Quick and Dirty High V Boost Driver

[pullbangdead]

Alright, so I know there are people working on various higher-voltage boost drivers, especially with the 8x and 12x Blu-ray diodes in mind.

But I need a quick and dirty solution to get higher voltage output from a boost driver with adjustable constant current drive mode. Doesn't need to be small, or fancy, or elegant, or efficient for the time being. Ideally, at some point in the near future, I'll have time to work on a switching regulator or someone else will come out with a pre-made driver that can do higher voltages, but for the time-being I need something quick and easy.

So like I mentioned, I need the variable current control mode also. So, would I be able to, say build an LM317 driver in constant-current mode like everyone does now, and instead of putting Vin as the battery voltage, throw a single IC or VERY simple circuit that I can grab at Radio Shack tonight on the input side and kick the input voltage up to, say, 20V? Is there some easy way to take the 7.2V output from 2 rechargeable lithiums, kick it up to 20V or so, and feed that higher voltage to an LM317, which will then regulate my current nicely while supplying anywhere from 10 to 18 or so V to the diode?

Seems there should be some very simple circuit that could accomplish this, to feed the LM317 a higher voltage than the batteries allow?

Anyone have a very simple idea like this?

I figure this is easier than doing all the legwork to get a switching regulator working perfectly with clean current control on the output, it seems from my initial searching that getting such a thing working properly would take me a lot longer than just finding a way to bump the Vin to an LM317 up to a higher V. All that, and I'm just an idiot with circuits too.

 

[Bionic-Badger]

If you can find the parts, you can build something like this. Radio shack usually has a pretty shoddy collection of parts though, so good luck with that.

I would instead look at something like rectifying (another) AC power into DC and then using a regulator on that. That only requires some rectifier diodes and a transformer, and even then you can get by without a transformer if you're okay with a full 120V. If you have a crappy ol' LED light bulb from DX it should contain enough components to build one directly. Just watch out, since it's AC.

If you rectify the AC the output is not going to be clean at all. I'm not sure how well the regulator can handle this, but you could put a capacitor before it to try and smooth out the signal a bit.

 

[charliebruce]

I remember seeing discussion about this before, but the primary reason against this is that the efficiency will be incredibly low - whilst a good buck-boost driver could hit 85-90% on a good day, I doubt that a circuit built along these lines would be able to hit more than 10-30% efficiency. Plus, it wouldn't be as compact either, it would make just as much sense to design a boost driver with heavy-duty components which are able to take the strain of the higher currents - if you look on Linear's site, there are several "power system" designers, where you specify input and output ranges, and it does most of the work for you.

 

[pullbangdead]

K.

Yeah, I don't care about efficiency at this point, I just wanted something quick that I could throw together this weekend, without having to order parts. Just something I could put together quickly and throw in a project box, size and efficiency be damned. If I can't get something fast, then I might as well put it off and design a circuit that works well, like a good switching boost drivers.

But it turns out some of the other parts aren't ready yet this week either, so oh well. Looks like it'll be January for this project.

 

[pseudolobster]

How about a joule thief?
Make a Joule thief.

If there's a simpler or cheaper boost driver out there, it'd be news to me.