MarkIV

For best results, set the color temperature of your monitor to 6500K (Red=40, Green=40, Blue=33).

I enjoy designing and building telescopes. I spend a significant portion of my free time exploring new ideas and like creating unique structures. Minimizing weight is a common theme. Consequently, when I discovered some new (to me) lightweight structural composite panels, I immediately began thinking how I would use them. At the very least, panels can provide significant weight savings just by substituting them for solid plywood in existing designs. However, new materials often offer new design possibilities in addition to design challenges. In the back of my mind I thought how nice it would be to take a 12-1/2" telescope on an observing vacation to Australia. If the telescope were light enough, i.e. total weight with packaging is less than 50 lbs, it easily could be taken as checked baggage. This article is about how I used composite panels to construct a 12-1/2" f/5 portable Dobsonian incorporating some new features and weighing only 40 lbs.

Figure 1

Portable 12.5" f/5 telescope made with lightweight composite panels has a total weight of only 40 lbs.

Composite panels have been used as lightweight structural members for a long time. Aluminum sheets laminated to an aluminum honeycomb core probably are the most well known lightweight composite. Aluminum honeycomb is commonly used in the construction of airplanes because of its very high stiffness to weight ratio. I've avoided these materials because of their high cost and construction challenges.

A friend, Alan Adler, recently introduced me to Tricel Honeycomb panels. Searching the Internet turned up similar products available from other manufacturers. Used extensively in boat building, lightweight structural panels can be purchased with a variety of skins and core materials. Lightweight cores include end-grain balsa wood, closed cell PVC foam, and honeycombs of polypropylene and impregnated paper. Skins include plywood, plastic laminates, and fiber reinforced resins. Figure 2 shows a representative sample of available materials.

Figure 2

(lower right) Fiber reinforced polyester and fiber reinforced wood veneer skins over Balsa wood cores. DecoLite Composite Panels available from Alcan/Baltek Corporation.

(upper center) Fiber reinforced polyester skins over foamed PVC cores from Baltek (Airlite DecoLite Panels) and DIAB (ProLamina Sandwich Panels).

(lower center) fiberglass and plywood skins over polypropylene honeycomb from Nida-Core.

(left and sheet along bottom) thin plywood skins over impregnated paper honeycomb from Tricel Corporation.

Composite panels have high stiffness to weight ratios because bending forces are concentrated near the surface. Material farther from the surface contributes significantly less to stiffness. Consequently, a large fraction of dense core material can be replaced by less dense, weaker materials without sacrificing much stiffness. The core only needs to resist shear between skins on opposite sides. For example, a composite panel of given thickness may weigh 25% of one made of solid plywood while retaining 70% of its stiffness. A composite panel slightly thicker than solid plywood can be as stiff yet weigh only 30% as much.

Based on cost, weight, availability, and ease-of-use, I decided to use materials available from Tricel Corporation. All of their composite panels use a moisture-resistant impregnated paper honeycomb core (Figure 3). Tricel honeycomb panels are available in 4' X 8' sheets with skins made with different types and thickness of plywood (Figure 4). I used the ½" thick panels with mahogany plywood skins, known as Tripanel Marine, in constructing the mirror box. ¾" and 1" thick panels with lauan plywood skins were used for the sides and bottom, respectively, of the rocker box.

Figure 3

The structure of Tricel's impregnated paper honeycomb is not the familiar hexagonal pattern. Cells are roughly triangular.

Figure 4

Representative samples available from Tricel Honeycomb.

(left) ¾" and 1" thick panels with lauan plywood skins.

(right) ½", ¾", and 1" thick panels with thin mahogany plywood skins.

Most composite panels can be cut with standard woodworking tools. In fact, I found them a joy to work with. For example, a 4' X 8' sheet of 1" thick Tripanel Marine weighs 20 lbs, and is easy to lift. In comparison, a solid plywood sheet 1" thick can weigh 75-100 lbs and is a challenge to drag across the floor. Because a saw blade passes through a couple of thin veneers, a little paper, and mostly air, cutting panels requires little effort.

On the other hand, structural composites have some drawbacks that require special, time-consuming construction techniques to overcome. Thin veneer skins are easily damaged along panel edges, provide a narrow glue line to hold pieces together, don't provide enough material for screws to hold onto, and can't support the forces of a bolt tightened across the panel.

All areas where a screw or through bolt are used must be reinforced. One approach is to drill out the weak core material and insert a solid plug. Figure 5 shows how this was done for the rocker box pivot bolt. A 1-3/4" diameter Forstner bit was used to drill through the bottom plywood skin in the center of the rocker box. The drill removes the paper core down to the interior surface of the top plywood skin, which remains intact. A disk of solid wood was cut from a 1-3/4" diameter hardwood dowel. It is slightly thicker than the depth of the hole it fills. I glued it in place with epoxy. After the epoxy cured, I sanded the disk flush with the surface and then glued Ebony Star laminate to the bottom. Afterwards, the hole for the center pivot bolt hardware was drilled through the solid wood.

Figure 5

A 1-3/4" diameter Forstner bit was used to drill through one plywood skin and remove the paper core. The remaining plywood skin was left intact. A disk of solid wood, cut from a 1-3/4" diameter wood dowel will plug the hole and provide solid support for the pivot bolt mounting hardware.

To prevent moisture ingress and protect the fragile veneer, solid wood should replace the paper core around the panel edges. I start by cutting the 4' X 8' sheet into smaller panels the approximate size to be used. A table saw works well to remove the core quickly and cleanly (see Figure 6). Set the blade height to the desired depth. With the panel on its edge and its face held firmly against the rip fence set to the proper position, make repeated passes to cut out the core. Be sure to use push blocks and keep your fingers away from the blade. The finished depth of solid wood around the perimeter should be at least the thickness of the panel. For the ¾" thick vertical rocker box panels, I typically used strips 1" wide and cut and/or sanded to width to fit the space between the veneer.

Figure 6

A table saw was used to remove the paper core around the edge, in preparation for gluing wood strips around the perimeter.

The type of wood used to reinforce the edges depends on individual goals and tastes. I glued Finnish Birch plywood around the perimeter because it is strong, I like the look, and I happen to have a lot of scrap from prior projects.

In addition to sealing the edges, the solid wood edging allows the use of traditional joining techniques. Figures 7 and 8 show the tongue and groove shapes cut into the edges of the rocker box panels. When glued together, the joint will be as strong as if the entire panel were solid wood.

Figure 7

After gluing in wood strips, a tongue is cut into the edges of the rocker box front and rear panels.

Figure 8

The rocker box side panels have corresponding grooves to receive the tongues. ¾" edge around the perimeter of the rocker box bottom panel is removed, leaving the bottom skin.

Figure 8 also shows how the bottom of the rocker box was cut to mate with the front and side panels. A ¾" wide section of top plywood skin and core material was removed. The side panels were glued to the bottom skin and the side of the top skin.

One particularly tricky construction challenge was creating solid wood along the entire length of the cut outs for the altitude bearings. I couldn't think of a good way to remove the paper core along the edge if I cut the arcs first. Instead, I removed material along the top edge down to a depth of 3-1/2" and glued in a wide strip of plywood. Working with an exposed blade 3-1/2" high is quite dangerous. If you try it, be very careful. With solid wood glued along the top 3-1/2", I was able to rout out the curved section, leaving a solid surface along the arc. Note that this construction method leaves a wide portion of solid wood in the upper corners of the rocker box. Although the amount of weight in the corners is small, I decided to drill a 1-3/4" diameter hole in each corner to remove a few more ounces (Figure 9). More importantly, the holes serve as convenient carrying handles.

Figure 9

Telescope packed up for transport. Note the holes in the upper corners of the rocker box. This is the only location with a substantial width of solid material. Holes were drilled to remove a small amount of weight. More importantly, they serve as convenient lifting handles. Easy to do since this configuration only weighs 35 lbs.

Although these materials offer the promise of lighter weight telescopes, that isn't necessarily their only advantage. New materials often offer new design possibilities. For example, thin mahogany plywood is one of the few woods that can be bent into a tight radius. The composite panel, by design, is too stiff to permit bending. However, if the inside skin is removed or separated into short, disconnected segments, the panel becomes flexible (Figures 10 and 11). This is one of the reasons mahogany is found on luxury yachts. Curved sections of the dark wood are very attractive. I wanted to incorporate this feature into the mirror box.

Figure 10

The lower panel with nine cuts spaced ¾" apart will allow the panel to be bent 90º, closing up the gaps.

Figure 11

The lower panel with nine cuts spaced ¾" apart allows the panel to be bent 90º, closing up the gaps.

It is straightforward to determine the number and spacing of cuts necessary to make a given radius knowing the panel thickness and blade width. For the ½" thick mahogany panel, I needed nine cuts spaced ¾" apart (lower panel in Figure 10). When that panel was bent 90º, the gaps closed up (Figure 11). When glued together, the curved structure once again is very stiff. However, the outside surface of the resulting panel was not a smooth curve. The individual sections were long enough so that the outside surface was a series of corresponding facets. In order to achieve the smooth curved surface I wanted, I found I had to space cuts no more than ½" apart (top panel in Figure 10). The larger number of closer-spaced cuts allows the panel to bend in a much tighter curve. However, when bent 90º, the gaps do not close completely. Wood glue normally does not bridge gaps well, resulting in weak structures. Epoxy can fill a gap and produce a very strong structure. However, the viscosity of typical epoxy formulations is too low. Epoxy would just run out of the gaps and into the cells of the core. To prevent the epoxy from running, I added a thickening agent - Cabosil. Adding a few percent by weight of Cabosil makes the epoxy the consistency of peanut butter. It stays where it is put until it cures. Alternatively, the interior surface can be fiber glassed, surface tension working to hold the epoxy in place while it cures.

The panel for the mirror box wraps around the sides, making a closed loop. However, the top and bottom edges have to be sealed. Figure 12 shows the last test piece I prepared before cutting the final pieces for the mirror box.

Figure 12

Pattern of cuts on the inside of the mirror box.

The finished mirror box is 11" high. As can be seen in Figure 12, a ¼" wide groove, ¼" deep, was cut around the inside surface approximately 4" from the top. This groove accepts a ¼" thick plywood shelf containing matching strut holes in the corners and a central opening (Figure 13). The primary purpose of this shelf is to prevent the struts from accidentally striking the primary mirror when they are inserted. In addition, the shelf stiffens the structure and provides a secondary baffle for the primary mirror. The inside veneer and core of the honeycomb panel were removed along the top ¼" and lower ½", leaving the outer skin (can be seen in Figure 12). This allows the ¼" thick plywood top and ½" thick plywood bottom pieces to drop in place, sealing the paper core. In Figure 14, the bottom 1/2" plywood panel can be seen recessed into the bottom of the mirror box.

Figure 13

View of the interior of the mirror box without the primary installed. Interior ¼" plywood shelf provides guide holes for the struts, stiffens the structure, and acts as secondary baffle.

Figure 14

View of the mirror box bottom. The solid plywood supports the mirror cell and seals the lower edge of the honeycomb panel. Four legs prevent damage to the bottom and adjustment screws when the mirror box is placed on the ground. Power connection and switch for the fans are located on the left.

The honeycomb panel on the sides of the mirror box is not strong enough to hold screws used to attach the altitude bearings. One approach is to install solid core plugs as was done for the rocker box pivot bolt. However, I decided to construct the mirror box in steps, creating top and bottom "U" shaped sections that are subsequently joined together with flat side panels. Two of the vertical seams are visible in Figure 15. The core along the adjoining edges was removed and a solid wood spline (Figure 16) was glued into place. The spline creates a strong joint and provides solid material to support the bearing attachment screws. ¼"-20 screws pass through the mirror box and screw into T-nuts inserted on the inside.

Figure 15

Each altitude bearing is attached via two ¼"-20 screws passing through the side of the mirror box and into T-nuts inserted on the inside. Note the two seams aligned with the screw holes.

Figure 16

Each rocker box seam is held together by a plywood spline glued between the panel plywood skins. The solid wood provides the support for the T-nut and attachment screws.

I spent a great deal of time conducting experiments determining the optimal size, number, and placement of fans for blowing across the surface of moderate sized mirrors. The 1" thick mirror used in this scope cools quickly, especially in the low profile mirror box. However, I was determined to implement the results of my testing and decided to install fans in this structure. The interior skin and core were removed in the areas of the fans and exhaust holes. ¼" thick plywood was glued in place to allow fan and exhaust holes to be drilled. I think fans sitting on the outside is unattractive, so I mounted them out of sight on the interior surface (Figures 13 and 17). Rather than make the mirror box larger than necessary, I used fans that were 15mm thick vs. the standard 25mm. Two 80mm diameter fans are mounted side-by-side on individual vibration damping gaskets and are directed slightly downward and across the face of the mirror.

Figure 17

Two thin 80mm diameter fans are mounted on the inside of the mirror box and direct flow across the surface of the mirror. The fans are angled slightly downward.

The rest of the scope is a minimalist design using a single upper ring instead of an upper cage assembly (Figure 18). The upper ring and mirror box can be joined together in a variety of means. The traditional approach is to use eight angled struts to form a truss. That would work well. However, I chose to continue the parallel strut theme I've used in prior scopes. The square mirror box suggests placing one in each of the four corners (Figure 19). However, symmetry isn't necessary. Three points define a plane. Three struts provide adequate alignment of the upper ring to the mirror box.

Figure 18

A single upper ring supports the focuser, spider, and baffle and is supported by three parallel struts.

Figure 19

The obvious location for the struts are the corners of the mirror box.

On previous parallel strut scopes, I've used wooden clamps to attach the struts. The clamps are time-consuming to make. Consequently, I wanted to explore simpler attachment methods. The Mark IV uses the same size struts as the Mark III. Each has a diameter of 1-1/2" and wall thickness of 0.049". Threaded inserts are placed inside both ends of all three struts. ¼"-20 screws protrude through the bottom of the mirror box (visible in Figure 13) and are permanently held in place with T-nuts on the inside. When assembling the scope, the struts are inserted into the corner holes and screwed into place. Likewise, ¼"-20 screws attach the upper ring to the tops of the struts. The screws are held captive in the upper ring, but require a tool for installation. Substituting clamp knobs with a threaded stud would make assembly tool-less, but I prefer the look of cap head screws.

Attaching the struts at only the ends creates a long unsupported length. This is the easiest arrangement to implement, but it allows the struts to flex the most. However, if the spacing of the struts on the upper ring is different than on the mirror box, the struts will be deflected inward (or outward) and force the struts against the walls of the holes on the top of the mirror box. It is almost as if the struts are clamped into place at the top and bottom of the mirror box and the upper ring. The effective length decreases by the height of the mirror box. Since stiffness is proportional to the cube of the inverse length, the stiffness increase can be substantial.

There are basically two choices for the location of the focuser. It can be placed on the side where there is one strut or on the side where there are two struts (Figure 20). If placed on the side with only one strut, there are more options for focuser position. For example, on smaller scopes, it is often more comfortable to have the focuser located at an angle approaching 45º vs. horizontal. However, since the distance between the struts is great, the upper ring should be stiffer (i.e. thicker).

If the focuser is placed on the side with two struts, the upper strut (at 45º) restricts where the focuser can be mounted. On the other hand, the upper ring does not need to be as stiff (i.e. thick) since the focuser attaches to a short span between struts. The finders also can be conveniently located on the top strut, near the focuser (Figure 21).

Figure 20

The long arc between struts at the top right and bottom left offer lots of space to mount the focuser. The short arc between struts on the top right and bottom right offers little space to mount the focuser.

Figure 21

The base of a JMI RCF mini 1 focuser was removed, permitting the focuser to be rotated 45º and mounted directly onto the mounting bracket, close to the strut which also carries the finders.

On this 12-1/2" f/5 scope, I wanted the focuser mounted near 30º. The focus knobs of most focusers further limit how close the focuser can be mounted to the strut. Generally, rotating the focuser so that the focus knob is farther away doesn't work because the corner of the square focuser mounting plate gets in the way. However, JMI's RCF mini 1 focuser offered the opportunity to do exactly what I wanted. The entire RCF focusing mechanism attaches to its mounting plate with only two screws. I discarded the mounting plate and attached the focusing mechanism directly to my home built mounting bracket (figure 21). I was able to rotate the focuser 45º so that the focus knob cleared the strut, while placing the focuser at approximately 25º from vertical - providing a comfortable eyepiece viewing angle for most elevations. Eyepiece height at Zenith is 60".

Some people might point out that I could have used a helical focuser or ask why I didn't use a 2" focuser. One of the subtle benefits of the 1-1/4" focuser used is its 3" long drawtube. A long, narrow drawtube allows a smaller cone of light to reach the focal plane than shorter and/or wider drawtubes. Consequently, the baffle on the other side of the ring can be smaller, lighter, and catch less wind. In addition, an f/5 scope does not sacrifice much field of view by using only 1-1/4" eyepieces. On the other hand, helical and 2" focusers are available with finer focus, which I would prefer.

When creating a new design there is always risk that it won't work as planned. In particular, there was considerable risk with this new scope since I implemented four new elements simultaneously:

Well, OK, installing the fans didn't entail much risk. They were unlikely to jeopardize mechanical or optical performance. The thin primary mirror cools quickly, and has plenty of time to do so since I usually set up before dark. To see any obvious improvements from the fans, I will need excellent seeing conditions coupled with immediate use of the scope after moving it between significantly different thermal environments. However, I see no deleterious effect of the fans while viewing through the eyepiece at high magnification. The damping gasket and damping characteristics of the honeycomb panels effectively eliminate fan vibration being transmitted to the eyepiece.

I am extremely pleased with the scope. I've received numerous compliments about its appearance and performance. In particular, the curved mahogany mirror box attracts a lot of attention. Because of its stiffness and plywood edging, most people assume the rocker box is made of solid wood. People are amazed that it weighs only 9.8 lbs and the entire scope, including finder, weighs 40 lbs. When observing near Zenith, vibrations damp out in about 2-3 seconds. At lower elevations, vibrations damp out within 1-2 seconds.

I very much notice and appreciate the light weight when packing up. On the other hand, at star parties, I am now more concerned about the scope being tipped over during wind gusts. This isn't a problem when the scope is in use because the minimalist design presents a small profile to the wind. When other scopes are being pushed around like wind vanes, this scope is steady. My concern arises when the scope is covered during the day.

The composite panels reduce overall weight, perform very well, and the mahogany skin is particularly attractive. There are downsides. Materials are available only through a handful of specialty distributors or directly from the manufacturer. Often minimum quantities must be purchased. Materials with the lauan skins cost no more than typical hardwood-faced plywood, but the mahogany panels cost 2X - 3X more. Unfortunately, the mahogany panels are the only ones that can be curved.

The honeycomb panels were not the only key elements to making the lightweight scope. For example, it was necessary to minimize weight of the primary mirror (11 lbs) and six-point floatation mirror cell (1.4 lbs). Top and bottom views of the custom lightweight cell are shown in Figures 22 and 23.

Figure 22

Top of lightweight six point floatation cell.

Figure 23

Bottom of lightweight six point floatation cell.

Is it possible to build a lighter scope? Yes, but I think the labor and materials costs soon reach a point of diminishing returns. Except for bragging rights, I see little need (even disadvantages) to building a lighter scope. This scope is now light enough to take as checked baggage to Australia - a trip I hope to be taking soon.