What's supposed to happen when PHD2 loses the guide star?

[kevi...@gmail.com]

In my case, the search box begins appearing at seemingly random locations, even when the guide star seems perfectly apparent to my eye.  I assume that what's supposed to happen is that PHD2 automatically locks on to the brightest "star" and pulls it onto the crosshairs, right?  I'm guessing that, with a planetary cam like my ASI120 mono, even with the bad pixel maps I regularly make, there's just too much noise for PHD2 to find the guide star again.  Does that all sound correct?

I'm asking because I'm starting to think it might be time to move to a Lodestar-type guide cam.  The ASI120 has given me three years of good service, but I'm starting to think it might be the source of most of the "what do you mean, 'star lost???'" moments I often have when looking at my logs the next morning.  :) 

Kevin

[Andy Galasso]

Hi Kevin,

PHD2 does not automatically locate a guide star within the camera frame unless explicitly told to do so. Here are the ways it will select a guide star:

  - if you click on a star while looping exposures

  - if you give the command to auto-select a star (menu: Tools => Auto-select star  or  Alt-S)

  - your imaging app directs PHD2 to auto-select a star.

There are many factors that can cause PHD2 to report "star lost".  We would need to take a look at your guide log and debug log to see what might be the cause. You could also post a sample FITS image from the camera (menu: File => Save Image) so we can look at that too.

Regarding bad-pixel maps: you should not have to regularly build them. It is sufficient to build one once, then, if necessary, add individual hot pixels if the camera ever develops any new ones over time.  Bad-pixel maps have no effect on camera noise, and camera noise is typically not the cause for a lost guide star.

Andy

[bw_msgboard]

Hi Kevin.  The explanation for what happens with a ‘lost star’ event is in the help content:  http://openphdguiding.org/manual/?section=Basic_use.htm#Guiding

 

PHD2 doesn’t start searching the entire camera frame for a guide star, only for a usable guide star within the original search region.

 

Without knowing a lot more about your specific situation – i.e. exposure time, SNR, all the things contained in the debug log file – we probably can’t advise you on whether a new camera is your best option.  That said, we don’t hear many complaints from people who opt to buy a high-quality guide camera. J

 

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.

[kevi...@gmail.com]

Thanks, Andy and Bruce.  Debug and guide logs from my last session are here:

https://dl.dropboxusercontent.com/u/58468743/PHD2_GuideLog_2016-06-29_221543.zip

You'll see that the star is lost a number of times, although I get it back on track - when I'm awake.  Around 03:21:28, after I'm asleep, it loses the guide star again, and eventually SGP directs it to abort.

Let me be a little more precise about what I think is happening in my case.  The search box can turn to a dotted liine - which I think means "star lost."  And in that case, no guide command is given.  Or it can stay solild, but to my eye the guide star is obviously outside the box.  In that case, PHD2 has locked on to something that isn't the guide star, and it issues a move command if that something is far enough off the crosshairs to require one.  All correct so far?  If the latter situation persists, the scope can wander far off the target, and the guide star never gets recovered.

Kevin

[Sander Pool]

It has been my experience for years that PHD does not hold on to guide stars very well. This seems to be especially true for OAG users where star shapes can deviate quite a bit from the perfect PSF shape PHD expects (or at least used to expect a long time ago). It will grab a hot pixel instead of a perfectly nice glob of star pixels. It got to the point where I just had to leave it be for a while and do some planetary imaging, swapping one set of frustrations for another :-)

Now that I have a handle on planetary imaging I'll have to try the latest PHD again and see what it does. Like you I see the star selection box move away from where I put it around the target star even though PHD supposedly isn't meant to do that. Long duration imaging sessions are a thing of the long ago past. I need to keep on watching and put PHD back on track, kick it off the hot pixel back on the guide star. I still blame the PSF based start selection routine for being too picky with long focal length stars being distorted by an OAG and seeing.

  Sander

[bw_msgboard]

Hi Kevin.  There are a couple of things I recommend changing.  In the first half of the session you were using 1-second exposures at a long focal length, which is a recipe for problems because it usually results in lots of seeing-induced artifacts.  Right off the bat, I’d recommend keeping this in the 2-3 second range as the norm for a long focal length.  If you’re thinking your can somehow defeat seeing with short exposure times, you should probably abandon that view – it doesn’t work.  Second, the vast majority of your “lost star” events are being triggered by the star-mass detection level of 50%.  This is probably also not a good choice.  Unless you happen to have two suitable guide stars that are close together, you should try disabling the star-mass detection altogether (Brain dialog/Guiding tab).  If you want to keep it enabled for some reason, try moving the tolerance up substantially, 70-90% for example.

 

Let us know if these changes improve your situation.  By the way, you should also completely ignore any advice along the lines of PHD2 can’t guide well with an OAG at long focal lengths.  Many users routinely do it, Andy and myself included. 

 

Good luck,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of kevi...@gmail.com

Sent: Tuesday, July 05, 2016 5:33 AM
To: Open PHD Guiding

--

[kevi...@gmail.com]

Thank you, Bruce!  The part about disabling/adjusting the mass detection level is an insight for me.  All these years using PHD, and there are still things that I keep learning.  BTW, there must have been some brief issue with the list yesterday.  I actually posted that I've had good experiences with long focal length, OAG and PHD - that's all I've ever used, and I usually have no troubles - but that message didn't show up.

I'll let you know how things work out.

Kevin