Can I use a volt meter to tell when a nicad is charged?
Michael Palmer
Is checking to see how the charging of a nicad is going as easy as slapping
a volt meter on it (i.e. unplug the charger, check to see if voltage is up
to where it is when I leave it overnight)? If it matters, the subject
battery is a Niterider Cyclops Pro bottle format. I just don't want to
overcharge it after using it for short periods, since I understand that is
bad for them.
--
Michael Palmer (mpa...@mo.net)
Kirk R. Darling
Michael,
Early nicads had "discharge memory" problems which caused them not to
provide any more power than they were "accustomed" to provide. That is, a
battery with a nominal 3-hour charge would begin to provide only an hour of
power after a few times of one-hour use before recharge. Modern nicads,
however, don't have that problem. Also, modern nicad chargers taper or
halt the charge when the cells are fully charged.
--
KRD
"Leap boldly. You can't cross a chasm in two easy steps."
Michael Palmer <mpa...@mo.net> wrote in article
<01bc3f0c$79a5d5c0$bde78bcd@mpalmer>...
Todd O. Morken
Michael Palmer <mpa...@mo.net> wrote in article
<01bc3f0c$79a5d5c0$bde78bcd@mpalmer>...
> Is checking to see how the charging of a nicad is going as easy as
slapping
> a volt meter on it (i.e. unplug the charger, check to see if voltage is
up
> to where it is when I leave it overnight)? If it matters, the subject
> battery is a Niterider Cyclops Pro bottle format. I just don't want to
> overcharge it after using it for short periods, since I understand that
is
> bad for them.
Nope, won't work. Since a Nicad will put out consistent voltage for 90%
of its charge capacity, you really can't tell how charged it is even under
load with a simple voltage meter.
using a Nicad battery for only a short time then recharging is bad for it.
It should be discharged to the point where you notice a decrease in
performance of whatever your using it in, then fully recharged. In this
case you should use the battery till you notice your lights dimming and
then give it a full recharge (12-15 hours usually, but check your manual).
True this is more of a hassle than just using it a short time and putting
it back on the charger, if your only using it a short time maybe cut the
full charge time in say 1/4 or 1/2 and charge it that long to cut down on
the overcharge time. Also do not discharge a Nicad below this "decrease
in performance" that can ruin a Nicad in one shot. One of the bad things
about Nicads is once your lights start going dim, you have to shut them
down almost immediately or you seriously harm the battery life.
But they are lighter, and of course more expensive.
Todd Morken
sni...@btigate.com
Jeffrey L. Bell
Todd O. Morken <sni...@btigate.com> wrote:
>using a Nicad battery for only a short time then recharging is bad for it.
Is this the Nicad Memory Myth resurfacing again?
Now there is such a thing as nicad memory, but it happens when you
discharge to the exact same point every time for a lot of cycles.
Unless you use it for exactly the same amount every time, you aren't
going to suffer from nicad memory with modern nicads.
The real problem is overcharging.
The problem isn't that you only slightly discharged it, but that
it gets overcharged afterwards.
Many chargers are oblivious to the state of the battery.
The realy cheapo chargers work at a constant rate forever.
Slightly better chargers use a timer, but that still cooks
any battery that doesn't need a full charge. In the second case
the "Run it down, and then recharge" regimen will work, since the
timer is set to handle low batteries.
A better approach is to use a charger that's set at a rate that won't
harm the battery if left on indefinitely. The main downside of this
is that it takes a long time to charge.
The best way is to use a smart charger. Most of them monitor voltage
and/or current draw to know when to switch over to maintenence charge.
For an example of a Lead-Acid battery charger controller, look at
http://www.benchmarq.com/prod/bq2031.html . It can be set to monitor
the second derivative of the voltage and compensate for temerature
while current-pulse charging. It also does a quick battery diagnostic
test before charging. The chip itself costs about $6.
Todd was right about not running the battery pack too low,
but whether or not to run it down before recharging depends on the
charger you have.
-Jeff Bell
Jason E. Knight
Excerpts from netnews.rec.bicycles.tech: 2-Apr-97 Can I use a volt meter
to t.. by "Michael Palmer"@mo.net
> a volt meter on it (i.e. unplug the charger, check to see if voltage is up
> to where it is when I leave it overnight)? If it matters, the subject
> battery is a Niterider Cyclops Pro bottle format. I just don't want to
> overcharge it after using it for short periods, since I understand that is
> bad for them.
Yep, voltmeter oughta do it, providing you know what the voltage is
supposed to be for the fully charged battery. I have a niterider too,
but I forget what voltage it is... 6V? If you're really concerned about
charging you can always go to a hobby shop and get a trickle charger (my
dad uses about a zillion of these for charging his r/c airplane
batteries).
but one night of charging after short usage seems unlikely to damage
the battery.
jason knight
jk...@andrew.cmu.edu
ak...@cmu.edu
***********************************************************************
Linguistics.Cycling.DIYaudio.Linux/MkLinux.Schizophrenia.Schizophrenia.
Dan Grunberg
In article <01bc3f22$28d336e0$2a28b7cd@toddmork>,
>>slapping a volt meter on it (i.e. unplug the charger, check to see if
>>voltage is up to where it is when I leave it overnight)? If it
>>matters, the subject battery is a Niterider Cyclops Pro bottle
>>format. I just don't want to
>>overcharge it after using it for short periods, since I understand that
>>is bad for them.
>
>of its charge capacity, you really can't tell how charged it is even under
>load with a simple voltage meter.
>
Fully charged NiCads should read in excess of 1.3 Volts PER CELL. IOW,
either 1.3 Vots, 2.6 Volts, 3.9 Volts, etc, depending on the number of
cells in the battery. A NiCad battery is APPROACHING the end of its
useful discharge at 1.0 Volts per cell, is probably at the end of useful
discharge at 0.9 Volts per cell (+/-).
To avoid charge reversal on the least charged of the several cells in the
battery, discharge should be halted at 0.8 Volts per cell (+/-).
>using a Nicad battery for only a short time then recharging is bad for it.
>It should be discharged to the point where you notice a decrease in
>performance of whatever your using it in, then fully recharged.
According to the Gates Rechargeable Battery Handbook, charging at rates
smaller than C/10 (where C is a number of milliamperes equal to the rated
capacity of the battery in milliampere-hours) may be carried on
continuously, REGARDLESS OF THE PREVIOUS STATE OF CHARGE OF THE BATTERY.
In practical terms this means that a charger that takes a minimum of
15 hours or so to charge a dead battery, can be used to trickle charge a
fully charged battery. I'D CHECK WITH THE MANUFACTURER OF THE CHARGER ON
THIS ONE.
>In this case you should use the battery till you notice your lights
>dimming and
Then turn the light off IMMEDIATELY to prevent reverse charging one of the
cells in the battery with undesireable results.
>then give it a full recharge (12-15 hours usually, but check your manual).
> True this is more of a hassle than just using it a short time and putting
>it back on the charger, if your only using it a short time maybe cut the
>full charge time in say 1/4 or 1/2 and charge it that long to cut down on
>the overcharge time.
The correct amount of time isn't easy to estimate, because charge
remaining vs. discharge time after charging isn't a linear function.
>Also do not discharge a Nicad below this "decrease
>in performance" that can ruin a Nicad in one shot. One of the bad things
>about Nicads is once your lights start going dim, you have to shut them
>down almost immediately or you seriously harm the battery life.
I couldn't agree more. FWIW, well designed, expensive portable radios
are designed to shut themselves off as their battery approaches the end
of its useful charge, so that they cannot reverse charge one of the
battery's cells.
>But they are lighter, and of course more expensive.
--
Daniel Grunberg Kensington MD USA
Todd O. Morken
Jeffrey L. Bell <jlb...@presto.eecs.umich.edu> wrote in article
<5htsn2$o32$1...@news.eecs.umich.edu>...
> Todd O. Morken <sni...@btigate.com> wrote:
> >using a Nicad battery for only a short time then recharging is bad for
it.
>
> Is this the Nicad Memory Myth resurfacing again?
>
> Now there is such a thing as nicad memory, but it happens when you
> discharge to the exact same point every time for a lot of cycles.
> Unless you use it for exactly the same amount every time, you aren't
> going to suffer from nicad memory with modern nicads.
>
>
Yeah, I should have clarified my statement, my thought was if he was
commuting or riding almost the same route all the time it could hurt the
battery over time by discharging it to the same point and
recharging/overcharging frequently.
> Many chargers are oblivious to the state of the battery.
> The realy cheapo chargers work at a constant rate forever.
>
Of the lights I've used, TurboCat, Nightsun, Specialized, Vista, Cateye,
they all use a cheapo constant rate charger, in fact the only one that
might not is the new nightrider XL.
> A better approach is to use a charger that's set at a rate that won't
> harm the battery if left on indefinitely. The main downside of this
> is that it takes a long time to charge.
>
> The best way is to use a smart charger. Most of them monitor voltage
> and/or current draw to know when to switch over to maintenence charge.
> For an example of a Lead-Acid battery charger controller, look at
> http://www.benchmarq.com/prod/bq2031.html . It can be set to monitor
> the second derivative of the voltage and compensate for temerature
> while current-pulse charging. It also does a quick battery diagnostic
> test before charging. The chip itself costs about $6.
How about one for a Nicad? since it's much harder to tell when a Nicad is
really fully charged, I would guess it would probably either very complex
or very expensive.
Todd Morken
sni...@btigate.com
Jeffrey L. Bell
In article <01bc3f9b$9661bdc0$f7b1a7cc@toddmork>,...
>> >using a Nicad battery for only a short time then recharging is bad for
>> Is this the Nicad Memory Myth resurfacing again?
>Yeah, I should have clarified my statement, my thought was if he was
>commuting or riding almost the same route all the time it could hurt the
>battery over time by discharging it to the same point and
>recharging/overcharging frequently.
True enough.
Sorry if I over reacted, but the "memory effect" story has
taken on a life of it's own. Electronics sales 'droids will
go on at length about it; it gives them something to talk about.
For a further discussion of the memory effect (in camcorder batteries)
see http://igubu.saix.net/speed/batter.htm
>Of the lights I've used, TurboCat, Nightsun, Specialized, Vista, Cateye,
>they all use a cheapo constant rate charger, in fact the only one that
>might not is the new nightrider XL.
So not even a two stage process?
I suppose that a semi-cheap way to prevent overcharging would be to
build a really slow seperate trickle charger and put the battery on
that after a few hours on the fast charger.
>> The best way is to use a smart charger.
...
>How about one for a Nicad? since it's much harder to tell when a Nicad is
>really fully charged, I would guess it would probably either very complex
>or very expensive.
Take a look at http://www.benchmarq.com/FCIC.html for fast charge ICs.
They've got half a dozen for Nicads.
I'm told that unitrode and philips electronics also make charge
controllers but I haven't found them on the web.
For a charge controller product with nice graphs of the charging process
see http://twins.co.uk/business/lawson/icams.html
In regards to the original question
It turns out that they do make a battery "gas gauge" chip. You put it in
the box with the battery so that it can do temperature compensation.
It learns about the charge cpacity over time, and it has a two wire
serial line for downloading information. It needs an EEprom to save
the data though. See http://www.benchmarq.com/prod/bq2091.html
-Jeff Bell
Andy Dingley
The moving finger of "Michael Palmer" <mpa...@mo.net> having written:
>Is checking to see how the charging of a nicad is going as easy as slapping
>a volt meter on it
No.
Buy an intelligent NiCd charger. For the last couple of years there
have been a number of cheap chipsets available to do this, so good
chargers are now very reasonably priced. A good charger (especially
with cell temperature monitoring) will charge faster, give you greater
capacity per charged cell, and give you a longer lifetime in terms of
number of cycles.
--
Smert' Spamionem
Michael Palmer
Thanks for the input, but I meant my question to be much simpler, viz., can
I tell if a battery is charged by unplugging it from the charger and
checking it with a volt meter? I'm wanting to avoid over charging and am
willing to check on charging progress by hand (I haven't the will or the
skill to build something automatic).
As for circumstances of use, I am riding a 1 hour 20 minute route for which
I need the light less each night as it stays lighter longer. I'm trying to
figure out how best to charge the battery when I end up only using the
light for 20 or 30 minutes per night.
(Todd may have answered that question, but so far I can only see Jeffrey's
replies.)
--
Michael Palmer (mpa...@mo.net)
Jeffrey L. Bell <jlb...@presto.eecs.umich.edu> wrote in article
<5hugn3$ah6$1...@news.eecs.umich.edu>...
Todd O. Morken
Michael Palmer <mpa...@mo.net> wrote in article
<01bc3fe4$3d165b60$a9e78bcd@mpalmer>...
> Thanks for the input, but I meant my question to be much simpler, viz.,
can
> I tell if a battery is charged by unplugging it from the charger and
> checking it with a volt meter? I'm wanting to avoid over charging and am
> willing to check on charging progress by hand (I haven't the will or the
> skill to build something automatic).
>
You cannot tell how charged your Nicad is for your system with a simple
voltmeter, maybe if you could measure each cell individually and had
detailed info on the cells used. But in general on a 13.2v pack like
nigthsun or nightrider uses it is virtually impossible to tell making a
measurement during charging.
This is due to the fact that Nicads will put you a consistant voltage, even
under load, for 90% or so of thier charge, your battery could have only say
a 20% charge and give off as high of volt reading as it would *or extremely
close* at 100% charge.
While you may be able to buy or make a smart charger that would handle this
for you I would guess it would be expensive and quite a undertaking,
although I have not looked into it personally.
Karl Edmark
snip
> > battery is a Niterider Cyclops Pro bottle format. I just don't want to
> > overcharge it after using it for short periods, since I understand that
is
> > bad for them.
> jason knight
> jk...@andrew.cmu.edu
> ak...@cmu.edu
> ***********************************************************************
> Linguistics.Cycling.DIYaudio.Linux/MkLinux.Schizophrenia.Schizophrenia.
>
>
Using your nicad for short periods and then recharging it is exactly what
you should NOT be doing. Do not charge it after every ride. Use it until
the light starts to go dim or turn yellow, then leave the light on until
the battery is completely dead. Now, plug the battery into the charger
when you go to bed. I charge mine next to my alarm clock so I won't forget
it. When you wake up, wrap your hands around the battery. If it is
slightly warm, then it is now completely charged. If it is not warm, it is
not fully charged. Leave it plugged in to the charger until slightly warm,
BUT NO LONGER than this, or you will over-charge it! Mine takes about 12
hours.
This procedure is called deep cycling your battery. Email me if you would
like the technical reasons for the importance of doing this.
Yes, I know, you may from time-to-time end up with the light dying out
during a ride. However, this is a small price to pay for saving the life
of your battery. If you keep charging it after short rides, the battery
will eventually have no life for long rides. I routinely get 6 hours of
battery life on my NiteRider cyclops using a 10 watt lamp and bottle
battery.
Todd O. Morken
Karl Edmark <Ka...@msn.com> wrote in article
<01bc4046$ab6bfa40$34e32399@default>...
>
> Using your nicad for short periods and then recharging it is exactly what
> you should NOT be doing. Do not charge it after every ride. Use it
until
> the light starts to go dim or turn yellow, then leave the light on until
> the battery is completely dead.
Everything I've ever read or been told says you should NEVER drain a Nicad
battery completely. If they use a sealed lead acid battery in the cyclops
that would be one thing, but NEVER deep cycle a Nicad battery, your just
asking for problems.
>Now, plug the battery into the charger
> when you go to bed. I charge mine next to my alarm clock so I won't
forget
> it. When you wake up, wrap your hands around the battery. If it is
> slightly warm, then it is now completely charged. If it is not warm, it
is
> not fully charged. Leave it plugged in to the charger until slightly
warm,
> BUT NO LONGER than this, or you will over-charge it! Mine takes about 12
> hours.
If your battery is warm at all your cooking it, granted if it's not long
its min. but anytime you can feel heat it's harmful to the battery.
Dan Grunberg
In article <tomkE82...@netcom.com>,
Thomas H. Kunich <to...@netcom.com> wrote:
>In article <01bc4046$ab6bfa40$34e32399@default>,
>Karl Edmark <Ka...@msn.com> wrote:
>
>>This procedure is called deep cycling your battery. Email me if you would
>>like the technical reasons for the importance of doing this.
>
>and some insist that the only way to handle nicads is to keep them on
>constant charge.
>
>Deep cycling a nicad was once necessary for mechanical reasons. It
>isn't anymore.
>
Whether you should "deep cycle" or whether you should "constant charge"
depends on the design of the CHARGER, not on the manufacturer of the
NiCad, and not on "mechanical reasons."
When, as in the case of a rechargeable bike light system, the charger,
the battery and the load (i.e. the light) are purchased together from
the same manufacturer, it would be best to follow the manufacturer's
instructions.
When you home brew a system, keep in mind that:
1. Constant charging at rates of C/10 (where C is a current in
milliamperes equal to the battery's capacity in milliampere hours) or
less is OK, regardless of how charged (or discharged) the battery is.
2. Charging rates greater than C/10 will save time, but the battery
must be conditioned (discharged to less than 0.8 volts per cell) prior to
charging, the charge rate must be controlled to prevent overheating the
battery, and the charger must disconnected (or, assuming that battery
temperature is well controlled, the charge rate must be lowered to C/10)
when the battery is fully charged.
Stephen Tu
>you should NOT be doing. Do not charge it after every ride. Use it until
>the light starts to go dim or turn yellow, then leave the light on until
This is a bad idea. Leaving the light on until the battery is
completely dead risks reversing the polarity of one of the cells of the
battery, making the battery unrechargeable.
Charge after the light starts to turn yellow.
>of your battery. If you keep charging it after short rides, the battery
>will eventually have no life for long rides. I routinely get 6 hours of
This is not true if you don't overcharge. Other posters have
explained the NiCd memory myth.
Stephen Tu
ste...@concentric.net
Thomas H. Kunich
Michael Palmer
Jason E. Knight wrote:
> <snip>
> Yep, voltmeter oughta do it, providing you know what the voltage is
> supposed to be for the fully charged battery. I have a niterider too,
> but I forget what voltage it is... 6V?
>
It reads 6.75-6.77v or so after charging overnight. It reads about 5.4v
when the light "goes yellow." But I only started checking it recently,
and it's over a year old. I don't know what it read when new.
--
Michael Palmer (jmpa...@worldnet.att.net)
Karl Edmark
Todd O. Morken <sni...@btigate.com> wrote in article
<01bc4067$25865120$a1b1a7cc@toddmork>...
>
snip
> Everything I've ever read or been told says you should NEVER drain a
Nicad
> battery completely. If they use a sealed lead acid battery in the
cyclops
> that would be one thing, but NEVER deep cycle a Nicad battery, your just
> asking for problems.
snip
Sorry Todd, but you obviously do not know what the fuck you are talking
about : )
Jeffrey L. Bell
In a previous article, Michael Palmer <jmpa...@worldnet.att.net> wrote:
>It reads 6.75-6.77v or so after charging overnight. It reads about 5.4v
>when the light "goes yellow." But I only started checking it recently,
>and it's over a year old. I don't know what it read when new.
Just for curiosity's sake, measure the voltage after you've let it sit
for a couple hours after you stop charging it. I expect it to drop back
a bit.
Try running it down to slightly yellow, then let it sit for a while.
I bet it recovers somewhat.
That's the problem with going solely by volt meter readings to tell
how charged it is. The voltage depends on the recent history of the
battery and on the temperature in addition to the state of charge.
The flat discharge curves of nicads make it all the harder.
I hear of some designs that switch off charging due to temperature.
When the nicad battery hits 45C the charge switches into trickle
charge. That might be a much more practical system for manually
avoiding overcharge.
-Jeff Bell
PS: (Somebody correct me if I'm wrong for nicads. My experience is with
Cyclon Gel Cells, but I read up on Nicads a few years back. If you do
the measurements, I would love to hear the real numbers.)
Henrik Münster
Karl Edmark <Ka...@msn.com> skrev:
> Yes, I know, you may from time-to-time end up with the light dying out
> during a ride. However, this is a small price to pay for saving the life
> of your battery.
Keeping your battery well charged asuring good light and risking
shortening the life of the battery, is a small price to pay for saving
your own life.
Henrik "It can be dangerous if a light dies out in the middle of a ride"
Münster
--
__o Anna-Marie & Henrik Münster Den Elektroniske Lægebrevkasse
-\<, http://home4.inet.tele.dk/munster/
(_)/(_) mailto:mun...@post4.tele.dk
Dan Grunberg
In article <199704041...@ppp23.esb.tele.dk>,
=?ISO-8859-1?Q?Henrik_M=FCnster?= <mun...@post4.tele.dk> wrote:
>Karl Edmark <Ka...@msn.com> skrev:
>
>> Yes, I know, you may from time-to-time end up with the light dying out
>> during a ride. However, this is a small price to pay for saving the life
>> of your battery.
>
> Keeping your battery well charged asuring good light and risking
>shortening the life of the battery, is a small price to pay for saving
>your own life.
>
>Henrik "It can be dangerous if a light dies out in the middle of a ride"
>Münster
Yes, but the ultimate result of overcharging is a shortened useful
discharge time.
If the charger is designed for constant charging, by all means keep the
battery charging whenever you're not using it. OTOH, if the charger
isn't designed for constant charging, don't constantly charge your
battery. FOLLOW THE INSTRUCTIONS IN YOUR OWNERS MANUAL.
Anders Olsson
A Nicad battery should never be deeply discharged for the reason that the
cells are never exactly the same, and discharging the battery deep enough
will almost certainly lead to polarity reversal of one cell. This is a fatal
condition. (A single cell is not harmed by complete discharging). The basic
rule is to never discharge a Nicad battery to less than 1 volt/cell.
Anders Olsson.
David Walls
Michael,
To test a battery with a volt meter the battery should be measured
on load i.e. connected to a bulb or resistor . What you are checking for
is the if the battery is capable of delivering the voltage with enough
current . Voltmeters themselves have a high impedance (resistance) hence
just to see enough volts with the battery disconnected doesn't mean it
won't die immediately it sees your bulb.
Dave Walls
Cambrian Timing
http://ourworld.compuserve.com/homepages/cambriantiming/
Karl Edmark
O.K., gentleman. Can we all agree that nicad battery management is a pain?
Would you believe me if I told you I replaced the nicad starting batteries
in my jet with lead acids for this very reason? If you think a weak
NiteRider battery is dangerous on your bike, email me for what can happen
with nicads on a jet, if interested.
All I simply tried to do is offer Mr. Knight a simple recipe for maximizing
the useful operational life of his NiteRider light. If any of you experts
have a better nicad charging recipe, or improvements to mine, stop arguing
and POST'EM!!!!!!!!!!!!!!!!
Henrik Münster
Dan Grunberg <dgru...@nyx10.cs.du.edu> skrev:
> Yes, but the ultimate result of overcharging is a shortened useful
> discharge time.
Having the light die out in the middle of a ride was mentioned
as a small inconvenience. I just wanted to point out, that it can be
dangerous to ride without lights. Personally I'd rather buy a new
battery more often than risk being hit by a car.
Kind regards
Henrik Münster
Andy Dingley
The moving finger of eka...@eraj.ericsson.se (Anders Olsson) having
written:
>A Nicad battery
Is anyone using NiCd batteries on bikes ? Mine are just loose cells,
and I charge them as cells. The only battery I've ever used on a bike
was a gel lead acid.
BTW Todd - You're an idiot.
--
A drowning surrealist will not appreciate a concrete lifebelt.
Jeffrey L. Bell
In article <01bc3fe4$3d165b60$a9e78bcd@mpalmer>,
Michael Palmer <mpa...@mo.net> wrote:
>Thanks for the input, but I meant my question to be much simpler, viz., can
>I tell if a battery is charged by unplugging it from the charger and
>checking it with a volt meter? I'm wanting to avoid over charging and am
>willing to check on charging progress by hand (I haven't the will or the
>skill to build something automatic).
>
It's not the voltage that smart chargers use, but the slope. When the
slope starts to drop it's charged. There are temperature and other
effects that make it tricky to use a voltage level, otherwise you
could just hook it to a voltage source and current would drop off as
it approached full charge.
I don't know whose page this, but look at the graph at:
http://twins.co.uk/business/lawson/icams.html#6
That blip that signifies end of charge is going to be hard to catch.
Another approach might be to switch build a slowcharger attachment to
your current system and switch to it once you get partway charged.
Measure the current that the regular charger puts into the battery.
When you use the battery, try to remember the amount of time you used it.
At charging time put the same number of amp-hours back in,
then switch over to a slower charger, and trickle charge until the next ride.
I think I read somewhere that most batteries require at least 1.4 Ah
back in for each Ah taken out, so you could probably shoot for 1.2 times.
(Does anybody out there have more accurate numbers????)
You might be able to make a slow charger out of your fast charger. If
your supply is not very well regulated you may be able to get away
with a resister in series to drop the current. See what happens to
the current if you put a resistor in the loop.
If your charger happens to be well regulated (ha!), you might have to
put the resistor in parallel to shunt the extra current, but then you
have to really careful not to let it discharge the battery if the
power goes out, perhaps a diode would be good.
For trickle charging you want something less than C/10 rate to avoid
overcharging, perhaps C/15. C/15 is the capacity of the battery
divided by 15 hours, so a 3 Ah battery would be given 0.2 Amps
for a trickle charge.
I am not an expert in batteries and would love to hear a better way to
do this.
-Jeff
Jeffrey L. Bell
Todd O. Morken <sni...@btigate.com> wrote:
>...
>Everything I've ever read or been told says you should NEVER drain a Nicad
>battery completely. If they use a sealed lead acid battery in the cyclops
>asking for problems.
>...
You might want to rephrase that.
A sealed lead acid is another matter .... it has even MORE problems.
You should never run a lead acid cell or battery to zero.
When the battery is run too low it refuses to conduct much
current when you go to recharge. I guess the electrolyte
starts to look like water near the end. You have to charge
really slow for a while until you get enough ions into the solution.
When I was first starting out I killed a number of Cyclon cells in
mistaken belief that they would experience memory effect. Doh!
Thank goodness I had gotten the batteries for free.
It's not too bad to run a single NiCad CELL to zero. They come back
pretty well, although I've read that for some designs the seal can go
bad if they are left completely dead for too long.
NiCad BATTERIES are a different matter, being composed of a bunch of
cells. (Remember the original meaning of the word battery.)
Try to avoid running batteries down.
If all the cells in a battery were really identical capacity it would
work OK, but there's always going to be a weaker cell.
Once it gets to zero, it gets reverse charged until you shut the
current off, so the next time it's an even weaker cell.
It's a sort of battery Darwinism I guess.
-Jeff Bell
Dan Grunberg
In article <5iehj2$rp1$1...@news.eecs.umich.edu>,
Jeffrey L. Bell <jlb...@presto.eecs.umich.edu> wrote:
>In article <01bc3fe4$3d165b60$a9e78bcd@mpalmer>,
>Michael Palmer <mpa...@mo.net> wrote:
>>Thanks for the input, but I meant my question to be much simpler, viz., can
>>I tell if a battery is charged by unplugging it from the charger and
>>checking it with a volt meter? I'm wanting to avoid over charging and am
>>willing to check on charging progress by hand (I haven't the will or the
>>skill to build something automatic).
>>
>It's not the voltage that smart chargers use, but the slope. When the
>slope starts to drop it's charged. There are temperature and other
>effects that make it tricky to use a voltage level, otherwise you
>could just hook it to a voltage source and current would drop off as
>it approached full charge.
That's true, but as you (Jeffrey) imply below, you can charge NiCads at
rates of C/10 or less for an indefinitely long period of time, without
hurting the battery.
>I don't know whose page this, but look at the graph at:
>http://twins.co.uk/business/lawson/icams.html#6
>That blip that signifies end of charge is going to be hard to catch.
That's true. A good "catcher" is a charge-controller chip that's made
for the purpose (by Maxim and others). Of course you have to sense the
temperature of the battery and input a voltage, that varies with the
temperature, to the chip.
>Another approach might be to switch build a slowcharger attachment to
>your current system and switch to it once you get partway charged.
>Measure the current that the regular charger puts into the battery.
>When you use the battery, try to remember the amount of time you used it.
>At charging time put the same number of amp-hours back in,
>then switch over to a slower charger, and trickle charge until the next
>ride.
This is a good plan, so long as you:
1. Don't overcharge the battery by (accidentally, perhaps)
starting to charge when the battery's too close to fully charged.
2. Don't allow the battery to overheat.
3. Remember to throw the switch (perhaps a switching timer can help
here).
An even better plan might be to keep two batteries: one to use and one
to charge. Charge at C/10 or less, and allow enough time for the
battery in the charger to charge fully (say 16 hours or more at C/10)
before swapping the batteries between the light and the charger.
>I think I read somewhere that most batteries require at least 1.4 Ah
>back in for each Ah taken out, so you could probably shoot for 1.2 times.
>(Does anybody out there have more accurate numbers????)
I seem to recall reading 14 to 16 hours from full discharge to full
charge, at a charge rate of C/10, but I may be wrong.
>You might be able to make a slow charger out of your fast charger. If
>your supply is not very well regulated you may be able to get away
>with a resister in series to drop the current. See what happens to
>the current if you put a resistor in the loop.
>
>If your charger happens to be well regulated (ha!), you might have to
>put the resistor in parallel to shunt the extra current, but then you
>have to really careful not to let it discharge the battery if the
>power goes out, perhaps a diode would be good.
No, I wouldn't shunt the charger. Measure the open circuit voltage of
the charger (i.e. the voltage present when the charger is plugged in, but
the battery is out of the charger).
DIGRESSION. Although you never would ask for a NiCad cell at the store,
the single, 1.25 Volt (nominal) NiCad "battery" that you buy in the
store is, more formally, a NiCad cell. Strictly speaking, two or more
NiCad cells must be connected together, usually (and I'm assuming it
here) in series, to form a battery. END OF DIGRESSION.
Let: V = measured open circuit voltage [Volts].
N = number of cells in the battery
R = series resistor resistance [Ohms]
W = series resistor power rating [Watts]
I = desired charge current [Amperes]
Then, for NiCad cells,
N = <Fully charged battery voltage>/1.3, rounded to the nearest WHOLE
number.
If we assume a fully discharged voltage of 1 Volt per cell, then a
good resistor to start with would be:
R = (V-N)/I, rated at W > 1.5 * ( (V-N)^2 )/R
Connect the resistor in series with the charger and the battery. Measure
the current. Select another resistor if the current is too large or
if the current is more than five or ten percent too small.
>For trickle charging you want something less than C/10 rate to avoid
>overcharging, perhaps C/15. C/15 is the capacity of the battery
>divided by 15 hours, so a 3 Ah battery would be given 0.2 Amps
>for a trickle charge.
>I am not an expert in batteries and would love to hear a better way to
>do this.
I've done some reading, and I've fooled with NiCads in my shortwave radios,
but I'm no an expert either. That said...
If you insist on fast charging, the following way might be a better way
to fast charge.
1. To avoid overcharging, the battery should be discharged
(sometimes called conditioned) before it is charged at rates in
excess of C/10. To avoid reverse charging the least charged of the
cells, each of the cells of the battery should be discharged
individually. Therefore, remove the cells from your battery.
2. Connect a separate (but equal) discharge resistor across each
cell, and discharge the cells (say to 0.7 volts or less), being
careful not to overheat the cells. Check the manufacturer's
specifications to determine the maximum safe discharge rate and
choose a discharge resistor that will not exceed this rate for
fully charged (say 1.4 Volts) cells. Continue to discharge for
the time it would take to discharge the battery, if the battery
voltage remained constant at 1.0 Volts. Discharge all of the cells
for the same time.
3. Charge the cells in series, being careful neither to overheat nor
to overcharge them. Consider limiting the charge rate to the
maximum safe DISCHARGE rate (perhaps, even less). In any case,
consider limiting the charge time to 90% of the time that you
calculate would be required for full charge. Then, if you allow
the battery many minutes to cool (to the touch?), thereafter you
probably can charge the battery at C/10 or less, long enough to
charge the battery fully, or to charge it for as long as you care
to charge it.
Let me emphasize that you're NOT trying to avoid the dreaded (but for
practical purposes fictitious) memory effect. What you are trying to
avoid is damaging the battery by overheating and/or overcharging the
battery.
IMHO, fast charging is too much trouble, and, in my experience, fast
charging tends to shorten the useful life of NiCad batteries. Therefore,
I would use a series resistor, to limit the charge rate to less than
C/10. Then I could charge the battery for as long as I needed or wanted
to, regardless of the previous state of the battery when I started
charging it.
Michael Palmer
Jeffrey L. Bell wrote:
>
><snip>
>
> Just for curiosity's sake, measure the voltage after you've let it sit
> for a couple hours after you stop charging it. I expect it to drop back
> a bit.
>
Yep. It does. Approx 6.7 --> 6.4
> Try running it down to slightly yellow, then let it sit for a while.
> I bet it recovers somewhat.
Yep, again. If you hold the meter on it you can watch it.
<snip>
> -Jeff Bell
>
> PS: (Somebody correct me if I'm wrong for nicads. My experience is with
> Cyclon Gel Cells, but I read up on Nicads a few years back. If you do
> the measurements, I would love to hear the real numbers.)
--
Michael Palmer (jmpa...@worldnet.att.net)
Dan Grunberg
In article <334C5A...@worldnet.att.net>,
Michael Palmer <jmpa...@worldnet.att.net> wrote:
>Jeffrey L. Bell wrote:
>>
>><snip>
>>
>> Just for curiosity's sake, measure the voltage after you've let it sit
>> for a couple hours after you stop charging it. I expect it to drop back
>> a bit.
>>
>
>Yep. It does. Approx 6.7 --> 6.4
The NOMINAL NiCad cell voltage is 1.25 Volts. Since 6.7/1.25 = 5.36, and
6.4/1.25 = 5.12, it seems reasonable to assume that you have a five cell
NiCad battery. [Note that, in this usage, a cell is what you get when you
got to the store and ask for one "battery," but more precisely a battery is
two or more cells, usually (and I'm assuming so here, in series.] The
voltage decay that you're seeing after a couple of hours is 0.3 Volts
for the full battery, but only 0.3/5 = .06 Volts/cell. That isn't
enormous. What's happening is that the charged cell voltage is falling
from 1.34 to 1.28 volts. Since the lowest USEFUL charge is probably
between 0.9 and 1.0 Volts/cell (possibly even less), which for a
five-cell battery corresponds to a terminal voltage of 4.5 to 5.0 Volts,
I can can say confidently that your battery is useful and close to fully
charged, regardless of whether you read the battery's terminal voltage
is 6.7 Volts or 6.4 Volts.
>> Try running it down to slightly yellow, then let it sit for a while.
>> I bet it recovers somewhat.
>
>Yep, again. If you hold the meter on it you can watch it.
Sure the battery will recover SOMEWHAT with rest, but how much more
lighting time can you get from it. Also, what is the LOADED terminal
voltage when you first notice the slight yellowing, and again after
a couple of hours' rest when you first switch the light on again? And
what PER CELL voltage do those terminal voltages correspond to?
>> PS: (Somebody correct me if I'm wrong for nicads. My experience is with
>> Cyclon Gel Cells, but I read up on Nicads a few years back. If you
>> do the measurements, I would love to hear the real numbers.)
For a full discussion, try:
"Rechargeable Batteries Applications Handbook" by the Technical
Marketing Staff of Gates Energy Products, Butterworth-Heinemann, 1992
Cicli Milano
>Using your nicad for short periods and then recharging it is exactly what
>you should NOT be doing. Do not charge it after every ride. Use it until
>the light starts to go dim or turn yellow, then leave the light on until
>the battery is completely dead. Now, plug the battery into the charger
>when you go to bed. I charge mine next to my alarm clock so I won't forget
>it. When you wake up, wrap your hands around the battery. If it is
>slightly warm, then it is now completely charged. If it is not warm, it is
>not fully charged. Leave it plugged in to the charger until slightly warm,
>BUT NO LONGER than this, or you will over-charge it! Mine takes about 12
>hours.
Correct me if I'm wrong, but my understanding is that you should not
run it until it's completely dead, only until it starts to go quite
dim. I've always been of the understanding that running it down
completely is one of the primary causes of premature battery failure.
However, you are right in saying that it shouldn't be frequently
recharged after only short rides, and that it also shouldn't be
over-charged.
Larry
Dan Grunberg
In article <5ikn9a$5...@camel3.mindspring.com>,
Cicli Milano <ma...@pipeline.com> wrote:
>
>
>>Using your nicad for short periods and then recharging it is exactly what
>>you should NOT be doing. Do not charge it after every ride. Use it until
>>the light starts to go dim or turn yellow, then leave the light on until
>>the battery is completely dead. Now, plug the battery into the charger
>>when you go to bed. I charge mine next to my alarm clock so I won't forget
>>it. When you wake up, wrap your hands around the battery. If it is
>>slightly warm, then it is now completely charged. If it is not warm, it is
>>not fully charged. Leave it plugged in to the charger until slightly warm,
>>BUT NO LONGER than this, or you will over-charge it! Mine takes about 12
>>hours.
>
Whether it is safe to charge frequently after short rides depends upon
the design of the charger. All NiCad cells have a rated capacity, C, in
milliampere-hours (or possibly, but not usually, in Ampere-hours). If the
charging current in milliamperes is limited to a current in milliamperes
that is numerically equal to or less than C/10, then it is safe (not to
mention convenient) to put your battery back onto the charger for as
long as you like, regardless of how fully charged the battery was to
start with. Unfortunately, the only ways to know for sure are to read
your instruction book or to check with the manufacturer. OTOH, if the
charger is rated to charge your battery fully from 0 Volts in 14 hours
or more, the charger's charging current PROBABLY is C/10 or less.
>Correct me if I'm wrong, but my understanding is that you should not
>run it until it's completely dead, only until it starts to go quite
>dim. I've always been of the understanding that running it down
>completely is one of the primary causes of premature battery failure.
You are correct. As the NiCad battery approaches "completely dead,"
each of its individual cells voltages falls precipitously from
0.9 Volts (or so) toward 0.7 Volts and then towards 0 Volts. The
problem is that one of the cells may be the "winner" and reach 0 Volts
first. When that happens, the other cells will continue to supply
current to the lamp and they will tend to charge the "winner" so its
+ terminal's voltage becomes negative voltage with respect to its
- terminal. Sometimes this damages the "winner" irreversibly.
>However, you are right in saying that it shouldn't be frequently
>recharged after only short rides, and that it also shouldn't be
>over-charged.
What you are saying is true ONLY IF the charger charges the battery at
a rate greater than C/10 (see above). Follow the charging instructions
in your instruction book or check with your manufacturer.
Jeffrey L. Bell
In article <5igfhc$n...@nyx10.cs.du.edu>,
Dan Grunberg <dgru...@nyx10.cs.du.edu> wrote:
[lot's of good advice snipped]
>IMHO, fast charging is too much trouble, and, in my experience, fast
>charging tends to shorten the useful life of NiCad batteries. Therefore,
>I would use a series resistor, to limit the charge rate to less than
>C/10. Then I could charge the battery for as long as I needed or wanted
>to, regardless of the previous state of the battery when I started
>charging it.
I agree.
If you really need to have a faster charge on a few ocaisions,
add a switch to bypass the series resistor. Then you can do a
faster charge for a little while.
-Jeff Bell
Chris Cote
A voltmeter will tell you the "open circuit" voltage of a battery,
which is not a very good indication of its condition. Without getting
into a lecture about internal resistance and such, a more meaningful
measurement would be the load voltage, or the voltmeter reading across
the terminals of a battery when it is connected to a light bulb or
something. That is what good battery testers do, they use a resistor
to act as a dummy load, and then tell you the corresponding loaded
voltage of the battery. Anyhow, today's NiCad batteries aren't really
damaged if you charge them before they are discharged, so I wouldn't
worry about it.
Chris Cote
cjc...@wpi.edu
Worcester Polytechnic Institute
Electrical Engineering
On 2 Apr 1997 02:24:48 GMT, "Michael Palmer" <mpa...@mo.net> wrote:
>Is checking to see how the charging of a nicad is going as easy as slapping
>a volt meter on it (i.e. unplug the charger, check to see if voltage is up
>to where it is when I leave it overnight)? If it matters, the subject
>battery is a Niterider Cyclops Pro bottle format. I just don't want to
>overcharge it after using it for short periods, since I understand that is
>bad for them.
>--
>
>Michael Palmer (mpa...@mo.net)