When does SGP tell PHD2 to look for a guide star after a slew?

I think the way the process should work is:

  1. Slew
  2. Platesolve, and re-slew if necessary
  3. Autofocus (if you’ve set SGP to do an AF run after a slew, which I always do)
  4. Set the selected filter, and adjust focus
  5. Restart PHD2 and autoselect a guide star
  6. Wait for PHD2 to settle before starting the exposure

However, what seems to happen is:

  1. Slew
  2. Platesolve, and re-slew if necessary
  3. Restart PHD2 and autoselect a guide star
  4. Autofocus (which pauses PHD2, since that’s what I’ve set it for)
  5. Set the selected filter, and adjust focus
  6. Restart PHD2, wait for it to settle before starting the exposure.

The problem is that in step 3, PHD2 sets a search box where it found the guide star. It looks in
that search box when it restarts in step 6, and with drift, wind, tracking errors, etc. the guide star might have drifted out of the search box during the AF run (I use 9 x 15 seconds) and filter change. I don’t often lose the guide star during this process, but sometimes. Like at 1:30 this morning, when I was asleep and conditions were really, really good. :slight_smile:

Is there a way around this? I used to guide during the AF run, but with my gear (moving mirror SCT) I would often lose the guide star during backlash comp moves, so I gave that up. Can SGP be modified to allow the user to tell PHD2 not to restart until the AF run and filter change are done?

Ken and Jared: A single-bullet fix that should solve many of the lost-guide-star issues I’ve raised might be to allow the user to tell SGP to pause guiding during autofocus backlash comp moves. It’s really the mirror shift during backlash comp that is at the root of all these problems. The problem I’ve outlined above would then be solved simply by guiding during the AF run. (BTW, I just now saw the release of 2.4.3 – apologies if these issues are already addressed there, as I’ve been using 2.4.2 until now.)



This is just what I wanted to ask right now :sunglasses: I do not understand this workflow SGP proposes?
+1 for Kevin on this. Especially after Meridian flips this can be tedious and lead to now where in certain cases. I also lost some precious imaging hours due to a lost in space setup trying to recover several times.
The other part that the session didn’t succeed was that after Meridian Flip my dome had to slew to the Scope position. The cam did then take plate solve images of the dome interior. ( I started an other treat on this; so enough for here :wink:
I also read somewhere that dev3 of PHD is out covering some bugs linked with pierflips…

Very best 'n CS!!!


I think I also encountered this last weekend when outing my RASA (moving mirror) for the first time. I used LPS-P2 and RGB filters and had configured the LPS with a different focus offset so changing to it would trigger the focus backlash compensation. The sequence did complete successfully, but most of the LPS frames went through sequence recovery due to guiding failure.

I resolved this issue by increasing the number of pixels in the setting for the search region in PHD2’s guiding tab. I increased mine to 35 when I switched to an OAG in my new QSI683wsg-8.

Previously I used a 60mm guidescope mounted on the OTA and continued to guide through AF. Now I pause during AF.

The increased search region made quite a difference and - thus far - has always been in the square after AF.



Unfortunately there are a lot of different ways to do this flow. This means we either have to pick something that works for the majority of folks or add lots of options to let you choose your own destiny.

We choose to start guiding after auto center because most people can guide during auto focus and if your auto focus frames are fairly long this may be required. Unfortunately it doesn’t work for SCTs so we have the option to pause guiding during AF which is why you’re seeing the restart in step 6.

I’m guessing you didn’t have recovery mode turned on? It should have caught the lost guide star and started guiding again.

Also like Barry mentioned you can up the search box in PHD which should help.


Hi Jared,

Understood; you guys have to choose something. Any thoughts about the option to pause guiding during backlash comp? I really think that would be the ultimate fix for all of this.

Recovery mode has only helped where PHD2 didn’t start guiding on hot pixels in the meantime. I guide with an ASI120MM (ie, video camera, meaning lots of noise). In my experience, PHD2 always finds a hot pixel with my camera, no matter how diligent I am with darks or bad pixel maps. As far as increasing the search box size, I’m at max (50) already. At my focal length, that’s only 45 arcsecs. The star is usually there when PHD2 restarts, but not always.


Seems like it is a pretty popular request. I will take a look at it. This thread has been moved into the 2.4.X list (unless it becomes extremely invasive… then it will be moved to 2.5):

Great! That would be really helpful.

This has been implemented in It is more invasive than I would have liked. Beta users please watch your sequence for a bit before you leave it unattended. The guider pause for backlash (and subsequencet resume) will take place anytime backlash occurs. The only exception to this is if:

  • Auto focus has paused the guider for the entire AF run
  • Backlash is invoked while a sequence is not running (manual actions)