I started designing my own LED drivers using some ICs that are specifically built for the task. I began over the summer while I was taking three summer classes, which won out over finishing this, and now that I am done with what everyone describe as the most difficult semester in the ECE department (I hope they are right, I don't want to see anything worse than that), I should finally have time to get back to this and finish the project up.
I am going to run a few different channels, basically one for each color/spectrum. This is going to be a pico light, so the blue/near ultra violet and white channel are on a 750mA chip. Red/orange, and green will be on a 500mA chip, and the UV will be on a 350mA chip (won't run anywhere near that. I will limit it to far less after some experimentation, and will most likely end up eliminating it, but I wanted to atleast test it out). I am debating splitting up the blue/near UV as well as the red/orange channels as this is pretty inexpensive to build. and I am not sure how everything will balance out with them on the same channel driven at the same current.
I don't have the spec sheets with me and I don't remember the exact chips and dies I am using (no stars, I will most likely be ordering a MCPCB), but the chips less than $1 each and they will allow me to dim with a potentiometer. I also grabbed some that will allow me to dim via PWM, which will have a more advanced controller for version 2 (sunrise, sunset, storms, and a few other features that may be disclosed later). Those are just over $1 each.
I have a couple of questions.
Should I run the other colors (other than UV) at the same 750mA? The LEDs, other than the blue and white, will be run without optics as they are more to balance the overall color and will have fewer dies used in the fixture, so I want more spread from them.
Are the UV worth experimenting with? Theoretically with the hula shift (yes, that's a technical term) pigments, the UV should allow for more pop in the colors as they shift the spectrum to get more usable light to the zooxanthellia and reject the remaining light, typically in the form of fluorescing colors, which should allow me to get those brighter colors without going into the washed out blue/blacklight range. Does anybody have experience with UV? Opinions?
What are people using to attach optics to their LEDs? I've seen silicone mentioned, but I am not sure if that is the ideal option. I am currently undecided if this will have a lens over it. If it does, the lens will retain the optics.
Has anybody else built their own drivers from scratch? What would you have done differently? What are some features/options you wish you added?
I haven't delved too far into any of the DIY LED threads, so any input on color blending/balancing, and different ideas on color combinations would be appreciated, as well as technical resources.
I'm partial to a bright white appearance (natural white sunlight), but like the effect that the blue and near UV spectrum adds. I am trying to accomplish a good balance that doesn't show the blues, but gives the fluorescing colors without it looking like a 70's flashback.
I am going to run a few different channels, basically one for each color/spectrum. This is going to be a pico light, so the blue/near ultra violet and white channel are on a 750mA chip. Red/orange, and green will be on a 500mA chip, and the UV will be on a 350mA chip (won't run anywhere near that. I will limit it to far less after some experimentation, and will most likely end up eliminating it, but I wanted to atleast test it out). I am debating splitting up the blue/near UV as well as the red/orange channels as this is pretty inexpensive to build. and I am not sure how everything will balance out with them on the same channel driven at the same current.
I don't have the spec sheets with me and I don't remember the exact chips and dies I am using (no stars, I will most likely be ordering a MCPCB), but the chips less than $1 each and they will allow me to dim with a potentiometer. I also grabbed some that will allow me to dim via PWM, which will have a more advanced controller for version 2 (sunrise, sunset, storms, and a few other features that may be disclosed later). Those are just over $1 each.
I have a couple of questions.
Should I run the other colors (other than UV) at the same 750mA? The LEDs, other than the blue and white, will be run without optics as they are more to balance the overall color and will have fewer dies used in the fixture, so I want more spread from them.
Are the UV worth experimenting with? Theoretically with the hula shift (yes, that's a technical term) pigments, the UV should allow for more pop in the colors as they shift the spectrum to get more usable light to the zooxanthellia and reject the remaining light, typically in the form of fluorescing colors, which should allow me to get those brighter colors without going into the washed out blue/blacklight range. Does anybody have experience with UV? Opinions?
What are people using to attach optics to their LEDs? I've seen silicone mentioned, but I am not sure if that is the ideal option. I am currently undecided if this will have a lens over it. If it does, the lens will retain the optics.
Has anybody else built their own drivers from scratch? What would you have done differently? What are some features/options you wish you added?
I haven't delved too far into any of the DIY LED threads, so any input on color blending/balancing, and different ideas on color combinations would be appreciated, as well as technical resources.
I'm partial to a bright white appearance (natural white sunlight), but like the effect that the blue and near UV spectrum adds. I am trying to accomplish a good balance that doesn't show the blues, but gives the fluorescing colors without it looking like a 70's flashback.
