A 3D-Printed Equatorial Platform
While using Artemis in its original form, I rapidly got used to having to track objects by hand. Since most of my views were wide field at lower power, this was not a huge inconvenience. I did ultimately want tracking but it was not my number 1. However a goal for Artemis has always been Electronic Assisted Astronomy (EAA), and my first foray into that made it abundantly clear that tracking was essential. So this project rocketed up to the top of the list.
The following is a work-in-progress and is published to allow for feedback and easy sharing of progress. This notice will be removed when the project is complete.
Given that Artemis is first and foremost an alt-az scope by design, I had two options for tracking:
- Direct drive on both axes, fully computerized. Go-to heaven.
- Putting the alt-az onto an equatorial (EQ) platform. Limited tracking, no go-to.
Clearly option 1 is a gold plated solution to tracking, and one that I can totally see happening with unlimited time and money. However option 2 is a very good alternative for its simplicity and ease of use. I don’t plan on long duration astrophotography with Artemis, so an EQ platform would fit the bill nicely.
Here are the principal design requirements for this project:
- I can easily place the platform between my ground ring and tripod base, so I can use the platform on the base or other suitable flat surface (like a table);
- Have detachable north bearings (the curved parts that stick out below the platform), so I can still use the ground ring untracked;
- Design the bearings in Fusion360 so I can print them and avoid any complex math or fabrication issues.
Theory
I won’t go into the details of how an equatorial platform works (a great resource is here). But all the theory and math boil down to these steps:
- Determine your latitude.
- Determine the axis through your scope’s Centre of Gravity (CG) that is at an elevation of your latitude.
- Determine the profile of your bearing surfaces based on the circle whose centre is along the axis in (2) that intersects the azimuth bearings.
- the profile will be the projection of this circle onto the vertical plane going through the bearings, which will be an ellipse
The math behind this is complex. However, modelling this is not and since I wanted to 3D print the bearings to make them as accurate as possible I could easily get Fusion360 to do the work for me. The geometry looks something like this:
When designing the components, it turned out (luckily) that the axis through the CG more or less came out the backside of the ground ring. While it would be hard for me to determine the CG location exactly given the unusual shape of Artemis, the design assumes this to be true. The impact of the axis not passing through the CG is more work on the motors to turn the scope, and so the expectation is that small variations in this location will not be materially impactful. So for modelling purposes, that is where the pivot point is placed (the black dot on the left edge of the ground ring, above).
For this design, it was also convenient to place the bearings and motors 120 degrees apart when viewed through the z-axis. The angle is less, of course, from the pivot point, but by fixing the location of these points, the remaining geometry of the EQ platform could be easily modelled. The profile of the bearings is the purple curve (above), which is the projection of the tilted blue circle on the plane normal to the bearings. This profile is then used to extrude bearing “blocks”, as described below.
Mechanical Design
As noted before, one of the key aspects of the mechanical design had to be that the platform components were highly portable and detachable. As seen in Jerry Oltion’s design, the north bearings can be attached with posts since gravity will keep them where they need to be on the ground ring. I also envisioned the drives and controller to be on a single platform that can be placed between the ground ring and base, and that the south bearing would be “hung” onto the back of the base. That was my original mental model.
North bearings
Here is a close-up of the model of one of the bearings. Note that it rests on a brass roller which, in turn, will be driven by a stepper motor (details below). A friend generously machined the roller and bracket for me, but as we’ll see shortly, I ended up printing this too.
The length of the bearing is somewhat arbitrary, but the goal is to have about an hour’s worth of tracking, which corresponds to 15 degrees of the circle calculated above in Theory step #3 (above). The printed version of this bearing looks like this. Note the holes where 3/8 aluminum posts are inserted, matching holes on the base of the ground ring to allow them to be quickly inserted or removed:
This was printed (as were all the parts) with 50% infill in PETG, and is hard as heck. But I will still follow Jerry’s advice and clad the bearing surfaces with steel rulers that I picked up at the Dollar Store. The right image shows the cladded surface, with a printed stop to prevent the platform from being pulled off the rollers.
South bearing
While the north bearings are very important, an equally important component is the south pivot. This needs to provide free rotation around the axis elevated to your latitude (44 degrees, in my case). Again, I wanted this to be easily added or removed from my existing base and ground ring, so I modelled it like this:
Note the pivot attaching to the face of the ground ring, and the bearing attaches to the tripod base. The design is liberally borrowed from Jerry, and here is how it comes together:
First, the main pivot component attaches to the ground ring using a 1/4-20 bolt that screws into a captured nut. I have fallen in love with captured nuts ever since I build my Prusa printer. The pivot itself is a 5/16 bolt, partially threaded, that has had its head sawed off. The head shows up again shortly. The sawn end is ground down to a smooth, round tip, and then threaded into another captured nut, while a second nut tightens it all up. The 5/16 bolt was selected because it fits snugly into the 8mm bore of a standard skateboard bearing.
The south bearing assembly screws into the tripod base using another 1/4-20 bolt. But the key is the bearing itself and, again, this is where a 3D printer really makes light work of a design.
The pivot passes through 2 skateboard bearings, and then butts up against a piece of metal. If you recognize the shape, it is indeed the head of the pivot bolt that was sawed off, ground smooth. The pivot forces are therefore supported by the skateboard bearings (laterally) and the bolt head (axially), giving a buttery smooth rotation. The placement of the bolt head leaves about a 1mm gap between the two components of the assembly.
Mechanical assembly
With the parts printed, assembly is relatively straightforward (assuming you can make straight cuts and drill holes precisely). In this design, the south bearing is screwed into one leg of the tripod base, with a sleeve nut (those Ikea things) catching the 1/4-20 bolt and holding the bearing snug. The printed part appears to be plenty stiff to hold the weight of the scope.
However, that part about “drill holes precisely” turned out to be a bit elusive. Add to that the fact that by attaching the south bearing assembly to the tripod leg meant that this could not be used on a tabletop. So I revised the design in two ways. First, the platform base, which was originally envisioned as a “Y” shape riding on top of the tripod, was modified to be a “delta” shape instead. Second, the south bearing assembly was remodelled to screw into this platform base, making the whole platform completely self-contained. Here is how it looks from the top, and a close-up of the south bearing:
Yes, the south bearing looks like it is at an extreme angle, but it is exaggerated by the photo. It is not nearly that bad, and is very solidly attached to the platform base.
The north bearings required holes to be drilled on the underside of the ground ring, and 3/8″ aluminum rods were cut to fit into the bearings, and the ground ring. The holes should be snug enough to keep the bearings in place as you lift the platform and move it around, but otherwise nothing but gravity is needed to keep them where they need to be during operation.
One other thing to note in the mechanical design is that the platform base is attached to the tripod base using printed brackets that slip onto the legs and hold the “delta” snugly. This makes for rapid yet rigid field setup:
The image above also shows the printed bearing roller to replace the machined version. This was because the machined bracket geometry was a bit too large for the base and didn’t match the general printed aesthetic.
Electronic Design
The electronics for the platform are based on an Arduino Nano driving a single stepper motor that is attached to one of the brass rollers:
In the image at left, the Nano is the component at top left, with the stepper driver below it. The red circuit to the right of it is a breadboard power supply that takes as input 5-12V and provide two stable 5V rails that power the Arduino and the stepper separately.
The stepper motor is the commonly-available 28BYJ-48, and the details of how to control it can be found here. My code for the Arduino can be downloaded from the following repo:
https://github.com/tomotvos/equatorialHalfStepper
The code is very simple to understand should you need to modify it, with the most likely changes being to account for your particular platform geometry (“great circle” radius, and roller diameters).
The power supply board allowed me to conveniently power this project from a phone charging bank and USB cable. The supply has inputs for USB, micro-USB, and DC barrel jack, as well as a convenient on/off switch. The enclosure I designed for everything allows any of those inputs to be used, and extends the on/off switch which would, otherwise, be unreachable when the box was closed up. My power bank seems to have ample storage for this, although I need to do a formal test to see exactly how much. I can say it is many, many hours, at least one night if not an entire star party weekend.
One unknown was whether or not a single motor would be sufficient to drive Artemis and, surprisingly, it turned out that the answer was “yes”. This was largely a function of ensuring the axis of rotation of the platform goes through the CG of the telescope. But because of the asymmetry of Artemis, I was concerned that that might not have been enough since, technically, it is the moment of inertia that affects how much torque is required to move everything. In tests, this concern was unfounded.
In the Field
Initial fields tests of the platform exceeded my expectations. When turned on, it kept stars centred in the field for 10 minutes or more with no appreciable drift. And as noted earlier, one motor seemed to be enough to move Artemis (which weighs in at 40 lbs). However there were two refinements that were immediately obvious that needed to be done. The first was that the bearings were “grounding out” on the platform base, yielding a total tracking time of only 35 minutes. While this was pretty good (I am told) it fell short of my 1 hour goal. Second, the bearings need a “hard stop” to keep from accidentally pulling them off the rollers.
It turned out the solution to both issues could be achieved by notching the platform base so the bearings could travel further without ground out. This notch also provided a hard stop that prevented the bearings from rolling off the rollers. As an extra security measure, some 1/4-20 nuts were hot glued to the ends of the bearing surfaces although, in practice, the bearings can’t travel that far.
The notching gave substantially more travel to the bearings, yielding just over 60 minutes of tracking from stop-to-stop, with not much room to spare at either end of the bearing. However, in the field the notch as a hard stop was a problem in that it created too much friction at the start of the tracking to allow it to move: it worked too well. So the notch was sanded more to not impede travel at all, and the hard stop revert to what was glued onto the bearing surfaces. And since the nuts didn’t look very nice, I ended up printing stops, like this:
The final field test…TBD
Improvements
Notwithstanding the success of this project, there are still a couple of improvements I’d like to make. First, an equatorial platform is sensitive to polar alignment and level. If either are off, then your platform won’t be rotating around the same axis as the Earth, manifesting in drift. So I would like to add a small bubble level to the platform base. Yes, my iPhone has a level, but still. Second, I would like to add a compass to help polar align. Yes, my iPhone has a compass, but still.
The other significant feature to add is an electronic “hard stop”. While the existing stops prevent the platform from accidentally being pulled off the rollers, and should also prevent the drive from driving off the rollers, it would be nicer to have a micro-switch cut the power when the bearing reaches its limits. This will prevent the roller from spinning under the stopped bearing, causing wear on the fairly soft brass.
Other than that, I am really happy with how this turned out. It adds a new dimension to what I can do with Artemis, visually and digitally.
If you would like to try your hand at this project (modifying to your latitude, perhaps) here is my latest Fusion360 project file:
Or, download it here:
https://drive.google.com/file/d/1GlEnqHlRzPI7Z94NYQj0M2kLN8aHFg-e/view?usp=sharing
