Originally Posted by General Lee
Buck, boost, same difference :tonka: (sorry, I'm tired!)
I understand that you guys want to save money at the cost of more power wasted, but I build more for efficiency.
I want to have rock lights that are as bright as the incandescents that I can leave on for several hours without frying my battery (in case I need to crawl under the jeep and do some work or something of that nature).
Again, though, you're losing out on some functionality. My set-up allows me the option of a PWM brightness control, or a DC signal voltage level to control brightness (and, of course, there are more advantages to this), which is flippin' convenient.
Congrats on your promotion.
I don't deal with LEDs too much. I've only recently picked up an interest with them when I was assigned to a project where I use infrared LEDs and infrared receivers to communicate and build a virtual 3D grid in free space (they can also communicate several other things, but that is the neatest part).
Kinda' cool stuff!
When we have heat specific, we can and do use buck regulators.
Thing is that for high power (InGaN) parts, we can't use current control for changing intensity, only the PWMing. (A typical LED circuit would have a PNP transistor no the high side for PWM control, the LED, then current limiting resistor, then an NPN switch on the low side for on/off control. ).
Resistors are binned at a specific current, which is the only point their characteristics are guaranteed, usually. What happens is that as the current changes, the X-Y color coordinates change as well. Its especially noticeable in the whites, and we have tight specs for the color coordiantes.
But, in the production world, cost is key. one of the projects I'm doing now, I took 3 cents out per LED, which at our volumes saves us $1.7M per year, so little things add up. I think my current prorgam is using 41M LEDs per year...