<HOME

My Ultracompact, Ultralight 200 mm (8") Dobsonian

---->


(Do you just want to see pictures?)

1. Introduction

I built myself a 205 mm (8-inch) Dobsonian scope from ready-made optics in 1997. It turned out really well, but it was way too heavy. 16 kg for the scope itself, plus 8 kg for the mount. A total of 24 kg! Can you guess how eager I was to carry it around? I would have to rebuild it some time and make it a bit lighter. A truss-tube design probably.

Then I saw an article in Sky and Telescope magazine describing Gary Wolanski's ultralight telescope. A truss-tube Dobsonian that could be taken apart for easy transportation! That got me thinking, and I realized I could not only make my scope lighter, but really portable as well! That would be my ticket to deep-sky objects of the Southern hemisphere also! Thus were set my criteria:

  1. The scope must be light enough to carry around easily, and it must not exceed (at least by much) the weight limit of hand baggage on airplanes. (I don't want to let my scope be banged around in the cargo hold.)
  2. The whole thing (excluding the trusses) must fit into a package small enough to take onto an airplane as hand baggage.
  3. The scope has to be easy and fast to set up—and collimate!

2. Construction

As is evident from the pictures, my telescope is based very much on Wolanski's design. The upper tube assembly (figures 1–2) is made from two aluminum rings (I cut them with a jigsaw without too much trouble) connected with spokes of 10x10 mm aluminum tube. The lower tube assembly (figures 3–6) is 30x30 mm aluminum angle stock, held together with "Pop" rivets. The trusses are also 10x10 mm square aluminum tube, just over 80 cm in length.

What is different in my design is that the UTA fits inside the LTA for transportation (figure 6). The low-profile rocker box fits on top like a lid. The only components left outside the package are the trusses and collimation rods. Since the scope is compact enough to transport as a whole, and thus rarely needs setting up in the field, I chose a conventional nuts-and-bolts solution for truss attachment, instead of Wolanski's more exotic approach.

I had been warned that collimation would be a problem with a telescope designed to be taken apart and reassembled. The secondary mirror is easy enough to adjust, since its collimation screws are conveniently located at the top end of the telescope tube. The primary mirror, on the other hand, is usually a different matter, and requires two persons for a comfortable adjustment procedure. I have seen somewhere (Sky and Telescope perhaps?) a solution which allows primary collimation from knobs located at the top end. My own design (figures 7-9) uses collimation nuts in front of the mirror cell and detachable collimation rods to provide the same convenience. As an additional benefit, the heavy primary mirror is located closer to the end of the tube. In balancing a telescope on a low-profile Dobsonian mount this can be of great advantage.

3. Finder optics

Located at the top of the tube is a very lightweight Telrad-type finder (figure 14) made of square aluminum tubing. I used a piece of microscope slide glass, a plastic magnifier lens, and a first-surface mirror of aluminized silicon wafer (which is way easier to cut than glass, and I just happen to have the stuff available. On second thought, I really wouldn't have needed even to aluminize it, but I was aluminizing a right-angle prism anyway, so it was no extra trouble). The bulls-eye pattern was photographed from a laser printout onto black-and-white negative film.

I also demolished a pair of second-hand binoculars (figure 15) to obtain parts for a finder scope. I only used up one half of the binoculars, and the other half is still intact and fully functional. Who knows when I'll find time to actually build the 10x50 finder... It will have to be located at the bottom end of the scope in order not to upset its balance, and that means using a right-angle prism in it, despite the inconvenience of it producing a mirror-image view. (The prism from the binoculars had an anti-reflection coating on its hypotenuse side. I could just barely get total reflection from it, so I aluminized it to be safe.)

This aluminization, by the way, was done by thermally evaporating aluminum under high vacuum conditions. Of course, every amateur scientist has a self-built vacuum coating system in their basement, but working at a semiconductor research lab where evaporators are standard equipment does have certain advantages... :)

4. Final thoughts

At about 10 kg, this scope is way easier to carry around than the previous version, which weighed in at 24 kg total. Now it is actually possible to carry this outside all by myself, which I was very thankful for when Comet Ikeya-Zhang was visible in March–April 2002. Since this was my first light-weight design and my first truss-tube design, I overdesigned some parts and ended up with perhaps a kilo or two of extra weight (especially the LTA materials are sturdy enough for a 300 mm primary!). But the structure did turn out sound, and I lost absolutely nothing in stability compared to the old, heavier monster. When I build a bigger scope, I will definitely use the same design.

I flew to Australia to see the Leonid meteor storm in 2001, and I managed to get this scope in usable condition just in time. Yep, the glorious Southern sky made it all worth while. :)

5. Pictures

The Upper Tube Assembly

[PICTURE]
Figure 1: The upper tube assembly rings were cut from 1.5 mm thick aluminum with a jig saw.
[PICTURE]
Figure 2: The complete UTA

The Lower Tube Assembly

[PICTURE]
Figure 3: The lower tube assembly frames are made of 30x30 mm, 3 mm thick aluminum bracket.

[PICTURE]
Figure 4: The primary mirror cell was cut from scrap 10 mm thick aluminum. The jig saw could just barely hack this.

[PICTURE]
Figure 5: The completed LTA with trusses attached.

[PICTURE]
Figure 6: The UTA fits inside the LTA for transportation

Primary mirror collimation

[PICTURE]
Figure 7: The mirror cell attached to the LTA frame. The two larger nuts will be used for collimation.

[PICTURE]
Figure 8: The collimation mechanism: a novel approach?

[PICTURE]
Figure 9: Schematic of the collimation mechanism.

A threaded rod is attached to a threaded hole in the primary cell, using a nut to fix it in place. This is supported with a conventional spring-and-nut system, from above. The collimation nut is a "long" nut often used to connect two threaded rods together. The whole thing is supported from the lower LTA-frame with a piece of sturdy, 5 mm thick aluminum bracket.

The advantage of this arrangement is easy collimation with rods reaching up to the top end of the telescope, as well as locating the heavy primary mirror as close to the bottom end of the telescope as possible.

The Dobsonian mount

[PICTURE]
Figure 10: The altitude bearings are aluminum/plastic composite. These are designed also to fit inside the LTA, but I rather leave them out to reduce weight of the package.

[PICTURE]
Figure 11: The bearings are attached to the LTA so that their center of curvature matches the scope's center of gravity.

[PICTURE]
Figure 12: The ground board is made of aluminum tube and sits on three junior-sized hockey pucks.

[PICTURE]
Figure 13: The rocker box. When packing the scope for transportation, the rocker box fits like a lid on top of the LTA.

Finder optics

[PICTURE]
Figure 14: The Telrad-like finder weighs just 150 g. Batteries included.

[PICTURE]
Figure 15: Senseless destruction. 10x50 binoculars slaughtered in the name of finder optics. I haven't completed this finder yet.

The completed telescope (Click to enlarge these pictures)

[PICTURE]
Figure 16: The completed telescope. The pink thingy is a reflective arm-band, meant to make the scope less invisible to approaching vehicles.

[PICTURE]
Figure 17: Remove the shroud...

[PICTURE]
Figure 18: ...take it apart...

[PICTURE]
Figure 19: ...and ready to travel!


Antti J. Niskanen <uuki@iki.fi>