1
$\begingroup$

I'm working on an experiment that have visible spectrum light source (RGB LED at e.g. 495–570 nm wavelength), holder with sample, narrow bandpass filter (eg: CWL:520nm, FWHM:5nm) and a photodiode to detect the signal. As light source is RGB LED I can easily change it's wavelength but simultaneously I also have to change bandpass filter (different filter for different wavelength).

Now problem is every time I have to replace filter according to light source wavelength which is not right way and these filters are costly too.

Is there any alternative for these bandpass filter?

$\endgroup$
2
  • $\begingroup$ Any device that "passes a particular wavelength of light" would be, by definition, a bandpass filter. Are you asking if there are alternate technologies to implement a bandpass filter? In that case, tell us what technology your current filter uses, otherwise we don't know what's different from it. $\endgroup$
    – The Photon
    Commented Dec 21, 2017 at 17:24
  • $\begingroup$ If you're trying to solve a particular problem, tell us clearly what the problem is. We only know as much about it as you tell us. For example, could you use a tunable laser rather than an LED with a bandpass filter? Without knowing the actual problem you are trying to solve, we can't know whether this is a reasonable suggestion or not. $\endgroup$
    – The Photon
    Commented Dec 21, 2017 at 17:26

3 Answers 3

1
$\begingroup$

You could use a grating. The different wavelengths of light would leave the grating under different angles. By changing the position of the diode you can thus measure different wavelengths.

You are effectively creating a spectrometer.

$\endgroup$
5
  • $\begingroup$ Problem with grating is I'm not able to get enough light at the detection end. Even if I do this I also need to replace photo-diode with other detection thing like photo-diode array. Is it possible to get 10nm to 20nm wavelength step difference in grating. $\endgroup$ Commented Dec 21, 2017 at 14:55
  • $\begingroup$ What I'm asking is "replacement of bandpass filter" $\endgroup$ Commented Dec 21, 2017 at 14:59
  • $\begingroup$ @KapilSinghRawat, and this answer gives you exactly that. If you have other requirements, you need to edit the question to state them. (But then this starts to sound like an engineering question rather than physics) $\endgroup$
    – The Photon
    Commented Dec 21, 2017 at 17:25
  • $\begingroup$ @KapilSinghRawat, You can open up the slits on the grating, (more light), pick a grating with less lines/mm, and then change the grating angle for different wavelengths.... A spectrometer run this ways is called a monochromator. $\endgroup$ Commented Dec 21, 2017 at 18:51
  • $\begingroup$ @GeorgeHerold thank you so much for this details I will try this method too. $\endgroup$ Commented Dec 22, 2017 at 5:29
0
$\begingroup$

One can get tunable bandpass filters where the filter frequencies are set by the tilt. I think I've used ones in this product line before:

https://www.semrock.com/Data/Sites/1/semrockpdfs/whitepaper_versachrometunablebandpassfilter.pdf .

They are pretty neat.

But, as usual with these things, the right solution will depend on details that only you know, like acceptable bandwidths and losses and prices, so you shouldn't expect a very useful answer in a medium like this one.

$\endgroup$
0
0
$\begingroup$

Firstly an RGB led is not really a variable wavelength source. It has 3 broad color LEDs and varies the brightness of each to fool your eye into seeing a continually varying color.

If you are only using 520nm in your detector you would do better with a single brighter source at 520nm. If your wavelength requirements aren't strict you can get 532nm laser diodes very easily, 520nm diode lasers are available but rarer/more expensive. Most green LEDs peak at 568nm.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.