Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers



DIY spectrometer and 633nm diode

RA_pierce

Well-known member
Joined
Sep 16, 2007
Messages
3,569
Points
113
This weekend I decided I would measure the wavelengths of a few lasers I have.
Using a diffraction grating, some scrap plywood, and spare hardware, I assembled a simple setup for measuring the wavelength of any visible laser.

Construction:
A 1.15 meter length of plywood was drilled to hold 4 right-angle brackets and two machine screws between each pair of brackets.
The first pair of brackets will hold the diffraction grating by using magnets to clamp the grating to the brackets.
The second pair of brackets is superfluous but I installed these in case I wanted to place material at the other end of the spectrometer.
The machine screws are threaded into the center line of the wood base. I cut through the screw heads with a Dremel cutting wheel so that they would accept a small washer, which was adhered to the screw head slit with epoxy. The device was primed and then painted black to avoid intense reflections from metal and light-colored wood surfaces.
The washers act as eyelets which guide alignment of the laser so that very consistent results may be obtained. The diameter of the hole through the washer is about 7mm.
The first set of brackets were placed so that the grating would be approximately 1 meter from the end of the base.

50896653621_1215b2d189_c.jpg


Operation:
A laser at one end nearest the diffraction grating is aimed through both eyelets.
The spectrometer is placed against a beam-stop (a sheet of carboard) at the far end.
The distance between the 0-order and 1st-order maxima is marked against the beam-stop with a marker.
This distance is then measured and recorded.

Calibration:
I used 3 solid state lasers of known wavelength to perform calibration.
The variables required to record accurate measurements include: distance from grating to beam-stop, distance between maxima, and the line spacing of the diffraction grating.
The measurement of the device and the distance between maxima can be measured easily. However, the line spacing is unknown, despite the specification given on the grating (1000 lines/mm).
Calibrating the device against 3 known lasers allows me to determine the empirical line spacing for higher accuracy in test measurements.
I used solid state 532 nm, 561 nm, and 589 nm lasers to calibrate the lasers because these are the only fully assembled solid state wavelengths I own.
I chose solid state lasers as the calibration lasers because their emission wavelengths are less likely to vary from design specification than diode lasers. Some diode lasers, especially those in the red spectrum, can vary greatly in their emission spectra with temperature.
To calibrate, I determined the distance from the grating to the beam-stop and measured the distance between maxima.
With these two measurements, I calculated the angle at which the first order maximum deviates from the grating.
Using the design wavelength of each solid-state laser, I computed the empirical line spacing of the diffraction grating.
For test measurements, I used the empirical line spacing to compute the wavelength.

dsin(theta) = m(lambda)

50895940378_2cd53f7d58_c.jpg


Results:
The calibration produced an average empirical line spacing of 1004 lines per millimeter. The variation between calibration attempts was very little - about 0.15% of the design specification for the grating.
Check the table for the results of the test measurements.
I examined most of my lasers, excluding a "505 nm" diode because it was not fully assembled into a portable format.
I found that three lasers measured at exactly their design wavelengths. These include the Sharp 185 mW 638 nm diode, a beautiful 495 nm diode (Sharp?), and a Ushio 633 nm diode. The 633 is one I've sought for several years. The difference between design wavelength and measured wavelength for the other lasers varied from 2-5 nm.

50895940993_ebcd56b97b_c.jpg


About the 633 nm laser:
This diode produces the best beam quality I've seen from any laser diode.
Many red laser diodes in this power range, including the Sharp 185 mW diode, and many of the 650-660 nm DVD burner diodes produce a line-artifact through the beam. Many of the blue and green diodes produce rectangular artifacts that are very distracting from the main beam. This diode produces neither artifact type. The beam is slightly elliptical. This laser produces the best beam characteristics with a high-quality short focal-length aspheric lens (AR coated for red). The G-8 lens produces strong diffraction patterns and alignment is poor, resulting in a distorted profile.
This diode is designed for 100 mW output, which it achieves at a low operating current of about 170 mA. The high efficiency of this diode, coupled with its low wavelength, makes it the most efficient red diode - in terms of brightness.
This diode at 95 mW is brighter than 300 mW of 665 nm and nearly as bright as 130 mW of 638 nm.
The color difference between 633 and 665 nm is obvious. The difference between 633 and 638 nm is subtle but noticeable. The orange-tinted red is exceptional. Those of you with 632 nm HeNe lasers will know, but HeNe lasers cannot achieve 100 mW except with a very large tube and extremely high power consumption.
I temporarily installed the 633 nm diode into a chrome pen. Later I will produce something special to house this diode.
Below, the three red wavelengths I own are shown. Can you tell which is 633 nm? The camera doesn't show the difference quite as well as my eye.

50896627156_a6c36d68b8_c.jpg
 
Last edited:



CurtisOliver

Well-known member
LPF Site Supporter
Joined
Jun 12, 2015
Messages
7,276
Points
113
Based on the pictures varying orange tints, they look like they are in order from left to right. 633,638 and 660.
Not easy to tell from a single photo however.

Do you own a HeNe as they are by far the best calibrators? Solid states are a close second to gas lasers but can still have some variation, but not in the same way as diodes.
 

RA_pierce

Well-known member
Joined
Sep 16, 2007
Messages
3,569
Points
113
The center pen is the 633?
The one on the left is. The 638 in the middle is slightly brighter than the other two, which increases its exposure in the photo, making it appear "lighter."
Based on the pictures varying orange tints, they look like they are in order from left to right. 633,638 and 660.
Not easy to tell from a single photo however.

Do you own a HeNe as they are by far the best calibrators? Solid states are a close second to gas lasers but can still have some variation, but not in the same way as diodes.
No, I don't have a HeNe. One of these days I'll get one.

I forgot to mention in the first post that the calibration laser with the least reliable measurement was the 561. I noticed that the m1 of this laser was more of a short "line" than a dot, which indicates to me that the line-width of this laser is much greater than any of the other lasers I own. The band was about 8 mm long, which translates to a ~4 nm band. I used the center of this band to compute the line spacing. This may not translate to the peak intensity wavelength of the laser but I figured the precision of this DIY spectrometer is not high enough that it would matter much anyway.
 

Giannis_TDM

Well-known member
Joined
Apr 27, 2019
Messages
665
Points
93
The one on the left is. The 638 in the middle is slightly brighter than the other two, which increases its exposure in the photo, making it appear "lighter."

No, I don't have a HeNe. One of these days I'll get one.

I forgot to mention in the first post that the calibration laser with the least reliable measurement was the 561. I noticed that the m1 of this laser was more of a short "line" than a dot, which indicates to me that the line-width of this laser is much greater than any of the other lasers I own. The band was about 8 mm long, which translates to a ~4 nm band. I used the center of this band to compute the line spacing. This may not translate to the peak intensity wavelength of the laser but I figured the precision of this DIY spectrometer is not high enough that it would matter much anyway.
I should do the same with my hene, got a lab-grade 1000lines per mm grating and a 594 hene...
 




Top