Building an RGB LED Transilluminator?

Question

I am attempting to build a gel transilluminator using RGB LEDs instead of UV fluorescent bulbs. Blue/Green LEDs can excite many fluorescent dyes similarly to UV without the harmful effects of UV. I would also like to make white light an option so we can image non-fluorescent gels. While I'm at it, I can include a blue only option because blue light illumination is also common.

Anyway, I thought the simplest way to do that would be a large common anode RGB LED array. Rather than trying to control each LED separately, I will wire them all up in parallel, with three switches to turn on each color all at once, so I can choose blue/green, blue, or white by turning on different combinations of switches.

I have a 12 V 5 A DC power supply. I know I need a resistor on each color channel, but is one resistor on each channel enough for what I want to do? Or will I need 3 resistors per LED? I also need to run the LEDs at a high current so that they are very bright, so "recommended" resistance levels might be too high, I'm not sure yet. I am okay with reducing the lifetime of the LEDs a little because transilluminators are only turned on for a few seconds to minutes at a time. I am very new to circuit building so I'm sure there is something I'm missing.

I have an example circuit diagram. In reality, I will likely have closer to 100 LEDs instead of 16, but the overall idea should be the same.

Answer 1

There are LED's that alreday have the PWM circuitry built in, like the ws2812b. They are available at adafruit and sparkfun in sheets and strips (and controllers are available). The problem with LED's and transilumnation is you need LED's that have the specific wavelength needed for florescence and I'd rather know the wavelengths ahead of time, and the ws2812b does not have the measured spectrum for its LED's listed in the datasheet, so you would have to measure it yourself.

Or you could roll your own with three LED's that you know the spectrum of, if you do this you'll want to use a circuit like this:

Source: https://www.powerelectronictips.com/solenoids-relays-part-1-2/

(you could switch them with a relay either before or after the LED's or a pmos before the resistor.)

Here is some info on spectra for a LED's

Source: http://www.phos-4.com/

Answer 2

If you wire in parallel, you need one resistor per LED. If the package has 3 LEDs on it, you still need 3 resistors. Usually LEDs are put in series to reduce the number of resistors needed.

If you're aiming for DNA, I'd use 405nm rather than an RGB diode. That way you can excite ethidum bromide, dapi, etc with one diode. Filtering the 405nn light will be much easier. With an RGB diode you will need a filter with a sharp cut off or possibly multiple filters depending on the fluorophore. Conversely with 405 you can just use short wavelength absorptive plastic and see the whole visible spectrum.

I built something very similar to this, but in pulsed operation with a camera. You can buy strips of LEDs that include the resistors, put a sheet of polycarbonate over them (spacer), then tape diffuser sheets over the polycarbonate to give a uniform illumination pattern.