Ultra-Widefield Dual Imaging Rig with Remotely Rotatable Cameras

[Gina]

I decided to change my current widefield narrowband imaging rig to ultra-wide FoV with 28mm lenses from 200mm lenses.  This project is just at the thinking stage ATM not even design yet.

Here is a photo of the current rig with 28mm lenses fitted, before camera rotation added, incomplete focus systems and one dewshield missing - there is a lot to do.

 I am having trouble getting focus with these very wideangle lenses

[Gina]

Having set the scene, I can discuss some ideas. 

Firstly, with the very wide view of 36x27 degrees the guiding system on the left in the photo is not needed - tracking with the EQ8 mount is quite adequate.  This will leave room for the additions required for camera rotation.

Secondly, with the guiding system removed the imaging sets can be further apart allowing more rotation - though I have yet to work out just how much would be needed.

[Gina]

More important than the camera rotation is getting the lenses to focus.  I don't have suitable T2 spacers to get the focus within the focussing range of the lenses so I have ordered some more. 

The rotation will involve the whole light path with the lenses firmly attached to the cameras and the remote focus systems attached to the lenses so they will rotate as entities with the cameras. With the whole lot rotating as one unit focus is maintained.

With two imaging systems required to rotate in unison, so that the the FoV matches between images, there are two possibilities.  Ether the cameras can have separate stepper motors synchronised in code or the cameras could be geared together.  I think the former is easier though that would involve two stepper motors and the electronics to drive them.  Nothing is "set in stone" ATM.

I am happy and indeed pleased to have any suggestions or comments.

[Gina]

Without the dewshield of the guider, the main imaging systems can rotate 180 degrees or more.  180 degrees will cover all orientations.  Here is a diagram.

Initially it would seem that 90 degrees might be enough to cover all cases but the rotation diagram shows that it isn't.  I have shown the case of turning the field of view 135° anticlockwise.  It can be seen that this would be covered by a rotation of either 135° or -45°.  A rotation of 45° would swap the X and Y axes.

This shows screenshots from KStars of zero and 135° (or -45 and flip).

To show a more realistic case here are 120° and -60°.

 

[Gina]

For the dual imaging rig the separation of the imaging systems is dictated by the size of the EFW.  The EFW protrudes 100mm from the centre so the spacing needs to be 100mm plus the width of the camera bracket.

 

[Gina]

The cameras will rotate quite easily in PLA but I have to find how to drive them from the stepper motors.  Needs a gear ring clamped on somewhere clear of the cooling vents.

The cameras have an 18mm clear band at the front (sensor end) and 15mm at the back.  With the main weight at the front - EFW and lens - it makes sense to make the support the full width with a thinner support and drive wheel at the back.  Also, the EFW would make adding a drive wheel at that end very difficult.

[Gina]

Starting point for the design.  Looking from the lens side.  The circles at the back of the diagram show the size of drive gears.  A stepper motor mounted on the rear plate could drive both lenses.

[Gina]

This shows the  gears as full- circle but only just over 180 degrees is needed.

[Gina]

Redesign of camera gears to provide clamping (with M4 screws in the lugs).  This is the base position - motor pinion rotates clockwise and cameras anticlockwise. Parts are coloured to aid clarity - in practice all parts will be printed in white PLA.  I may change the design to prevent rotating past the end.

 

[Gina]

[Gina]

The gear ratio from stepper motor to cameras is currently 11:1 but I'm thinking it might be better to make it 9:1 which would give 5 full steps per degree - a nice whole number.  This would result in a modulus of just over 2 to get the right size gears.

[Gina]

Bought a selection of T2 extension tubes/rings to supplement what I already had and been trying to get focus with the 28mm f3.5 lens on the filter wheel and no.1 camera.  15mm is not enough and 17.5mm is too much to get within the focus range of the lens. The lengths I've got in fixed rings are, :- 5mm, 7.5mm,10mm, 15mm, 20mm and longer. I also now have a 12-14mm variable.  I want between 15 and 17.5mm, so around 16mm should be right. The variable plus 5 gives 17-19mm so that's no good.

Yesterday late afternoon I had delivery of a set of M42 extension tubes of 9mm, 16mm, 30mm so the middle one may be OK or perhaps the 9mm plus 7.5mm T2 = 16.5mm. Got too dark for testing on the far trees by then though so I'm hoping the mist will clear enough to try again today.

The mist cleared for a while enough to check focus range and I have been able to get the focus withing the range available from the lend focussing sleeve. The thicknesses of the M42 extenders turned out to be different from the spec.  What was supposed to be 16mm was actually 14mm and the 9mm is actually 6mm.  So I used the 6mm M42 on the lens and a 10mm T2 onto the EFW adding 16mm altogether.  That worked and I was able to get focus somewhere around half way on the focus sleeve.

[Gina]

I should be able to work out the spacing required for the other lens.  The ZWO EFWmini is 20mm and the extra spacing is 16mm giving 36mm.  The second camera has an 11mm thick T2 adapter that takes the OIII filter in another adapter, meaning the extra length needed is 36-11=25mm.  A 10mm and a 15mm.  When the mist clears I can check that I'm right (or not).

Later...  The visibility improved for a few minutes and I was able to check that the second lens could focus.

[Gina]

A slight delay on the focussing - a wire has broken off one of the focus stepper motors.

I've been looking at the possibility of a mosaic covering the Milky Way.  These screenshots of KStars show some of the panels   Camera angles of 120°, 150°, 30° and 70° for Cygnus, Cepheus, Cassiopoeia and finishing with Orion.  One or two panels to fit between Cassiopoeia and Orion.  These screenshots were taken with the Time set to 9pm Saturday 12 December 2020.  It shows that this would be when Cygnus starts to go too low to be fully visible from here.

[Gina]

Just checked the base or default camera orientation for my imaging rig and it urns out it is -90 degrees in terms of FOV in KStars. This means the "FOV Symbols" settings will go from -90° to +90° rather than 0-180°.   Or to put it another way - 0° camera rotation corresponds to -90° in KStars, 90° rotation to 0° and 180° to +90° in KStars "FOV Symbols" settings.

Camera angles of 120°, 150°, 30° and 70° become 30°, 60°, 120° and 160° rotation angles.