Compact Harmonic Equatorial Astro Platform (“CHEAP”)

It weighs approximately 1.5 kg and can reliably slew (GOTO) and track telescope setups of up to about 5 kg. For astrophotography with a telescope, active guiding is required.
Cost of all the parts for the mount is around 200 € without wedge and tripod.  

It consists of two load-bearing, additively manufactured parts, two low-cost harmonic drive gearboxes for NEMA 17 stepper motors, an inexpensive 3D-printer controller board, a few screws and cables, and an easily color-customizable housing.

Its advantages are the very low price, compact size, low weight, and functionality. A drawback is that during unguided tracking it is only suitable for very wide-field imaging with a pixel scale of around 3 arcseconds. Comparable commercial tracking mounts cost roughly twice as much or more.

I built it for my small, lightweight travel setup. Realistically achievable tracking accuracy is 1.2–2.5 arcseconds RMS with full control via PHD2. Without PPEC, about 3″ RMS was possible using very aggressive guiding.

It can easily be expanded to integrate an EAF (electronic autofocus) and a rotator. This is my third iteration of this travel mount and I am still in the process of updating parts of it.

Parts list:

In the PDF file.

I recommend using stainless steel screws throughout, stepper motors with a 0.9° step angle and moderate inductance (no “high-speed” motors from 3D printing), and motors of medium length. Due to the gear reduction, pancake steppers are also possible; however, a second reduction stage as used in other mounts is omitted. Therefore, good and smooth microstepping is important, which is easier to achieve with longer (= more powerful) motors.

For operation, a polar wedge for alignment (approx. €50–80), a tripod, and a Vixen dovetail clamp are also required.

Printing:

Make sure to use a strong material with low creep tendency for the two load-bearing parts. I used Prusa PC Blend with carbon fiber (CF).

I recommend 4–5 perimeters and a high infill—this is not an area where you should cut corners.

The remaining parts are non-critical. For the cable strain reliefs, a more ductile material is recommended, while for the covers any UV-resistant material is sufficient.

For the adapter plate to the Vixen clamp, it must be ensured that it is tall enough so that the clamp does not interfere or snag in any position while rotating. If necessary, print a suitable spacer and use longer M3 screws. The spacer provided is 1 cm, just scale it in Z only to get the distance you need.

 Make sure the Mounting Plate Version fits your Vixen Clamp.

Assembly:

Assembly is relatively straightforward. Depending on the stepper motors used, some soldering may be required. When assembling the harmonic drive gearboxes, special attention must be paid to cleanliness to prevent foreign particles from entering the gearboxes. Such contamination could, in the best case, degrade tracking accuracy and, in the worst case, cause damage to the gearboxes.

More detailed instructions with images can be found in the PDF file.

Usage:

Excellent images can be produced with this mount, provided its limitations are taken into account. The harmonic drive gearboxes are excellent for their price—but far from perfect. I installed a total of four of these gearboxes, and each one exhibited a strong periodic error.

This error is caused by an eccentric positioning of the wave generator and results in a periodic error that can be compensated very effectively in software. For this reason, it is recommended to use PPEC (Predictive Periodic Error Correction) on the RA axis.

For sidereal tracking, the error period is 435 seconds and can be almost completely mitigated by PHD2 after approximately 14 minutes.

In addition, the corrections should be set fairly aggressively and the guide camera exposure times should be kept short.
I use an SVbony 30 mm f/4 guide scope with a 1-second exposure time, a predictive weight of 90% and a reactive weight of 60%. Due to the periodic error, it may be necessary to perform the calibration multiple times until the relatively linear region between under-correction and over-correction is found.