Andy Mitchell

Please, someone tell me I'm not the only one who looked at the link and thought of emperor palpatine. You can not imagine the power of the dark site!

Does the 26mm one have a 1.25" or 2" barrel? The picture suggests the former, and the text the latter.

The primary colours, that match the receptors in the human eye, are RGB. However, if you look at an inkjet printer you are dealing with a subtractive colour model. You need just Cyan, yellow, and magenta inks. Each colour absorbs one of the RGB colours only, and allows other colours to be reflected by the paper. So think of yellow as "no blue", magenta as "no green" and cyan as "no red". I presume when we were all five and our school teachers told us in art class that yellow was a primary colour its a similar situation.

Do you have a tracking mount? Without one stacking software can struggle with high lens distortion, and you don't get any more distorted than a fisheye It's not a problem I've ever hit myself, but people I know have, when using zoom lenses with higher distortion. The short focal length will of course help to mitigate these problems.

CLS like filters are incredibly effective at suppressing the effect of the orange light from low pressure sodium street lamps. Should still be some help with high pressure sodium (yellowish white) lamps. Against white CFL/LED type light pollution they are pretty much useless. Narrowband filters help greatly with white light pollution on suitable targets. Taking mono Ha pictures of emission nebulas is the least hassle from my light polluted garden (white CFL lamps).

You will almost certainly want an appropriate field flatner if you want to do astrophotography with a refractor, and these often also act as a reducers. So you need to factor the speed increase and field widening of this when deciding if the scope is fast enough and gives you the right field of view. Many of the trendy small ED refractors have a native f6-f7ish, get reduced by 0.8ish reducer to about f5. f5 is pretty fast in telescope terms.

Could you combine this with an OTA from one of the mini dobsonians with fast mirrors for a super wide angle newtonian for astro photography, at least for sub APS sized CCD sensors? Something like the mead lightbridge 114 (4.5" 450mm f3.9). Suspect trying to do this with my camera would overload the focuser's weight bearing ability.

I've got the 10x50 version of those, they seem solid, ergonomic, they are sealed so the inside cant fog up. Optically I'm happy. They work well with or without my specs on. Only thing I dont like is the cover for the front lenses which has a habbit of falling off.

You can get a right angle bracket thingy like this:https://www.firstlightoptics.com/alt-azimuth/skywatcher-l-bracket-dovetail.html Usefull for binoculars, cameras, and telescopes with fixed mount/finder positions. I've got something like that for my porta ii, I think mine was celestron branded and has a wider horizontal bar. edit: Might have actually been an Orion one. Looks a lot like this one: https://www.telescopehouse.com/orion-dovetail-l-bracket.html

In a well collimated telescope coma should be zero in the centre and increase with distance from the centre. Is the image you posted cropped? If not you may also need to adjust your collimation. I would try prime focus as others have suggested.

Sounds like what many refer to as Synta* Grease (or Synta Glue). Exactly what sort of grease it is no one seems to know. Probably because people either avoid going near it or replace it with something utterly different * They make kit for lots of people including Skywatcher, Orion and Celestron.

When you read out binned pixels from a CCD you only get one set of noise from readout related sources. If you pull out four separate pixels one at a time and add them, you still get 4 times the signal. You also get 4 lots of readout/quantisation/whatever noise. You don't get 4 times the total noise as the noise signals are uncorrelated noise so they will in part cancel each other out. I think they sum something like: squareroot(noise_1^2+noise_2^2+......+noise_n^2) So comparing a single noise source to n of equal magnitude, we get an average total magnitude of squareroot(n). So for a single sub, on chip real 2x2 binning should have twice the SNR compared to down sampling to do the "binning" in software. Its been over 20 years since I studied this sort of stuff, so please forgive me if I've messed up the sums/theory Andy

If we imagine we have a 0.5 reducer and a mono CCD, cause it makes my example simple. Then imagine we bin 2x2 the unreduced case, and compare with 1x1 with the reducer. For the same exposure time we get virtually identical results with respect to SNR and resolution etc. Which I think is Olly's point. But.... If we do the "binning" by down sampling the final image (or subs) I'm pretty sure the SNR wont be quite as good. Anyone fancy doing a ton of maths?

This is how I've done it, but I hope someone knows an easier way... Load all your red light frames, right click on one and set it as the reference frame. Make a note of which frame you used. Add darks etc, and stack. Make sure the stacking mode is set to "standard". Save the final result. The trick is when you do green and blue, to also add the red reference frame we used earlier, and set again as the reference frame. Uncheck it though, so it isn't actually stacked. At the end you should have nicely aligned red, green and blue images. You can combine these in GIMP. Or, have a play with atik's dawn program, its probably still considered experimental mind and it seemed a bit flaky when I tried it. Seems to be fairly good at stacking and merging colour channels, but not so good at rejecting bad frames before stacking.

You can buy telescope mirrors, you can even grind your own at home (not quick), but I've not heard of anyone repurposing a dressing table mirror. Best of all, you can just buy a telescope for not much I believe you can save quite a lot if you want a huge telescope and you grind your own mirror, a video you may find interesting on grinding mirrors and building scopes: