search

ColibriKBO / ColibriObservatory

GNU General Public License v3.0
2 stars 0 forks source link

Blue images bad headers #8

Closed RomanAkhmetshyn closed 2 years ago

RomanAkhmetshyn commented 2 years ago

Image data starting from 2022-06-24 have incorrect time headers. Both bias and sky images have 2016-00- year and month, which causes an error during timingplots analysis. Frames in 06-22 are correct as well as some frames from 06-24, other telescopes also have correct dates. image

mjmazur commented 2 years ago

It looks like the GPS was not connected for some reason. Can you see whether the GPSLocked field in the metadata is True or not? If it's false for these images, a quick fix is to output this field to our fits files when doing the RCD to fits conversion. If it's the GPS, then we need to scan our images and look for this field and document when it's false and how often.

Mike

From: RomanAkhmetshyn @.> Sent: July 6, 2022 11:02 AM To: ColibriKBO/ColibriObservatory @.> Cc: Subscribed @.***> Subject: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8)

Image data starting from 2022-06-24 have incorrect time headers. Both bias and sky images have 2016-00- year and month, which causes an error during timingplots analysis. Frames in 06-22 are correct as well as some frames from 06-24, other telescopes also have correct dates. [image]https://user-images.githubusercontent.com/108425347/177581488-b7273b4d-d978-4e9c-82a8-28f7c06a5c73.png

— Reply to this email directly, view it on GitHubhttps://github.com/ColibriKBO/ColibriObservatory/issues/8, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYN2Z5PKF2OGKGNVIIDVSWN2FANCNFSM522DAH4Q. You are receiving this because you are subscribed to this thread.Message ID: @.***>

mjmazur commented 2 years ago

This is what it should look like.

[cid:36058e6d-20ba-4336-8868-8183710410f6]

From: RomanAkhmetshyn @.> Sent: July 6, 2022 11:02 AM To: ColibriKBO/ColibriObservatory @.> Cc: Subscribed @.***> Subject: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8)

Image data starting from 2022-06-24 have incorrect time headers. Both bias and sky images have 2016-00- year and month, which causes an error during timingplots analysis. Frames in 06-22 are correct as well as some frames from 06-24, other telescopes also have correct dates. [image]https://user-images.githubusercontent.com/108425347/177581488-b7273b4d-d978-4e9c-82a8-28f7c06a5c73.png

— Reply to this email directly, view it on GitHubhttps://github.com/ColibriKBO/ColibriObservatory/issues/8, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYN2Z5PKF2OGKGNVIIDVSWN2FANCNFSM522DAH4Q. You are receiving this because you are subscribed to this thread.Message ID: @.***>

RomanAkhmetshyn commented 2 years ago

GPSLocked is set to False in metadata

mjmazur commented 2 years ago

Okay, so that's the problem. I can see that the GPS is not attached at the moment. I'll go unplug it and plug it in again. When it's attached, the lower right corner of FLIPilot should look like,

[cid:66785ddc-c5cc-4b8f-92fb-7fb4ef6cf2e0]

Possibly the GPS has died. We don't have a exact spare, but could hack one together from some other receivers that I have here until we can replace it.

Mike

From: RomanAkhmetshyn @.> Sent: July 6, 2022 11:44 AM To: ColibriKBO/ColibriObservatory @.> Cc: Michael Mazur @.>; Comment @.> Subject: Re: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8)

GPSLocked is set to False in metadata

— Reply to this email directly, view it on GitHubhttps://github.com/ColibriKBO/ColibriObservatory/issues/8#issuecomment-1176381332, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYK2CPBIDE25NBN6LU3VSWSWRANCNFSM522DAH4Q. You are receiving this because you commented.Message ID: @.***>

RomanAkhmetshyn commented 2 years ago

It looks like the GPS was not connected for some reason. Can you see whether the GPSLocked field in the metadata is True or not? If it's false for these images, a quick fix is to output this field to our fits files when doing the RCD to fits conversion. If it's the GPS, then we need to scan our images and look for this field and document when it's false and how often. Mike … ____ From: RomanAkhmetshyn @.> Sent: July 6, 2022 11:02 AM To: ColibriKBO/ColibriObservatory @.> Cc: Subscribed @.> Subject: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8) Image data starting from 2022-06-24 have incorrect time headers. Both bias and sky images have 2016-00- year and month, which causes an error during timingplots analysis. Frames in 06-22 are correct as well as some frames from 06-24, other telescopes also have correct dates. [image]https://user-images.githubusercontent.com/108425347/177581488-b7273b4d-d978-4e9c-82a8-28f7c06a5c73.png — Reply to this email directly, view it on GitHub<#8>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYN2Z5PKF2OGKGNVIIDVSWN2FANCNFSM522DAH4Q. You are receiving this because you are subscribed to this thread.Message ID: @.>

Do you mean changing headers for rcd images after converting them to .fits?

mjmazur commented 2 years ago

I was thinking of just modifying the RCDtoFITS.py code to write the GPS status to the fits header. Off the top of my head, I'm not sure which byte location corresponds to the GPS status, though, so it won't happen right away.

Mike

From: RomanAkhmetshyn @.> Sent: July 6, 2022 1:58 PM To: ColibriKBO/ColibriObservatory @.> Cc: Michael Mazur @.>; Comment @.> Subject: Re: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8)

It looks like the GPS was not connected for some reason. Can you see whether the GPSLocked field in the metadata is True or not? If it's false for these images, a quick fix is to output this field to our fits files when doing the RCD to fits conversion. If it's the GPS, then we need to scan our images and look for this field and document when it's false and how often. Mike … ____ From: RomanAkhmetshyn @.> Sent: July 6, 2022 11:02 AM To: ColibriKBO/ColibriObservatory @.> Cc: Subscribed @.> Subject: [ColibriKBO/ColibriObservatory] Blue images bad headers (Issue #8https://github.com/ColibriKBO/ColibriObservatory/issues/8) Image data starting from 2022-06-24 have incorrect time headers. Both bias and sky images have 2016-00- year and month, which causes an error during timingplots analysis. Frames in 06-22 are correct as well as some frames from 06-24, other telescopes also have correct dates. [image]https://user-images.githubusercontent.com/108425347/177581488-b7273b4d-d978-4e9c-82a8-28f7c06a5c73.png — Reply to this email directly, view it on GitHub<#8https://github.com/ColibriKBO/ColibriObservatory/issues/8>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYN2Z5PKF2OGKGNVIIDVSWN2FANCNFSM522DAH4Q. You are receiving this because you are subscribed to this thread.Message ID: @.>

Do you mean changing headers for rcd images after converting them to .fits?

— Reply to this email directly, view it on GitHubhttps://github.com/ColibriKBO/ColibriObservatory/issues/8#issuecomment-1176517358, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABPGQYNJF2KK4I2EUWWHBOLVSXCNDANCNFSM522DAH4Q. You are receiving this because you commented.Message ID: @.***>

mjmazur commented 2 years ago

The GPS cable was loose. It has been fixed now.