andrewb
0
- Joined
- Sep 7, 2011
- Messages
- 144
- Points
- 28
(25 Dec 2013 - Added new plots for ML101U29-25, S06J, PL450B, and ML501P73 diodes.)
Hi all,
(Note this is my first 'review' here, though I've been mostly a lurker for the past couple of years. Please be gentle.)
This weekend I was able to accomplish something I've wanted to do for a while, which was to build an automated laser diode measurement system, in order to automatically measure/generate PIV curves for different diodes. I accomplished this using a programmable bench power supply (Tektronix PS2521G), and programmable benchtop DMM (Fluke 45), both of which are controlled via GPIB interfaces, using MATLAB.
The general idea is simply to sweep the drive current from the PSU, and measure the laser output power using the DMM + Ophir head as the LPM, exactly as you would normally do it if you were doing it by hand. Except in this case the data can be read from the instruments directly into MATLAB, and you can capture a lot of data points automatically and quickly, with arbitrary resolution, in order to generate more detailed curves. Also, once the data is in MATLAB, it's relatively easy to manipulate, analyze, compare, and plot many sets of data, with relatively little effort.
My setup is shown below. Basically the PSU and DMM are controlled from the laptop, using a USB-to-GPIB adapter from National Instruments. All code is written in MATLAB to control the instruments and plot the data. The Ophir head is powered by two 9V batteries, and is connected to the DMM, which serves as the LPM.
My motivation for building this system was not just to characterize diodes, but to evaluate lens performance as well, as I've casually noticed the differences between various lenses in different builds with my LPM, however I wanted to quantify the differences to determine how much better or worse a given lens will perform with a given diode, as there did not seem to be a strictly consistent pattern.
I know a lot has been discussed and presented on this topic already, so this isn't exactly groundbreaking stuff, however I don't think anyone has done the measurements exactly as I have (i.e., automated), and presented the data for these diode/lens combinations, so I figure I'd share what I have with the community for what it's worth, as it was a fun exercise, and weekend project.
I was able to measure three different diodes (a PL520, 9mm 445, and LPC-826), along with five different lenses (G9, G2, 3-element 455 AR coated, 3-element 650 AR coated, and Aixiz acrylic), all of which I obtained from DTR.
The only combination that couldn't be run was the 9mm + acrylic (as it would melt), so there are 14 curves total. Each curve contains about 50-60 points, and took about 2 minutes to capture. (Each point took about two seconds, giving the LPM enough time to stabilize.) 5-minute cooldown breaks were taken between runs, after swapping out lenses. The results are shown below for the 3 separate diodes.
Diode #1: PL520
Observations: At the peak power point (~400mA), the G9 outperformed the G2 by 5%, and beat the two 3-elements by approximately 45% (both of which performed similarly if not slightly better than the acrylic).
Diode #2: 9mm 445
Observations: At peak power, G9 beat G2 by 6%, and beat the 3-elements by approximately 55%. The 445-coated 3-element was only slightly better than the 650-coated 3-element.
Also note that even though G9 beat the G2 by only 6%, the 200mW difference was enough to make the difference between a 2.8W build, and a 3W+ build.
Diode #3: LPC-826
Observations: At peak power, G9 beat G2 by 14%, and beat the 3-element 445 by 49%. The 3-element 650 beat the 3-element 445 by 13%, however the acrylic outperformed the 3-element 650 by 3%.
Diode #4: ML101U29-25
Observations: G9 beats G2 by 9%, and beats 3-element by 23%.
Diode #5: S06J
Observations: G9 beats G2 by 7%, and beats 3-element 445 by 23%. 3-element 445 and acrylic are approximately the same, and noticeably better than 3-element 650.
Diode #6: PL450B
Observations: G9 beats G2 by 10%, and beats 3-element 445 by 27%. Similar to S06J above, 3-element 445 and acrylic are similar, and both better than 3-element 650.
Diode #7: ML501P73
Observations: G9 beats G2 by 5%, and beats 3-element 445 by 34%. Unlike single-mode reds above, acrylic does much worse than 3-element 445, being almost identical to 3-element 650.
Summary: Based solely on output power (and measured under peak power conditions), lens performance varies according to lens AND diode, with acrylic being superior to 3-element glass for 650nm, but not 520nm, where acrylic and 3-element are approximately equal. G9 typically outperforms G2 by anywhere from 5-14% (with the biggest difference being at 650nm), and can add more than 50% compared to 3-element glass.
Comments welcome. Thanks for reading!
-Andy
Hi all,
(Note this is my first 'review' here, though I've been mostly a lurker for the past couple of years. Please be gentle.)
This weekend I was able to accomplish something I've wanted to do for a while, which was to build an automated laser diode measurement system, in order to automatically measure/generate PIV curves for different diodes. I accomplished this using a programmable bench power supply (Tektronix PS2521G), and programmable benchtop DMM (Fluke 45), both of which are controlled via GPIB interfaces, using MATLAB.
The general idea is simply to sweep the drive current from the PSU, and measure the laser output power using the DMM + Ophir head as the LPM, exactly as you would normally do it if you were doing it by hand. Except in this case the data can be read from the instruments directly into MATLAB, and you can capture a lot of data points automatically and quickly, with arbitrary resolution, in order to generate more detailed curves. Also, once the data is in MATLAB, it's relatively easy to manipulate, analyze, compare, and plot many sets of data, with relatively little effort.
My setup is shown below. Basically the PSU and DMM are controlled from the laptop, using a USB-to-GPIB adapter from National Instruments. All code is written in MATLAB to control the instruments and plot the data. The Ophir head is powered by two 9V batteries, and is connected to the DMM, which serves as the LPM.
My motivation for building this system was not just to characterize diodes, but to evaluate lens performance as well, as I've casually noticed the differences between various lenses in different builds with my LPM, however I wanted to quantify the differences to determine how much better or worse a given lens will perform with a given diode, as there did not seem to be a strictly consistent pattern.
I know a lot has been discussed and presented on this topic already, so this isn't exactly groundbreaking stuff, however I don't think anyone has done the measurements exactly as I have (i.e., automated), and presented the data for these diode/lens combinations, so I figure I'd share what I have with the community for what it's worth, as it was a fun exercise, and weekend project.
I was able to measure three different diodes (a PL520, 9mm 445, and LPC-826), along with five different lenses (G9, G2, 3-element 455 AR coated, 3-element 650 AR coated, and Aixiz acrylic), all of which I obtained from DTR.
The only combination that couldn't be run was the 9mm + acrylic (as it would melt), so there are 14 curves total. Each curve contains about 50-60 points, and took about 2 minutes to capture. (Each point took about two seconds, giving the LPM enough time to stabilize.) 5-minute cooldown breaks were taken between runs, after swapping out lenses. The results are shown below for the 3 separate diodes.
Diode #1: PL520
Observations: At the peak power point (~400mA), the G9 outperformed the G2 by 5%, and beat the two 3-elements by approximately 45% (both of which performed similarly if not slightly better than the acrylic).
Diode #2: 9mm 445
Observations: At peak power, G9 beat G2 by 6%, and beat the 3-elements by approximately 55%. The 445-coated 3-element was only slightly better than the 650-coated 3-element.
Also note that even though G9 beat the G2 by only 6%, the 200mW difference was enough to make the difference between a 2.8W build, and a 3W+ build.
Diode #3: LPC-826
Observations: At peak power, G9 beat G2 by 14%, and beat the 3-element 445 by 49%. The 3-element 650 beat the 3-element 445 by 13%, however the acrylic outperformed the 3-element 650 by 3%.
Diode #4: ML101U29-25
Observations: G9 beats G2 by 9%, and beats 3-element by 23%.
Diode #5: S06J
Observations: G9 beats G2 by 7%, and beats 3-element 445 by 23%. 3-element 445 and acrylic are approximately the same, and noticeably better than 3-element 650.
Diode #6: PL450B
Observations: G9 beats G2 by 10%, and beats 3-element 445 by 27%. Similar to S06J above, 3-element 445 and acrylic are similar, and both better than 3-element 650.
Diode #7: ML501P73
Observations: G9 beats G2 by 5%, and beats 3-element 445 by 34%. Unlike single-mode reds above, acrylic does much worse than 3-element 445, being almost identical to 3-element 650.
Summary: Based solely on output power (and measured under peak power conditions), lens performance varies according to lens AND diode, with acrylic being superior to 3-element glass for 650nm, but not 520nm, where acrylic and 3-element are approximately equal. G9 typically outperforms G2 by anywhere from 5-14% (with the biggest difference being at 650nm), and can add more than 50% compared to 3-element glass.
Comments welcome. Thanks for reading!
-Andy
Last edited:
