Thoughts on which imaging rigs to concentrate on

[Gina]

Looks like we just might get some clear sky tonight   So... what to do with it.  I think it will be several days yet before I get a viable Linux system working for astro use, though I seem to be making progress.  Trying to do proper imaging with my present Win7 laptop and apps is just not worth the hassle!  So I think I'll just do more playing about - much easier on the old nerves   Maybe I'll try the extra widefield 28mm lens and cover a pretty huge area of sky above

[Gina]

28mm f3.5 lens set up but no fine focussing yet - this is just with manual focus.  120s Ha sub at -29°C and gain of 500.  Flipped image but no other processing - saved in PNG format full frame.  This is the raw, untouched image.  There are signs of image curvature around the edges.

Edited October 2, 2016 by Gina

[Gina]

Whilst this image is far from perfect, it does show the relative positions of the DSOs in Cygnus.  Near centre are the NAN and Pelican, and the Cygnus Wall.  Just above centre left is the Cygnus Loop and top right IC1396.  The Crescent Nebula is also in there - very small - about 7 o'clock from the Cygnus Wall towards the bottom right of centre.

[Gina]

The sky clarity seems to have improved.  Here is another sub, same parameters as before but this time histogram stretched a little.  I shall need to take some flats for this lens and also darks at this temperature and exposure.  I shall be interested in seeing what I get in OIII and SII but I need the remote focuser connected up before I can change filters.  These lenses are not apochromatic in the red end of the spectrum.  SII focus is usually not very different from Ha but the OIII is something like 20 odd focus steps different.

[Gina]

The sky is looking even better now so will continue with Ha imaging having just completed a meridian flip and re-centred.  I'll add the focus quadrant to the lens tomorrow ready for the night in case the forecast of clear skies materialises.  Here is another flipped only image - no histogram stretch.

[Gina]

I've examined a small portion of the image in the brightest part - the centre of the NAN and the white level is just touching saturation so this is the maximum exposure at this gain setting (500 = 50db).  If the sky continues clear I should have 3+ hours of imaging time.  100 x 2m = 200m.  OTOH I think Cygnus will have set below the observatory roof by then.  Oh well, have to see what I've got tomorrow morning.  So far so good

[Demonperformer]

8 hours ago, Gina said:

Whilst this image is far from perfect, it does show the relative positions of the DSOs in Cygnus.  Near centre are the NAN and Pelican, and the Cygnus Wall.  Just above centre left is the Cygnus Loop and top right IC1396.  The Crescent Nebula is also in there - very small - about 7 o'clock from the Cygnus Wall towards the bottom right of centre.

Not only the relative positions, but also the relative brightnesses, the veil being very faint in comparison to NAN/Pelican complex.

[tekkydave]

I wish I had your clear sky here. I looked out at 11pm last night and could only see a few stars despite zero cloud.

[Gina]

Sorry to read that Dave

Yes, it was a good imaging night   Roof started cutting into the frame at about 2am but with subs captured both before and after the meridian flip I reckon I should have over 100 viable frames for processing out of a total of 130 up to 2am.  We had a very heavy dew but I had no dewing problems with the rig - probably due to the heat given off by the camera and mount.  No dew shield or even lens hood on the lens.

The orientation of the observatory and the roll off roof is not ideal for the current widefield imaging and a better option will be the planned 2nd observatory - a miniature one, maybe with a dome.  One thing that determined the full sized observatory orientation was the neighbour's billion candlepower flood light in addition to the lie of the land and the gradient - I live on the side of a hill.

I have two options for the mini observatory - an NEQ6 mount ready to go or a DIY mini fork mount.  Both have pros and cons.  The big problem with the NEQ6 is it's size.  In particularly the large volume required to accommodate the GEM system - this is total overkill for a widefield rig with the mini filter wheel and camera lenses.  Of course, the EQ8 mount looks even more ridiculous with this "pimple" of a rig sat on it

I know I've got enough projects on the books ATM but several of these are being shelved following the amazing results form the ASI1600MM-Cool camera.  They might even be ditched altogether though I might go for a triple imaging rig again later on.  The benefits of the compact and lightweight single ZWO camera and mini FW make for a much more viable mini fork mount in a small dome.  Also, the thought of building my own mount is more appealing than a much more bulky and heavy triple rig.

Edited October 3, 2016 by Gina

[Gina]

The RPi makes a mini obsy with dome and DIY fork mount a definite option.  I could run a 12v feed from my main observatory for the power or maybe just a car battery in the base of the mini obsy.  I would need to work out the maximum current required.  An RPi could also provide the mini mount drive control.  I would use NEMA stepper motors as the gearbox on the 28BYJ-48 motor has a peculiar ratio - it is only approximately 64:1 and I would want precise steps.  I guess a mount is not as precise as a clock but avoiding a peculiar ratio gearbox is still best I reckon.

With a widefield rig that is well polar aligned (I have the PoleMaster for that) the mount should only need RA tracking at 15 degrees per hour so sidereal tracking should be quite easy.  A DEC drive would be needed for slewing of course.  I would also use a NEMA stepper motor for the dome azimuth drive.  Not sure about the shutter but I guess another NEMA would make things simple.  I can see my large 3D printer having some work to do

[Gina]

Been out this morning and lunchtime but now back and have added the focuser quadrant gear which clamps onto the focus sleeve on the lens so now ready for remote focussing is the sky is clear tonight.

Had some deliveries while I was out including another 64GB micro SD card and used medium format SLR camera lens - 45mm FL and f2.8 - quite some piece of glass   I shall need to sort out the mounting as this has a bayonet mount.

[Gina]

If clear tonight as seems a possibility for a couple of hours, I shall be imaging with the 28mm lens in OIII.

Meanwhile, I'm going to have a theoretical look at a fork mount for ASI1600MM-Cool + ZWO FW mini + SLR 35mm film camera lenses of 28mm to 200mm focal length.  I estimate the total weight as around 1Kg and fork depth around 100mm or so.

The mounting would consist of a 3D printed clamp system like a double sided version of the fitting I'm using with the dovetail mounting.  I'll draw some diagrams in SketchUp later.  The imaging rig mounting would have axles coming out of the sides with a drive wheel on one side of say 100mm diameter - this will form the DEC axis drive.  For the RA drive the shaft of the fork is turned and the fork framework would have another large wheel for the RA drive.

Edited October 3, 2016 by Gina

[Gina]

Drat - my post with all the calculations has disappeared

[Gina]

Anyway, without re-writing all the arithmetic, it works out that with a NEMA stepper motor motor with 1.8° step angle and 16x micro-stepping, a ratio of 100:1 would be required - very difficult to achieve in one go without using worm and worm gear and I haven't got the facilities to produce those.  At first sight a NEMA stepper motor with built-in gearbox would look like a good idea but further investigation shows that their gearboxes are also not an exact integer ratio just like the 28BYJ-48!   

OTOH there are alternatives to the Pololu drivers that will do 32x micro-stepping meaning a ratio of 50:1 - a bit easier perhaps.  I don't think 3D printed gears would be accurate enough and an alternative is belt drive with MXL timing belt and pulleys

[Gina]

The smallest MXL timing pulley with 5mm bore is a 15t with pitch diameter just under 10mm meaning the big pulley would have to be a whopping 500mm diameter.  Even the smallest pulley with 10t has a 6.7mm pitch diameter and hence 330mm big pulley - that's over a foot!!  Not on.  So the step down will have to be in two steps.  Two steps of 5:1 and 10:1 would be possible, but for the pulleys available a 15t motor pulley would have to go with a 60t on intermediate shaft giving 4:1 leaving 25:2 for the final drive step.  With another 15t pulley and 10mm pitch diameter the large pulley would be 125mm diameter (5").  This is quite feasible

[Demonperformer]

5 hours ago, Gina said:

I shall need to sort out the mounting as this has a bayonet mount.

I know you like making your own , but, if a canon bayonet mount sounds like a lot of trouble, there is an adapter for canon lenses onto the 1600. With your mechanical aptitude, I'm sure you would not run into the sort of problem I managed to create:

 

[Gina]

Ah yes - I have read that    I can buy a Mamiya 645 lens to EOS adapter and that could by connected to an EOS to T2 adapter to solve the problem in two steps.

Edited October 3, 2016 by Gina

[Gina]

Clear sky and I'm imaging in OIII with 28MM lens.  Still using 120s subs though I could have gone up a bit before saturating but since the final result is more dependent on total time I'm sticking with 120s so that I can use the same darks for calibration.  

Here is an OIII sub histogram stretched in Photoshop.

[Gina]

Captured best part of 100 OIII subs last night then it was meridian flip time so I changed to the Ha filter for alignment and rough focussing the SII filter, fine focussed and then went over to APT for SII imaging but by that time the cloud had thickened.  Left it running overnight but on checking this morning saw no usable SII subs so I'll do those next clear night we get.

[Gina]

In view of the moon phase and moonrise/moonset times I think it's time to have a go at M31 the Andromeda Galaxy in RGB or LRGB before the moon gives problems for RGB.  The wider field NB stuff cam wait.  Clear Outside is showing a nice clear sky for tonight   Not sure when M31 will be above the trees tonight but should get several hours in in the forecast is right.

[Gina]

I now have the Linux based astro capture and control system working remotely with the RPi 3.  The latter has INDI installed and connected to the ASI178MM camera in my all sky cam.  I have KStars/Ekos installed on Linux Mint laptop and communicating with the RPi via WiFi and displaying the all sky image in Ekos, so apart from changing the DSO imaging rig to RGB and 135mm lens, I shall be working on my all sky camera project.

[Gina]

Something else has cropped up - the strong wind has got at my observatory fold-down window/flap and torn it off it's brackets so I need to make replacement brackets.  Fortunately there's no rain in the forecast for a few days.

[Gina]

RGB imaging on M31 - the Great Andromeda Galaxy.  Here a 10s Red sub with gain of 150 and temperature of -29°C.  Gamma adjusted in Photoshop to show the galaxy by not saturate the whites.  The shadow in the corner is my weather station mast which will soon disappear from the frame. Click to choose f

[Gina]

Another sub - same parameters but the shadow has gone.  Full size and cropped/rotated to put north upwards (I think).

[Gina]

Now onto Green subs - here's one.  I'm taking 200 subs in each colour.  Not sure if this is enough - only 2000s total time in each colour - just over 33m.  Maybe it would be better to reduce the gain so that longer exposures are possible.  With a gain of 150 = 15dB (actually 148 = 14.8dB) 10s exposure is the absolute maximum without white clipping.  If I went to 60s ie. 6x I would need to reduce the gain by 6x = 7.8dB so a gain setting of 148 - 78 = 140 (or 14dB) should give the same image brightness.  I'll finish this set and then maybe try the new settings - or something in between.