Auto Star Selection Failure

[Bob Mimiaga]

I normally use PHD2 manually and always manually select the guiding star (which happens to be the highest SNR star on the loop map) to guide with. It works fine this way. But last night I attempted to auto start PHD2 with SGPro and PHD2 automatically used a lessor SNR rates star which results in a "lost star" status within a minute.

What setting do I have to adjust in PHD2 to have it auto select the best or highest SNR star in the loop display?

Thanks for your help.

Bob

[Andy Galasso]

Bob,

The auto-star selection is usually pretty reliable, so we'll need more information to figure out what went wrong in your situation. Please upload your PHD2 Debug Log and Guide log from the session where the guide star was lost. Log upload instructions here: https://openphdguiding.org/getting-help/

Andy

[Bob Mimiaga]

Andy,

Thanks for the quick reply. I'm setting up my gear for a test tonight and should have some data later in the evening. Thanks for your help!

BTW, When I run PHD2 manually I have to select "Auto-Select Star" from the Tools pulldown to have the program auto select a star. Yet when it's controlled by SGPro it seems to turn on this feature automatically. Is that the way it should work?

Thanks,

Bob

[Andy Galasso]

On Thu, Nov 8, 2018 at 8:02 PM Bob wrote:

When I run PHD2 manually I have to select "Auto-Select Star" from the Tools pulldown to have the program auto select a star. Yet when it's controlled by SGPro it seems to turn on this feature automatically. Is that the way it should work?

Auto-select Star is not a mode that is on or off, rather it is an action that executes immediately when you select it from the menu (or press Alt-S.)

When SGP interacts with PHD2 it directs PHD2 to start guiding at the appropriate time in the imaging sequence. When PHD2 gets the command from SGP it starts the camera looping exposures, waits for a camera frame, selects the guide star (as if you had done it from the Tools menu), starts guiding, waits for guiding to settle, and reports back to SGP that guiding is ready and settled.

Andy

[Bob Mimiaga]

Hi Andy,

Sorry for the delay on posting my log file. I had all kinds of problems last night but was finally able to get the software and hardware to work together. This attached log file had some good tracking but that was after I manually selected the star with the highest SNR. Anytime I allowed PHD2 to auto select a star it would select star with a lower SNR (about 15-20) which would result in the "lost star" condition. Stars with a much higher SNR (about 40) seem to guide much better.

Thanks for looking at this log file for me. With SGPro automatically invoking PHD2 it can't seem to select a star that provides a stable guide.

Bob

[Andy Galasso]

Bob,

We'll need to see both the guide log and the debug log.  You can follow the log upload instructions here: https://openphdguiding.org/getting-help/. That should be easier than locating the files and attaching them.

Andy

[Bob Mimiaga]

Andy,

Here's the link to my log and debug files. https://openphdguiding.org/logs/dl/PHD2_logs_8DaL.zip

Part of the session was effected by calibration issues. I was able to find some "lost star" messages in the debug file but couldn't determine the cause. Thanks again for your help!

Bob

[Andy Galasso]

Hi Bob,

Thanks for the log files.  I had a chance to take a look at them and have some recommendations.

1. rebuild your dark library

Your current dark library has darks for 1.5 and 2.0 second exposures, but you were using 1.0 second exposures. PHD2 was using the 1.5 second dark frame and this had a negative effect on the star auto-finder.  Build a new dark library being sure to include the range of exposure durations that you will use when guiding -- be sure to include 1.0 second in the range.

2. improve your OAG focus

Your guide star HFD of ~ 7 arc-sec (6 px x 1.2 arc-sec per pixel) indicates the focus is not very tight.  You should be able to tweak your focus to get that down to about 4-5 pixels (as shown in the star profile window).   Good focus is critical -- especially with an OAG -- to get the best SNR and to allow guiding on fainter guide stars.

3. Decrease your search region

Your search region of 35 pixels is very wide, perhaps that was done to compensate for the fat guide stars, and this is also contributing to the lost stars and star auto-selection problems.  After you tighten your focus, you should be able to reduce the search region back to the default value of 15 pixels, or maybe 20 pixels max; larger than that should not be necessary.

4. some other settings changes that may help

Here are some other suggestions that are not necessarily related to the lost star events, but could help in your guide configuration.

  a) use an explicit saturation threshold - select Saturation by ADU on the Camera tab in the brain and enter a saturation value of about 250 or 255.

  b) disable star mass change detection - disable it on the Guiding tab in the brain; there's too much variation in the star mass due to the focus; once you tighten the focus the mass change detection may work better, but it's safe to turn it off

  c) you could try operating the camera in 16-bit mode.  Click the camera setup button in the Connect Equipment window and select 16-bit mode.  This will require a new dark library so you should create a new equipment profile for the 16-bit mode camera connection. (The 16-bit profile will need a different star saturation ADU value, more like around 65000)

  d) enable sub-frames (brain, camera tab) - the camera supports subframes so you may as well use them; this should decrease the time for downloading image data from the camera

If you continue to see star lost messages after making the recommended changes, please enable diagnostic image logging for all star lost frames (brain, Global tab). That will cause PHD2 to capture the camera frames around each star lost event which you can post along with your debug and guide logs.

Good luck and let us know how it goes.

Andy

[Bob Mimiaga]

Andy,

Thanks for all your recommendations. I'll implement them and let you know how it comes out. 

Bob

[Bob Mimiaga]

Andy,

2. improve your OAG focus

Your guide star HFD of ~ 7 arc-sec (6 px x 1.2 arc-sec per pixel) indicates the focus is not very tight.  You should be able to tweak your focus to get that down to about 4-5 pixels (as shown in the star profile window).   Good focus is critical -- especially with an OAG -- to get the best SNR and to allow guiding on fainter guide stars.

A couple of questions about your comment above;

(1) can you explain how the (6px x 1.2 arc-sec per pixel) parenthetical remark was derived? 

(2) Do I tweak my OAG focus by adjusting the OAG focus knob to a lower HFD value on the Star Profile window when I'm guiding on a star?

Thanks,

Bob

[Andy Galasso]

On Sun, Nov 11, 2018 at 8:17 PM Bob M wrote:

(1) can you explain how the (6px x 1.2 arc-sec per pixel) parenthetical remark was derived? 

The debug log records the HFD value for the guide star for each camera frame and on average it was about 6 pixels; the debug log also shows that your guide camera pixel scale is 1.2 arc-seconds per pixel. I wanted to convert the HFD value from units of guide camera pixels to arc seconds for comparison with other gear configurations, thus 6 pixels x 1.2 arc-sec per pixel gives approx 7 arc-seconds for the HFD.

When you open the Star Profile window you can see the HFD value directly in units of both pixels and arc-seconds.

(2) Do I tweak my OAG focus by adjusting the OAG focus knob to a lower HFD value on the Star Profile window when I'm guiding on a star?

The key is to get your OAG par-focal with the imaging camera.  First make sure your imaging camera is as precisely focused as you can get it -- use your imaging app's auto-focus routine or use a Bahtinov mask.  Then once the imaging camera is focused, tune the OAG focus to get the guide star to have the minimum HFD value as you look at the Star Profile window.  You don't need to be guiding, just looping exposures with a star selected.  Start with a very bright star, then when you get close to perfect focus switch to a dimmer star if the star you started with becomes saturated (flat top).  Tip: you can resize the Star Profile window by dragging the corner and the HFD display will get very large so you can see it from far away while you are standing at the scope.

Hope that helps,

Andy

[Bob Mimiaga]

Hi Andy,

Here's my latest log files after (1) creating new dark files, (2) reducing the search area region to 15px and (3) getting a better focus on my OAG. I ran two simple steps directly from PHD2. The first was to start the looping process and select auto-select star from the TOOLS pulldown menu. It resulted in PHD2 selecting a star that was too faint and I lost track of it. Second, I manually selected a higher SNR star and it seemed to track without a problem. 

https://openphdguiding.org/logs/dl/PHD2_logs_xikB.zip

Can you please look at this again and let me know what else I'm doing wrong? I'd like to use SGPro to command PHD2 to auto select stars without any problems.

Thanks,

bob

[Andy Galasso]

Bob,

Nice job getting the OAG focus improved.  Now the primary problem with star selection in that session was the saturation detection. The automatic saturation detection does not work so great with an 8-bit camera and OAG, but that's why we have the Saturation by ADU setting available. Try switching to Saturation by ADU and set an ADU value of 250-255. That should make a big difference.  You can also disable star mass detection since that is probably not needed and may do more harm than good.

If you still have trouble, switch the camera to 16-bit mode as described in the earlier post (and build a new dark library for 16-bit mode); enabling subframes will be more important if you go to 16-bit mode since the amount of data downloaded from the camera for each frame will double.

Andy

[Bob Mimiaga]

Andy, I was able to switch the saturation and disable the star mass detection and do a quick test before the clouds covered the sky. It might be working now. I'll run some more tests when the wx clears up which may not be for a week here in Bend OR.  I'll keep you posted and thanks again for all your help. PHD2 is an amazing tool!

Bob

--
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.

--

Bob Mimiaga

[Bob Mimiaga]

 c) you could try operating the camera in 16-bit mode.  Click the camera setup button in the Connect Equipment window and select 16-bit mode.  This will require a new dark library so you should create a new equipment profile for the 16-bit mode camera connection. (The 16-bit profile will need a different star saturation ADU value, more like around 65000)

Andy,

On your previous post you suggested operating my ASI244MC camera in 16-bit mode. But when I select the ZWO ASI camera from the camera pull-down menu in the Connect Equipment dialog the Camera Setup button is desensitized. Does this mean I cannot run my camea in 16 bit mode?

Bob

[Andy Galasso]

On Tue, Nov 13, 2018 at 2:12 PM Bob Mimiaga <bmim...@gmail.com> wrote:

On your previous post you suggested operating my ASI244MC camera in 16-bit mode. But when I select the ZWO ASI camera from the camera pull-down menu in the Connect Equipment dialog the Camera Setup button is desensitized. Does this mean I cannot run my camea in 16 bit mode?

I see that you are running version 2.6.5 and the ZWO 16-bit functionality was added in a newer version, v2.6.5dev4.  You'll need to upgrade to the latest dev version if you want to try it.

[Bob Mimiaga]

Andy,

Thanks! I've downloaded it. BTW Does Dev 6 have all the functionality of be previous versions?

Bob

[Andy Galasso]

On Tue, Nov 13, 2018 at 7:52 PM Bob Mimiaga <bmim...@gmail.com> wrote:

Does Dev 6 have all the functionality of be previous versions?

The downloads page lists the changes in each release so you can review what went into each new version.

[Bob Mimiaga]

Andy,

Last night the sky cleared up for about three hours so I was able to test my PHD2 settings again.

I adjusted the following according to your recommendations;

1. Updated my Dark files to include the exposure time I am using (2 secs).

2. Fine tuned my OAG focus.

3. Decreased my guiding track search region to 15 px.

4. Selected my camera's Saturation by Max ADU value to 255.

5. Disabled my Star Mass detection.

PHD2 worked perfectly! I had guiding stars averaging a HFD of 3 to 4.5 and Total RMS Errors of .81 to 1.2px. Given the conditions of the sky I'm very pleased. I'm now able to have SGPro start a sequence to slew to an object and automatically engage PHD2 to guide on a nearby track without any problems.

Andy thanks again for help. Without the help of people like yourself it would be very frustrating. You guys have developed an amazing tool for us amateur astrophotographers to use. I've also learned a little more about your PHD2 Log Viewer in the process.

One thing I didn't do was change my camera bit mode from 8 to 16 bit. Given that things are working well do you feel I need to make this change as well?

Thanks,

Bob

[Andy Galasso]

Hi Bob,

Thanks for reporting back. I'm glad to hear it is working well now.

No need to bother switching to 16 bit mode, I don't see there being any benefit to that now that you have things working.

Andy