Hot pixel and not able to remove it ......

73 views
Skip to first unread message

Sara Wager

unread,
Apr 1, 2015, 7:00:05 AM4/1/15
to open-phd...@googlegroups.com
I hope that someone can make some suggestions that will work. My lodestar seems to have developed a hot pixel in the middle of the field of view for PHD guiding. No problem I thought and did a Bad pixel map, using the default settings. It was still there. So then I manually added it. I know that it's manually added as when it ask it to show me all hot pixels it is red.

But.... it still shows and when a star is auto selected for guiding it selects this particular pixel. I have also looked at the SNR figure of the pixel compared to the stars and tried to set the recommended star selection SNR in 'the brain' to pick a proper star over this hot pixel ..... nope, still no joy.

As it stands at the minute, I can't so any automation with SGL (which I was doing fine) as the programme will always select this hot pixel.

I am using PHD vf.

I hope that someone can help as I really don't know what else to try.

Andy Galasso

unread,
Apr 1, 2015, 11:54:42 AM4/1/15
to Sara Wager, OpenPHD Guiding
Sara,

Are you using the ASCOM driver to connect to you Lodestar or the built-in driver "Starlight Xpress SXV"?

If you are using the built-in driver, please update to the most recent version and rebuild your bad-pixel map.  Or, try the SX camera ASCOM driver and see if that makes a difference.

The Lodestar uses an interlaced sensor and software has the option to either read the pixels in two passes (even rows then odd rows), or to vertically bin the pixels and read them all at once.  The ASCOM driver uses the former method, and the native driver uses the latter but then "deinterlaces" the data, resampling to produce an image with the same dimensions as if it had not been binned. Until recently the native driver was incorrectly doing dark subtraction and bad-pixel removal *after* deinterlacing. PHD2 v2.4.1i fixes this and does the dark subtraction before deinterlacing. (The dark frames are now the raw, binned pixels thus half the height of the deinterlaced frame.)

If neither of those ideas help, please post a sample image showing the hot pixel (could be a frame taken in the daytime with the camera covered). You can get the sample image by doing File => Save Image. Also please post your phd2 debug log and a zip file of your darks_defects folder. The darks_defects folder is located at C:\Users\YOUR_USERANAME\AppData\Local\phd2\darks_defects

You mentioned setting a SNR value in the brain. Just to clarify, there is no selection in the brain controlling star selection. The only setting in the brain related to SNR is the setting for the Target SNR for Automatic exposure duration.

Andy

Sara Wager

unread,
Apr 1, 2015, 12:47:25 PM4/1/15
to open-phd...@googlegroups.com, saraw...@gmail.com
Thanks Andy - I will update PHD to the latest version.

Many thanks for helping and providing a comprehensive answer.
Reply all
Reply to author
Forward
0 new messages