Username:

Password:

Altair AstroDIO DehumidifiersAtik CamerasModern AstronomyDavid HindsNe3 Filters
UKAI is moving to Facebook

Author Topic: PHD Guardian: new automation software for PHD (anyone interested?)  (Read 1473 times)

0 Members and 1 Guest are viewing this topic.

Offline rofus

  • Full Member
  • ***
  • Posts: 365
  • Astrophotography between a song and an album
    • The Ski.es Astrophotography Gallery
Hi all,

I'm sure I'm not the only one with this problem with PHD...loosing guide star for no apparent reason (even with mass detection disabled) while the star is in reality perfectly visible where it always was. I get this sometimes, as well as some weird glitches after/before star lost message (whole image from guide cam shifted and star not there anymore for a second), like a graphical bug.

Anyway usually is enough for me to stop guiding and restart it again and PHD magically decides the star is not longer lost.

So I wrote a new piece of software called 'PHD GUARDIAN', that I'll probably expand to other functions but is at the moment an automatic software that does:

1) print on screen verbose output from PHD automatically every X seconds depending on your exposure time (useful to always know what's happening in details)
2) detect if there's a (new) star selected and VERIFY it is locked and if PHD is looping or not
3) if PHD is looping (and there was a confirmed locked star) or guiding, it DETECTS automatically if STAR LOST happens
4) within 1 second, it STOPS/RESTARTS looping/guiding to catch again the guide star asap without any manual procedure

The point 4 action at the moment is performed with 'auto select star' because of a bug of PHD, in theory my software remembers the GOOD LAST GUIDE STAR and after restarting the guide starts guiding on the last locked guide star. When PHD dev will fix the bug it will work with last selected star.

If there's interest I'm thinking about adding a notification system by email, so if star is lost the program can automatically email. I might also implement an SMS/TEXT MESSAGE alert but that has a cost so I cannot do it for free :) Because I can get any event happening in PHD and trigger actions after it, I might expand it with more automations/alerts.



I've done this for myself and it works quite well (PHD2 but should work with PHD1 as well), if anyone is interested let me know I might do a distributable public version.
AA 8" RC + AT Flattener, SW Esprit 100 + SW Flattener, Daystar Quark Chromosphere
Canon 6D modded, Nikon D750 unmodded, ASI174MM-COOL, SX Lodestar X2, ASI120MM
Canon & Nikon various prime lens
SW HEQ5 Rowan Belt Mod, AA 60mm Guidescope, Astrotrac, SW Star Adventurer
Mac 15" Retina Parallels: EQMOD, PHD2, SGP, PlateSolve2, BYN, CDC, Pixinsight, Photoshop
William Optics binos, Baader Hyperion eyepieces, Meade & Celestron binoculars
http://theski.es

Offline rofus

  • Full Member
  • ***
  • Posts: 365
  • Astrophotography between a song and an album
    • The Ski.es Astrophotography Gallery
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #1 on: 20:37:27, 12 October, 2015 »
To make it act faster, I ported it as native C++ app in Windows so that now if PHD for any reason looses guide star, within 0.5 secs PHD Guardian is able to

1) stop/restart the images loop
2) make PHD auto-select best visible star in that moment
3) start guiding immediately using previous calibration data

That is usually what in most cases, for me, make PHD keep guiding if problem is just passing light clouds, noise, light pollution, bad seeing etc.

AA 8" RC + AT Flattener, SW Esprit 100 + SW Flattener, Daystar Quark Chromosphere
Canon 6D modded, Nikon D750 unmodded, ASI174MM-COOL, SX Lodestar X2, ASI120MM
Canon & Nikon various prime lens
SW HEQ5 Rowan Belt Mod, AA 60mm Guidescope, Astrotrac, SW Star Adventurer
Mac 15" Retina Parallels: EQMOD, PHD2, SGP, PlateSolve2, BYN, CDC, Pixinsight, Photoshop
William Optics binos, Baader Hyperion eyepieces, Meade & Celestron binoculars
http://theski.es

Online psjshep

  • Database dude... for the last breath of UKAI
  • Administrator
  • Galactic Poster
  • *****
  • Posts: 10206
  • My god! It's full of stars.... just often enough!
    • Some astro Pics
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #2 on: 21:35:36, 12 October, 2015 »
This looks intriguing, Rufos....

I think I'm going to need a few minutes to digest it all..... but in principle.... intriguing

Phil
--------------------------------------------------------------------------
Web Images

Offline rofus

  • Full Member
  • ***
  • Posts: 365
  • Astrophotography between a song and an album
    • The Ski.es Astrophotography Gallery
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #3 on: 22:39:22, 12 October, 2015 »
Well, putting it in a simple way, you just use PHD as your usual (as I do).

Then if you start PHD Guardian, it will just keep an eye on what happens in PHD. In case of star lost, it just immediately stop/restart image capture & guiding of PHD, using the best guiding star available in that moment. As this happens all very quickly (0.5 to 1 second total time depending on your hardware) is quicker than any manual procedure, and in most cases stopping / restarting guiding with auto best star solves most problems of Star Lost due to passing clouds/seeing etc., and you don't have the keep an eye on it (and 1s unguided is not enough to ruin an exposure).

Unless there's a big cloud (or a house/tree!) covering the whole FOV of the guide camera, this way I can be sure that PHD will not stay for way too many seconds flashing red saying Star Lost but actually restart guiding immediately with a new best star (all without calibration because it uses the last valid one).
AA 8" RC + AT Flattener, SW Esprit 100 + SW Flattener, Daystar Quark Chromosphere
Canon 6D modded, Nikon D750 unmodded, ASI174MM-COOL, SX Lodestar X2, ASI120MM
Canon & Nikon various prime lens
SW HEQ5 Rowan Belt Mod, AA 60mm Guidescope, Astrotrac, SW Star Adventurer
Mac 15" Retina Parallels: EQMOD, PHD2, SGP, PlateSolve2, BYN, CDC, Pixinsight, Photoshop
William Optics binos, Baader Hyperion eyepieces, Meade & Celestron binoculars
http://theski.es

Offline rofus

  • Full Member
  • ***
  • Posts: 365
  • Astrophotography between a song and an album
    • The Ski.es Astrophotography Gallery
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #4 on: 21:25:51, 22 October, 2015 »
Finally after few weeks of bad weather, tonight is a good night here and earlier with some clouds passing by, ideal setup for testing PHD Guardian in its latest update.

A few bug fixes and optimisations, and more than once starting PHD Guardian after all is setup and PHD is guiding saved my images: as soon as star was lost within 2s the software stopped/restarted the images looping and guiding, issuing commands to PHD to also reselect the best available star at that moment.

That allowed for guide to not be affected, and more than once allowed PHD to select another better star, like in case when there's a bloated star with clear sky and PHD selects another one, but if there are some lights cloud the previous one becomes too faint and PHD selects the one that before was bloated and now is good.

Also it helped a lot when for some reasons the images from webcam start arriving shifted or not right and PHD again looses the guide star: stopping and restarting the image loop basicly stops and restarts the camera for a fraction of second, so the problem is solved and within 2s PHD is guiding again.

Very glad I managed to do this, it solves the problem of having to monitor PHD for problems, and having the log in the window I can also go back and see if star was lost or not.

AA 8" RC + AT Flattener, SW Esprit 100 + SW Flattener, Daystar Quark Chromosphere
Canon 6D modded, Nikon D750 unmodded, ASI174MM-COOL, SX Lodestar X2, ASI120MM
Canon & Nikon various prime lens
SW HEQ5 Rowan Belt Mod, AA 60mm Guidescope, Astrotrac, SW Star Adventurer
Mac 15" Retina Parallels: EQMOD, PHD2, SGP, PlateSolve2, BYN, CDC, Pixinsight, Photoshop
William Optics binos, Baader Hyperion eyepieces, Meade & Celestron binoculars
http://theski.es

Offline Phillyo

  • Poster God
  • *****
  • Posts: 1492
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #5 on: 08:45:32, 23 October, 2015 »
This looks very useful and promising! I will watch this thread and see how you get on.

Phil
Astrotrac | Avalon Linear Fast Reverse with EQmod | WO Star 71 | Astrodon 3nm Ha, 5nm SII and 5nm OIII | Astro-modded Canon 450D | QSI 583wsg | Lodestar X2 Guidecam | Sequence Generator Pro | Pixinsight

Offline rofus

  • Full Member
  • ***
  • Posts: 365
  • Astrophotography between a song and an album
    • The Ski.es Astrophotography Gallery
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #6 on: 10:29:20, 23 October, 2015 »
Whoever wants to try it just let me know, for me useful to test it in other configurations (for example other cameras, guiding exposures of more than 1s, slower computer etc.). The only requirement is that it needs the latest beta of PHD2 (the dev ones, the same needed for SGP basically).

My computer is a very fast Macbook Retina SSD so all is tuned for it, on another configuration I might tune the software to wait 2 secs between stopping and restarting the guiding etc.

Yesterday during a 6h session, with mixed conditions, I lost star 10 times for different reasons (clouds, bug of my camera/driver/PHD and shifted image, no reason at all!) and 8 times out of 10 I did not even notice it if not afterwards reading the output, images were perfect. The other 2 times somehow the 1s pause I set between stop/restart did not give PHD enough time to properly restart/select another star, so it did not work properly and it stopped and restarted again twice and so image was affected. I expect this and if I can keep fails at 20% then I'm happy to trade it with a faster stop/restart so that the remaining times (80%) are recovered quickly without affecting final image.

Anyone wanting to test it just let me know, not a problem, if it works for others I might just release it publicly.
AA 8" RC + AT Flattener, SW Esprit 100 + SW Flattener, Daystar Quark Chromosphere
Canon 6D modded, Nikon D750 unmodded, ASI174MM-COOL, SX Lodestar X2, ASI120MM
Canon & Nikon various prime lens
SW HEQ5 Rowan Belt Mod, AA 60mm Guidescope, Astrotrac, SW Star Adventurer
Mac 15" Retina Parallels: EQMOD, PHD2, SGP, PlateSolve2, BYN, CDC, Pixinsight, Photoshop
William Optics binos, Baader Hyperion eyepieces, Meade & Celestron binoculars
http://theski.es

Offline Oscar The Grouch

  • New Member
  • *
  • Posts: 30
Re: PHD Guardian: new automation software for PHD (anyone interested?)
« Reply #7 on: 13:23:09, 16 February, 2016 »
Hi Rofus,

I'm using a QHY5 camera attached to my ST80 and running PHD off a Windows tablet.  It's only a 1.2 processor but running Windows 10 of it's SSD it copes really well.  I was using it for guiding at the same time as running EQMod and Stellarium, but that started to tax it a bit!!  My mount is a Skywatcher HEQ5 Pro.

If you'd like me to test your prog to see if it works ok on that, as well as my laptop (a bonkers fast HP) that I'm using for controlling and processing, please let me know!!

 

ukbuysellRemote Imaging from AustraliaSharpSkyblank APTUKAI on Facebook
Powered by SMF 2.0.15 | SMF © 2006, Simple Machines LLC
DarkBreak by DzinerStudio. Theme modified by The UKAI Team

Page created in 0.172 seconds with 38 queries.
TinyPortal © 2005-2012