Yet another Samyang 135mm mounting with Stepper (V2)

[pharscape]

Yet another 135mm mounting solution. This one is enclosed with main electronics attached in a side box.  Most other solutions are open plan with separate front and rear rings. My idea is to make something that is self-contained, hopefully doesn't suffer flexure, robust enough to travel as one unit and keeping wiring in place and condensation off the lens and electronics.

The base is fitted with a NEMA17 stepper + 16 tooth wheel to drive the Samyang focuser via a clamped on larger diameter (98mm) ring. It provides about 11:1 reduction - full focus range is about 1100 steps. A belt tensioner (not shown) is attached to the stepper motor and is adjustable via access holes at the front of the base.

Daylight testing gives me infinity focus at around step 37 (0 is the infinity stop of the lens), with defocus obvious at +- 3 steps so not a lot of margin for fine tuning.  The forecast clear skies tonight should allow first light to check if the step resolution is good enough for a bhatinov mask focus test.

The side enclosure contains an RPI4 running Astroberry , Waveshare stepper controller and 12V to 5V @5A buck converter. The Rpi boots from the SSD velcroed on the side (it's also the removable media for the images). The enclosure slots into the side of the base with only wires connecting to the stepper.

 

 

To provide grip I superglued a section of timing belt to the ring. Since the 135mm focuser's full rotation is 180 degrees this is adequate.  The lens is sandwiched with some 2mm crafting foam to pack the rings.

Once the concept is proven I may refine the styling. Currently its printed in four parts: Bottom, top, side panel, focuser ring and belt tensioner. 

Edited June 1, 2022 by pharscape

[pharscape]

Well it's working pretty much as planned. Managed an hour or so of the Sadr region as a bonus. Manual focus with ekos and bhatinov was easy. Automated focus worked once, I think backlash was increasing.

Lum, Ha and Oiii seem to have very different focal points, I guess that is an F2 lens effect. I have a few issues to resolve around rotating the lens. I'll probably fit timing belt all the way round the ring. 

Most of my problems are with learning ekos and trying to read the fine text at night. I need to attack the gnome configuration. It isn't out of the box ready for small screens and  night time old eyes IMO. 

[StarryEyed]

Thats a great job. Impressive you must have invested some time on this. 

[pharscape]

6 hours ago, StarryEyed said:

Thats a great job. Impressive you must have invested some time on this. 

Thanks, its my first attempt at using Fusion 360, it's not pretty but it is robust 😄.

I just need to work out why auto-focus only worked once (I suspect belt tension/backlash) then I can begin full Ekos automation. 

I will be changing the design a little to use captive nuts. Heat-staking the knurled inserts is very unsatisfactory, it is too easy to go wrong, variable in its bonding and alignment

[Len1257]

Captive nuts are easy to implement, wouldn't do it any other way except tapping the PLA+ / PetG directly. For enclosures that works fine, nuts for mechanicals. Don't know what you have used but I have had very poor results with some of the cheapo ebay buck converters and moved to Pololu devices. More cost but better reliability. 

[pharscape]

The cheapest ebay/amazon brass inserts have no pull strength so I chose the slightly more expensive cross gnurled  version. When placed well they have good strength, trouble is the brass/plastic keeps the heat and they move about easily before the plastic sets. I was thinking about making a jig but nuts will max the pull strength and is a lot simpler. Just not relishing the 15hr print 😬 so I won't make that change until I see a need for another design tweak.

The buck converter is a generic design, XL4015 based, with two trim pots. Ordered from Az-delivery. They advertised the unit only having the voltage trimmer but these seemed to be a revised design which not all their docs have been updated for. The second trim pot is the current limiter which I have set to max.

 

I found the reason for the variable focus/ auto focus failure - my false economy. I didn't zero the stepper before rotating the lens and the timing belt came off the short piece of timing-belt I used for the ring teeth. The simple solution is to just fit a band of timing belt around the whole focus ring.

Judging by my first night experience, my enclosed design keeps the Samyang warm and dew free whereas the guide-scope was suffering (I hadn't intended a long session so didn't bother to fit the dew strap to the guider).

And proof of the pudding is in the eating: 56 mins Ha and 30 mins Oiii. There is some star distortion in the top right quadrant and Sadr is not round - tilt I think so I've ordered the Samyang M48 adapter from FLO.

Edited June 3, 2022 by pharscape

[pharscape]

Here is the belt tensioner: