Hi, with apologies for doing this as a new post; the repo for the new Teensy 4 driven lccd sensor board is up. There is a lot of information and some very interesting results.
github.com
One of the main points is achieving linear reproducible response in a linear CCD sensor. There are a lot of these from commercial sources that show up in the scientific literature all the time. So, guess what. Some of them are actually not reliable, The README for the repo explains why.
In a short version, CCD signals are different, they never met Nyquist. The large dV/dt in an image or spectrum with sharp lines gives all sorts of interesting behaviors.
What is not described in the repo yet, is the use of the FlexPWM to accurately time the gates for the sensor. This an important part of the design, along with the gate drivers. Will probably add it to the readme some time next week.
The point there is that the Teensy is an important part of why this works so well. We plan to puiblish in a journal, and it seems likely to get attention since it ties into an ongoing scandal in reproduicibility. But we need more results from a wider range of commercial instruments. If you have one of those and would like to contribute some data, please contact me. We are especially interested in new models from the big companies.
So, all in all, it might be an interesting read. Do let me know what you think, in a constructive way please.
GitHub - drmcnelson/TCD1304-Sensor-Device-with-Linear-Response-and-16-Bit-Differential-ADC: TCD1304 device with kicad, firmware, library and user app, designed for highly reproducible, linear response for precision photo-/radiometric spectrometry and
TCD1304 device with kicad, firmware, library and user app, designed for highly reproducible, linear response for precision photo-/radiometric spectrometry and holographic imaging. - drmcnelson/TCD1...
github.com
One of the main points is achieving linear reproducible response in a linear CCD sensor. There are a lot of these from commercial sources that show up in the scientific literature all the time. So, guess what. Some of them are actually not reliable, The README for the repo explains why.
In a short version, CCD signals are different, they never met Nyquist. The large dV/dt in an image or spectrum with sharp lines gives all sorts of interesting behaviors.
What is not described in the repo yet, is the use of the FlexPWM to accurately time the gates for the sensor. This an important part of the design, along with the gate drivers. Will probably add it to the readme some time next week.
The point there is that the Teensy is an important part of why this works so well. We plan to puiblish in a journal, and it seems likely to get attention since it ties into an ongoing scandal in reproduicibility. But we need more results from a wider range of commercial instruments. If you have one of those and would like to contribute some data, please contact me. We are especially interested in new models from the big companies.
So, all in all, it might be an interesting read. Do let me know what you think, in a constructive way please.