Another USB charger!

[mat_the_cat]

...but with a difference. It's something that's been in the back of my mind for a while now, but I want to build an emergency USB power supply. Primarily for when I go camping away from mains electricity - I know you can buy AA battery powered USB chargers but I'd prefer to build one powered by CR123A cells as that's what I use for my A2 Aviator EDC. The long storage life is ideal for a backup charger and they are capable of low temperature operation; I've just returned from a skiing holiday where some of the people I was with suffered from flat phone batteries. They were most surprised when I didn't have something in my bag to solve the problem - I seem to have acquired a reputation for being prepared for anything!

But then I had a revision of the basic idea - it struck me that sometimes I would be using it as a main power supply (i.e. to charge my MP3 player) rather than an emergency backup phone charger - in which case it would get fairly expensive to use CR123A batteries. So my current (

) plan is to use CR123As in conjunction with 2 AA cells. That way, whenever possible I will use the cheap to recharge Eneloops of which I already have plenty; but can still switch to an alternative source if too cold for NiMH chemistry, or if they are discharged. And as a third resort I can simply replace the AAs with alkalines bought just about anywhere!

Of course, this doesn't stop me from losing the damn thing!

Or dropping it in a river, although I'm looking at an IP67 case to cover that eventuality...

Watch this space!

 

[Illum]

so... 5V at what? 500mA? What are you planning to charge and for how long a charge time?
The answers to these questions could turn your project from "simple" to "impossible"

 

[mat_the_cat]

I'm thinking of 500 mA output...as for what to charge then I have a Sonim XP2.10 phone which is pretty good on battery life so rarely needs charging on trips. It would be purely as a backup for that. But my MP3 player doesn't last a full day, so the ability to regularly charge that would be handy if not critical. I like the idea of being able to provide power from an easily replaceable source (AA cells) but still having the back up of the CR123As for more important needs. So I'd probably use the AAs as long as was possible/needed, but if down to the lithiums I'd eke things out as much as possible because if I'm in the situation of needing to use them then things won't have been going to plan!

Also not ignoring the nice warm feeling you get from being prepared, even if that preparation is not needed. There's a (smug?) sense of satisfaction of being able to help others too.

I'm rarely going to be away from civilisation for more than a few days, otherwise I'd probably be thinking along different lines. I'm thinking of the LT1302 switching IC, although I notice a minimum operating temperature of 0 degrees C. I assume that it will still work outside this range but perhaps not as efficiently until self heating brings its temperature up. It's probably safe to assume that if the temperature is much below -15 deg C I'll be heading for a bar rather than a piste though...

 

[Steve K]

My "to do" list includes making a USB charger like this, although I'm planning to stick with AA's of either the nimh or alkaline variety. This is a common enough need/desire that there is a nice on-line project called "Minty Boost", where this sort of charger is packaged in an Altoids tin:
http://www.ladyada.net/make/mintyboost/

It uses the LT1302, so it's a nice confirmation that you have chosen well.

I'm a little surprised that Linear Technology doesn't list a version of the LT1302 rated for the industrial temperature range, but the datasheet does say "consult factory for industrial and military grade parts". That usually means that they can sort out the ones that do well over the broader temperature range of those applications. If it makes you feel better, the datasheet does show plenty of graphs indicating how some parameters vary between -50C to 100C, so I don't think your LT1302 will stop working at -1C. If you are worried, just keep it in your pocket and keep it warm.

 

[mat_the_cat]

Interesting link there, Steve! I'd seen the cheap versions on eBay, but I enjoy a little bit of dabbling with electronics even if it does cost a little more. I may incorporate their findings on charging iPhones into my circuit, assuming that voltage on the data lines does not affect the charging of other devices. It would be useful to be able to charge as many things as possible - my wife has an iPhone so I can test on that.

Revision 2 is the idea of a built in charger for the Eneloops - in order to get an IP rated AA battery holder I can access from the outside I would have to go up a case size, which I don't really want to do. So if I'm finding that I'm having to open the lid frequently to charge/replace cells I may build in a basic charging circuit so I can charge from a USB port. It would then have the rather cool (if completely pointless) ability to charge itself!

To test everything I'll probably put the whole shebang in a freezer for a while...I wonder whether it's a thermal expansion factor that is the reason for the temperature limit? In which case the rare instances of extreme cold may shorten its life slightly but I doubt will be a real issue at all. I've now ordered the LT1302 and the case, so will update with pictures once I start building.

 

[Steve K]

Since there's a distinct advantage to being able to use whatever AA's are available, it seems like you'd give up some flexibility if you sealed up the AA's in the device (although it would give you an excuse to build a buck converter for charging the AA's from the 5v usb power!).

My approach towards achieving a degree of moisture immunity would be to coat the board with a couple of layers of Plasti-Dip....
http://www.plastidip.com/
...specifically, the tins of goop intended for coating tool handles. For about $7 locally, I can get a tin that has more goop than I can use in a couple of years. I've used it on my bicycle electronics as the only sealant for a small circuit board with 4 mosfets on it, and it has provided years of all-weather service.

 

[mat_the_cat]

Ah, I might not have explained very well - I wasn't thinking of totally sealing in the Eneloops, just that to change/charge them I would have to undo 4 screws. I think time will tell whether the frequency I perform that is worth the effort of squeezing in a charging circuit.

Plasti-Dip is good stuff though, and I'd never even thought of encapsulating circuits in it :thumbsup: At the risk of digressing, can I ask what circuit you have on your bike requiring FETs?

 

[Steve K]

The fets are for a dynamo LED light where I used them to full wave rectify the AC power. It's a little complicated, but it does cut the voltage drop. Not a big deal until you need to optimize the light output at slow speeds, such as grunting a heavy commuting bike up a 14% grade in the dark.

 

[mat_the_cat]

Interesting stuff - I just use a standard (wasteful) regulator on my commuter bike, although not used it for a few years now I live 25 miles from work!

Made some progress with the charger - mostly assembled although just with cheap alkalines for now.

Looks like I will be able to fit in a USB charging circuit, and hopefully a voltmeter...

I've gone for a trailing socket simply as that's what I had in stock, and an IP rated socket would be both costly and bulky. Sure, it won't be waterproof when in use but even with a proper socket it wouldn't be unless I got matching plugs.