Guide Log Data Analysis Question

[Bill Richards]

I'm trying to analyze data from the PHD2 Guide Log to gain a better understanding of what may be causing various guiding issues.  But I'm having some difficulty understanding how the "DEC Raw Distance" value is computed.

My calibration run yielded the following:

    West calibration complete. Angle = -75.2 deg

    North calibration complete. Angle = -174.9 deg

The first 10 steps in the first guiding run show this:

The "RA RAW Distance" numbers can all be computed accurately using the standard coordinate rotation formulas:

    RA Raw Distance = dx*cos(-75.2 deg) + dy*sin(-75.2 deg)

 

The corresponding "DEC Raw Distance" values should be derived using the corresponding formula:

    DEC Raw Distance = (-dx)*sin(-75.2 deg) + dy*cos(-75.2 deg)

But none of the values match those in the PHD2 Debug Log file.  Even if I change the angle to the north calibration angle (-174.9 deg), the values do not match.  Does anyone know how these values are derived?

[Bruce Waddington]

Hi Bill, I don't really know what's behind all this, it seems like you're making things much harder than they need to be.  So here's a series of answers for you to consider:

1.  If you need help analyzing your guiding results, we can provide that - it's what we spend most of the time on the forum doing.  To pursue that option, just follow the instructions here: https://openphdguiding.org/getting-help/

2.  If you want to analyze the logs yourself, that's fine.  But the best tool for doing that is this one: https://openphdguiding.org/phd2-log-viewer/ and a tutorial for getting started analyzing logs is here:

https://openphdguiding.org/tutorial-analyzing-phd2-guiding-results/

3.  If you have some reason you need to use a spreadsheet or some other app, you should be getting the data from the PHD2 guide log not the debug log.  The guide log already includes the correct transforms between the camera and mount coordinate spaces.  Those transforms have been in place for nearly 10 years, they've been vetted by multiple experienced programmers, and they are correct.

4.  The equations you've listed don't appear to apply to the problem at hand.  The guide star position offsets are represented by a vector of (dx, dy).  We compute the hypotenuse of that vector (hypo) and apply simple cos/sin scalars to create a corresponding offset vector in the mount coordinate space: (dRA, dDec)  where dRA = hypo * cos(x_angle) and dDec = hypo * sin(y_angle).  By convention, the camera x-axis is mapped to RA, the y-axis is mapped to Dec.

if you have other programming questions like this,  we think they are best answered by a study of the source code - it is, after all, open-source.

Regards,

Bruce

[Bill Richards]

Hi Bruce,

Thanks for the detailed reply. 

> 1.  If you need help analyzing your guiding results, we can provide that - it's what we spend most of the time on the forum doing. 

I've been getting a lot of help on various forums to improve my guiding, but I'd like to become better at diagnosing the issues myself so I can attack them in the field and not have to wait until my next monthly viewing session - hence, the detailed questions.

> 2.  If you want to analyze the logs yourself, that's fine.  But the best tool for doing that is this one: https://openphdguiding.org/phd2-log-viewer/ 

I already have the PHD2 log viewer and use it often.  But I'm trying to diagnoze issues that require "looking under the hood" and that requires obtaining a better handle on what is actually happening (with my mount and how PHD2 is responding) so I can determine whether the root cause is something related to the mount/payload or if I need to tune some PHD2 parameters.  (Full disclosure - yes, I'm an engineer.)

> 3.  If you have some reason you need to use a spreadsheet or some other app, you should be getting the data from the 

> PHD2 guide log not the debug log.

The spreadsheet table the I included in my original post was simply an easy way to display the data.  I'ma ctually writing Python scripts to analyze the PHD2 Log data.  And I am using the Guide Log, not the Debug Log.

> 4.  The equations you've listed don't appear to apply to the problem at hand...

OK, thanks for the redirection on that... I'll dig into that some more.  It's curious that the equation I listed DOES work for the RA Raw Distance, however.  Perhaps a coincidence of the geometry.