Sun in Ha - 20th Apr. Lessons

[jezhughes]

Hi all

This is my first session with my Solarmax so be nice!

Saturday was a great day, i spent a while just setting up for the first time and getting used to tune the new scope. I then gave some imaging a go. Focusing seemed to be the hardest thing, couldn't really get it that crisp but the seeing wasn't too great.

None the less, i took about half an hour of data by the end.

Next was trying to work our Registax, which was tough.. I need to find some good tutorial videos really (suggestions?). Anyway, this is about 250 frames from 790. It was more of a 'find my feet' session and learning some ropes. I took some flare clips too but not sure how to process them to a usable state, do you just take 1 frame and use that? The seeing was causing a lot of distortion on the flares so stacking them just made a blurry mess..

As always, i walk away with more questions than answers for now.

Oh, and sorry about the crop. Annoyingly my sensor on the ASI is a tiny bit too small despite me checking before i bought it. So that's a problem... The top gap is added.

Any advise would be great, especially any Registax masters

[michael.h.f.wilkinson]

Good start. Focusing is difficult, and Registax often borks on solar tracking. I find I need to select points for tracking manually before getting it right.

[Montana]

That's a great start

On the Registax site there are plenty of tutorials to help, I learned from there. Even simpler is to use Autostakkert 2, only 2 buttons to press on that one. Again I think there is a link to a good tutorial from there too.

Alexandra

[michael.h.f.wilkinson]

Autostakkert is also good on solar, and by now I prefer that. I do sometimes need to hand-pick alignment points. This may be due to the banding I sometimes get with the LS35THa/ASI130 combo, which means that the program starts aligning the bands rather than the true image features. I am not sure if the banding is an effect of the tilt-tuned etalon or of the camera 9or the combination).

[DrRobin]

Hi Jez,

What a great start, well done on your first image. I am not sure if you use Registax 5 or 6, the alignment process is different. I use both and Autostakkert (AS!2), so the following are my guidelines.

Registax 5 or 5.1. - Use a medium sized alignment box and centre on the most prominent feature, probably one of the sunspot groups. You have to scroll through the frames to select a good one or the best first. Registax 5.1 works well on all images.

Registax 6 - Scroll through the frames first to find the best frame. If you use auto-alignment placements, make sure you deselect the weakest points. Better to select around 15-20 alignment points manually. Go for the black sunspots and any other fine detail. When the quality/stability of the originals are very good then Reg6 produces a better stack, otherwise it produces stacking artifacts.

Registax stacking - Stack above 90% or about 1/4 to 1/5 of the frames. If I take 1200 frames I try to stack around 300, but if the quality is really good, might stack a little more to reduce noise. I use drizzle to reduce pixel size.

Autostakkert (AS!2) - select planet for full (or nealy full disc) and surface for the close ups. I select noise robust between 4 and 6 depending on the quality of the original. Hit analyse and then manually select around 15 to 20 alignment points, more if you can manage it. I try to select the sunspots and any other fine detail. More alignment points won't produce a better stack if they aren't consistent. If the quality is low then you will get stacking errors so go back to Reg5.1.

Like all things the better your original the better the stack, so double check focus and tuning before hitting the record button. Take as many AVIs as you can manage and hopefully you will get one or two good ones.

Robin

[sologuitarist61]

Nice image

[Knight of Clear Skies]

Nice image. Thanks for the Registax tips DrRobin, I'll try those settings the next time I get a chance to image the Sun.

Can't offer much in the way of processing tips myself, but for false-colouring you might want to try adjusting the colour curves rather than just applying a tint. Here is a solar mosaic I put together the other day:

[markt]

Autostakkert (AS!2) - select planet for full (or nealy full disc) and surface for the close ups. I select noise robust between 4 and 6 depending on the quality of the original. Hit analyse and then manually select around 15 to 20 alignment points, more if you can manage it. I try to select the sunspots and any other fine detail. More alignment points won't produce a better stack if they aren't consistent. If the quality is low then you will get stacking errors so go back to Reg5.1.

Hi Robin. Sorry, completely disagree with most of this. I always select surface for all solar shots. Noise I run on 3, 2 if it is good seeing, in terms of alignment points I go for around 2000+, maybe around 1000 if I have a shed load of avis to process and time is of the essence. If there are stacking errors it's more than likely the settings you are using.

[DrRobin]

Hi Mark,

Interesting. I always select surface unless I take the full disc in one frame, then it is the same as a planet, yes? If you are using 1000 plus alignment points are these manually placed ot auto placed? I have found with the auto placed, it just places them on a grid and as such doesn't always get the strongest features. Either way I have just tried a full disc with around 20 strong points (planet) and 1600 points auto placed (surface) and both pictures had stacking errors, yet when aligned and stacked in Registax they came out just fine.

I would love to know what I am doing wrong, or what you are doing right, but that is my experience.

Robin

[Montana]

Wow! 2000 alignment points!!!

I always tick surface whether full disc or close up for solar (only because Emil told me so) and it works every time unless there is a cloud. For alignment points I have 30 selected, then click set alignment points. I find the lower the number is it gives more alignment points for me, rather than a higher number? Yes, you should have a grid of points covering everything, this is important as it aligns the whole area not just regions of interest and this is what gives you an even processing across. If it has placed some points on an area you don't like, noise or a dust bunny, then right click on it to remove that point. For noise I keep at default 3 all the time.

Regards

Alexandra

[anthony]

Fine start there! Shame about not managing full frame. There is a pretty cheap reducer from modern astronomy that might be worth a go.

I too have found that AS!2 performs better with manually selected alignment points on recognisable features. I get a poor (unusable) stack if i let it auto select a load of them. This is more of an issue when I have poor polar alignment and appreciable drift and have to use the hand controller to keep the sun on chip (LS60/DMK8618 mono).

[markt]

Just to check we're all talking about the same thing here: I'm talking about having ~2000 alignment points as a result of having the AP SIZE set between 25-40.

[jezhughes]

Thank you for everyone's tips! I'll be sure to give AS!2 a go.

I actually went back to Registax and realised i didn't adjust the wavelets upon processing... . I must have missed it and not realised that's evidently a critical step. Anyway.. Here is a version where i've taken people's advise on board, using only 20-30 align points on detail and about 30% of best frames. I've also added the prom layer on to this, im not entirely sure how you do those in Regstax so i'll research that.

I used a photo filter for the colour, i'll look in to doing this via channel adjustment in the future but i felt like a photo filter works out quite well anyway. Rest of processing was done in photoshop. Added a very subtle high pass and tweaked shadows.

I hope this is an improvement on the same data, it looks as so for me.

[DrRobin]

Hi Jez,

Very nice indeed, a steep learning curve but well worth it. The proms look okay good as well and I would be very pleased with that. I have just compared your latest picture with the first one and now that you have said that the first wasn't sharpened, I think the stacking was fine on the first, just needed sharpening.

Out of all the pictures I take, I have the least amount of success with AS!2 on proms, I just don't seem to be able to get it right (about one of of ten works for me) so even if I use AS!2 for the surface detail, I nearly always use Registax 5.1 for the proms. I just draw a large rectangle round the most distinct prom and let it run. I do often watch it and if the prom is drifting in the box, stop it and re-start.

I tend to find a stacking program and settings for a particular picture that work for me and then use that until I figure out something better. Given an infinite amount of time, you can always fiddle with settings and re-stack pictures and get them better, but it becomes one of diminishing returns.

Now that you are hooked, your next pictures will just get better.

Robin