Newbie - Understanding RA Variability

[NY2KW]

I am imaging at a remote site in Texas.  Last night was one of my first clear sky nights.  Imaging M106 at astro dawn and a few hours later (02:00a) switching to NGC7000. I dithered every 2.  Its hard to know what the high cloud cover may have been.  The mount is a SW Wave150i with 120mm guidescope / ASI120MM Mini on a RedCat 71 WIFD.  I found on previous test runs that RA algorithm Hysteresis gave me better averages than PPEC (after running PPEC for > 60 min).  Can I get some advice how to interpret or gauge the variability in terms of what I could do to improve guiding or at least my expectations :-)

 https://openphdguiding.org/logs/dl/PHD2_logs_bicw.zip  

[Brian Valente]

can you be more specific where you are concerned? What are you looking at to come to this conclusion?

Looking at your two longest runs, RA was 0.44"RMS and then 0.36" RMS, so i don't really see a problem. the total RMS was 0.58" and 0.60"

.

your first run had quite a bit of variable guidestar SNR, which typically signals passing clouds or similar

On Tue, Jun 2, 2026 at 6:35 AM NY2KW <jost...@gmail.com> wrote:

I am imaging at a remote site in Texas.  Last night was one of my first clear sky nights.  Imaging M106 at astro dawn and a few hours later (02:00a) switching to NGC7000. I dithered every 2.  Its hard to know what the high cloud cover may have been.  The mount is a SW Wave150i with 120mm guidescope / ASI120MM Mini on a RedCat 71 WIFD.  I found on previous test runs that RA algorithm Hysteresis gave me better averages than PPEC (after running PPEC for > 60 min).  Can I get some advice how to interpret or gauge the variability in terms of what I could do to improve guiding or at least my expectations :-)

 https://openphdguiding.org/logs/dl/PHD2_logs_bicw.zip  

--
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 visit https://groups.google.com/d/msgid/open-phd-guiding/676072d4-6f0c-4c59-96f0-4240bb111f9en%40googlegroups.com.

--

Brian 

Brian Valente

astronomy portfolio https://www.astrobin.com/users/bvalente/ 

[Brian Valente]

Here's another view of RA, showing the residual (uncorrectable) RA in green, vs. the raw RA with corrections backed out (red silhouette). you can see PHD is doing an excellent job of reducing periods to below 0.20"

Brian  

On Tue, Jun 2, 2026 at 6:35 AM NY2KW <jost...@gmail.com> wrote:

I am imaging at a remote site in Texas.  Last night was one of my first clear sky nights.  Imaging M106 at astro dawn and a few hours later (02:00a) switching to NGC7000. I dithered every 2.  Its hard to know what the high cloud cover may have been.  The mount is a SW Wave150i with 120mm guidescope / ASI120MM Mini on a RedCat 71 WIFD.  I found on previous test runs that RA algorithm Hysteresis gave me better averages than PPEC (after running PPEC for > 60 min).  Can I get some advice how to interpret or gauge the variability in terms of what I could do to improve guiding or at least my expectations :-)

 https://openphdguiding.org/logs/dl/PHD2_logs_bicw.zip  

--

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 visit https://groups.google.com/d/msgid/open-phd-guiding/676072d4-6f0c-4c59-96f0-4240bb111f9en%40googlegroups.com.

[Bruce Waddington]

The guiding looks fine for the coarse image scale you're using.  I think you should set your expectations based on the results of your imaging - looking at star shape and star size to see if they are good enough to produce images you're happy with.  Just chasing the guiding numbers can be a real waste of time, especially in low-altitude locations where the seeing is probably not consistently good.

Regards,

Bruce

[NY2KW]

I don't have the experience under my belt to correlate what these guiding graphs should look like with respect to changes occurring during imaging that impact guiding. Clouds, temperature affecting AF, stars SNR, HFR, FWHM, etc.  For example, I see lots of spikes (after dither settles) that exceed +/-1 " about 2-4 x / minute.  Is that considered high frequency errors caused by air turbulence/seeing related?  I know these are very basic questions, and if this isn't the right place to ask them, that's fine and I'll go search some YouTube :-)  

[Brian Valente]

(btw, not sure your name? you aren't signing your posts) The best method to evaluate seeing conditions is unguided output for a few minutes via the guiding assistant. You can view the dec axis and judge for yourself the current conditions for your sky position. 

Brian

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/85b8d84c-2ef5-415d-9049-5199c482bc77n%40googlegroups.com.

[NY2KW]

I see repeating DEC patterns similar to those shown in Bruce's "Analyzing PHD2 Guiding Results – A Basic Tutorial" .After running the Guiding Asst, I accepted the recommendations for the algorithm and maybe Aggression =100 is too high and Min-mov 0.10 too low? 

Thanks in advance

Jerry 

  

[NY2KW]

Same oscillating pattern as on page 11 in Bruce's tutorial pdf.

[Brian Valente]

Jerry - There isn't a "here's the values you need" answer. It depends.

This is the part where you take your hunches and experiment. If you find it improves, go with that. Looking at your graphs (and again, the log files [not screen captures]) it may be more of a backlash compensation value, which will settle in over time, you need to be patient and let it adjust itself. 

Strainwave mounts are known for needing aggressive guiding and can be 'noisy'. As Bruce said, you have to decide how much time you want to spend fiddling with guiding, or imaging. Did you look at your images and were the stars acceptable? I'm guessing yes based on your image scale and the already low RMS but you didn't answer that

Brian 

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/61cb1a40-c629-4590-9d85-faa9d1ea1cd3n%40googlegroups.com.

[NY2KW]

Thanks Brian, I'm using a wide field scope but the stars magnified still look round and not bloated to my novice 'imaging' eyes. I dont know the best FWHM or HFR but over the past 2 months (about 8 nights of imaging) the last session whose log I uploaded seems to give me the lowest values in NINA.  I guess I'm at the "if it's not broken..." stage so will not do anything but keep going as is and just compare images over the next few weeks.  Please clarify " may be more of a backlash compensation value, which will settle in over time,..".  IS this a mechanical 'wearing in " process?  Sounds like I should run guiding assistant every week or so and see if it suggests further changes to the algorithms?   

Jerry

[Bruce Waddington]

As we've suggested, I think you're past the point where staring at the guide logs is useful.  The 5-minute section of the guide log in your screen shot is characterized by a higher rate of Dec direction reversals compared to the short sections before and after that.  Why did that happen?  Almost certainly because of changes in the local seeing conditions which can be quite volatile even on short times scales .  Here's an example:

Look at what happened shortly after midnight and then again shortly after 02:00.  Seeing monitors like this like to report mean values but look at the variances around the mean values in the two time periods I mentioned - all of that "bad stuff" is embedded in both your guide data and in your main camera exposures.  This example came from a high-altitude dark-sky site known for good seeing - depending on where you're located in Texas, your absolute seeing measurements will probably be higher than this and may be much higher.  The point here is to understand the seeing volatility and the futility of thinking that guiding could somehow make all this go away.

Getting back to the guiding session you were looking at, the higher rate of direction reversals triggered a higher rate of Dec backlash compensation adjustments and some incidents of over-correction while the algorithm analyzed the behavior and decreased the size of the pulses.  At the start of this guiding sequence, the compensation amount was 325 ms.  By the end of the sequence, the amount had been adjusted downward to 145 ms, all of which happened within 5 minutes.  This is a naturally occurring process and depends both on the seeing conditions and the behavior of the payload in different parts of the sky.  None of it had any effect on your overall results and isn't worth worrying about.  To add one other point, your total Dec guiding rms for the whole 3 hour session was 0.4 arc-sec while the RA rms for the same period was 0.44 arc-sec.  So what would have happened if we had done something magical and somehow reduced the Dec rms by 50%?  You would probably have ended up with elongated stars in your images because you couldn't make corresponding improvements in RA.

If you want to get an objective baseline of what matters and what doesn't, you can analyze the FWHM (in arc-sec) and elongation of the stars in your main camera images.  You can do that on a frame-by-frame basis using an average of all the stars in each frame.  Most people find that a 10% aspect ratio on star images becomes invisible in the final result, particularly with the advent of tools like BlurXTerminator.  Your minimal star FWHM is going to depend on your optics and the seeing conditions for that particular night.  If you take a sequence of 10-sec unguided exposures using the main camera,  you can get an average FWHM that probably represents the best you could hope to do on that night for those specific seeing conditions.  If you're doing automated focusing, these FWHM results might be available as a side-effect of the focusing procedure.  Then you can compare the results in your guided main camera images to see how close you are coming to the "ideal".  Of course, this assumes you are using a fine image scale with the main scope and are capable of critically sampling 2-3 arc-sec stars (Airy disks).  From your comment about using a "wide field scope", I suspect you probably have a coarse image scale and your star sizes may then have little or nothing to do with guiding.

Regards,

Bruce