23.8.15

7" f/12 iStar refractor 2: Alternative mountings for a classical refractor:

*
Anyone hoping for advice on how to spend £oodles of money on a commercial mounting should look elsewhere. Unless you are firmly within the $AP1200/Titan income bracket then you should probably be looking at older 16" reflector mountings of yesteryear.

A minimum of 2" shafts [for 6"-8" refractors] should be your goal. 3" shafts for instruments above that size. The AP uses pre-loaded bearings to ensure rigidity with low friction. The majority of vintage mountings have shafts which just slide through the bearings. No end loading of conical bearings is possible. Those of us who cannot afford AstroPhysics quality should be searching the small astro ads for a really massive old mounting instead. You can't afford portability on a shoestring. Not unless you completely rethink conventional mounting wisdom. Even then you must [usually] sacrifice appearance and [quite possibly] a degree of portability.


Expanding the shafts will deny you conventional pillow block bearings. The sheer size and weight of these will not allow your mounting to be moved about much. If at all! So expand the shafts but leave the truly massive bearings safely back in the illustrated catalogue. Make the large axes out of hollow pipes and run them directly on slippery bearing material instead. Even if you did use ball bearings you'd only have to add friction somewhere. Usually at the clutch between the shaft and the drive's wormwheel. The silky smooth, light and near constant friction of Teflon against PVC is your pocket money saviour. You can have a 4", or even larger, polar axis pipe with much nicer movement than ball bearings on a wimpy and flexible 1" shaft.

An image of a 2m x 160mm tube mounted on the MkIV on its massive pier. A comfortable balance of scale for a 6" f/12. The MkIV has 1.25" shafts and 2' long saddle. Fullerscopes sold their 6" f/13 refactor on the MkIV. Considering the MkIV was sometimes used for heavy 16" reflectors it should be able to cope easily with a classical 6" refractor.

The diameter of the PVC bearing surfaces will dictate the degree of resistance to movement applied at the eyepiece. Fortunately pipes readily lend themselves to concentric construction. If a 4" pipe moves too freely then make it a 6" with a spacer ring between the two diameters. The ring can be cut from plywood, using a lathe, a router or even sawn out. Electric routers are available for very little money these days. Add a circle jig and an accurately concentric ring scan be cut out in a few minutes.

Want even more bearing resistance? Then make your bearing diameters even larger. Short stumps of PVC pipe should be available somewhere. Even if you have to ask the chaps laying pipes in the road. Pipes get crushed regularly by contractors machines but sometimes leave enough to make a bearing, or three. Short stumps are considered scrap and may even be burnt on a bonfire! Check if there is a well-boring company locally. They usually have masses of pipe of every imaginable diameter [and colour] which can be cut off with a hacksaw or even a carpenter's hand saw. You can tidy things up when you get home.

If  bearing pipes are not readily available then make your bearings out of plywood disks and wrap them with Formica counter-top edging strip. Only testing will dictate the diameter you need for the applied load on the bearings and the leverage you can apply at the eyepiece. Do not underestimate the latter. That foolishly long telescope tube can really be your friend in applying fine and smooth, hand control. With the right construction there will be no familiar oscillations. As the OTA moves effortlessly instead of the jerky snatching of an overloaded commercial mounting struggling to control your long OTA.

The mounting wants to be well triangulated or box-like in form to resist flexure of your chosen materials. 18mm or 3/4" waterproof, exterior plywood is popular and will last for years if painted and protected from the worst of the weather. Spacing the bearings well apart is always a useful ploy to avoid rocking and play. Your greatly expanded axes won't flex so don't introduce play elsewhere by trying to make everything too compact. Well spaced bearings don't throw heavy local loads on the construction. Small mountings have to be made of metal to resist local flexure. Your massively overgrown variety has no local loading to worry about if you design and build it well.

An image of the MkIV carrying a 20cm x 2m [8" x 6'6"] tube. I have had to order some 20cm/8" rings so had to make do with cord for this image. The scale still looks reasonable for a 180mm/7" f/12. Though the MkIV now seems to have shrunk even without a long dewshield on the 'OTA' mock-up. With a matching 27" overhang, on either side of the saddle, the ground clearance is 37" when pointing at the zenith. A bit low, but not impossible to reach, from a normal height, kitchen chair when an eyepiece is fitted in a star diagonal. It would be much more comfortable with a lower seat or 4' ground clearance. Which would mean raising the MkIV 30cm/12" from its present height. Most easily achieved by standing the pier on 12" high foot extensions. Or making  a massive plywood box for the MkIV to sit on top of the present pier flange.

The problem is that the pier is so heavy that it is all but immovable from its present spot. So it would not give me the portability I crave for whole sky observation. Perhaps the answer for the MkIV is a very tall wooden tripod built of substantial timber. Metal is prone to vibration but wood damps vibrations rapidly. Tripods are a bulky method of avoiding flexure in the support for the mounting and the legs often get in the way when viewing overhead.  They are also incredibly difficult to move with a heavy mounting and potential 30lb OTA sitting up on top!

A form of over-sized "wheelbarrow" using puncture proof, barrow wheels could lift the whole pier vertically. It would need very long handles [for leverage] and a very high lifting point to avoid all risk of tipping during movement. The pier + MkIV + OTA could then be moved around bodily to a new site as demanded by the object under study. Moving the MkIV between several permanently installed piers is a complete non-starter. It is far too heavy on its own to be handled that way.

Several piers equipped to take a plywood "Berry" offset fork would work best. The OTA can be "simply" lifted out of the altitude trunnions. The bare fork is then carried over and dropped onto the next pier, the OTA replaced in the fork and one can continue observation. How likely is this to happen in real life? Better ask somebody who could get away with planting several 8' tall steel pipes in large concrete blocks in several places in a rural garden.

On the same subject: Steel pipe piers set in concrete want to be up to the job because they have to be so tall to mount a long refractor. Leverage applied by the OTA to the mounting, or by the wind to the long OTA, will easily find the flexure point of a slim pier. My old, 6' high x 4" diameter, steel pipe pier was far too heavy to lift comfortably but still flexed just above the ground in the slightest breeze. The stiffness of a pipe rises as the square of the diameter. If you decide to use a pipe pier then make it at least 6" in diameter and preferably thick wall.

I just hope you have the clear view of the sky to make the most of where you sink it into that huge block of concrete. Too small a foundation will turn the instrument on its mounting into a compound pendulum. Look up "compound pendulum" online and be very afraid! You may think that filling the pier with concrete would help but it will only lower the resonant frequency of the oscillations. So don't stint on your pier pipe nor the foundation block!

An alternative to a tripod is to make the entire mounting into a tripod. English Yoke or Cross-axis mounting, anybody? The Yoke provides by far the best support at the cost of sheer size and difficulty of transport. Like a fork mounting, it offers support to both sides of the telescope tube. All the bearings are widely spaced for maximum rigidity and lack of play. The length of the yoke arms will be dictated by the length of the OTA. Unless you want a large circle of the north polar sky to be inaccessible then the tube must fit inside the yoke when pointing north.

The cross axis is  like a German mounting but with greatly extended polar axis. The north bearing can be a simple Teflon/PVC pipe bearing. The south bearing can be the same but will want some low friction, end resistance. A single bearing ball, or pointed pin, in a suitable retaining dimple will often suffice. A seriously heavy mounting can use a junk, taper roller bearing.

The downside of the Cross-axis is the need for a counterweight. Though this can be applied almost literally anywhere along the extended polar axis. It need not be directly opposite the telescope tube. Like it must usually be in a compact German mounting. Getting the counterweight out of the way provides much more room in a compact observatory. Walking into the counterweight in the dark is not a pleasant experience. Even worse is standing up under it! The very long polar structure of the Cross-axis allows the counterweight to be placed permanently out of the way. Either down at the southern foot or above the observers head at the north bearing.

A hybrid of the Yoke lifts the OTA above the yoke to allow full polar access. This also requires counterweights to balance the OTA assembly. There are offset equatorial forks too which will also allow a long fork mounted refractor to access the Pole. Again, the necessary counterweights can be fitted at either end of the extended polar axis well away from the observer's likely need for space around the telescope.

The main problem with  these more exotic mountings is the need to provide ground and support clearance for the observer at the eyepiece. With a long refractor this can mean a very tall north bearing support and very long yoke arms. Or an even longer polar axis for a Cross-axis. Moving a Cross axis or Yoke is probably as difficult as a MkIV on a pier. There is no natural point to lift the whole construction however light one might manage to make it.

Allowing the observer to sit on an adjustable height, observing chair will reduce the required scale quite dramatically. Particularly when  compared with having him, or her, stand while observing at the zenith. Reducing the eyepiece ground clearance offers major benefits at most other pointing altitudes too. The use of a star diagonal is taken for granted at higher pointing altitudes but may force the use of a ladder at others. The straight-through view may well be more appropriate at lower pointing angles of the telescope tube.

Try facing the stepladder and looking over the top rung, or around the side. This is often the most comfortable and safest way to observe when you really do have to be off the ground. Don't ever use a flimsy "indoor" stepladder. Builder's stepladders are tough and don't flex and often have extended crossbars joining the feet for greater stability. The folding/locking variety are both cheap and stable and have seriously big, rubber feet. Useful for those bumpy lawns and when snow is lying. Moving them around slightly, to find the most stable position will usually provide a safe perch out in the darkness.

You really don't want to be found lying, frozen solid, perhaps a week after you fell off, because you were observing alone on a shaky ladder! An indoor [domestic] stepladder will sink into the grass, gravel or soil and flex all over the place through piss-poor design and dirt cheap material choice. I have fallen off them enough times to be something of an expert and that was in broad daylight while picking apples!

Having read all this you way wonder why you'd want to reinvent the equatorial mounting when a "Berry style," offset/counter weighted, altazimuth fork is so easy to set up and use. Well, if you are really clever you could become the next John Dobson in equatorial refractor mountings. He broke the stranglehold of heavy equatorial mountings on large, reflecting telescopes. His quiet revolution can still be seen in ever-larger, amateur instruments. Now pushing well beyond the 1 meter / 40" scale. All rely on his sliding bearing principles to support unheard of apertures in deliberately portable form.

Instrument sizes which were once only found in great observatories are now regularly carted around the countryside, for hundreds of miles, in quite ordinary vehicles and set up in fields and parking spaces to share the views and increasingly rare, dark skies. Lightweight, fast mirrors became the norm. As did truss tubes and multi-point, mirror support cells. There are now far more company's making telescopes and their vital components than at any time in the past. Sophistication in design has revolutionised component choice. The startling astro images produced by clever amateurs have long eclipsed those of the professional observatories of only a few decades ago. It takes a multi-million dollar, orbiting satellite with a full instrument platform to provide similar images today. Who knows where amateur imaging progress will lead as computing power, clever software and optical sensors follow a steeply climbing path towards the stars themselves. 

Meanwhile, refractors have hardly been touched by Dobson's great sea change in ATM. A six inch, long focus, "classical" refractor is still considered a huge instrument and remarkably few affordable, modern, commercial, equatorial mountings can possibly cope. It's not just a matter of shaft  size but controlling the moment arm of the heavy refractor lens in its cell on the far end of a 6' or [much] longer tube. Which usually means fitting large enough wormwheels to provide the necessary torque and vital braking forces of that huge, unwieldy tube. If you demand compactness, lightness and portability too, then [quite literally] it "doesn't compute!"

The trend to short dovetails on mountings does the refractor no favours at all. The long refractor tube wants to be supported on sturdy, widely-spaced rings from a long and rigid saddle fixed firmly to a really massive, declination axis. The rest of the commercial "dinky toy" mountings and their tripods may cope with a tiny APO but not much else. Many imagers now use motor driven focusers to avoid contacting the OTA. Portability and Goto have become the hype factors in commercial mounting choice. Or, to misquote the popular Irish expression: [If I owned a "real man's" refractor] "I [really] wouldn't start from here!" ;ΓΈ)
 
*

No comments: