N.I.N.A and Hocus Focus

[PRejto]

Has anyone used this software to adjust for tilt and Back Focus? I'm attempting this with my TEC180 and SX-46 camera (KAF6200 ccd 21.8 x 27.2mm)

I've got it running and it is pretty impressive, however there are aspects that I don't understand. The tilt adjustments are easy to understand (unless you have a tilt plate that doesn't have adjusting screws at the 4 corners). What is confusing is back Focus. Supposedly once tilt is mostly gone BF should be quantified and can be dialed in, however, in my case all I get is a recommendation to try moving out with no recommendation as to distance. So I've been moving out in .5mm increments. As I do the curve radius is increasing. I'm unsure what curve radius is but assume longer is better. So I started with a radius of 3057, 3885, and then a jump to 6854. When I look at CCD Inspector abberation reports, the best one is at the 6854 radius however the FWHM values are too high and seem to be increasing with distance away. Are the abberation reports just better because the stars are more out of focus?

I'm wondering how you have adjusted flattener distance? Looking at corner stars is one thing, but it doesn't tell you whether to go in or out. I've seen those diagrams (attached) that show how a field might look but I cannot see these patterns in any of my images. I suspect that because there is a bit of tilt left to expunge I'm getting non symmetrical corners. Advice?

Thanks,
Peter

[ChrisD]

I've been using Hocus Focus for a while now, maybe I can help.
If you've setup the focuser step size in Aberration Inspector you should be getting the backfocus error in microns. However, this value really isn't helpful as how far you need to physically increase or decrease the backfocus is dependant on the telescope and flattener optics.

Looking at the corner stars does, in fact, tell you how you need to change the backfocus. See image below. Can you post an image of what you're seeing?

What I do is use the take an in focus image and using the diagram to determine if I need to move in or out. Adjust and recheck. When backfocus is close then use Hocus Focus to fine tune.

I use the hyperbolic curve graph in Hocus Focus to adjust backfocus. Use the centre focus point as reference. When you have all 4 corner focus points around the centre focus backfocus is good. If the 4 corner focus points are grouped left or right of the centre focus point you need to adjust.

For tilt, the further the corner focus points are spread out, the more tilt you have.

If the corner focus points are tightly grouped and close to the centre focus, backfocus and tilt are ideal.

Just a warning for others, if you're using an SCT the star shapes in the diagram are reversed, move in becomes move out.

Chris.

[Retrograde]

Hi Peter,

I haven't set up my focuser step size (as per Chris's comment: )

but I'm still getting backfocus/tilt errors in "steps" so it still works as a relative measure between adjustments.
If you haven't already done so, make sure you update NINA and Hocus Focus to the latest versions as new updates are arriving quite rapidly at present.

[PRejto]

Thanks for your answers Chris and Retrograde.

I'm getting there slowly! This group has been very helpful: https://discord.com/channels/436650817295089664/967399798426706050

I now understand the concepts better. When I adjust BF it messes with my tilt so my aim is to hget reasonably close in tilt and then concentrate on BF, then go back to tilt and hopefully be done.

One initial mistake I made was to not have "sensor model" enabled in options. It seems to be off by default. It seems to improve accuracy. The other odd thing is I can run HF twice or more without touching anything, get decent R2 values but rather different results. One exprerienced guy on the discord group said that because of this he does 5 runs for every position and the takes the average! This is dedication, but perhaps necessary as I've seen too often that HF often doesn't agree with ASTAP or CCD Inspector.