From: Phil Hobbs
Subject: Re: Anyone Need A Laser Power Meter?
Date: Tue, 08 Oct 2002 15:12:13 -0400
Organization: IBM T. J. Watson Research Center
References: <3DA2D172.F3230CD4@webaccess.net> <firstname.lastname@example.org>
NNTP-Posting-Date: 8 Oct 2002 19:12:16 GMT
X-Mailer: Mozilla 4.61 [en] (OS/2; U)
Ed Edmondson wrote:
> Surely, one of these detectors would do what I want to do? What I would really
> like is something with the following specs.
> Ranges: 9.99 uW, 99.9 uW, 999 uW, 9.99 mW, 99.9 mW, 999 mW, and 9.99 W.
> Spectral response from 254 nm to 1,100 nm
> Accuracy 5%
Hmm. That's quite a specification for a homebrew project, especially
for a beginner at design. I was thinking along the lines of a DVM
measuring the photocurrent, and a calibration chart or calculator
program to convert amps to watts.
1. The wavelength meter part
For 5% accuracy for the wavelength meter, you can use your eyeball over
much of the visible range. You could do something over a narrow range
with a coarse grating and a lateral effect cell, but it hardly seems
worth the trouble. To do a good job, you need either of two things.
The first possibility is a grating spectrometer with either
cross-dispersion or a series of filters to take care of the grating
order overlap, which happens any time your range is more than 2:1. You
might be able to make this yourself, but will have a hard time
automating it. The second possibility is a motorized Michelson
interferometer and a He-Ne for fringe counting, which is easier to
automate but much harder to build.
You could potentially use two crossed gratings and a webcam, plus a
bunch of software, but you won't find an inexpensive camera that works
over this wavelength range.
2. The spectral range part
You're going to have real trouble getting to 1.1 um with a silicon
photodetector of any description, and most diodes have responses that
fall off very badly in the UV due to the thickness of their epitaxial
3. The power range part
The high power end is going to be very hard to do even to 5% accuracy
due to the requirement for a high power neutral density filter of some
4. The controller part
All that calibration and digital display is going to require a
microcontroller and a bunch of flash memory, which is a whole other
Overall, I think getting this project to work is probably way beyond the
capability of someone who doesn't know how to design it, sorry. I'd try
the DVM approach, and sneak up on the fancy version by stages.