×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Operational issues with DIY flat panel - It works, sort of...

  • Posts: 67
  • Thank you received: 6
Let me preface this by pointing out my understanding of how this should work... When an exposure sequence of flats (with various filters) is initiated, the flat panel illumination is set to the preset brightness level for the filter currently in use and the panel turned on before the exposures commence. With a desired ADU level set, test exposures are made to determine the correct exposure, then the required number of flat frames taken. These steps are repeated until all the flats for the various filters are made.  (Not sure whether to expect the flat panel to be turned off between each filter and then back on after the brightness is set or just left on until the entire sequence is finished...)

I am building a flats panel using LEDs controlled by an Arduino using the Alnitak Generic Commands (Rev 44). It is just an illumination panel without being a "flip" cover, a "Flat-Man" emulation. The Arduino program (based upon several previous designs published on the internet) appears to respond appropriately to the indi_flipflat driver when operated manually via the Main Control tab of the "Flip Man" tab in the Indi Control Panel window.  On/Off and brightness changes happen as expected. Filter brightness presets are reflected correctly in the log file.When a test exposure sequence is run (w/ CCD, filter wheel and telescope simulators along with the infi_flipflat driver) the debug files indicate filter changes initiate the appropriate brightness changes (according to the presets). Polling (1000ms setting) occurs with the appropriate responses being returned and the exposures are made after initial test exposures to achieve the selected ADU setting (set in the Calibration tab of the Capture Module). <strong>BUT</strong> the light panel is never turned on or off during the sequence.

I have attached an edited copy of the debug log file. I have added some notes to it and cut out a lot of non-relevant entries.

What am I missing here? 
 

File Attachment:

File Name: Logcapture...9-18.txt
File Size:38 KB
Last edit: 2 years 1 month ago by David Swinnard. Reason: spelling
2 years 6 months ago #75661
Attachments:

Please Log in or Create an account to join the conversation.

Thank you for the report, turns out to be a bug in Ekos itself. I submitted an MR for a fix: invent.kde.org/education/kstars/-/merge_requests/434
The following user(s) said Thank You: David Swinnard
2 years 6 months ago #75667

Please Log in or Create an account to join the conversation.

  • Posts: 67
  • Thank you received: 6
Jasem, I have some more information. Using a Kubuntu install with Kstars 3.5.6 beta Build: 2021-09-28T22:15:36Z. I get different operational results from the 3.5.5 trial in the first post.

As you can see from the attached (edited for length) log in each of the LRGB flat frame sequences (of 5 exposures with a 1 second starting exposure) I get the first (1 sec) exposure saved, then a couple of "test" frames to determine the ADU value (set 28000 as target) which are NOT saved, then the 4 remaining "correct" exposures are saved. So for each 5 frame flat sequence I get one frame with the starting exposure and four "correct" exposure frames saved.

In version 3.5.5 all the saved frames were of the "correct" exposure.Additionally, the exposure light is still not being turned on or off. The lamp responds to manual control from Indi as before.

Note: in the log file attached I only included the L exposure sequence and not the RGB ones as they were identical.

 

File Attachment:

File Name: Flattest3....4-31.txt
File Size:33 KB
2 years 5 months ago #76147
Attachments:

Please Log in or Create an account to join the conversation.

Thank you for the report, I found the issue and submitted a fix. It's in GIT and should be in the next nightly.
The following user(s) said Thank You: David Swinnard
2 years 5 months ago #76152

Please Log in or Create an account to join the conversation.

  • Posts: 67
  • Thank you received: 6
I have tried the new nightly build and found it does in fact properly save the "correct" ADU flat frame exposures.

Flat panel brightness is still only set just once during the setup of the driver (to value for filter slot 1) and is not changed during a sequence where each filter calls for a different brightness setting.

Is the flat panel supposed to be turned off between different filter sequences?  It turned on at the start of the sequence for a filter, but not off at the end, resulting in a flat panel that stays on until manually switched off.

Thanks for these swift updates!
 

File Attachment:

File Name: Flattest3....9-41.txt
File Size:33 KB
Last edit: 2 years 5 months ago by David Swinnard. Reason: forgot the file...
2 years 5 months ago #76174
Attachments:

Please Log in or Create an account to join the conversation.

Just pushed a possible fix for lightbox not switching off. Regarding brightness this is controlled by the INDI driver and not by Ekos. In the driver you need to configure the brightness for each filter.
2 years 5 months ago #76198

Please Log in or Create an account to join the conversation.

  • Posts: 67
  • Thank you received: 6
A quick test with my prototype light panel seems to be working as expected now. (using Oct 01 nightly build.) Now to finish the LP build.

Many thanks for this Jasem.
The following user(s) said Thank You: Jasem Mutlaq
2 years 5 months ago #76371

Please Log in or Create an account to join the conversation.

  • Posts: 14
  • Thank you received: 0
I came across this post, and I see that after a year, with the relise 3.6.1 oct 2022 the bug exposed in this post has not been solved.

I report the problems with the operation of automation for flat fields.

I state that the indi driver works well with the device, which in this case is the same as David Swinnard: an arduino with Flat Man or Flip Flat firmware.

1) the panel is not turned on before the capture of the flat frame and is not turned off after. This must be done manually via indi panel in the flat man driver.

2) the dust cover is neither lowered nor raised automatically, this must also be done through the driver then manually.

3) once the panel is switched on with zero brightness, the brightness of the first selected filter is not activated. The second filter selected activates the brightness set in the driver presets. After that the filter that did not take the brightness data also starts working.
So, the first flat must be taken only to unlock the bug, a second flat must be taken with another filter, then it works until a new ekos session is restarted.

4) in the windows version, if you set the coordinates for the position of the telescope towards the wall where the flat panel is positioned, the telescope does not move. in the linux version it works.

5) the values ​​set to zero in the presets are not read.

and then, it is not a bug but a very important function. should be implemented in the automatic calculation with ADU in the flat and dark exposure settings window, the possibility of adjusting the light intensity of the panel with the shutter speed priority set manually or for a given flat. otherwise it would make no sense to have a flat box with software adjustable brightness.
Last edit: 1 year 4 months ago by Duccio.
1 year 4 months ago #87606

Please Log in or Create an account to join the conversation.

  • Posts: 67
  • Thank you received: 6
Addressing your points. This is how I have evolved my approach to flats.
1. Yes, I manually turn on the light and adjust brightness to that of the first filter I want flats for (via Indi panel).
2. I don't have a dust cover. My flat illumination assembly fits over the dew shield like a lens cap. (71mm refactor is my only OTA)
3. By manually turning on the light and adjusting to the first required brightness I have bypassed this.
4. I only run Linux and manually point the telescope to the zenith before setting the flat illuminator on tube.
5. I have not noticed this. All my values are above 0 and the two empty filter positions are left blank.

In order to get the flat panel off and the lens cap on manually I attend the sequence and stop if after the last flat is taken. I then take off the flat panel and replace the lens cap before starting the sequence again to take the flat-darks. I take my flats and flat darks during the day after the previous night's imaging session. I use an profile containing only the camera and the auxiliary device flat-man.
Last edit: 1 year 4 months ago by David Swinnard. Reason: spelling
1 year 4 months ago #87615

Please Log in or Create an account to join the conversation.

Please attach the logs so we can take a look.
1 year 4 months ago #87629

Please Log in or Create an account to join the conversation.

Time to create page: 1.285 seconds