Searching \ for '[EE] making a simple battery charger ?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/power/batterys.htm?key=battery
Search entire site for: 'making a simple battery charger ?'.

Exact match. Not showing close matches.
PICList Thread
'[EE] making a simple battery charger ?'
2011\10\17@211816 by Jason White

picon face
part 1 1148 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

Hello, I am trying to design a simple battery charger. It is supposed to
charge 4x AA (2000mAh) rechargeable batteries off of a 5 volt supply. Right
now what I have designed has an op-amp is tied to a 4.76 volt reference and
the positive battery terminal. When the battery's voltage is below the
reference it switches on a NPN transistor with a 35 ohm resistor at its
emitter. When fully on the batteries will receive about 140mA. The charger
will continue to remain on until the voltage rises above the reference.

I believe I'll be using a dual op amp, which leaves room for expansion.

What do you suggest for fuses ? advice on resettable fuses, temperature
sensors, and other safety features would be appreciated.
What Is the recommended charge voltage cutoff ? (RTFM ?)
Is the voltage reference method described a viable mean of charging
batteries ?
What other cheap/low part count methods are there to safely charge batteries
?


Note: This is not a production design, it is meant to be simple and cheap.
Though I don't want it to destroy the batteries or leave them uncharged.

-- Jason White


part 2 4335 bytes content-type:image/png; name="charger-email.png" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\10\17@214620 by RussellMc

face picon face
part 1 1406 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

> Hello, I am trying to design a simple battery charger. It is supposed to
> charge 4x AA (2000mAh) rechargeable batteries off of a 5 volt supply.

It's good to have a diagram - but old tired eyees, and probably many
younger ones as well would prefer a monochrome one which is not so
hard to view. (See attached).

The batteries will never charge fully.
If only 5V was available you will get either only relatively uncharged
cells or you will need to charge portions of the pack separately in
some manner.
At under about 1.35v x 4 = 5.4V you probably don't need charge
termination as the cells will trickle to a full stop without being
fully charged.

A fully charged NimH cell has a voltage under charge of about 1.45V.
Slightly less is bearable but much less will result  in incomplete
charge. This is not necessarily bad as it can substantially increase
cycle lifetime.

5V/4 = 1.25V cell =  very low
4.76V/4 = 1.19V = very very low.

4 x 1.45V = 5.8 V
4 x 1.4     = 5.6 V
4 x 1.35V = 5.4V


This is not a recommended way of charging NimH, but is sometimes the
best available. The two main methods are to charge at about C/1 rate
and either look for a negative delta V or to look for a rise in delta
temperature rate or an absolute temperature limit. These methods are
not viable at charge rate substantially below C.



           Russell McMahon


part 2 26846 bytes content-type:image/jpeg; name="charger.jpg" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\10\17@215837 by RussellMc

face picon face
Note - battery polarity as shown in diagram is reversed.
Longer bar is positive, shorter bar is negative.


     Russel

2011\10\17@220128 by V G

picon face
On Mon, Oct 17, 2011 at 9:18 PM, Jason White <
spam_OUTwhitewaterssoftwareinfoTakeThisOuTspamgmail.com> wrote:

{Quote hidden}

You probably won't charge the NiMH batteries fully. Also, NiMH shouldn't be
charged with a constant-voltage method. You charge them with constant
current, and detect a -dV and/or +dt (rise in temperature). Also,
personally, I don't feel comfortable charging them in series like that since
slight differences in capacities might cause odd charging behavior over
time. I always charge in parallel, via single channels. Also note that you
don't know exactly how much current is going through the transistor since
the B-E current vs C-E current properties change over temperature and amount
of amplification. Also, I don't think you /really/ need that capacitor
there

2011\10\17@221014 by V G

picon face
On Mon, Oct 17, 2011 at 10:01 PM, V G <.....x.solarwind.xKILLspamspam@spam@gmail.com> wrote:

> You probably won't charge the NiMH batteries fully. Also, NiMH shouldn't be
> charged with a constant-voltage method. You charge them with constant
> current, and detect a -dV and/or +dt (rise in temperature). Also,
> personally, I don't feel comfortable charging them in series like that since
> slight differences in capacities might cause odd charging behavior over
> time. I always charge in parallel, via single channels. Also note that you
> don't know exactly how much current is going through the transistor since
> the B-E current vs C-E current properties change over temperature and amount
> of amplification. Also, I don't think you /really/ need that capacitor
> there.
>

Since this is a simple low current application, you can get awesome battery
charger ICs from Linear and National that do exactly what you're looking for
for less than the price of a coffee.

Check these out this linear NiMH charger (I love this one, it rocks):

LTC4060 http://www.linear.com/product/LTC4060 ~$4.50 in singles, exactly
what you need. Just order a sample if one is all you need

2011\10\18@030826 by Michael Watterson

face picon face
On 18/10/2011 02:45, RussellMc wrote:
> The batteries will never charge fully.
> If only 5V was available you will get either only relatively uncharged
> cells or you will need to charge portions of the pack separately in
> some manner.
> At under about 1.35v x 4 = 5.4V you probably don't need charge
> termination as the cells will trickle to a full stop without being
> fully charged.

At 4.76V as Russell says the batteries hardly charge at all.
For quite a while on 24mA load I measure NiMH AA as being > 1.3V  (four is thus 5.2v)

Likely 5.6V approximately is a minimum charge voltage. But for any reasonable charging maybe over 6V supply is needed.

It's somewhat temperature related.

The "traditional" 1.2V quoted is
1) A nominal kind of average over most of discharge time of NiCd. Not an actual voltage of fresh battery.
2) NiMH are slightly higher terminal voltage. Maybe about 0.1V to 0.15V, but it depends on where you decide the "end point" is.

You'll get two different voltage readings for no current and charging current for a nearly charged cell

2011\10\18@032327 by Michael Watterson

face picon face
On 18/10/2011 02:18, Jason White wrote:
{Quote hidden}

Years ago before charger ICs existed I made a charger just using transistor, green LED as voltage reference and emitter resistor to set current. Batteries went on collector. So 45mA Constant current.
Higher capacity NiMH don't like a trickle charge after charged.

I've implemented a "turbo" charger with a PIC. But a dedicated charger IC is way to go

2011\10\18@063726 by Jason White

picon face
Thank you, I'll take a look into battery charge ICs.


-- Jason Whit

2011\10\18@084451 by Mark Hanchey

flavicon
face
I went the PIC route for a battery charger. It wasn't the lowest part count though.  Each battery is charged to get the max charge . It has  a routine where the batteries  charge for a period of time, about 10 minutes, then are switched off,  allowed to rest for 20 seconds and then the voltage checked with a load switched in by the PIC. Each battery is connected to a separate ADC on the PIC and if below the limit is charged more. The great thing about this setup is that it never heats up the batteries much during the charging routine and the run time on the batteries is longer than with any charger I had used before.

Mark





On 10/17/2011 9:18 PM, Jason White wrote:
> What other cheap/low part count methods are there to safely charge batteries
> ?
>

2011\10\18@143540 by V G

picon face
On Tue, Oct 18, 2011 at 8:33 AM, Mark Hanchey <markspamKILLspampixeltrickery.com>wrote:

> I went the PIC route for a battery charger. It wasn't the lowest part
> count though.  Each battery is charged to get the max charge . It has  a
> routine where the batteries  charge for a period of time, about 10
> minutes, then are switched off,  allowed to rest for 20 seconds and then
> the voltage checked with a load switched in by the PIC. Each battery is
> connected to a separate ADC on the PIC and if below the limit is charged
> more. The great thing about this setup is that it never heats up the
> batteries much during the charging routine and the run time on the
> batteries is longer than with any charger I had used before.
>
>
I thought about this method before, but never went through with it. I'm
assuming you're talking about NiMH. What voltage threshold did you use

2011\10\18@210423 by Mark Hanchey

flavicon
face
On 10/18/2011 2:35 PM, V G wrote:
> On Tue, Oct 18, 2011 at 8:33 AM, Mark Hanchey<.....markKILLspamspam.....pixeltrickery.com>wrote:
>
>> I went the PIC route for a battery charger. It wasn't the lowest part
>> count though.  Each battery is charged to get the max charge . It has  a
>> routine where the batteries  charge for a period of time, about 10
>> minutes, then are switched off,  allowed to rest for 20 seconds and then
>> the voltage checked with a load switched in by the PIC. Each battery is
>> connected to a separate ADC on the PIC and if below the limit is charged
>> more. The great thing about this setup is that it never heats up the
>> batteries much during the charging routine and the run time on the
>> batteries is longer than with any charger I had used before.
>>
>>
> I thought about this method before, but never went through with it. I'm
> assuming you're talking about NiMH. What voltage threshold did you use?

I used 1.28 for the voltage under load with a 100ohm resistor switched through a FET.
The method of charging is called burp charging, there is a NASA paper on it where they use very short
charge and rest periods , only ms long.
http://dnd.hu/letoltes/AdvanceTec_whitepaper.pdf

It works well and the batteries don't wear out as often.
Mark

2011\10\19@101000 by RussellMc

face picon face
Re "burp charging".

NASA burp charging report  1996.
1st author subsequently did a PhD on this.
Scanned report is VERY badly damaged in a number of critical places.
(Houston. We have a problem ... )
What is left seems self contradictory -  eg cycle life is/isn't affected.


http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19970013743_1997021557..pdf

___________________________

PhD. Darcy 1998
INVESTIGATION OF THE RESPONSE OF NIMH CELLS TO BURP CHARGING
University of Houston

    http://corsair.flugmodellbau.de/files/elektron/NASA-II.PDF

Ah - thesis makes sense of missing parts of paper.
Have skimmed thesis - may have missed crucial points.

- Method 1: Initially NimH cells were charged and discharged at 1C and
charge-terminated when Tcell_absolute = some_limit.
Cell capacity rises from 90% x Cnominal to100% in about 120 cycles and then
fell back to 90% after 200 total cycles and was down to 60% at 400 cycles.
[Their fig 3]

- Method 2: Cells were then charged / discharged at 1C/1C with negative
delta voltage termination.
Capacity took 130+ cycles to rise to value of method 1 capacity rose withe
cycle count, reaching within a few % of nominal capacity at 300+ cycles.
This method plus original in fig3

- Method 3 = Method 2 plus "burping". All 3 graphed in fig 4.
Burping adds about +2%  capacity across the range.


Note that if input energy is low cost and available (eg grid powered) then
burping may cost slightly nor energy wise.
A 2.5C discharge pulse every 1.075 seconds (sample as given) then the
discharge represents lost energy at
2.5C x 5 mS / 10 75 mS cycle ~~= 1.16% C discharge rate. If charge rate is C
then about 1.16% energy is lost.

If charge rate is slower then more energy % is lost.
eg a C/5 charger with pulses as shown would lose almost 6% of the charge
energy in the burp pulses.
The burp pulses may not have to be as long or as high current to be
"effective".

If the thesis is indicative of what can typically be achieved then a say
2.5% increase in capacity is useful but not stunning.
eg
2000 mAh gets 50 mAh added.
2500 mAh gets 62 mAh added.
A battery that takes 400 photos may take an extra 10.
A motor that runs for 4 hours may run for 6 minutes longer.

The thesis seems to indicate that the overall cycle life is NOT improved.


      Russell McMaho

2011\10\19@121738 by Mark Hanchey

flavicon
face
On 10/19/2011 10:09 AM, RussellMc wrote:
> Re "burp charging".
>
> NASA burp charging report  1996.
> 1st author subsequently did a PhD on this.
> Scanned report is VERY badly damaged in a number of critical places.
> (Houston. We have a problem ... )
> What is left seems self contradictory -  eg cycle life is/isn't affected.
> .
>
> The thesis seems to indicate that the overall cycle life is NOT improved.
>
>
>         Russell McMahon

The problem with the thesis is that the methodology used was where they charged and discharged batteries repeatedly over a short period of time and extrapolated that to mean that the same results would be gained from a year or two usage of the same cells. I can say that it definitely made a difference for me. I built the charger because a canon digital camera I have was going through batteries  quickly.  I started trying different charging methods to see what would give the camera the longest run times and burp charging exceeded everything else by a fairly large margin. I could get about 30 minutes of video recording with normal charging, with burp charging that went to 43 minutes and I got several photo shots in as well.

Mark

2011\10\19@125919 by Jim Higgins

flavicon
face
Received from RussellMc at 10/19/11 14:09 UTC:

{Quote hidden}

Yes.

And the number of battery packs tested was rather low - 10 or less - and the researcher was using only an 80% confidence interval in making his conclusions.  Also he doesn't provide either raw data or summary statistics that give even a hint to the variation between the relatively small number of battery packs involved in the research past claiming a result with 80% confidence.  Where are data showing variation between battery packs charged by the same method?  We got summary stats for pore and particle size, why not for capacity performance between groups and between packs in the same group?

In hindsight, a far more interesting result might have been achieved by running this experiment in duplicate and switching charging methods between the two groups of batteries at about 200 cycles.

2011\10\19@125919 by Jim Higgins

flavicon
face
Received from Mark Hanchey at 10/19/11 01:03 UTC:

{Quote hidden}

This isn't a NASA paper.  It's promotional literature from the manufacturer of the burp charger.

2011\10\19@132122 by doug metzler

picon face
I see a PICList group project in our near future :-)

I'd love to have a well-tuned 4@ single channel AA/AAA NiMH charger.

DougM


On Wed, Oct 19, 2011 at 9:58 AM, Jim Higgins <HigginsJspamspam_OUTsc.rr.com> wrote:
{Quote hidden}

>

2011\10\19@153641 by Mark Hanchey

flavicon
face
On 10/19/2011 12:59 PM, Jim Higgins wrote:
> Received from Mark Hanchey at 10/19/11 01:03 UTC:
>
>> e talking about NiMH. What voltage threshold did you use?
>> I used 1.28 for the voltage under load with a 100ohm resistor switched
>> through a FET.
>> The method of charging is called burp charging, there is a NASA paper on
>> it where they use very short
>> charge and rest periods , only ms long.
>> dnd.hu/letoltes/AdvanceTec_whitepaper.pdf
>>
>> It works well and the batteries don't wear out as often.
>> Mark
>
> This isn't a NASA paper.  It's promotional literature from the manufacturer
> of the burp charger.

I know . The NASA paper isn't available online but the information they give in their whitepaper for the advancetec charger is almost the same as the NASA docs and illustrates what I was referring to in the charger design.  The NASA docs are available from NASA for a small fee.

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...