Spectral filtering of RGB components -

[rpswift]

My application involves narrow-band notch filtering of primary red, blue, and green wavelengths simultaneously from a white-light source. Will this yield greyscale output, black, or white? What if the light source is a RGB display? I understand that this may be a naive question - but it's outside my area and my E-M & Light course was long ago... Thanks!

 

[Andy Resnick]

I don't understand your setup, exactly... for example, what's the detector(s)?

 

[rpswift]

The detectors would be the eyes - sorry for the omission. This is still at the proof-of-concept stage, but we envision eyeglasses, goggles, or some other lense for this heads-up display. We want to filter the three primary colors (if possible) so that only the other visible regions can be "seen".

 

[Andy Resnick]

I'm still having trouble- you are using *notch* filters to selectively tune the output of a broadband source?

How wide (spectrally) are the filters?

 

[rpswift]

Thanks, Andy, for your repsonse and apologies for my late reply.
Correct - notch filters with FWHM tbd - but say 20 nm or so; this is both to tune the broadband source (we don't care about any spectral frequencies outside the range of the human eye) and in a second application to filter RGB from common computer displays or LEDs. We are looking at unique/proprietary organometallic compounds with suitable optical properties.
So - if we apply the notch filter to RGB components of a white-light source, I expect that we'd see the range of complimentary colors? Similarly - it seems intuitive that the filter(s) to red, green, and blue LEDs, we would not see any visible light (depending upon the true spectral purity of the LED emissions?
One application may be in quality control of LED and CRT displays.
Thanks!

 

[Andy Resnick]

Interesting application...

If you applied the notch filters to a broadband source, I suspect it would still look white- the spectral response of your cones is quite broad.

I think using them as quality control elements for a tricolor display is a *very* intriguing idea. I don't know the spectral characteristics of the pixel elements, tho.