NimH AAA to 5V, suggestions?
in [Design]
|
Prev: passive rc-bandpass with 0dB attenuation in the passband
Next: increasing employment by firing people
From: Tim Williams on 3 Aug 2010 01:32 The trick is to put a regulator around it. The worst thing about a fixed bias circuit is it cooks the transistor if your load opens up! This works, if a bit cheezy: http://myweb.msoe.edu/williamstm/Images/Blocking%20Oscillator%20Supply.gif Once the output voltage comes up, all sorts of regulation schemes can be applied using slightly more voltage-hungry approaches (TLV431, etc.). You could even use a small one to provide bias for a FET circuit, which could keep working to arbitrarily low voltages (not that you can get any current out of a AAA at 0.2V). Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms "Jan Panteltje" <pNaonStpealmtje(a)yahoo.com> wrote in message news:i36oqj$kq$1(a)news.albasani.net... >I was looking for some 9V rechargeable blocks to power my sign_led display, > http://panteltje.com/panteltje/pic/sign_pic/ > to make it portable (put it in the car side window for example). > Those are expensive, I had one but cannot find it. > So I decided to build a small converter 1.2 to 5 V DC, > singe AA cell about 2.2Ah should give about 10 hours.. > ______________|\|_________ +5 > | |/| | > ---------------------------|-- Schottky | > | | | 5.6_______ || ( | 5t | > | | [ ] | 2t )|| (___| | > | | |--- ---- || ( 2t __ __ | > | | | | || | | | | | | > 1.2 V - === | | | --- ----- === > AA --- --- | | c NPN Si --- > NiMH | |22u | ------ b |150u > | | | e BC547 | > | | === | | > | | --- | | > | | |1u | | > ------------------------------------------------------- > Orignal Hand drawn ASCII > That works OK, does not get hot, but maybe it can be done better. > Input current is about 220mA, output about 30mA > Waveform at the collector is a nice assymetrical square wave, but it is a bit a 'soft' > output, loading it lowers the voltage a lot, high Zi. > Also for the ferrite core I grabbed some old small filter core. > no idea what stuff it is made of. > I do use the battery voltage in series to improve efficiency, > > I know there are little chips for 1.2 to 5V or 3.3V, > what is a good jellybean one? > Preferably one I can see with the naked eye and can buy in Europe :-) >
From: Grant on 3 Aug 2010 04:47 On Tue, 3 Aug 2010 00:32:18 -0500, "Tim Williams" <tmoranwms(a)charter.net> wrote: >The trick is to put a regulator around it. The worst thing about a fixed bias circuit is it cooks the transistor if your load opens up! > >This works, if a bit cheezy: >http://myweb.msoe.edu/williamstm/Images/Blocking%20Oscillator%20Supply.gif Ah! It's you! I'm trying to organise some of these circuits, now I have a name to put on the folder with this style circuit :) >Once the output voltage comes up, all sorts of regulation schemes can be applied using slightly more voltage-hungry approaches (TLV431, etc.). You could even use a small one to provide bias for a FET circuit, which could keep working to arbitrarily low voltages (not that you can get any current out of a AAA at 0.2V). OT: Does Agent have a line rewrap function? I can't find it, would save me propagating Tim's endless lines when I'm too lazy to manually reformat (no I don't mean the line length setting, I need to break that sometimes too). Maybe a solar panel too, get the most out of one. Though now there's people trying to power sensors from environment, be alright if there was much energy to recover, and, if there was, it'd be taxed! Grant.
From: Jan Panteltje on 3 Aug 2010 08:50 On a sunny day (Tue, 03 Aug 2010 10:04:48 +1000) it happened David Eather <eather(a)tpg.com.au> wrote in <OfqdnUbHooK-wMrRnZ2dnUVZ_oudnZ2d(a)giganews.com>: >On 3/08/2010 1:38 AM, Jan Panteltje wrote: >> I was looking for some 9V rechargeable blocks to power my sign_led display, >> http://panteltje.com/panteltje/pic/sign_pic/ >> to make it portable (put it in the car side window for example). >> Those are expensive, I had one but cannot find it. >> So I decided to build a small converter 1.2 to 5 V DC, >> singe AA cell about 2.2Ah should give about 10 hours.. >> ______________|\|_________ +5 >> | |/| | >> ---------------------------|-- Schottky | >> | | | 5.6_______ || ( | 5t | >> | | [ ] | 2t )|| (___| | >> | | |--- ---- || ( 2t __ __ | >> | | | | || | | | | | | >> 1.2 V - === | | | --- ----- === >> AA --- --- | | c NPN Si --- >> NiMH | |22u | ------ b |150u >> | | | e BC547 | >> | | === | | >> | | --- | | >> | | |1u | | >> ------------------------------------------------------- >> Orignal Hand drawn ASCII >> That works OK, does not get hot, but maybe it can be done better. >> Input current is about 220mA, output about 30mA >> Waveform at the collector is a nice assymetrical square wave, but it is a bit a 'soft' >> output, loading it lowers the voltage a lot, high Zi. >> Also for the ferrite core I grabbed some old small filter core. >> no idea what stuff it is made of. >> I do use the battery voltage in series to improve efficiency, >> >> I know there are little chips for 1.2 to 5V or 3.3V, >> what is a good jellybean one? >> Preferably one I can see with the naked eye and can buy in Europe :-) >> > > >LT1110 is nice, simple and seems like it will do everything you're after >(supplied by arrow and farnell). It also comes in a SO or PDIP package >so perfect for chucking on a lump of veroboard. That 8 pin DIP looks just what I need :-) >(I just wish I could find the couple I have sitting around somewhere. I >know I will find them - just as soon as their replacements arrive by >courier :-( >
From: Jan Panteltje on 3 Aug 2010 08:56 On Aug 3, 7:32 am, "Tim Williams" <tmoran...(a)charter.net> wrote: > The trick is to put a regulator around it. The worst thing about a fixed bias circuit is it cooks the transistor if your load opens up! > > This works, if a bit cheezy:http://myweb.msoe.edu/williamstm/Images/Blocking%20Oscillator%20Suppl... > > Once the output voltage comes up, all sorts of regulation schemes can be applied using slightly more voltage-hungry approaches (TLV431, etc.). You could even use a small one to provide bias for a FET circuit, which could keep working to arbitrarily low voltages (not that you can get any current out of aAAAat 0.2V). > > Tim Tim, great, nice regulator. Did not see your reply in my newsreader, so I did see Grant reply, looked it up with google.
From: Jan Panteltje on 3 Aug 2010 10:55
On a sunny day (Tue, 3 Aug 2010 05:56:40 -0700 (PDT)) it happened Jan Panteltje <panteltje(a)gmail.com> wrote in <081bd806-cb72-4733-b3ae-2ffde3fa08e2(a)w30g2000yqw.googlegroups.com>: >On Aug 3, 7:32�am, "Tim Williams" <tmoran...(a)charter.net> wrote: >> The trick is to put a regulator around it. �The worst thing about a fix= >ed bias circuit is it cooks the transistor if your load opens up! >> >> This works, if a bit cheezy:http://myweb.msoe.edu/williamstm/Images/Block= >ing%20Oscillator%20Suppl... >> >> Once the output voltage comes up, all sorts of regulation schemes can be = >applied using slightly more voltage-hungry approaches (TLV431, etc.). �Yo= >u could even use a small one to provide bias for a FET circuit, which could= > keep working to arbitrarily low voltages (not that you can get any current= > out of aAAAat 0.2V). >> >> Tim > >Tim, great, nice regulator. >Did not see your reply in my newsreader, so I did see Grant reply, >looked it up with google. PS I build your circuit, and it works perfectly, even with almost empty NiMHs. Will make a little PCB for it, and use that, it seems more efficient then my version. No need to order some chip :-) Here is the setup, I use the box to switch between full and almost empty NiMH batteries. ftp://panteltje.com/pub/1.2_to_5V_converter_test_img_2249.jpg This is the collector waveform: ftp://panteltje.com/pub/1.2_to_5V_converter_Vce_img_2250.jpg The 7805 on the LED PCB is disconnected, I go in after that. Project page: http://panteltje.com/panteltje/pic/sign_pic/index.html Thanks :-) |