I was running a sequence last night with darks at the end of the night and noted that CCD temp is missing from the fits file header on some, but not all, files. I think these were the only missing fits items as the fits list was just two items less in the files where temp info was missing.
Specifically the second target was M81, I took 7 luminance frames of M81 followed by four each RGB. The luminance frames had the CCD-Temp and Set-Temp, the RGB did not. All targets after that were missing the temp information.
Will this temp information show up if the cooler was somehow off or does the cooler have to be on for the fits temp information to show up? I don’t know why the cooler would have turned off itself off (it is working now) but it now appears that is what happened.
FYI, the image where the temp info stopped starts at around line 34100.
UPDATE: I am running the darks again now and the information is there and based on a quick dark subtract, it does appear the cooler somehow turned itself off. I am at a loss to know how as I was just letting it run. I suppose I could have bumped the mouse just wrong while working on the other adjacent PC but this seems very unlikely. I guess a related question is if I did somehow click by accident and turn the cooler off, would that show in the log?
So I tested on a bunch of stuff. SGPro does not care what filter you’re on… it just keeps track of frame type. Assuming that you selected light frame for all of these images (i.e. did not accidentally select flat), there are only 4 possible causes of this intermittent behavior (more in a minute). I tested:
Using an ASCOM camera on every type of frame (CCD-TEMP always present)
Using an SBIG camera on every type of frame (CCD-TEMP always present)
This leaves only the condition in which SGPro had difficulty getting the temperature from the camera. There are three conditions in which the SBIG camera can return an error temperature (and these are not written to the headers):
SGPro thinks the camera is not connected (not really possible since the camera continued to image… even if it was in error, SGPro would not continue to take images if it thought the camera was not connected)
SGPro failed to query the camera for temperature info (not possible because this logs an error).
So… that leaves 2 scenarios that can explain this:
The cooler is queried successfully, but it is reporting that it is off
We have an error in the code when the Camera and internal guider are are operating simultaneously
I am not sure how likely the second of these two is… this is because your L frames came first. These are the ones you say successfully write to the headers. The behavior of the Camera (when operating in tandem with the internal guider) is that if the guider is busy reading off the CCD (not integrating), the camera will skip the temperature read… BUT this means that the temperature would read as the last valid temperature (from your LUM frames) and not as an invalid temp (like you are getting). Is it possible that the cooler was tuned off somehow (did end of sequence options run at an inopportune time?)
It is possible but I can’t say for sure. I did get up and check things a few times at night but don’t know if that was one of them. Two possibilities along those lines:
I did abort the sequence in order to make a change and did not notice the end of sequence was checked. Since I did not have “park telescope when sequence completes” checked, it would not have afflicted the scope or roof. It might argue for leaving “park telescope when sequence completes” checked, come to think of it, as one would notice that whereas one might not notice the cooler.
In moving the windows around on my small screen I am using to test the beta, I might have clicked the cooler while trying to move a window. That should not happen now as I have just installed the beta on my main machine with four monitors and lots of screen space.
Just wondering along these lines: Is there a warning if one tries to turn off the cooler or change temperature settings during a sequence? If not, it might be a nice feature for pickle fingered users.
Would a user turn off of the cooler show in the log?
Note that many of these missing temps were on darks after the lights at the end of the night so prolly not that. It looks like the most likely scenario is that the cooler was off somehow. Do I understand you correctly that if the cooler was off we would see this?
Correction: The temp did not disappear between the Lum and the RGB on M81, it actually went away after the second green frame, about in the middle of the RGB part of that target. After line 59659 in the log.
OK, don’t think the logs were on as I never needed them before. I have moved all imaging operations back to my main PC from the beta test machine now with all current versions there as well. I did try to enable logging on the main machine but SBIG driver checker crashes as soon as you try to enable the logging. Great.
I guess I will just have to hope it does not happen again. I am at a total loss as to how it turned off.
If we ignore how the cooler got turned off for the moment, would you expect to see the temp data missing if it was off? If that is the case, at least I know to check to be sure it is on once in a while.
For what it is worth, this does argue for a browser based or other mobile app that one could use to monitor status of the systems. GNS is nice but pretty limited in what it tells you. It might save some night’s data.
Yes, regardless of camera, if the cooler is off, we will not add the temperature to the headers list. If you continue to see this, I will give you a version with better logging (but obnoxious). This will let us know, with certainty, why SGPro does not grab your image temp.
Thanks. Skies are clear but poor tonite so I am going to try beta 12 but not try to get keeper images. No difference for this issue even if it is SGP related (I would guess it is probably not). I will let you know if this happens again, could have been some weird power thing (power company not known for reliability) or user error.
Update on this. I just ran the whole night (beta 12) w/o seeing this issue so right now will have to put this into the “random mysteries” column. I will let the forum know if I see it again.
