Very high guide star SNR
Larry Couture
Larry
bw_msgboard
Hi Larry. There’s nothing very important about SNR as long as it’s well above the noise threshold and the star isn’t saturated. The calculation has changed a bit with successive releases and I don’t know what you’re comparing against. Are you choosing the stars manually, picking stars that look nice and bright? If so, you should try letting PHD2 choose them automatically. What you see in real-time is the result of a display stretch operation and you are quite possibly not seeing many faint stars that would work well as guide stars. The auto-select function isn’t affected by stretching so it might do a better job for you. In any case, if you’re worried about high SNR levels, use the star profile tool to see if the star is flat-topped or isn’t showing a consistent profile. And you should generally not reduce exposures times in an effort to lower SNR without strong evidence that you’re having guiding problems.
Beyond that general advice, we can’t say much more about what you’re seeing without having more detailed information about your set-up and seeing some log files.
Good luck,
Bruce
--
You received this message because you are subscribed to the Google Groups
"Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to open-phd-guidi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Larry A Couture
Bruce, Thanks. From this and a post on CN, I’m getting the sense the number is not a concern.
I just updated to 2.6.1, but think had the previous version since late last year. I get the high SNR numbers with either auto-select or manual. I normally only adjust exposure to address the star profile and guiding, but tried short exposures here to see if it would affect SNR – it did not. PHD2 has never auto-selected a star that was not visible in the stretched display, though sometimes I override its selection for what I think is a better star. Star profile is always my criteria and I always pick a star that is small but bright, but no so bright that it flat tops the profile, though I can get excellent guiding if the profile is slightly flat-topped.
Over on CN, it’s been suggested that others are seeing higher SNR numbers. The number itself doesn’t matter to me but there was a significant increase in the number after I built up this new laptop, so I was (am) concerned that there is something going on between the camera driver and PHD2. If others are seeing higher numbers with 2.6.1, then there is no problem.
Once I get clear skies again I’ll save and upload a guide log.
Thanks,
Larry
bw_msgboard
Hi Larry. People are probably just noticing the effects of our having changed the SNR calculation. We did that because we were trying to improve PHD2’s ability to detect faint but usable guide stars, something that’s important for people using OAGs and automation packages. As I recall, it’s been adjusted twice since the 2.5.0 release and it could change again at some point because we’re using it mostly for internal purposes. It’s not a photometric SNR calculation so it isn’t meaningful on an absolute scale. Users should probably worry about “too-faint” or “too-bright” and not worry about the actual SNR value. It sounds to me like you already know this and have developed a good feel for how to pick a guide star. Like you, I generally let PHD2 auto-select the star but then may override that if I don’t like the result – mostly I leave it alone.
It figures that you’d notice the change right after putting together a new system. <g> If you’re getting consistent guide camera images, you’re good to go - I would ignore the SNR number and just look at the guiding results. There’s probably not much point in uploading a log at this point unless you think the guiding results are bad or that PHD2 is choosing inappropriate guide stars. In the latter case, we’d also need FITs versions of the guide camera images to understand what went on.
Have fun,
Bruce