mds82,
There are a few drivers that can have both their inputs and outputs put in parallel without the electronics interfering. I heard of someone hooking two drivers from DX in parallel and he successfully had combined current outputs. I bought the same drivers and tried it myself. I have two of them in parallel, running from 2 li-ion cells, and it powered the LED at ~2000mA. Here is the driver:
http://www.dealextreme.com/details.dx/sku.3256
It is rated for 800mA according to the site, but we have found that the current varies quite a bit between each one (from 700mA to 1000mA). I was fortunate to get two that were both providing ~1000mA to the LED.
Theoretically, any buck circuit should work this way, but there is often conflicts with the signals from the driver's ICs and component values that may cause instability and reduced efficiency.
You could try connecting several different types of drivers from DX/KD in parallel to test which work in multiples. Fortunately, those drivers are relatively cheap. You may be able to study the designs of some to see if component values interfere (increase or decrease due to being connected in parallel).
With the driver I mentioned above, you may need three or four together to get ~2.8A. If the current is too high, but sort of close, I heard that you can sand/file/scrape the top of the sense resistor to reduce current some.
Oh, and the only driver (or actually just IC) that is designed to be attached in parallel is the AMC 7135 ICs. They are basic linear current regulators, but you can 8 attached in parallel to get the needed 2.8A. They require an input voltage of 3.6-4.5V, so they would be great for li-ion, or three NiMH cells. I am unsure about efficiency, but they are simple and easy to use. Here is a link for their 1400mA version (four AMC ICs in parallel on each driver). You could just attach two of the boards in parallel...
http://www.dealextreme.com/details.dx/sku.1886
I, too, am exploring circuits for the Seoul P7. I am expecting one to come in the mail soon, then I will start testing it with my CC power supply and begin ordering flashlight parts for it. I am stuck between using 2 li-ion cells and the first driver I mentioned (sku 2356 -3 in parallel), or just direct driving the emitter from a li-ion or 3 NiMH cells (if the voltages will allow). With using the lower voltages (maybe in DD), I can then use George's D2DIM PWM driver to dim the Seoul P7. I wonder how efficient the AMC 7135 drivers are, because I can always hook 8 in parallel. So, I hope you can find drivers; good luck. I will be on the search as well...
-Tony
:thinking:
