Infinitely variable 6amp driver?

Hello everyone,

I am working on a project where I need to power 4 3-3.5v leds at aprox 1.5amps each wired in parallel. I can't wire them in series because they are non isolated and mounted to the same mcpcb.

I need the driver to be very high current and adjustable via a potentiometer or some other sort of dial control. I would also like it to be possible to run on several 18650s in parallel. I could do series but parallel is preferred if it's possible.

I considered running several lm317s but the input voltage wouldn't work and efficiency would be pretty bad as far as I can tell.

Anyone have suggestions either diy or modifying some available current regulator?

Yeah, LM317's need a lot of overhead voltage, and efficiency would suck - in the vicinity of 50%.

If you can run 3 18650's in series, a Taskled H6CC would work. That would be my first choice.

If you go to eBay and search "CVCC" you'll get some hits on regulators that can do the output current and voltage you want, but I don't know that they will work on an input voltage as low as 1 cell - if you can do 2 or 3 cells in series, you can probably find one that will work. Not the quality or small size of the Taskled driver, but okay if the specs work out.

If you really want to run on LiIon cells in parallel but NOT in series, check out this thing that I make: http://www.candlepowerforums.com/vb...driver-fits-D-cell-tubes!-thermally-protected

This would only be a good choice if you want to run on a single cell or you need to run cells in parallel but not in series.

Needless to say, I don't consider 6A to be 'very high current'.

If you aren't finding what you want, keep posting here and we'll keep helping you.

Since your goal is a light with infinite variability, perhaps consider thinking in analog terms instead of digital control.

What do you think about:
- Set up the desired 18650s in a parallel pack
- Run the current through a variable resistor (potentiometer) rated for the current level (6 amps or so)
- Select a resistance range suitable for the desired current range. Probably 1-100 ohms would cover it, maybe even 1-10 ohms.

The efficiency will be less than a perfect buck converter, but no as bad as you might think. Look at the battery voltage over the main discharge curve, not just the voltage when fully charged and turned on.

One extra cell can make up for imperfect efficiency.

The nice thing is that you can mock this up on your desk before committing to a complete build and see if it meets your needs or not.

Potentiometers are really impractical at those current levels. A 6A, 0-10 ohm potentiometer would be bigger than a tea saucer and thicker than a tea sandwich. If you could actually find one, it would cost well over $100 new.

That also would give poor control, as most of the change you want is in the first 1/10th of a turn, which means most of that massive size would be wasted.

Actually now that I think about it, 0-2 ohms would give you a fairly decent range of control, and would be a lot smaller. I found a datasheet, and it's 2.75" (70mm) diameter, and costs $88. The bad part is that you have to order 5 and wait 8 weeks. Oh, wait! Mouser stocks that one, and you can buy just one for $84!

Seriously, I suspect you'd be happier with a proper driver.

There ARE infinetely variable drivers around liek the ones with 3 AMC7135 chips from intl-outdoor.com
And similar (soldout) drivers from Kaidomain.
Thne you just have to parralel another AMC7135 drivers (many info if you search the forum)