detect light in a specific range with no filter

Thread Starter

Spange-Bob

Joined Feb 24, 2019
5
I am new in electronic science
is there a possibility of detect light in a specific range without using filters? (via changing the voltage of the sensor or somehow)


Thanks In advance Mrs/Mr engineers
 

pmd34

Joined Feb 22, 2014
518
Hi Bob, it depends a little how specific you want to be. Different kinds of semiconductor material are better at detecting different wavelengths and generally have a "low energy" cut off below which they will not respond, but the high energy cut off is not usually as sharp.
The most common sort of selective detector is a Red Green Blue (RGB) such as:
https://www.hamamatsu.com/jp/en/product/type/S9702/index.html

(Hamamatsu are experts in light detection, but they are the rolls royce, so look else where for affordability!!)
 

pmd34

Joined Feb 22, 2014
518
Then the answer is no then im afraid. The current / voltage output depends chiefly on the light intensity there will be some change with wavelength, but you would not be able to tell the difference between, for example; a low intensity blue light and a medium intensity red light with a standard sensor (assuming the sensor is most sensitive to blue light in that example).
 

jpanhalt

Joined Jan 18, 2008
11,088
Thanks for replay
But I don't want to use RGB sensors
just a normal sensor
From the datasheet of the adafruit sensor (TCS3472):
The 3 × 4 photodiode array is composed of red-filtered, green-filtered, blue-filtered, and clear (unfiltered) photodiodes.
While in search for unobtainium, perhaps you will provide some rationale for that specification?
 

MrChips

Joined Oct 2, 2009
22,911
You can separate the colour components of white light using refraction (prism) and diffraction (grating).
You can physically scan across the spectrum with a single optical sensor.
You can also measure colour components without having to physically scan using a multi-element linear optical array.
 

Yaakov

Joined Jan 27, 2019
1,683
Thanks for replay
But I don't want to use RGB sensors
just a normal sensor
There is no such thing as a "normal" sensor. Sensors just measure things and in that sense they are all "normal". No matter what method you choose, something will have to integrate multiple measurements because you are trying to measure something with at least three dimensions.

You don't want to use filters, so you can't use simple LDRs or the like. You don't want to use an integrated array so there is no device you can buy.

What you are asking is very similar to wanting to know the volume of an arbitrary cube but demanding that you can only measure the height of it.

It can't be done.
 

djsfantasi

Joined Apr 11, 2010
7,402
What you are asking is very similar to wanting to know the volume of an arbitrary cube but demanding that you can only measure the height of it.

It can't be done.
I get your point @Yaakov, but dislike your example

Um, not to be pedantic (but I’m going to be), by the definition of a cube H=W=D. Knowing the height, you have all the dimensions and the volume is (height)^3

Now if you said “box”... :)
 

Yaakov

Joined Jan 27, 2019
1,683
I get your point @Yaakov, but dislike your example

Um, not to be pedantic (but I’m going to be), by the definition of a cube H=W=D. Knowing the height, you have all the dimensions and the volume is (height)^3

Now if you said “box”... :)
By “arbitrary cube” I mean rectangular prism. But aside from that, my analogy is apt.
 

DickCappels

Joined Aug 21, 2008
6,972
I investigated this about 20 years ago but did not get to the experiment. From discussions with a optical physicist with whom I worked, you can vary the wavelength at which a reverse biased photodiode is most sensitive by modulating the depth of the depletion region, and that is done by changing the bias voltage. Sorry, that's all I have now.
 
Top