Turret Style Widefield Triple Imaging Rig

[Gina]

The large spur gear printed fine    Here are some photos.

I decided that the small gears ought to be bigger and I'll add flanges so that they retain the turret axially so here's a new diagram of the drive system.

[Gina]

Been printing the small gears and flanges.

[Gina]

Started printing a first go at the framework that will support the turret.  This just fits on the bed of my Titan 3D printer.  The whole print will take 2½ hours and I shall leave it printing overnight.  When printing has finished the printer turns off automatically.

This is a screenshot of the Sketchup model.

Top view having just started printing the first layer.  This is through the top clear acrylic panel so not a perfect image.

Side view with the printer door open for a better view.  A bit more printed.

[Davey-T]

Looking good Gina, must be nice having a reliable printer that you can leave running and go to bed, after all the trials and tribulations with your first one

Dave

[Gina]

Unfortunately the print went wrong and for some unknown reason crept towards the back of the printer.  This sort of thing happens sometimes with 3D printers.  I have now designed a new framework which I think will give better support by holding both sides of the bearing wheels.  I have also provided coupling between the two sides in between the bearing wheels.  This end of the framework will now be in two parts which will be solvent welded together to form one rigid structure.  To help with printing this maximum size object I have adjusted a couple of printing parameters - increasing the amount of infill and reducing speed slightly.  Estimated printing time - just over 3 hours.

[Gina]

Looking good Gina, must be nice having a reliable printer that you can leave running and go to bed, after all the trials and tribulations with your first one

Dave

It would be if t'were true   Actually, I think it is for smaller prints at least - I have been producing usable parts with this printer for a while now.  Although this object is theoretically within the capabilities of the printer this creep problem seems to happen with prints that approach the limits.  I've checked the drive on the printer and nothing is loose so if it is the mechanical parts it must be the Y stepper motor losing sync.  This can be caused by insufficient driver current for the required speed or the axis motion hitting the stops.  The latest attempt is still doing it so I'll try reducing speed first and if that works I can see if I can up the drive current.

EDIT - Looked at the slicer settings and wondered which speed (of several) wanted decreasing so I changed my mind and turned up the Y drive current a touch.  I know the driver can take it because it doesn't get too hot without the fan and I have a good airflow from the fan   Trying again...

[Gina]

Printing goes fine for the first three layers then goes wrong in the fourth where it starts the honeycomb infill.  So I surmise that the honeycomb infill is being printed too fast.  Next thing then is to reduce the speed and see if we can get this printed right.

[fireshipjohn]

Hi Gina,

Or switch to Rectilinear infill - I find it's less stressful and just as strong?

John

[Gina]

Thank you John - I'll try that

[Gina]

Update :-

I'm having a small problem with the Titan 3D printer which I need to fix before I can complete this project.  Also, I'm still waiting for extension tubes - must get onto that too.  I'm going through my projects in preparation for an end-of-year progress report which a couple of members have asked for - good idea anyway, I need to see where I'm at as it were.

Reading through this thread, I now know where my missing OIII filter is

[Gina]

This is another project I shall be getting back to now that I have decided to keep my Atik 460EX cameras.  It won't be for some time though as I want to get a couple of other projects out of the way first.  Also, I have thoughts for a slight change in design to make it easier to construct.

[Gina]

Whilst I'm having a cuppa break, I'm thinking about this project.  I'm having second thoughts about mounting the not inconsiderable weight.  Not so much the astro mount because I can put it on my EQ8 but the assembly of three CCD cameras and three real glass prime lenses.  The 460EX CCD cameras weigh 400g each and the lenses something like the same.  That makes a total of 2.4Kg.  I guess since these are for widefield the alignment (between imaging systems and also guiding) is rather less critical than with longer focal length rigs.

A problem I had with the arrangement of three imagers in a triangle and guider in the middle is that the guider could do with aligning to RA or Dec axis rather than rotating with the turret (or is this important for widefield?).  It would be so much easier if either the guider rotated with the turret or was mounted separately.  I'm thinking the guide camera rotation is less important than good close coupling to the imaging system.  I'm more used to guiding for a 1000mm FL scope than of the order of 100mm lens so I may be being pedantic with having the guider aligned with one of the mount axes.

Another thought is whether guiding is needed at all when mounted on the EQ8 which has good tracking on its own, but then the question of flexibility between imagers and mount comes into the discussion.  Maybe I'll try a simple guide scope in the middle of the turret fixed to it with Lodestar X2.  Since I can only use the 460EX cameras on one rig at a time I might as well use my best guide camera.  My current idea is to use the finderscope from the ST80 as guidescope for its small size.

Having said all that I guess the best idea is to make whatever is easiest and quickest and be ready to try it for real when I have time and clear night skies (plus ground dry enough to be safe to cross at night).

[Gina]

I'm thinking I might use a camera lens for the guider rather than the cheapo lens in the ST80 finder scope.  The anount of space between the imaging lenses precludes the use of many of my lenses but the 55mm Super Takumar fits in between easily.  Seems to me that with the relatively small sensor in the Lodestar, that this short focal length should be alright - worth a try anyway.  This will give a much wider view that the ST80 finderscope making direction adjustment unnecessary and the much greated aperture will give many more stars to choose to guide from.  If anyone thinks this would be the wrong choice then please say.

The guider would be fixed in the aluminium tube that goes up the middle of the turret and therefore firmly attached to the turret with close mechanical coupling.

With the improving weather I would like to get this imaging rig ready for testing.  Meanwhile, maybe I can see if I can get the MN190 taking images.

[Gina]

I think I now know what the problem was with my Titan 3D printer viz. the print bed supports and Z drive arrangement.  I'm now in the process of rebuilding the Z axis rail and carriage system with a wide extrusion and dual support brackets.  This should make the platform much more rigid. 

I'm getting keen to get the widefield imaging system up and running ready to capture the whole Veil Nebula system in one go.

[Gina]

My large Titan 3d printer seems pretty much working now and with the Cygnus Loop coming into view with images already posted by others eg. Veil Nebula in Ha by ChrisLX200  I think it's time I resurrected this project.

[Stu]

6 minutes ago, Gina said:

My large Titan 3d printer seems pretty much working now and with the Cygnus Loop coming into view with images already posted by others eg. Veil Nebula in Ha by ChrisLX200  I think it's time I resurrected this project.

Would be great to see you imaging agin Gina ??

[Gina]

Thank you Stu    I agree it's time    I have done a little bit with the all sky camera but not had a nighttime clear sky since I got it working.

Now to the triple turret rig then...  I really wonder if guiding is necessary particularly as from reading posts it seems some of the best astro imagers here are going for shorter exposures and more of them and getting great results.  If I can get the EQ8 mount set up well for polar alignment, I think it's worth a try.  The FOV with Asahi takumar 105mm FL f2.8 lenses and Atik 460EX cameras is 6.8° x 5.45° and resolution 8.91"/pixel according to 12 Dimentional String FOV Calculator. 

The Cygnus Loop looks like about 3° x 4.5° (or maybe 5°) so these lenses easily cover it though in my experience Wikipedia often seems to underestimate DSO sizes.  But if that's right the 135mm FL f2.5 lenses would be a better fit - FOV: 5.30° x 4.24° Resolution: 6.93"/pixel.  I'll try the 105mm lenses first - nothing worse that missing out on the peripheral parts of a DSO.  Actually, I've had a thought - I can check my mosaic of the area (if I can find it).

[Gina]

The mosaic turned out to be with DSLR and scope but I did find an image using the 105mm lens and Atik 314L camera.  So it seems the 135mm lenses will be more suitable.  Atik 314L+ has 1392x1040 sensor with 6.45µm pixels = 8.978.4mm x 6.708mm - call it 9mm x 6.7mm while the 460EX is 12.49mm Horizontal x 9.99mm Vertical - 12.5 x 10.  So horiz ratio = 12.5/9 = 1.4x and vertical = 10/6.7 = 1.5x. approx  Ratio of image sizes for 105mm & 135mm lenses = 135/105 = 1.3x approx.  So the 135mm lens image would be just a bit smaller than the attached.

[Gina]

The current design is for the 105mm lenses with a central guide camera but having had many months turning this over in my mind I have doubts about the current design.  As I said, I think I shall try without guiding and could hence make the design more compact.  My original idea was to make the rig able to take different size lenses but I think a better plan might be to have a rig for each size lens, optimised for that size lens.

With the current DSO in mind I shall work towards a design with just a rod up the middle rather than quite a large tube.  Also, I've gone off the idea of having the front of the turret running on three gears.  This is due to my experience with my Epicyclic Clock which also has a large gear running on three smaller gears to locate it.  This is not as easy to engineer as I thought - particularly using 3D printed plastic, and the imaging rig needs far better stability than needed for a clock.  So the clock production has helped with ideas for an imaging rig

There is no real need to have the turret easily rotated - more important for it to be really rigid.  Yes, I want to be able to turn the rig to get the best orientation of the DSO image on the camera sensors, but once set it doesn't want to change until the next DSO is to be imaged.

[Gina]

I think the Cygnus Loop is oriented with the long axis N-S or near enough so I'll build the imaging rig with that in mind.  I think each of the bigger DSOs that suit this rig can have their own build - after all, designing and printing plastic parts is quick and easy

[Gina]

Been looking for all the bits to measure up for designing the mounting but so far only found one 135mm lens.  I'll go on looking but if I can't find the others I might go for the 105mm lenses so that I can at least do some imaging when weather permits.

[Gina]

Found another 135mm lens - as guidescope for the 400mm FL triple rig so I've pinched it back.  I have another 135mm lens but it's got the wrong mounting - some sort of bayonet   It's a Hanimex rather than an Asahi Takumar.

[Gina]

Been looking through my widefield imaging threads and now I'm not sure I ever had three 135mm lenses with M42 mounts.  I have one Asahi Super Takumar 135mm f2.5, a Caligar 135mm f2.8 and the Hanimex with a bayonet mount.  I don't think I fancy fiddling about with the Hinimex lens though I guess it might be possible to do something.  I'll keep an eye on ebay and see if I can get a Takumar or Super Takumar 135mm f2.5 lens.  f3.5 versions are available but the f2.5 see rather scarce and relatively expensive.  Funds are low ATM.

All this means I shall go with the 105mm lenses for first design.  These are about the same diameter as the 460EX cmeras and should produce a nice compact design.  An earlier design used three 200mm f4 lenses - big and heavy.

[Gina]

Since each DSO to image will have it's own mounting I have decided to do away with the rotating turret but still keep the triangular arrangement for a compact and closely coupled mounting.  The rotating turret seemed a good idea but isn't reall practical for an imaging system unless made of metal for rigidity.

Here is a diagram showing a possible mounting plate system with the focussing system also shown.  Red are stepper motor pinions and green are straps round the focus sleeves of the lenses with a quadrant gear section.  I'm thinking of probably three plates - two to hold the cameras and another to take the focus motors.

[happy-kat]

You are busy and motivated, looking forward to the output. I could do with some of that :-)