EAA first light with new kit

[AKB]

New everything... observatory, mount, Hyperstar, ... it was also rather novel having a clear sky on the night of August 10, between about 11pm and 1am.  On the down side, there was quite a bright moon.  Still, you can't have everything.

This was a shake-down session with all the new gear, so nothing really adjusted or tweaked to any great extent, and I wasn't expecting anything special.  As usual, I just set the exposure to 60 seconds (except for M51 where I got it wrong) and pointed at a few things.  Came to an end as the laptop ran out of power (mains power since connected!)  Some findings from the session:

Observatory:

• What a joy not to have to set everything up (or break it down again.)

Mount:

• I'd done a rough PA on an earlier night, but checked it again with SharpCap (which says it got me to within  ~45")
• My previous mount (HEQ-5 with belt mod) was OK, but this one (Avalon M-Uno) is spectacular in stability, pointing accuracy, and zero backlash.
• Didn't bother with guiding

Hyperstar:

• I checked the collimation on Altair, and without any adjustment it looked fine
• focused with a Bahtinov mask - so easy to get spot on the with Avalon FOCS focuser drive controlled by the mount
• made no attempt to route the camera cable neatly (something for later)
• need to improve my fixing of a dew shield (think it led to some vignetting)

Here's a few processed images from the night.  Not very good - much room for improvement - I think I've remembered to do the left-right flip in each case.  When I checked everything in the morning, I was appalled to see how much grime was on the corrector plate (there was a heavy dew by the end of the session - need to investigate options to remove this).  All this with an Ultrastar mono camera and no filters (with bad pixel map, bias, but obviously no flats!)

• M51, 22x 30s. ... because I know what it looks like from previous equipment
• M101, 11x 60s. ... ditto
• M13, 5x 60s. ... lots of stars
• Pelican nebula, 11x 60s ... I had no idea what to expect here: quite pleased with the result.  Ha filter next time!

Anyway, it's a start.  Stars are not great - especially the bright ones - but adjustments yet to be done.

 

   

   

 

[andrew s]

A very good start. There is no getting away from the advantages of an observatory for getting up and running without delay. Same goes for closing up at the end of a long night.

Rehatds Andrew

[HiloDon]

Very impressive, AKB!

Are you using Paul's SLL?

Don

[AKB]

24 minutes ago, HiloDon said:

Are you using Paul's SLL?

Hello Don!  For the EAA session itself... but of course!  Here's a couple of the 'live' screenshots.

You'll see that the wish I made here...

On 3/27/2017 at 21:33, AKB said:

My ultimate platform would, I think, be either a Hyperstar for my C9.25 or a RASA, but they're a bit heavy (you have experience with both, yes?)  The mount I'd want to use is an Avalon M-Uno.  I just need to know it would be worth the investment.  I think an observatory comes first because setting up, although fun, is a big time-waster. 

...has come true in every respect!  Lucky me, I really now have no excuses.  The Ha filter is in place for this evening, but I don't know if the weather will play nicely.  We also have some meteors to look out for...!

 

 

 

 

[RobertI]

Very impressive results from a lovely setup. Well done and keep posting your progress.

[ollypenrice]

Promising start. Regarding the cables, there are two schools of thought. You can make a false spider and run them along one of the arms, so giving symmetrical Newt-like diff spikes. Or you can make the cables take a curved path across the corrector plate. This should spread the diffraction effects over the whole image making them insignificant. If you could contrive it you might be able to move the path of the cables round the plate a little between subs and then they'd Sigma out entirely in the stacking.

Olly

[AKB]

17 minutes ago, ollypenrice said:

If you could contrive it you might be able to move the path of the cables round the plate a little between subs and then they'd Sigma out entirely in the stacking.

Ooh, thanks for that!  Neither seen this suggested before, nor considered that it might be done.  There's a challenge... some sort of reciprocating rotator on the cables/dew shield.  

Perhaps @Gina could 3D-print me one?!

[ollypenrice]

9 minutes ago, AKB said:

Ooh, thanks for that!  Neither seen this suggested before, nor considered that it might be done.  There's a challenge... some sort of reciprocating rotator on the cables/dew shield.  

Perhaps @Gina could 3D-print me one?!

I think it would be fairly simple to make. It would probably be best if integrated into the dewsheild, which you will surely need anyway. A curved arm in flat alloy could be fitted to the inside of the dewsheild and the cables attached to that. The arm would be unsupported at the centre. You could then rotate the dewsheild between subs, moving the cables with it.

If you just run the cables in a fixed straight line you'll get pretty odd diff effects on stars.

Olly

[Martin Meredith]

Congrats on the new setup!  Great results -- the Pelican in particular. Does this mean you're going to pension off your 8" Quattro or will that get a turn on the mount at some point? As  a Quattro owner I'd be interested in hearing your opinions on the difference between the two scopes. Having a permanent setup frees you from the constant collimation requirements of the fast reflector...

When you say 'processed' images from the night, what processing did you do?

cheers

Martin

[AKB]

2 hours ago, Martin Meredith said:

Congrats on the new setup!  Great results -- the Pelican in particular.

Thanks, yes, I was pleasantly surprised by that result.

 

2 hours ago, Martin Meredith said:

Does this mean you're going to pension off your 8" Quattro or will that get a turn on the mount at some point?

No, and yes, respectively.  I like the Quattro and its collimation seems relatively stable.  It gives me a FL of 800mm, which is nicely between the 1575mm of the SCT with a x0.67 reducer, and the 540mm of the x0.23 Hyperstar.  I'm trying to bracket my 'final' destination, which I think is a 600mm FL from an F3 Riccardi-Honders 200mm.

 

2 hours ago, Martin Meredith said:

When you say 'processed' images from the night, what processing did you do?

Slightly different for each, since every image needs its own TLC.  For the Pelican, it was something like this:

• background extraction (rather important, see attached image of raw stack and calculated background clipped to the same level.  Combination of moon and Hyperstar?)
• stretching with curves
• small amount of noise suppression (GREYC, in this case)
• multi-scale unsharp mask (a version of APF-R)
• bilateral filter (for more noise suppression.)
• ONR ('Olly Noise Reduction' - a raising of the low-end curve towards the desired background level)

All processing done in a combination of Nebulosity and my own code in MATLAB.

[AKB]

An amazing thing... well, it is to me.  I thought I would practice taking flats with the Hyperstar lens and re-assembled all the kit, having stripped it all down yesterday (separating Hyperstar from 'scope and camera from Hyperstar.)  So I just re-assembled it all and tried it in the gloom of the obs'y, without any uniform light source, and got this...

...but what struck me was that the bunnies looked to be just about in the same place as for my images from the night before, so I gave it a go, and, lo and behold, spot on!  Or rather, "spots gone"!!

Before and after:

   

 

Received wisdom suggests that moving the camera would ruin the chances of this.  However, the Hyperstar is threaded to the corrector plate, the camera is threaded to the Hyperstar, so it must all go back together in the same way.  What's also surprising (to me) is that I had inserted an Ha filter in the chain (in anticipation of a clear night last night, which didn't happen) so the flats were taken with that in the chain too.

Wonders will never cease.

 

 

[Robrj]

If the dust was on the camera sensor, moving it wouldn't matter because the dust is over a specific pixel.  It will be over that same pixel regardless of the orientation of the camera (it moves with the camera).  Since the camera will readout the same pattern regardless of its orientation (it doesn't know it's upside down), the dust donuts will always line up with the dust donuts on a previous image.   If you took another image of M51 where the camera was 90° from the original image, the galaxy itself would be rotated (so it wouldn't stack) but the dust donuts would be in the same position.  However, any dust on a a filter would require the orientation to be exactly the same, since twisting a filter can move a dust speck over a different pixel.

[AKB]

13 minutes ago, Robrj said:

If the dust was on the camera sensor, moving it wouldn't matter because the dust is over a specific pixel.

That's certainly true, but dust on the sensor is so close that it makes dark dots.  I'm fairly sure that this is dirt on the corrector plate, or possibly primary, because of the size of the bunnies.  There's actually a calculator somewhere on line where you can put in details of your optical system and find out exactly the distance from the sensor.  Must find it and give it a go.

[HiloDon]

4 hours ago, AKB said:

That's certainly true, but dust on the sensor is so close that it makes dark dots.  I'm fairly sure that this is dirt on the corrector plate, or possibly primary, because of the size of the bunnies.  There's actually a calculator somewhere on line where you can put in details of your optical system and find out exactly the distance from the sensor.  Must find it and give it a go.

What Rob wrote is correct and I suspect that the dust is on the Ultrastar's protective window over the sensor.  Try a blast of compressed air and I bet some will disappear.  Stubborn ones may need a soft touch of a cotton swab.  If you use a can of compressed air, make sure to clear it of any propellant before shooting the camera.  If you use any cleaner such as alcohol, be very careful not to get any between the window and sensor.

I really doubt these are from the corrector or primary mirror.  You probably would not see them.

Don

[Martin Meredith]

10 hours ago, AKB said:

That's certainly true, but dust on the sensor is so close that it makes dark dots.  I'm fairly sure that this is dirt on the corrector plate, or possibly primary, because of the size of the bunnies.  There's actually a calculator somewhere on line where you can put in details of your optical system and find out exactly the distance from the sensor.  Must find it and give it a go.

Try this:

I must admit they look like regular (i.e. on the protective window) dust shadows to me.

Martin

[AKB]

13 hours ago, Robrj said:

 It will be over that same pixel regardless of the orientation of the camera

 

9 hours ago, HiloDon said:

What Rob wrote is correct and I suspect that the dust is on the Ultrastar's protective window over the sensor.

 

3 hours ago, Martin Meredith said:

I must admit they look like regular (i.e. on the protective window) dust shadows to me.

 

My thanks to all of you for correcting my fallacious opinion!  I have, indeed, previously cleaned my camera sensors and documented it here:

In that exercise, I used a pinhole to image the dust – certainly a different F-ratio than the Hyperstar – which appeared very differently from on my images here.  I will go back and redo this exercise, but you are undoubtedly correct about the source of the problem.

My misconception about the necessity to take flats without moving the camera obviously (now) relates to this issue, where if you do align the galaxy, the bunnies are not...

14 hours ago, Robrj said:

If you took another image of M51 where the camera was 90° from the original image, the galaxy itself would be rotated (so it wouldn't stack) but the dust donuts would be in the same position.

So, in fact, new bunnies aside, I can take flats whenever I like?

[AKB]

42 minutes ago, AKB said:

I will go back and redo this exercise, but you are undoubtedly correct about the source of the problem.

...I did, and you were.  Here's the result of the pinhole test on the sensor itself.

...with the dust motes in the same place as the previous bunnies.

Thanks, once more, for the education.  Time to clean the sensor again... ho hum.