Correct interpolation and dark pixel offset

[krisgry]

Hi,
I have tried to verify the calibration and dark-pixel offset in ramses_calibrate.py by testing with a RAMSES sensor in an optical lab with a mercury test tube. Below is a plot comparing the original pytrios code by a modification done by my former colleague Artur Zolich. The vertical lines are the expected peaks of mercury, at [365, 405, 435, 546] nm. As you can see, the code by Zolich fits better.

Let me know if you would like me to send you the numerical values used in the plot, and/or the raw data from the sensor.

To be clear: this fix was made by Artur by comparing the output from MSDA with pytrios. I know he has been in contact with you on that, via email. I have also been in contact with Triosm, who confirm that there is an error in the indexing in the interpolation in the manual (but frustratingly became radio-silent after that), and one of our research partners who also confirm having problems with this.

I hope that this PR can fix the problem, and clear the confusion.

Regards,
Kristoffer

[StefanSimis]

Hi and thanks for the PR. This is a long standing issue and it's great to know you have had some confirmation from the manufacturer. The reason this wasn't already implemented is that I have not yet been able to check how the instrument calibration laboratories interpret the indexing. Fixing the issue at one end but not the other could make things worse. I have acquired documentation from Zeiss that, along with these experiments, should get us closer to closure.

…

On Fri, 26 Sept 2025, 11:48 Kristoffer Gryte, ***@***.***> wrote: Hi, I have tried to verify the calibration and dark-pixel offset in ramses_calibrate.py by testing with a RAMSES sensor in an optical lab with a mercury test tube. Below is a plot comparing the original pytrios code by a modification done by my former colleague Artur Zolich. The vertical lines are the expected peaks of mercury, at [365, 405, 435, 546] nm. As you can see, the code by Zolich fits better. image.png (view on web) <https://github.com/user-attachments/assets/cd2ab7fd-7ad3-432b-b730-3784009b1d35> Let me know if you would like me to send you the numerical values used in the plot, and/or the raw data from the sensor. To be clear: this fix was made by Artur by comparing the output from MSDA with pytrios. I know he has been in contact with you on that, via email. I have also been in contact with Triosm, who confirm that there is an error in the indexing in the interpolation in the manual (but frustratingly became radio-silent after that), and one of our research partners who also confirm having problems with this. I hope that this PR can fix the problem, and clear the confusion. Regards, Kristoffer ------------------------------ You can view, comment on, or merge this pull request online at: #16 Commit Summary - 7b94b07 <7b94b07> Correct interpolation and dark pixel offset File Changes (1 file <https://github.com/StefanSimis/PyTrios/pull/16/files>) - *M* ramses_calibrate.py <https://github.com/StefanSimis/PyTrios/pull/16/files#diff-1fa705d0105ef5b083aa60a5f14ba52bcb0a4d321abadbec608029fb91be4bff> (8) Patch Links: - https://github.com/StefanSimis/PyTrios/pull/16.patch - https://github.com/StefanSimis/PyTrios/pull/16.diff — Reply to this email directly, view it on GitHub <#16>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABCMYXGC62V3AKR2SEGHDBL3UUKYNAVCNFSM6AAAAACHSN2ZT6VHI2DSMVQWIX3LMV43ASLTON2WKOZTGQ2TMOJZHAZDENA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[krisgry]

For completeness: below is how another repo that is dealing with calibration of trios ramses data in python handles the interpolation:

https://github.com/hbliu104/PyRamses/blob/154df21a2dac0041f3585cffdc6ab0d706001c52/PyRamses.py#L336

Later in the code, the calibrated spectrum is accessed by discarding the first element (cal[1:])
https://github.com/hbliu104/PyRamses/blob/154df21a2dac0041f3585cffdc6ab0d706001c52/PyRamses.py#L670

[StefanSimis]

Hi, could you email me on my pml.ac.uk address so we can share a few bits of documentation? Much appreciated!

…

On Fri, 3 Oct 2025, 13:36 Kristoffer Gryte, ***@***.***> wrote: *krisgry* left a comment (StefanSimis/PyTrios#16) <#16 (comment)> For completeness: below is how another repo that is dealing with calibration of trios ramses data in python handles the interpolation: https://github.com/hbliu104/PyRamses/blob/154df21a2dac0041f3585cffdc6ab0d706001c52/PyRamses.py#L336 Later in the code, the calibrated spectrum is accessed by discarding the first element (cal[1:]) https://github.com/hbliu104/PyRamses/blob/154df21a2dac0041f3585cffdc6ab0d706001c52/PyRamses.py#L670 — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABCMYXB7WYAHNYAJ5GVOI3D3VZUVTAVCNFSM6AAAAACHSN2ZT6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGNRVGUZDAOJWGE> . You are receiving this because you commented.Message ID: ***@***.***>