a lot of equipment to get any meaningful
data. I had fun doing it, but it's not clear that the data made any
real difference in the
applications.
Understood. Still, if I can do something toward protecting the cells with
what I already have then it'd be a big help.
Cheers,
BIG FAT DISCLAIMER.
Don't do this unless you know exactly what you're doing and have the
test equipment to verify your assumptions. YOU CAN SET YOURSELF ON FIRE
if the power supply fails or is set up wrong.

Based on the specs, that power supply will work fine.

But, you're putting two switchers in series. If either of them goes wonky,
you're in trouble.
Wonkiness often happens right at the threshold where current limiting
starts.
Also happens if your current limit on the second supply causes the
first one to current limit.

The input to the second supply is a negative resistance.
If the voltage goes down, the current goes up. If the first supply
goes into current limit, especially foldback, things can go bad quickly.

In a perfect world, this all works. But I've seen some seriously
wonky power supply designs from people who's job was power supply design
working for a custom power supply design house.

I've seen the whole thing come crashing down when you disconnect the input
or short the input or have noise on the input...you name it.
Make sure that the open circuit voltage is what you set.
Some supplies require a minimum load to regulate.
If not, you need a load resistor so you don't overcharge your cells.
You also need to verify the voltage regulates to the proper voltage
when the supply heats up.
And if it overloads and shorts the switcher, you can smoke your cells
in a hurry.

Messing with lithium chargers is NOT SAFE.

Design worked fine in the intended application, but may not in yours.

I'm assuming that you know better than to try to solder directly onto the
cells. People who claim it's safe just haven't hurt themselves YET.

So, here's one option.
Set your power supply to the max charge voltage, (maybe 4.2V depending
on the lithium technology). Set your power supply current limit to the
maximum charge current you want.
Put two diodes in series
supply -->|--->|-- to battery.
Pick a combination of regular silicon junction and Schottky diodes so
that the output
of the diodes is the low voltage limit you choose.

Use a discharge test current that is LESS than the current limit set above.

Run your discharge test.
When done, disconnect the bottom end of the load resistor
and hook it to the power supply
across the diodes and it will charge back up.
You can even remove the load and short the diodes and charge at the max
you set.
But you really must evaluate whether this all works with your
batteries and your power supplies.

Are we having fun yet?
~misfit~
2015-01-12 23:18:59 UTC
Permalink
Once upon a time on usenet mike wrote:
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I have a bit
of time to play with.

I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that used 3x
small NiCad cells in shrink-wrap, the total size of which was similar to a
single 18650 but had little 'power' and didn't hold much charge. I bought a
new solder-tab Panasonic cell for that job as I didn't know how good my
slavaged cells were and it's hard to open up the modded tool....

However it doesn't seem to like running a motor from the 'out' pads - maybe
it's something to do with the brushes and how it works? I tried three of
these PCBs for the job and they all didn't work right. In the end I had to
wire a small momentary contact switch between the B+ and OUT+ pads to start
the motor. Once it's started it runs fine from the 'OUT' contacts. <shrug>

As you can probably tell I only have basic knowledge (but try hard to not
take risks - I'm very aware of the inherant dangers with Li-Ion tech and the
potential for 'thermal runaway').

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-01-13 01:20:51 UTC
Permalink
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I have a bit
of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that used 3x
small NiCad cells in shrink-wrap, the total size of which was similar to a
single 18650 but had little 'power' and didn't hold much charge. I bought a
new solder-tab Panasonic cell for that job as I didn't know how good my
slavaged cells were and it's hard to open up the modded tool....
However it doesn't seem to like running a motor from the 'out' pads - maybe
it's something to do with the brushes and how it works? I tried three of
these PCBs for the job and they all didn't work right. In the end I had to
wire a small momentary contact switch between the B+ and OUT+ pads to start
the motor. Once it's started it runs fine from the 'OUT' contacts. <shrug>
As you can probably tell I only have basic knowledge (but try hard to not
take risks - I'm very aware of the inherant dangers with Li-Ion tech and the
potential for 'thermal runaway').
Cheers,
The website was translated into ChEnglish and doesn't make a lot of sense,
but if I were to guess...

How are you switching the power?

If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.

If you're switching between the board and the motor, the turn-on transient
from the motor may exceed the 3A current limit. Maybe a big cap
across the out pads will help.
~misfit~
2015-01-15 00:35:21 UTC
Permalink
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't hold
much charge. I bought a new solder-tab Panasonic cell for that job
as I didn't know how good my slavaged cells were and it's hard to
open up the modded tool.... However it doesn't seem to like running a
motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably tell I
only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
Sadly with that site you *have* to guess a lot (unless there are informed
feedbacks).
Post by mike
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
I'm leaving the board and cell connected all the time. There doesn't seem to
be much in the way of constant drain. I modded it a year ago and have only
had to charge once in that time (it only gets light use).
Post by mike
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Hmmm, yes that might indeed be the problem - thanks. (I told you I have only
basic knowledge of electronics. However I have a cheap rework station that I
haven't unpacked yet which was delivered today so I better learn more!)

There's not much room in the enclosure for a 'big cap' - or anything else
for that matter. I dismantled it several times, re-soldering and hot-gluing
it all in place each time. The last time /was/ the last time - I taped it up
good (as the plastic threads for the screws weren't made for repeated use
and stripped quickly). I'll use it as-is now I think but it's good to know
how to combat this issue if I use one of the other modules.

Silly me; Where it says "charge current 1,000mA max" I took that to mean
what the board will put into the cell - not what the board would take. I was
charging my 3,000 A/h cell from a 500mA source and it was taking hours so
hooked it up to my 2,000mA plugpack. Well, I shouldn't have walked away for
the ~30 minutes that I did! Seems it was dumping extra current as heat,
there was hot glue running out of it! I thought that it would just 'take'
what it could handle from the source? <shrug>

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-01-15 01:37:33 UTC
Permalink
Post by ~misfit~
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't hold
much charge. I bought a new solder-tab Panasonic cell for that job
as I didn't know how good my slavaged cells were and it's hard to
open up the modded tool.... However it doesn't seem to like running a
motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably tell I
only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
Sadly with that site you *have* to guess a lot (unless there are informed
feedbacks).
Post by mike
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
I'm leaving the board and cell connected all the time. There doesn't seem to
be much in the way of constant drain. I modded it a year ago and have only
had to charge once in that time (it only gets light use).
Post by mike
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Hmmm, yes that might indeed be the problem - thanks. (I told you I have only
basic knowledge of electronics. However I have a cheap rework station that I
haven't unpacked yet which was delivered today so I better learn more!)
There's not much room in the enclosure for a 'big cap' - or anything else
for that matter. I dismantled it several times, re-soldering and hot-gluing
it all in place each time. The last time /was/ the last time - I taped it up
good (as the plastic threads for the screws weren't made for repeated use
and stripped quickly). I'll use it as-is now I think but it's good to know
how to combat this issue if I use one of the other modules.
Silly me; Where it says "charge current 1,000mA max" I took that to mean
what the board will put into the cell - not what the board would take. I was
charging my 3,000 A/h cell from a 500mA source and it was taking hours so
hooked it up to my 2,000mA plugpack. Well, I shouldn't have walked away for
the ~30 minutes that I did! Seems it was dumping extra current as heat,
there was hot glue running out of it! I thought that it would just 'take'
what it could handle from the source? <shrug>
Cheers,
Not at all clear what happened. Too many "it" and not enough specificity.

Lithium cells have a specification that defines the absolute maximum
charge termination voltage and the maximum safe charge current.
Also has a low-voltage cutout spec that tells you how far you can
discharge it without damage. Two problems with that. You may not
be able to find a spec for the cell you have in your hand.
And if it's an EBAY cell, the capacity may be seriously exaggerated
and you can't rely on that to estimate what you should do.

Your charge board manages all that.
Typically, the voltage ramps up as the cell charges at the
set max current, but stops ramping
at the terminal voltage, typically 4.2V depending on the exact lithium
technology.
The current max is set by the charge board and current starts to
decrease when
the voltage hits 4.2V. But it will charge until the current drops below
some point determined by the setup of the charge board.

The supply that powers the charge board MUST have enough voltage and current
capacity to run the charge board under all conditions.
Once the source current limits, the voltage to the charge board
drops and you have no idea what it will do. DO NOT ASSUME
that it does anything safe. It might, but DON'T trust that it does.

Here's an analogy. Sit in your easy chair and breathe thru a straw.
It limits your supply. You will lose control. On second thought,
sit in the bathtub. You may lose control of stuff you don't want
in your easy chair.
~misfit~
2015-01-15 23:33:30 UTC
Permalink
Post by mike
Post by ~misfit~
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't
hold much charge. I bought a new solder-tab Panasonic cell for
that job as I didn't know how good my slavaged cells were and it's hard
to
open up the modded tool.... However it doesn't seem to like
running a motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably
tell I only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
Sadly with that site you *have* to guess a lot (unless there are
informed feedbacks).
Post by mike
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
I'm leaving the board and cell connected all the time. There doesn't
seem to be much in the way of constant drain. I modded it a year ago
and have only had to charge once in that time (it only gets light
use).
Post by mike
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Hmmm, yes that might indeed be the problem - thanks. (I told you I
have only basic knowledge of electronics. However I have a cheap
rework station that I haven't unpacked yet which was delivered today
so I better learn more!) There's not much room in the enclosure for a
'big cap' - or anything
else for that matter. I dismantled it several times, re-soldering
and hot-gluing it all in place each time. The last time /was/ the
last time - I taped it up good (as the plastic threads for the
screws weren't made for repeated use and stripped quickly). I'll use
it as-is now I think but it's good to know how to combat this issue
if I use one of the other modules. Silly me; Where it says "charge
current 1,000mA max" I took that to
mean what the board will put into the cell - not what the board
would take. I was charging my 3,000 A/h cell from a 500mA source and
it was taking hours so hooked it up to my 2,000mA plugpack. Well, I
shouldn't have walked away for the ~30 minutes that I did! Seems it
was dumping extra current as heat, there was hot glue running out of
it! I thought that it would just 'take' what it could handle from
the source? <shrug> Cheers,
Not at all clear what happened. Too many "it" and not enough
specificity.
The "it" was referring to the charging module - you prompted me to check the
specs when you mentioned 3 amps.
Post by mike
Lithium cells have a specification that defines the absolute maximum
charge termination voltage and the maximum safe charge current.
Also has a low-voltage cutout spec that tells you how far you can
discharge it without damage. Two problems with that. You may not
be able to find a spec for the cell you have in your hand.
And if it's an EBAY cell, the capacity may be seriously exaggerated
and you can't rely on that to estimate what you should do.
This cell in question is from a legit seller local to me and is genuine
Panasonic.
Post by mike
Your charge board manages all that.
Typically, the voltage ramps up as the cell charges at the
set max current, but stops ramping
at the terminal voltage, typically 4.2V depending on the exact lithium
technology.
The current max is set by the charge board and current starts to
decrease when
the voltage hits 4.2V. But it will charge until the current drops
below some point determined by the setup of the charge board.
The supply that powers the charge board MUST have enough voltage and
current capacity to run the charge board under all conditions.
Once the source current limits, the voltage to the charge board
drops and you have no idea what it will do. DO NOT ASSUME
that it does anything safe. It might, but DON'T trust that it does.
This is my whole point. I was using a USB plug-pack PSU capable of 2A yet
the specs for the charge board says, as above, "charge current 1,000mA max".
That's when the thing got so hot it melted hot glue used to hold it in
place. Very inelegant. I've used the same PSU for low-drain devices and they
work fine, no excess heat anywhere so it wasn't the fault of the PSU.
Post by mike
Here's an analogy. Sit in your easy chair and breathe thru a straw.
It limits your supply. You will lose control. On second thought,
sit in the bathtub. You may lose control of stuff you don't want
in your easy chair.
I haven't been able to get into the bathtub for years (well, out of it
actually), since my back got so much worse. I even need a plastic stool in
the shower as I can't stand long enough to wash properly....

..... but I get your analogy. I said I'm a little challenged w/r/t
electronics. I didn't say I had an IQ less than 80. (Actually I used to me a
member of Mensa until I couldn't keep up the membership fees due to my
injury and loss of income / life savings.)

Thanks,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
~misfit~
2015-02-08 01:24:32 UTC
Permalink
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't hold
much charge. I bought a new solder-tab Panasonic cell for that job
as I didn't know how good my slavaged cells were and it's hard to
open up the modded tool.... However it doesn't seem to like running a
motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably tell I
only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
Cheers,
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Mike, quick question as I'm going to use another of these modules for
controlled discharge of Li-Ion cells using a motor as a load... What value
do you mean when you say 'a big cap'?

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
mike
2015-02-08 05:02:46 UTC
Permalink
Post by ~misfit~
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't hold
much charge. I bought a new solder-tab Panasonic cell for that job
as I didn't know how good my slavaged cells were and it's hard to
open up the modded tool.... However it doesn't seem to like running a
motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably tell I
only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
Cheers,
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Mike, quick question as I'm going to use another of these modules for
controlled discharge of Li-Ion cells using a motor as a load... What value
do you mean when you say 'a big cap'?
Cheers,
That's not a simple question. Motor load is complex. I have no idea
how the control board works.
The spec alludes to a 3 amp protection limit. If your motor starting
current exceeds 3A, you need to limit that somehow. A cap can help,
but it might take a very big one. I'd start with 100uF and experiment
from there. But a cap won't help for a constant load like locked rotor.

The board says 3A Protection. It's designed for a Maximum of 1A load.
You can be sure if it would survive much over that, the spec would be
higher. Monitor the FET temperature to see if it will melt off the board.
~misfit~
2015-02-10 22:09:28 UTC
Permalink
Post by mike
Post by ~misfit~
Post by mike
Post by ~misfit~
[total snip]
Post by mike
Are we having fun yet?
Thanks Mike. I've copied that to notepad and saved it for when I
have a bit of time to play with.
I have a couple of these too
http://www.dx.com/p/mp1405-5v-1a-lithium-battery-charging-board-blue-black-219454#.VLRT1MlZqUc
and used one for re-powering a mini rotary tool I got cheap that
used 3x small NiCad cells in shrink-wrap, the total size of which
was similar to a single 18650 but had little 'power' and didn't
hold much charge. I bought a new solder-tab Panasonic cell for
that job as I didn't know how good my slavaged cells were and it's hard
to
open up the modded tool.... However it doesn't seem to like
running a motor from the 'out' pads
- maybe it's something to do with the brushes and how it works? I
tried three of these PCBs for the job and they all didn't work
right. In the end I had to wire a small momentary contact switch
between the B+ and OUT+ pads to start the motor. Once it's started
it runs fine from the 'OUT' contacts. <shrug> As you can probably
tell I only have basic knowledge (but try hard
to not take risks - I'm very aware of the inherant dangers with
Li-Ion tech and the potential for 'thermal runaway').
Cheers,
The website was translated into ChEnglish and doesn't make a lot of
sense, but if I were to guess...
How are you switching the power?
If you're switching between the battery and the board,
that may be your problem. Sounds like the board has to be powered
up continuously. I'd measure the drain from the battery in the
off state. May be inconsequential...maybe not.
If you're switching between the board and the motor, the turn-on
transient from the motor may exceed the 3A current limit. Maybe a
big cap across the out pads will help.
Mike, quick question as I'm going to use another of these modules for
controlled discharge of Li-Ion cells using a motor as a load... What
value do you mean when you say 'a big cap'?
Cheers,
That's not a simple question. Motor load is complex. I have no idea
how the control board works.
The spec alludes to a 3 amp protection limit. If your motor starting
current exceeds 3A, you need to limit that somehow. A cap can help,
but it might take a very big one. I'd start with 100uF and experiment
from there. But a cap won't help for a constant load like locked rotor.
Thanks again. I really appreciate your input.
Post by mike
The board says 3A Protection. It's designed for a Maximum of 1A load.
You can be sure if it would survive much over that, the spec would be
higher. Monitor the FET temperature to see if it will melt off the board.
That's what I have an infrared thermometer for. ;) Well, that and to see if
my LEDs are running at acceptable temps.

I just dismantled a completely dead Dell nine cell battery pack. It came
with a laptop that I reburbished and sold on a couple of years ago. The
laptop wouldn't have a bar of it, said battery dead and needs replacement.
When dismantled I saw it was wired 3s/3p and three of the cells have 0.00v
charge and simply won't take a charge no matter what I try - they just get
hot.

However the other six (Samsung 18650-26 cells, 2,600 mA/h) seem great. They
were sitting at 3.2v when the pack was dismantled and they came up really
quickly in the charger. There's lots and lots of capacity left in all of
them - a really good score. If I had to guess I'd say that the pack had been
left in storage in a discharged state for some time and the pack contol
circuitry took it's power from just that group of three cells. When it
completely drained those it flagged the pack as dead.

Oh, I upgraded my charger to a Nitecore D2. Now I have a reasonable
collection of usable Li-Ion cells it seemed like a good investment.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
~misfit~
2015-01-02 22:47:21 UTC
Permalink
Post by mike
Post by a***@gmail.com
Mike, your last post was entertaining. As I said, though, I'm not
as passionate about it as to make it a hobby. I'm OK with the gross
guidelines of discharging to 5% once a month. I appreciate the
rationale behind those guidelines as much as I appreciate the
guidelines (which are not trustworthy without rationale). Most of
the time, I'm on AC, and the battery is only a UPS. If it goes,
I'll buy another (accepting the risk that it might have been on the
shelf since the turn of the century). I'll just complain to the
vendor if it's crappy, and if they play hardball, I'll complain in
public. I know it will not make it above noise level because so
many people do that anyway, but I am riding the technology
improvement trend that is driven by capitalism. Batteries today are
heads and shoulders better than a decades ago...except to
enthusiasts who intend to crack them open to extend their lives. More
power to you if you do, though.
Consensus seems to be that you should take the battery out if using it
on AC.
Assuming the design of the charger stops charging when it gets to max,
there's still the issue of heat.
I use mine little, so just leave the battery in on AC all the time.
If I used it a lot, I'd take out the battery.
The problem with that is there are some laptops that refuse to ramp the CPU
up to full speed if there is no battery fitted - even on AC. The rationale
being that it acts as an energy buffer, helping to supply high ampage bursts
that would stress the power brick....
Post by mike
I had more than one laptop where the battery sits two plastic
thicknesses away from the processor heat sink.
When you take the case off the battery, the ones closest to the
heat are discolored...and much higher resistance than the others.
Some designer oughtabeshot.
Or paid a bonus when sales of batteries increase. ;)
Post by mike
It's been said that lithium cells degrade almost as fast on the shelf
as if you use 'em sensibly. A spare battery is great if you need to
use it. If it's sittin on the shelf justincase, it may not be a good
investment. Still might be better sittin' in your fridge than
having it sittin' in that container
in Arizona till you buy it. ;-)
That's my philosophy. I'm happy with my T60s (the last IBM designed
ThinkPads) and can't see me needing to change for years to come <touch
wood>. However as they get older battery packs will get harder to come by /
more expensive so I grabbed a couple when I could and store them in ideal
conditions / state of charge.

Incidently my T60s software will tell me a batterys date of manufacture,
cell manufacturer (although I'm sure this can be fudged by after-market
suppliers), date first used and number of cycles since new. Also it tells me
the obvious things like design capacity and current capacity. (Battery
currently fitted (my 'lazarus' battery); Manufacturer Sanyo. Manufacture
Date 2008-12-04. First Used Date 2009-02. Design Capacity 56.16 W/h.
Remaining Capacity 23.2 W/h.)

All after-market battery packs are rebuilt OEM units using original
(reprogrammed) electronics - or at least they were last I heard.

Instead of stocking up on a couple batterys I was going to just re-pack them
with panasonic 18650s but you pointed out some of the issues with that. With
these batteries you can also include the fact that you need to supply the
packs electronics with ~8v from a bench power supply while you have the
cells disconnected. If you don't the battery CPU decides the pack has
dropped below threshold and 'bricks' it (while running on a reserve
capacitor). Then the laptop won't complete handshake with the battery pack
CPU on boot up so will report no battery present.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)
Loading...