Thread Tools
Jul 12, 2004, 10:46 AM
mhmitchell's Avatar
Thread OP

multi purpose charger plans

I am looking to build a all purpose charger for NiMh, NiCad and at least two cells L-poly.

I would prefer one that does not use a PIC or any type controller since if I have to get an emulator and electronics to program the thing I might as well save my pennies for a commercial system.

It should also be capable of reasonably quick field charging.

Does anyone know of a tried and true charger schematic/plans that fits this description.

Sign up now
to remove ads between posts
Jul 12, 2004, 11:31 AM
now that's a wattmeter...
simingx's Avatar
Why not just buy something like the Hobbico Quick Field Charger II? Easier and will probably end up cheaper than building your own....
Jul 12, 2004, 01:36 PM
mhmitchell's Avatar
Thread OP
actually I have several hundred dollars wortth of electronic parts. My cost would be very low.
Jul 12, 2004, 08:10 PM
Hi Mike,

Originally Posted by mhmitchell
I would prefer one that does not use a PIC or any type controller since if I have to get an emulator and electronics to program the thing I might as well save my pennies for a commercial system.
I don’t do PIC’s, but I program AVR’s and it costs less than $10 to build a ‘Programmer’ and ‘Development board’ total. (including an ATTiny-15 microcontroller ($2.56)) Free Compilers and Assemblers, IDE, programming software and Program Simulators are all available.

If you go the AVR Freaks website ( ) there’s even an Appnote, AVR 450, ‘Battery Charger for SLA, NiCd, NiMH, and Li-Ion, batteries’

Originally Posted by mhmitchell
It should also be capable of reasonably quick field charging.
You didn’t say how many NiCad and NiMh cells you wanted to charge but it’s no problem charging a 2-cell Lithium battery pack from a 12 volt supply.

The chemistry, and how they’re charged, is totally different between NiMh / NiCad batteries and Lithium batteries. An analog charger that does both would require two different chargers in the one box. The only thing they would share is the DC supply

Here’s what you need for NiCad and NiMh batteries, or this one, .

And here’s a Lithium charger,

Doing a search in this group for ‘Lithium’ and ‘charger’ or ‘NiCad’ (or ‘NiMh’) and ‘charger’ gets results.

Jul 12, 2004, 09:54 PM
My member is not
Zlatko's Avatar
If you want to tinker have a look at LM3647.
Thats the "brain chip" from National Semiconductors.
Does the lot Li-Ion, Li-Po, NiMH and NiCd
I got the Eval board before there were any Li-Po cells available ( 4 years ago ) and it still works great but there are lots of jumpers and settings.
Cheers and Good Luck
Jul 13, 2004, 08:49 AM
mhmitchell's Avatar
Thread OP
thanks guys seems like I have some more researcg to do!
Jul 14, 2004, 04:17 AM
We want... Information!
Bruce Abbott's Avatar
I suggest you make two (or even three) chargers, one for each type of cell. That way there's less likelyhood of an unfortunate accident, and you might even be able to charge several batteries at once...

Here's my Nicad/NiMH charger which doesn't use any exotic chips:-

Simple Peak-detecting Nicad Charger
Jul 31, 2016, 09:56 PM
Registered User
Hi. Built your nicad charger. Thank you for the design. However it will not start upon pushing the momentary switch. Built the Pfet version. Has anyone else built the pfet version which works? All of the parts used are identical to your schematic except for the Fet. We cannot get the Hitachi part here. Instead used a Fairchild FQP47P06. The specs are close. Appreciate any insights. Thank you.
Oct 07, 2016, 02:18 PM
Registered User


Anybody? Please. Crawled through this numerous times. Is the switch momentary or on-off? If monetary the fet version does not work. Thank you.
Oct 07, 2016, 03:01 PM
Registered User
It's momentary, and: have you checked the given voltages and component selection section?
If the charge cycle won't lock in it seems either
- the charge current is not high enough so disconnect/end of charge is detected
- the peak detect does not reset properly

Maybe you have wrong LED colors, or the gate voltage on the FET is too low so it does not "open up" properly.
And of course there's always bad parts, missing connections, wrong polarity.
Oct 07, 2016, 03:35 PM
Registered User
Thank you very much for commenting. As mentioned cannot find 2Sj295 so tried FQP47P06. Noticed from data sheet that 2SJ295 has a 4V RDSon drive but FQP47P06 does not. Changed the FQP47P06 to a 4V drive part but still does not work. Been crawling over this ad nauseam for way too many months off and on but still no go. Have the correct voltage drop across R11 (.9V). Changed several LEDs to ensure this. Went over every solder joint numerous times and re-tested every component today again but still push switch, led lights but does not stay on. Using 1.5 ohm current resistor but circuit only draws 40mA and output to battery is ~5mA. I build a lot of uC circuits and occasionally have a bad solder joint but I think this is my all time worst build. Can't image that the 2SJ295 is so unique that a substitute with close specs will not work. 4V drive part I am using now is Rohm RSJ250P10TL. It also has the same schottkey body diodes found in the 2SJ295.
Oct 07, 2016, 03:56 PM
Registered User
Every FET has inherent Diodes, so you can't miss them.

Only 5mA on the Batt means the FET does not get driven properly, and as such the charge end detector triggers and ends charge on letting the switch go.
If you short C-E on Q1 the charger is fully active and should deliver full current to the batt.
Connect a battery or a resistance for proper current/voltage and check voltages in the top right corner then:
- R8 should have about 0,6V
- G-S should have a few volts

It looks like either the gate does not get enough potential or the current limiter resistor has a too high value/is interrupted.
Oct 07, 2016, 04:17 PM
Registered User
Thank you again. Shorted Q1 C-E: Charge led turns when shorting C-E on but still only 3-6ma to battery.
Oct 07, 2016, 04:27 PM
Registered User
Here is another curiosity. In the eagle board posted by Bruce Abbott the BC327package is "TO-92 EBC" The schematic shows Q3 E connected to VCC but if the package he is using is EBC then from the eagle .brd it appears that Q3 C is connected to VCC. Adding to the confusion I am using which shows leads as CBE. The old Phillips part, now NXP is EBC, Which do you think is correct? Thanks again.
Oct 07, 2016, 04:44 PM
Registered User
Ah, so you have the semiconductors polarity wrong.

Correct is what type you are using.
Personally, i prefer datasheets from the manufacturer, and transistors with reversed pinning usually have a different marking.
And you need to take a second look at the sheets... I see CBE standard pinning in both your linked sheets.

Quick Reply

Thread Tools