Detect image shift

[Rudi Bjørn Rasmussen]

Hi,

I am running PHD2 and it guides perfectly for hours on with sub arc sec RMS values.

But my main scope has an image shift of 1-2 arc sec per minute and I am trying to find out, where that shift comes from.

My mail telescope is a 10" f/5 newtonian, focuser pointing "inward" towards the EQ6 mount, a small 50mm F180mm guidescope is placed on a vixen doveail on the tube rings oposite the losmandy on the mount.

I suspect, that the imsge shift i due to flex in the system, but I would really like to verify if the PHD FOV is fixed or if it shifts.

Can I deduct this from the log files? I can't find any information for this.

For example, it would be nice to have the centroid x/y for the guide star(s) in the logfile for the whole guide session. I tried to save imaged (advanced settings), but it only allowed me to save 100 images. To analyse 4-5 hours og 2 sec images, I will need to save thousands of images.

Any sugestions?

[John Nagy]

Just a thought (maybe there is a better way) you might be able to reverse your setup. Guide through your main imaging camera and then image through your guide scope to capture that 4-5 hours.

--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/cd16466c-f1de-453a-b3e0-333dcfacd298n%40googlegroups.com.

[Bruce Waddington]

Perhaps I'm not understanding what you're saying.  My interpretation is this:

1.  Guiding is great for long periods of time

2.  During that time, you see consistent 1-2 arc-sec shifts in your main camera images

If so, that IS differential flexure, that's exactly how it's confirmed.  What that means is that PHD2 is keeping the star(s) correctly positioned on the guide chip but the main scope is sagging/moving relative to the guide scope.  So looking at the lock-point positions in PHD2 will show no significant changes - that's why the guiding is good.  The lock-point positions will change for dithering, whenever guiding is started/resumed, and whenever you do an auto-find.  All of the entries in the guide log are showing you offsets relative to the lock point.  If you doubt what I'm saying, you can find all the centroids for all the stars for every camera exposure in the debug log file - open the debug log file in a text editor and do searches for "Star::Find returns".  There will be a zillion of them although you could reduce the count by disabling multi-star guiding.  But really, I think this would all be a waste of time.  If there's a big shift that affects the guider, it will show up quickly in the LogViewer.

Good luck,

Bruce

[Rudi Bjørn Rasmussen]

Ok - I just wrote a small C# program, that plots the stars from the dubug log file. A small circle is painted with alpha=63 (transparant) for each stars, Stars on top of each other will appear brighter.

It is very obvious, that there's no drift at all in the 5,5 hours session