Calibration Step Size

[mj.w...@gmail.com]

My curiosity causes me to look at the GuideLogs posted here and on other on forums.

When I see a Calibration that has too few/too many steps despite sensible guide rates,  I sometimes suspect that the Cal Step is wrong.

Is there a formula for calculating it, without having to open up PHD2 and run the Wizard ?

Michael

Wiltshire UK

[Bruce Waddington]

The easiest way is to just run PHD2, don't connect to any of the gear, then use the 'Advanced' button on the Advanced Settings/Guiding tab.  That will bring up the calibration-stepsize calculator where you can enter whatever values you want for focal length, pixel size, guide speed, binning, etc.  Alternatively, the code is simple, something you could put in a spreadsheet if you are so inclined:

// Based on computed image scale, compute an RA calibration pulse direction that will result in DesiredSteps
//  for a "travel" distance of <distance> pixels in each direction, adjusted for declination.
//  Result will be rounded up to the nearest 50 ms and is constrained to insure Dec calibration is at least
//  MIN_STEPs
//
//  FocalLength = focal length in millimeters
//  PixelSize = pixel size in microns (un-binned)
//  Binning = hardware pixel binning factor
//  GuideSpeed = guide rate as fraction of sidereal rate
//  DesiredSteps = desired number of calibration steps
//  Declination = declination in degrees
//  ImageScale = computed image scale (arc-sec/pixel)
//  *pStepSize = computed calibration step size, milliseconds
//

    double ImageScale = MyFrame::GetPixelScale(PixelSize, FocalLength, binning); // arc-sec per pixel
    double totalDuration = (double) distance * ImageScale / (15.0 * GuideSpeed);           // 15 arc-sec/sec is sidereal rate
    double Pulse = totalDuration / DesiredSteps * 1000.0;                 // milliseconds at DEC=0
    double MaxPulse = totalDuration / MIN_STEPS * 1000.0;            // max pulse size to still get MIN steps
    Pulse = wxMin(MaxPulse, Pulse / cos(radians(Declination)));      // UI forces abs(Dec) <= 60 degrees
    *pStepSize = (int) ceil(Pulse / 50.0) * 50;                      // round up to nearest 50 ms, too-small pulses can lead to calibration problems

Problems with the calibration step-size are nearly always the result of user mistakes, especially trying to manually change configuration settings without running the new-profile-wizard.

Bruce

[mj.w...@gmail.com]

Hi Bruce

" The easiest way is to just run PHD2"

Err, yes, it would seem so.

I'll try 5arcsec/pixel,  0.5 GuideSpeed,  12 steps,  Dec = 10

 double totalDuration = (double) distance *5/ (15.0 * 0.5);       What's "distance" ?

double Pulse = totalDuration / 12 * 1000.0;     Okay

double MaxPulse = totalDuration / MIN_STEPS * 1000.0;         What figure should MIN_STEPS be ?

 Pulse = wxMin(MaxPulse, Pulse / cos(radians(10)))    What's wxMin ?     And  Pulse is the result AND in the formula ?

 *pStepSize = (int) ceil(Pulse / 50.0) * 50;    Okay

Michael

Wiltshire UK

[Bruce Waddington]

Well, you asked. J  Really, if you just use the calculator, your work will be done.  I just counted – it takes you 4 mouse-clicks to launch PHD2 and be in the middle of the calculator dialog. 

 

Distance is the number of pixels to be moved in each direction, MIN_STEPS is 6, wxMin is a Minimum function

 

Just use the calculator… J

 

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/eb2371cf-ca54-4d6a-a768-1fa55c7b0aa4n%40googlegroups.com.

[mj.w...@gmail.com]

Hi Bruce

I don't have PHD2 on my "indoors" PC that I read forums on, but installation is what I'll do.

The formula is an eye-opener to those who think guiding is simple.

Thanks

Michael

Wiltshire UK