PHD log viewer: Analyze selected; raw RA

[Al Moncayo]

Hello,

I’d like to understand why, in the PHD log viewer, upon issuing the “Analyze selected; raw RA” command, one always ends up with an RA waveform that has what appears to be exactly half of a sinusoidal period?

Attached is an example..

This seems to always be the case, no matter the equipment or length of log.

Al

[bulrichl]

Hi Al,

PHD2 Log Viewer does exactly what is described in the window title (that you cut off in your posted screen section):

"Analysis [b]** RA Corrections Removed **[/b]"

That means it represents how the RA deviations would appear whithout applying RA guiding corrections. These deviations are due to the periodic error of your mount.

Bernd

[bw_msgboard]

Hi Al.  I think you’re misinterpreting what you’re seeing.  First, the tool doesn’t behave in the way you’re suggesting, it works correctly.  Here’s a view of someone’s uncorrected RA over a period of almost 2 hours:

 

 

You can see many cycles of the mount’s PE.  

 

It’s hard to read the timeline on your screenshot but it looks like an interval of several hours.  I think what you’re seeing is uncorrected drift in RA, which is significantly larger than the uncorrected PE.  One common source of this is sag in the guiding assembly, which usually changes magnitude and direction depending on the pointing position of the scope.  This sort of drift can also come from polar alignment error, depending on where you’re pointing.  Without having a log and some real data to look at, it’s hard to say.  In any case, this isn’t likely to cause you much trouble, this sort of slow steady drift is usually guided out pretty easily.

 

Hope this helps,

Bruce

 

---

--
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/dd91357b-9bb9-423f-a17c-188071a498b6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Al Moncayo]

Hi Bruce,

Your comments are a segway into the detail of my inquiry. First some facts:

1. I don't have an issue, per se, with my guiding.

2. I use a standard mount with PE of about 7" pk-pk; I also use an encoder-based mount with presumably negligible PE (spec is 0.35" RMS)

With the non-encoder mount, I see the uncorrected PE over short-term, and it matches my expectations. Over a long time, as you suggested below, I would interpret the curve as RA low-frequency drift.

With the encoder mount, I do not see PE over short-term. Over a long time, I see similar curves to the other mount.

What is curious (perhaps coincidental), is that in pretty much in all of my guide logs that I've looked at over the past couple of months, the plots show essentially a half cycle of something that looks periodic and sinusoidal. I find it unlikely that no matter what my PA error is, what mount I use, over what long duration I look at this; all plots look kind of like what I first showed.

Another interesting thing is that I have one case where in the guide log corrections in RA were almost entirely West. However, in the log viewer analysis plot the same half-sine wave appears. If the tool is in fact computing the uncorrected RA, and if all guide corrections were West, then wouldn't the analysis plot show a line and not a curve? That case is attached.. (PHD2_GuideLog_2019-05-28_214801e.txt)

Al

On Tuesday, June 4, 2019 at 7:16:11 AM UTC-7, bw_msgboard wrote:

Hi Al.  I think you’re misinterpreting what you’re seeing.  First, the tool doesn’t behave in the way you’re suggesting, it works correctly.  Here’s a view of someone’s uncorrected RA over a period of almost 2 hours:

 

It’s hard to read the timeline on your screenshot but it looks like an interval of several hours.  I think what you’re seeing is uncorrected drift in RA, which is significantly larger than the uncorrected PE.  One common source of this is sag in the guiding assembly, which usually changes magnitude and direction depending on the pointing position of the scope.  This sort of drift can also come from polar alignment error, depending on where you’re pointing.  Without having a log and some real data to look at, it’s hard to say.  In any case, this isn’t likely to cause you much trouble, this sort of slow steady drift is usually guided out pretty easily.

 

Hope this helps,

Bruce

 

---

From: 'Al Moncayo' via Open PHD Guiding [mailto:open-phd...@googlegroups.com]
Sent: Monday, June 03, 2019 11:29 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] PHD log viewer: Analyze selected; raw RA

 

Hello,

 

I’d like to understand why, in the PHD log viewer, upon issuing the “Analyze selected; raw RA” command, one always ends up with an RA waveform that has what appears to be exactly half of a sinusoidal period?

 

Attached is an example..

 

This seems to always be the case, no matter the equipment or length of log.

 

Al

--
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...@googlegroups.com.

[Brian Valente]

Hi Al,

really nice guiding, btw  

i do see some variability to your guiding and resulting analysis

the second run on your first log ending 438e shows this:

it's interesting that the two segments in this log highlight your primary PE of around 230 sec, 

but the 7 hour log doesn't show that at all.

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/50a26b74-736d-46c9-be98-1e76ba07d771%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Brian 

Brian Valente

portfolio brianvalentephotography.com 

[Al Moncayo]

Thanks Brian.

The 7 hour log is with a CEM40EC. There should be no trace of PE, as the encoder eliminates it.

Indeed, the first one is a standard mount, a CEM120, therefore it has an expected PE at 240 second period.

Al

On Tuesday, June 4, 2019 at 10:12:32 AM UTC-7, Brian Valente wrote:

Hi Al,

really nice guiding, btw  

i do see some variability to your guiding and resulting analysis

the second run on your first log ending 438e shows this:

it's interesting that the two segments in this log highlight your primary PE of around 230 sec,

[Andy Galasso]

Al,

I agree with Bruce, the uncorrected RA motion is primarily due to polar alignment, though there may also be contributions from flexure and refraction. Check out the graph here: https://canburytech.net/DriftAlign/DriftAlign_1.html -- the graph is for Dec drift, but RA drifts in the same way (reference).

Congratulations on the wonderful guiding results!

Andy

[Al Moncayo]

Thanks for the references, Andy. I'll be reading them tonight.

Any response on the following?

"...I have one case where in the guide log corrections in RA were almost entirely West. However, in the log viewer analysis plot the same half-sine wave appears. If the tool is in fact computing the uncorrected RA, and if all guide corrections were West, then wouldn't the analysis plot show a line and not a curve? That case is attached.. (PHD2_GuideLog_2019-05-28_214801e.txt)"

Maybe I'm missing something silly...

Al

[Andy Galasso]

On Tue, Jun 4, 2019 at 2:01 PM Al wrote:

I have one case where in the guide log corrections in RA were almost entirely West. However, in the log viewer analysis plot the same half-sine wave appears. If the tool is in fact computing the uncorrected RA, and if all guide corrections were West, then wouldn't the analysis plot show a line and not a curve?

The analysis view is showing a drift-corrected result. Here's your RA data plotted with RA corrections removed but without the drift correction. The yellow line shows the linear drift, and the difference between the blue and yellow lines is what you see in the log viewer.

Andy

[Al Moncayo]

Andy,

Cool. That's what I want. How do I disable the linear fitting in the viewer (disable drift correction)? I don't see anything in the settings menu...

The drift correction is always just linear regression?

Al

On Tuesday, June 4, 2019 at 11:36:19 AM UTC-7, Andy Galasso wrote:

The analysis view is showing a drift-corrected result. Here's your RA data plotted with RA corrections removed but without the drift correction. The yellow line shows the linear drift, and the difference between the blue and yellow lines is what you see in the log viewer.

Andy

[Andy Galasso]

On Tue, Jun 4, 2019 at 3:27 PM Al wrote:

How do I disable the linear fitting in the viewer (disable drift correction)?

The drift correction is always just linear regression?

Currently the drift correction in the analysis windows is based on a simple linear regression and there is no option to disable it. I have added enhancement requests to the issue tracker (#10, #11)

Andy

[Al Moncayo]

Great! Thanks for the help Andy.

We can mark this item closed.

Al