Feature Tracking during Partial Solar Eclipses

[pier_dmx]

Dear all,

As you know, during a partial solar eclipse, the Sun is progressively cut by the Moon.
So, its center of mass varies as the eclipse progresses.
If I use the feature tracking with the option of Center of Mass, what will be centered over time is the center of mass of how the Sun looks like at that moment.
This would basically move the target.

Is there a solution to it?

Second question: before I buy the AZ GTi, is feature tracking fully supported or is field rotation affecting it? Thanks!

[admin]

Hi,

center of mass feature tracking will definitely be affected by the change in shape of the visible part of the sun's disc. SharpCap 4.2 has a new feature tracking mode where it uses edge-of-disc detection, but that is new, relatively untested and was really designed for partial solar/lunar views rather than a partially obstructed whole disc.

I have no direct info on the AZGTi - I have personally used feature tracking on a Skywatcher EQ6 mount using EQMOD, but obviously in EQ mode. I know people that have used in on AltAz mounts (like a Celestron SE mount), but no direct info about your proposed model.

I don't think field rotation will be an issue - what Feature tracking cares about is the relationship between axis movements and camera image movements, and those will remain consistent (unless you are using a derotator or similar).

cheers,

Robin

[pier_dmx]

Thanks for the answer, Robin!

From a mathematical standpoint, the software should:
1) detect the edge
2) compute an imaginary circle based on that edge
3) if the imaginary circle is NOT in the frame, return an error and stop; otherwise, continue
4) track and keep in center the center of that imaginary circle

Or something like this, done more efficiently.

I'd be happy to test a feature like this.
Maybe it could be the answer for the 12th of August!

[pier_dmx]

P.s.: the external edge is always the longer edge.

PENUP_20260712_222346.png (79.6 KiB) Viewed 675 times

[admin]

Hi,

the software looks for a part circular curve where the bright area is on the inside and the dark on the outside of the potential circle. It then works out the position of a point on the detected arc at a consistent 'clock angle' from the center of the circle and will keep that calculated point stationary. If you work through the initial 'auto alignment' procedure of the Solar Mosaic Planning tools then you can see the disc edge detection in action tracking the movements of the sun (or the edge of the sun's disc) in view as the mount moves - the same algorithm is used in the disc detection in feature tracking.

Feature tracking using the disc detection will probably loose tracking after the moon has passed across maximum coverage and need to be re-started (stop, then restart monitoring, then tracking - no need to recalibrate). That's because the angle picked for the reference point before maximum will no longer be part of the bright edge of the sun after maximum as the moon moves away from the sun in a direction that will cover the reference point.

This is all a bit experimental I'm afraid - using disc detection for this is brand new, so be sure to have a backup plan if it misbehaves during the eclipse.

cheers,

Robin

[pier_dmx]

Yes, for this eclipse:
Plan A would be to attempt polar alignment the night before on the top of the place where I'll stay. This should prevent the need for active tracking and allegedly save some FPS.

Plan B would be to go for active tracking. Even with a moving center of mass, the target should not get out of the frame.
Before totality, I would disable active track, jog manually, and let it go until the start of the next partial phase.

Field Rotation during Solar Eclipse 2026.png (593.28 KiB) Viewed 631 times

But for the future, do you think of implementing a geometric tracking feature for eclipses?

[admin]

Hi,

I have to admit that I don't really have any plans to add specific eclipse related functionality to SharpCap, although I am happy to general functionality that can also be useful during eclipses.

The problem is with eclipse related features is that you simply can't properly test them - or refine them while developing for that matter - under realistic conditions. Also, because the eclipse is basically a 'one chance' event, it's important that the imager is 'on the ball' ready to take corrective action if something goes wrong. You are thinking ahead about how to image the event, so if something does go wrong with the automated tracking, you will have a good idea of how to fix it.

Now imagine if there was some automated 'record my eclipse' tool in SharpCap... That would be a dangerous thing because people would inevitably try to use it without having given any thought to how the imaging of an eclipse works, and would have no idea how to correct things in a hurry if/when things went wrong.

cheers,

Robin

[pier_dmx]

Hello, understood.

I think I have one last question:
What is this option for?

Immagine 2026-07-14 200521.png (144.92 KiB) Viewed 586 times

What is the difference between guiding while it is ticked and while it is not?

Thanks,
Pierpaolo

[admin]

Hi,

if that is unticked then SharpCap will try to preserve the current position of the object being tracked in the frame. If it is unticked then SharpCap will try to move the object to the center of the frame. I think this only really applies when you are using the 'Center of Mass' option for movement monitoring.

The idea was that if you are maybe imaging Jupiter with 3 moons on the right, then you want to have Jupiter to the left of the frame with room for the moons on the right hand side. By ticking this option, SharpCap will keep Jupiter there to maintain the framing, rather than the normal default of moving the planet to the center of the field of view.

cheers,

Robin

[pier_dmx]

Thanks for the reply,

Then maybe this will be beneficial for the eclipse.
Anyway, I downloaded 4.2 and I will test edge tracking.
It's good to see the migration of settings and licence automatically.

Great job!