Switch Mode IPod Charger

switch mode power supply

Usually when someone asks me how to build a USB charger I point them to Jason Striegel’s USB battery v2. It’ll work for a lot of things that only need 5V. Unfortunately [ian] has a 3G firewire iPod so he decided to build a switch mode power supply to generate the 8-30V required by firewire. The device is powered by 3 AA batteries and uses a PIC 12F683 microcontroller to regulate the output voltage. The device can power a dead 3G iPod, but it can’t charge it, so Ian is looking for some input from the community. Some have already suggested using FETs with a lower turn on voltage.

33 thoughts on “Switch Mode IPod Charger

  1. My gut feeling also says that the microcontroller is major overkill for this. It also says that there should be more AAs than just three, but I’m assuming it wouldn’t fit into the Altoids tin at that point. ;)

    Seriously, though, switching up their voltage is what’s killing a lot of their current (P=IV, and all), so they could get better current from the device by bringing their starting voltage closer to their ending voltage. And not having to drive a whole microcontroller would also benefit them a lot.

  2. Oh. And, while I’m at it, certain rechargeable put out more power than their alkaline equivalents. I think on my GPS project I did a year ago, we used four NiMH batteries to power a rather beefy wireless transceiver and GPS receiver on the same board. When we tried to stick alkalines in for testing, the system wouldn’t work because it wasn’t getting enough power. Other types of rechargeables might give even better results.

  3. Simple fix for them:

    Ditch the AA batteries and move to a _single_ 9 volt.

    They probably dont want to buy the charger though. They want to use all AA because of the cheaper charger they bought. If they would bite the bullet they would make it easier on them.

  4. Okay… so the first post stuff sucks… yeah… but what sucks worse is that hackaday’s comments don’t seem to line up timewise. Earlier I checked this and the only comment was imweasel’s (which is now showing at 3 for me after two posts that claim to not care about being first, yet still manage to mention it anyway).

    I’ve noticed this on many comment threads. Why is that?

  5. That’s cool and all….The simpliest way i’ve made a iPod charger was with 4 AA and a 5V voltage regulator. works perfectly with r3echargalbes (gets a bit above 4.8V) and the 6V from regular AAs are cut down nicely to five. Runs all day well over 10 hours on new ipods with 2500 mAh rechargables. All fits perfectly into an altoids tin (the 4 AA caddys at Radioshack fit perfectly in there).

  6. o man this looks sweet.

    but i want, a 5v usb port one for my pda.

    and the guy above me i thought that a 5v regulator needs 7v minimum. and 4aa makes 6…or did u use a LM113 or w/e, those adjustable ones.

    and quick question, what would it take to put 5v down to 3v. like just a resistor? if so, how many ohms?

  7. #4 Driving a PIC or similar microcontroller consumes little power — I think they consume something like 30mW.

    #11. Heh. Or use a 3.3v regulator.
    As for the minimum voltage required to run a regulator, it varies as one would expect by make/model, but I tend to also think he’s running it a bit close, since NiMH are closer to 1.2V/cell, which would give an input to the regulator of 4.8V.

  8. I read parts of his full documentations posted on instructables. The Pic is doing its job perfectly, its the components he’s using that causing the design to fail.

    His inductor looks like it has a rather high resistance (for an inductor). If he wants to generate the higher output current he has to cause a higher input current spike through the inductor. Using less inductance and a less resistive inductor would probably fix this but it depends on the frequency he’s using.

    The fet ‘might’ be causing current limiting but I kind of dout that. A BJT would be more suitable since its low voltage and there a current controlled device (as opposed to a fets voltage control). It’s not going to be any less efficient than a half turned on fet.

  9. An interesting hack…
    Of course it would have been just way too easy to buy a 3 cell lipo pack… You could probably fit 15-20w/hr of salvaged lipo in an altoids tin… (double the 10w/hr he has…)

    But since you insist on the hard way…
    You have a few problems with the circuit… The first is that the darlington you are using is not suited to this circuit at all, 100v is about 10x what you need… It is probably because there is about a 2.5v base-emitter drop across the darlington, combined with the low starting voltage of the circuit leads to a really crappy converter.
    Using a nice mosfet will be ideal here, and yes you can find them that will turn on with low voltages. You will find them to work much better.

    As to your inductor, it is not the inductors fault the circuit sucks, it is the drivers. You should be running at close to a mhz to get decent efficiency out the these things. The inductor can only store so much energy, so the faster you charge/use it the more power you can get out.

    And I still don’t see why you need to use a pic for the control. You say that you did not use a boost controller because you did not want specialized parts only available from a certain manufacture, but you use a pic that is only available from microchip… A simple boost converter would just be soo much easier and have much higher efficiency (both space wise and power wise) than your converter will (by design)… Using a PIC for this is like replacing the simple float switch in your toilet with a level meter attached to a computer with a solenoid attached to fill the bowl, sure it doesn’t require the specific float switch but it requires 10x as much work, power, space, etc…


  10. Hi guys,

    Thanks for the input, this is my project.

    The primary problems are:
    A) yes, the darlington is not doing it. I ordered a 2.5 v Vgs FET to replace the darlington.
    B) The inductor is to small. I read a few comments about using higher frequencies etc – that won’t work. Inductors have an ideal on time and off time that can be mathematically derived – the PIC measures the input voltage and sets the duty/period to match these values. I think at this boost it is running at a few 10’s of Khz – moving to a Mhz or more will just give less power from the inductor (and run it in discontinous mode, wasting power as heat and ringing). There is a 3A inductor at digikey I am considering, but have not ordered. A 3 amp inductor will put out 500mA @ a 3.8 to 12 volt boost, 700mA @ a 3.8 to 8 volt boost (3g ipod low), and more than an amp (1140mA) @ 3.8 to 5 volt boost.

    As for the purpose of using a uC instead of a custom chip, aka “but you use a pic that is only available from microchip.” I use a pic that is only available from microchip, but ANY uC with 2 ADCs, Vref, and a PWM will drive a SMPS – A different PIC, an AVR, even a PC if you were dedicated. Using a custom IC just makes this an exercize in following the manufacture’s datasheet. Where’s the fun in that?

  11. #7: Ditch the AA batteries and move to a _single_ 9 volt.

    You are spot on about the charger. But also, 9 volt batteries only have 150mAh, while rechargable AAs have 2700+, thats a whole lot more watts of power in the 3 AA @ 2700mAh than a 9 volter at 150mAh.

  12. does it work on a 5g (video) ipod? I heard that these don’t always work with USB charger type things because they are usb highspeed only, and there’s some kind of handshake thing they gotta do first. Anybody know?

  13. sorry man, the higher the frequiency (until you saturate the core) the more power you will get out of a given amount of inductance. Have you looked at any of the complete boost converters? You can get 10w from a 1uh inductor running at .5hmz easy. Some chips can get ~90% eficiency putting out 20w from from 1uh(600khz)

  14. Resistors limit current, not voltage. You could use a voltage divider if the appearant resistance of the voltage inputs of your device is constant.

    Radioshack’s 7805 only needs about a volt more than 5 to output a regulated 5 volts. It’s high current too, up to 1A. $1.59

    Radioshack also sells the LM317T, which is adjustable with two resistors from 1.2 to 37 volts DC, currently $2.29

    Can you tell I just got home from work?

  15. i dont know how to hook um a LM317T.

    for my pocket dj, it charges via wall charger, and in the wire, long story, is a 3 and 5v line. and a gourund. a normal usb provides 5v. i want to hook the wall charger mini usb to the comp, but i cant i have to be getting 3v into that other line.

  16. heres the pin out on a IPOD wall battery charger, from apple. this is what I got with my multimeter
    | |___| |___| |___| |___| |
    1) 5.12
    2) 2.75
    3) 2.15
    4) 0

    anyone think they can design a circut taht woule mimick that?

  17. Easiest solution I found was to buy a RadioShack racecar recharger and battery pack for approximately $25. Add a homebrew connector from battery pack to USB and/or Firewire. I’ve been using this combo for several months now and I love it. I bought the 9V battery pack because I have a Gen 3 that takes 8-30V input but it works with newer models that take 5-30V input. No voltage regulator is needed. Even cheaper stillm, I also bought a 8 battery pack from RadioShack for only $2 or so and put my existing 1.2 to 1.9 batteries in it. I use this battery pack with my selection of NiMh batteries and the race car charger to charge 8 batteries at the same time to be used for any other purpose they are good for.

  18. i cant find a 5v regulater so i pulled one out from a pc power suplly, but when i plug it up with a 9v battery i get 5.4v.

    will 5.4v kill my ipod nano?
    if so how can i put the v down?
    mabey a weaker battery? (ex.3-4 aa/aaa batterys)

  19. I have about ten litium-ion batteries from a bunch of recent Nokia cell phones my company has thrown out (new phones came with the renewal contract). I’ve been thinking about finding a way to use some since the batteries are so thin – they’re like little stackable cookies LOL. I think they are 3.6 v and they hold a charge for quite a while sitting on the shelf (in the gadget bag et cetera.) Perfect rechargable solution – now if I had a clue about circuits and electronics :-)

  20. I bought this mini ipod charger it runs of 1 AA battery and it has 2 parts, the charger itself it looks like a short stubby marker and it has a mini mono headphone type jack on it 1/32 inch i think it is and the other part has the ipod sync plug and a mono headphone plug 1/32 inch it says turbo charge on the bottom and it has a blue light that lights up and gets brighter whan you hool ur ipod to it, oh and the packaging says MADE FOR IPOD on it, on another note i have a 12 volt car jumpstart box and i use my griffin powerjolt cigarette lighter plug and i can charge and run my ipod off that just like its plugged into 110v wall power

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.