ollypenrice

I use three eyepieces. That's it. In my SCT, with long focal length, I use only two. I never ever feel the need for more, but the ones I have are top quality. Olly

Something like this. I think it's quite often necessary to attach two items without the attachment interfering with their alignment but I don't know what term to use in a search. Olly PS @DaveS I dare say Dave might know.

We want to use a bar or plate to bridge between one scope on a dual rig and the other. The physical tubes are not necessarily perfectly parallel (optical axes, drawtubes etc may introduce slight tube misalignment) and they are offset relative to the saddle plate because one rides on a tilt pan adjuster. What we need are self aligning joints (some kind of ball section in a cup section) which would allow the tubes to be aligned before these joints were tightened. I can visualize the blessèd things and am pretty sure I've had hands on them at some time. I remember two thick washers, if you like, one machined into a cup shape and the other into a segment of a ball. They sat between the two items to be aligned and tightened. Does anyone know what they are called or where they might be ordered? Thanks, Olly

I did wonder about ball joints at the attachment points. That's what you need to avoid so many iterations, I think., if you could find the right kind. We don't need the range of angles provided by a camera ball head but a joint using just a section from a sphere would provide the range of angles required. I found some high end industrial self aligning joints on the net but nothing suited to a small application like this. They probably exist, though, since the engineering problem is highly predictable. We need someon in light mechanical engineering to enlighten us. Olly Edit:

@FLO What we need for dual rigs is a bracing arm with some kind of angle-adjustable joint at each end. It would pass from one scope tube ring to the other, find its own angle, and then be bolted up firmly. I think it would need only two simple components at manufacture, the L bracket and the arm. The customer would buy two L brackets, of course. (The tubes are naturally parallel but are offset relative to the saddle plate because one is on the tilt-pan adjuster. The arm would need slotted holes to allow for a range of tube separations. Slight variance in the fore-and-aft position of the L clamps would be taken care of by tightening them onto their tube rings last. Two potential sources of flexure remain intractable in dual rigs, though. Mirror movement and drawtube sag. The RASA might suffer mirror movement (or might not) but it has no drawtubes. Olly

Agreed. You can put your hand into a hot oven without being burned, even at atmospheric pressure, but you can't touch a metal object in the oven even though the gas and the metal have the same temperature. There are more atoms and molecules in metal than in gas... Olly

Like Mutley earlier on in the thread, I would look into a David Lukehurst Dob. He has a solid reputation. It can be very instructive to put a maker's name into Google followed by 'customer service...' You're right to uncouple visual observing from astrophotography because the requirements have astonishingly little in common. Olly

I think that worked well. You have have a distinction between red and blue stars which gives the image a more natural look and is truer to the target. Maybe the hue in the blues could have a tweak to make them slightly less green but being wary of going into magenta. I reckon that, for star colour, less is more because too much signal burns everything to white. This is the first image of the Lobster Claw which I've found engaging. Olly

I think Paul's going to try a focus routine which uses multiple stars. The old rule was always to choose a star at one of the four intersections of the 1/3 lines. This is my first time with autofocus as well. Olly

It really is a tricky one, I agree, but this is most attractive. Olly

What happened? Good question, but basically the improved collimation made the corners worse. Why? Not a clue. Last night we removed the Celestron window and it didn't make a huge difference. However, we then rotated the chip by 180 degrees and there was a vast improvement. My thinking on this is shown in this magnificently high quality sketch : The beam is directed by tilting the lens cell. On the left the tilt in the cell and the tilt on the chip go the same way, causing maximum deviation from orthogonality. On the right the camera has been rotated by 180 degrees so the tilt on the chip lies opposite to the tilt in the beam, correcting the angle between beam and chip. I suspect that both errors, in themselves, are very small but that they become significant when added together. In any event we think could live with what we have, especially if we could get a more convincing autofocus software than the one we're using now. That shouldn't be difficult and focus visibly improves the corners when it has no obviously visible effect in the middle. Comments on my theory are more than welcome! I think only testing will show what your dual rig can do. Because our dual Tak 106 just worked like a charm from night one, I hoped the dual TEC140 would do likewise, but it doesn't. The change in resolution brings differential flexure. The RASAs are physically short and, with one possible exception, are also physically very stiff, having no draw tube to sag. The uncertainty concerns the primary mirror. How much might it move and how much would such movement matter, given that it's spherical? In the end, there remains the possibility of simply using two mounts for two scopes. Thanks again for your input, it's very helpful and reassuring. Olly

Agreed. Great way to check! Thanks. I did something along those lines but yours is the way to do it. Thanks. Olly

Optimism proved premature! This is the current state of our collimation. Star in the centre of the image. What do you think of it? Close, bad, whatever you think? It is not giving us acceptable corners in either landscape or portrait, unfortunately. In fact we had better corners with worse collimation! This is not proving an easy instrument to set up... (I havevn't routed the cables carefully yet, hence the two dark lines.) Olly @gorann

I like the image. I've tried this in HaLRGB and it was as dull as ditchwater, just a uniformly red affair with one dimensional colour. I didn't even keep it, let alone post it. Here, though, the upper claw, with a significant OIII component, has modelling and three dimensionality which make it very attractive. Since HOO closely replicates RGB I agree with Martin that a handful of subs for RGB stars would be perfect. Alternatively you could just lose the greenish stars by lowering their saturation. Olly

Par for the course and very easy to fix in pre- or post-processing. Lots of software products have a hot pixel filter at the stacking stage and, with a DSLR, you really do want to use a large dither anyway to get rid of colour mottle so that will kill them a second time! Olly

Old post or not, this debate, and the misinformation it never fails to generate, is forever young! For instance, a few posts above this we see the assertion that two 6 inch scopes collect different amounts of light because they have different focal lengths and, therefore, different focal ratios. This cannot be true. When a six inch diameter beam of incident light arrives at the 6 inch objective of a telescope, how does it know the focal length of the objective it is about to pass through? And even if it discovers that the objective is a fast F2, how does it increase its photon count with this discovery? No, exactly the same beam passes into the objective whatever its focal ratio. I think the confusion has a single and simple origin, the camera lens. Regular photographers use the terms aperture and focal ratio interchangeably. They can do this because, when they do so, they are discussing a lens operating at a fixed focal length and they are changing the F ratio by changing the aperture. (They open or close a diaphragm in front of the objective.) When they do this, it is perfectly obvious that they alter the amount of light arriving at the chip and so they alter the required exposure time. This is precisely what we are not doing when we place a six inch F5 scope next to a six inch F10 for comparison. We are not changing the aperture, so there is neither more light nor less joining the party. Sure, we are changing what we do with it but here we move into sampling rate, or how many pixels we place under the photons from an object - but, for now, let's just de-mystify F ratio. Olly

I wouldn't underestimate the contribution of a cooled astro camera... Olly

My reason was pure ignorance! I'm only a PI dabbler. Thanks for this. Olly

I always try to get up to 100% luminance to avoid wasting its signal. It can be difficult but there's an iterative process in Ps whereby you add partial lum as a layer, slightly blur it, increase the saturation of the RGB, flatten and repeat several times, increasing the lum percentage as you go. On the last application of the luminance you should be able to get to 100% and, this time, you don't blur it. I also felt that the L to RGB ratio in the data was not going to find IFN colour. In my own IFN shot from years ago I deliberately set out to find colour if possible and shot a far higher proportion of RGB. I don't think this matters: the IFN (and other low level signal) in this data is really exciting. Olly

Very good. I particularly like the Cigar and your resolution in the Ha filaments. Perhaps, like me, you saw no point in using the RGB as an extra source of lum? Olly

We collimated the scope last night and the corner elongations are gone. Relief! (And, in my case, surprise, I must say. I'm still not sure why mis-collimation would affect only two diagonally opposed corners but it did. This confirms Goran's experience above.) I'm not going to say that I got the out of focus doughnut perfectly symmetrical but I was going round in circles after a while so called a halt at 'pretty good.' We can always go back to it once we have all the rest of the rig running well. To collimate the RASA you need access to the collimation screws on the central lens cell. A full size camera covers theses so you can use your guide camera for collimation. (Another good tip from Goran.) Thanks for the input on this thread, everyone. Olly

We're helping a beginner here so this is really for another thread, but let's just say that I don't agree. Actually a thread on assorted ready made and hand-made stretches would be a good one, I think. Olly

By PS, do you mean Photoshop? And are you referring to the Photoshop Levels stretch? If so, yes, it's a standard logarithmic stretch. However, By using Curves instead of Levels you can shape any stretch you like. There are also plenty of pre-packed stretches available in other programs. However, in my view the completely hand-made Curve stretch gives the finest control but is the hardest to execute. Olly

Software bin, yes. However, since the instrument will probably support the higher resolution, it might be a shame not to use it. Or, to put it another way, you could get the reduced resolution more cheaply and with more portability with a smaller scope and mount. My main point was to to draw the OP's attention to the changes brought by reduced pixel sizes since these can lead to both mount and scope being smaller. Olly

Let's look at the histogram of your image: We see the histogram peak is jammed up hard against the left hand edge of its frame. That peak contains the faint and interesting stuff in an astrophoto. The flat line to the right contains the stars. By moving the lower left hand slider in too far you have thrown away (or 'black clipped') much of you signal. Alacant has been careful not to do that so his histogram will look different and will have a short flat line left of the peak. Yup! If you subtract your histogram from Alacant's, what is left is what you threw away by black clipping. All that faint stuff is found on the left hand side of the histogram peak. That little flat line left of the peak must always be there. The temptation is to black clip in order to cut out gradients but resist that temptation! Olly