28.10.16

2" shaft mounting Pt.51: Saddle reinforcement.

*

Publishing one's ATM on a forum like Cloudy Nights is apt to throw up questions and suggestions.  The 100 x 50 x 5mm channel section saddle was questioned as to its stiffness. So I found some aluminium angle in 50 x 50 x 5mm and narrowed one web so that two lengths would fit side by side inside the saddle's channel section. The advantage of this arrangement is that the Tollok bush is bolted right through the now, 10mm thickness. As are the tube rings of course. The greater thickness should preclude twisting or other flexure modes without needing local reinforcement on the saddle.

If only life were that easy. There was no room for the Tollok bush in the narrower space without major surgery! I don't have any angle section large enough to meet on the outside of the saddle channel.

I considered placing an inverted channel on top of the original. If the two were bonded together the saddle would become H-section. The problem is that it looks rather ugly and contrived. Perhaps I should look for some flat plate or even box section which would fit between the webs of the original channel. Or I could contact an eBay aluminium stockist to see if they have an off-cut of something sturdier in channel section. They might have a short length which isn't worth listing.

Clamping the channel saddle at one end to the workbench showed it to be quite stiff except in rotation.  I then tried clamping one end of one of the 2m [6'6"] rectangular box, straight edges to the workbench. An 8" G-cramp [C-clamp] applied to the free end showed the beam  was very stiff in its deepest direction, remarkably stiff against rotation but rather floppy in the thinnest direction. All largely as expected.

Using this material in two or three layers for the saddle would both save weight and increase stiffness. The only problem is the ease of crushing with screwed fasteners. It really needs solid metal inserts to resist crushing forces. It would not be too difficult to push alloy packing strips down the channels to the center of the saddle. They could even have some epoxy added to keep them in place. This would give the Tollok bush screws and flange something hard to work against.

Beech inserts did not work when I used these beams for my 10" reflector. They did not have the necessary resistance to crushing forces. Which is a shame because I have lots of old beech bed slats saved for my construction projects. I could entirely fill the beams with those if it were worthwhile. I suppose I could have metal inserts at the center and ends with beech packing in between. Or, I could epoxy laminate a single beam on top of the existing channel section saddle to increase its stiffness. With the caveat that resistive packing would still be essential to avoid crushing.

Then I had the idea of adding a half pipe of aluminium to the back of the channel section saddle. My length of pipe managed to be 100mm in diameter. Exactly the same as the width of the saddle but not the 90mm I needed to fit nicely inside the channel. Grr? I wonder whether a half of the tube is capable of being gently squashed to a slightly smaller diameter? 

As can be seen from the images above I finally chose to add a slightly lower version of the existing [channel] saddle. The adjoining plates are now 10mm thick with ribs on both sides. The Tollok bush is bolted through both. As are the tube rings. Further fixings along the edges might be useful to reduce flexure until I can bond the two halves firmly together with epoxy. I have only run a 120 grit angle grinder wheel across both sections so far just to clean them up. Further smoothing work with finer abrasives will follow.


Click on any image for an enlargement.

*

26.10.16

2" shaft mounting Pt.50: Boxing in the forked base.

*

The height of the base fork blades is still annoying me slightly . The fork only has to be high enough to allow the 11.5" wormwheel to fit under it without contacting the base plate.  I'd like to lower the height of the fork base for cosmetic reasons alone. No doubt a shorter base height would also help to increase its stiffness.

It is now much too cold to be working with epoxy out of doors. So at least the pressure is off to finalize the fork length for the moment. I shan't be able to bond the doubled blades until I can arrange local heating in my normally unheated workshop. I don't want to risk the wrath of my long suffering wife by working with epoxy indoors. There has recently been a huge health problem for those working on epoxy laminations in the Danish windmill industry.

The image shows how the forked base will appear once it is boxed in [at the front and back.] Boxing is important to avoid the fork blades moving independently of each other around the pivot on the RA housing. When I lower the mounting, on the chain hoist, the mounting wants to lean drunkenly to left or right. Each fork blade wants to 'fold' around the polar altitude pivot. Eventually the polar altitude will be finely controlled by a stainless steel turnbuckle fitted between the base and the polar axis, bearing housing.


It took some time to saw out and file a neat circle in the 10mm, 3/8" alu. The hole will provide access to the azimuth nut for using a ratchet. I allowed a little extra freedom at the top of the stiffening plate between the fork blades. This was to ensure the altitude pivot clamping was still secure.

I also beveled off the top and bottom of the front plate so that it sits flush against the base plate and polar axis housing. The height of this stiffening plate will be determined by the PA altitude angle. As will the angle of the fork base where it rests on the base plate. Mine is set up for 55N.

A base plate between the forks proved to be more important for maintaining squareness of the fork blades than a rear sloping one. Though the polar axis housing at the top provides plenty of squareness there is no harm in further reinforcement at the base.

Soaping the 36 grade, angle grinder disks does seem to help the disks to continue cutting. Soap works on the files too provided they are cleaned with the file card [flat steel brush] in between efforts to remove metal. The mounting is finally beginning to look as if it actually means business. The next step is to drill the fork blades for cross studs and furniture nuts. Just as I used them on the bearing housings. Then the azimuth pivot arrangements need further scrutiny.

Click on any image for an enlargement.  

*

21.10.16

2" shaft mounting Pt.49: Counterweight retaining bush.

*

While I was on the lathe I turned a hefty, brass, counterweight retaining bush for the end of the declination shaft. However, it would require rather long screws on typical plastic tightening knobs. Unless, of course, I add a wide spacer to ensure finger clearance. Or used hex socket head screws which can be tightened with a hex wrench. Having struggled with hex wrenches on the Fullerscopes MkIV weight retaining bushes over the years I am not very keen. It can be critical to get the clamp on and tightened before one's strength fails and all the counterweights slide off onto one's feet! Which is why I would much prefer tool-free tightening and free-fitting weights and clamping bushes. It was always a struggle to remove the MkIV's shaft clamping bushes.

Simple thumb screws were used on the Fullerscope MkIII mounting bushes. They attracted rust like a magnet due to to both components being made of steel. Wing screws need only a little clearance for the fingers but aren't normal stock items in most DIY outlets. So the search begins again. I could use normal, plated, wing nuts and thread lock a stud into them. They would rust too unless I can obtain stainless steel wing nuts. I can feel another search coming on unless I buy them online.

As a temporary measure I drilled and tapped the bush at 120° intervals anyway. Three radial screws seemed like a better option than the usual one. Then I used plated steel M8 screws and wing nuts. The socket head screws needed the undersides of their heads tapered in the lathe to fit inside ordinary wing nuts. This gave the option of finger tight or hex wrench for extra grip, while still looking quite smart. Now I have had some practice I shall buy stainless steel screws and wing nuts in the same size for a more permanent [non-rusting] job.

Wing nuts provide plenty of torque so there is no hurry to get out the wrench to retain the counterweight. Though I would want to be absolutely certain the weights would stay in place when the declination shaft is vertical! I can see the good sense in radial drilling posher counterweights for individual screw locking. This would not be possible with my cheapo dished 'bodybuilder's' counterweights. I may need to dimple the shaft with a drill to provide extra security for the weight retaining bush screws. First I need to know how many and where the weights will be normally positioned to balance the OTA. An A-frame prop would aid weight fitting and removal.

To complete the day's efforts I drilled out all six altitude pivot holes to 16mm. A painless exercise using the lathe in back gear, the drill in the 3-jaw chuck and the tailstock providing the necessary pressure. Trying to drill large holes in under-powered, DeWalt rechargeable drills, or cheapo bench drills, is a complete waste of time and effort. The speed is far too high so they chatter and grab because there isn't enough pressure even if you lean on them.

Click on any image for an enlargement.
*

20.10.16

2" shaft mounting Pt.48: Declination/saddle bush reinforcement.

*
The Tollok bush which holds the saddle onto the end of the declination shaft had little or no room for an expansion ring. The supplied steel ring would not fit inside the saddle channel section without surgery. This would involve thinning the ring considerably with the potential for disaster. The Tollok bush relies on opposed tapers to grip the shaft and simultaneously expand into something tubular which offers serious resistance. The grip on the shaft relies on the physical containment of the outer, flanged bush. 

I decided to turn a full length cover for the Tollok designed to provide the vital resistance to expansion. Unfortunately I have no round aluminium stock in the necessary diameter. So I was forced to use brass. This is much heavier than I wanted for this situation and will require more counter-weighting. The advantage of the sleeve is that it protects the bare metal of the Tollok bush from unsightly rust. It also overlaps the flange bearing projection.

The images show the general idea. The piece of scrap, brass bar had been discarded by an educational establishment after a couple of knurling exercises. I left the knurling in place for decoration. The middle image shows the Tollok bush sunk flush with the brass sleeve. All much as I had done with the much large [7" diameter] aluminium cylinder on the polar axis.

Having the entire Tollok bush enclosed maximizes its ability to compress strongly onto the declination shaft without local stresses or flexure. I reasoned that since both halves of the bush are  split, the full length enclosure would remove any chance of local flexure. Far better, surely, than a short ring?

The narrow front lip, which accepts the inner cone flange, set the maximum and minimum diameter which would just go inside the saddle's channel section. I deliberately allowed a little extra clearance here to avoid stressing the thinner lip by internal expansion. The much thicker, main body was made a closer fit where there was over a cm of solid brass to resist expansion.

I had bought some oversized, stainless steel washers to fit the Tollok, clamping bush screws. The idea was to spread the loads from the ten replacement, longer, stainless steel screws. Due to the screw spacing I had to turn the washers down in diameter to just clear each other. This ensures the maximum area and of the saddle is sharing the pull from all ten screws. Adding a load spreading plate under the screw heads would be pointless unless the plate was bonded on. It would also demand much longer screws with the potential for flexure.


Click on any image for an enlargement.
*

11.10.16

2" shaft mounting Pt.47: Firmly planting the base fork.

*
My initial plan to add angle profiles on each side of the fork base would not provide vertical stability. It badly needs some downward pressure which might as well provide some rotation for polar alignment at the same time. This will require a division between the fork base and the large bottom plate. Unfortunately I only bought one of the these large 10mm [5/16"] plates from the scrap yard. So cannot simply duplicate the base plate with a central pivot and familiar screwed azimuth adjusters. I do have a little more of the 15cm x 10mm plate but wont have much in reserve if I duplicate the fork tines with doubled plates. Nor will the plate provide enough resistance from the compression of a large, central pivot screw.  

The fork is now sloping backwards by 20° to provide more clearance for the wormwheel. The cross studs and vertical pivot ought to be centrally placed relative to the fork tine base. Which doesn't provide much room the higher they are placed.

I was thinking about having a really sturdy crossbar between the fork tines to resist the compression loads. A large nut and washer would provide the downward pressure via a vertical pivot stud or bolt. I do have limited milling ability on my lathe but would rather avoid milling slots in the fork tines just to fix a crossbar between them.

The original idea was to bolt the entire mounting together without machining. The cylinder rather dented the original plan but was at least optional. I just happened to have a chunk of 7" round stock. Alternative means could have been found for effectively supporting the Declination axis.

Two sturdy, spaced, 15mm studs between the tines would allow a vertical stud to pass between them with a 6" plate resting on top and sandwiched between the tines. The stiff cross studs would help to support the cross plate and simultaneously compress the fork tines together against the edges of this horizontal plate. Or plates, if a bottom plate is added.

Here is an image of the mounting supported from the new 1T lifting strop and 1T chain hoist. The steel hook allows more rapid movement along the ceiling joist than the former, multi-looped cord.


Front and rear sandwiched plates would make a closed box out of the fork and it would all be under heavy compression from the twin, horizontal cross studs. Plus the vertical compression from the central, pivot stud. The tops of the tines are compressed against the sides of the Dec housing by the altitude pivot stud. All helping to avoid flexure anywhere in the fork.

However, first I have to decide on the final fork dimensions and angles. I must avoid contact between the large RA wormwheel and the base plate, fork or large diameter pier pipe.

Materials shortage over. Another visit to the scrap yard produced more aluminium plate in various thicknesses up to 10mm. I now have a second 10mm [5/8"] base plate to bond to the first. A pack of Loctite 'Metal' is on its way in the post.

I have received some useful information on filing and "grinding" aluminium. Files pick up the metal which causes scratches and slow removal of material. Using ordinary bar soap on the file is supposed to help. I've just bought a bar of soap to give this a try. Lamp oil had little or no effect despite being recommended [and highly beneficial] for turning aluminium. The lamp oil had an amazing effect when I was turning the 7" diameter cylinder yet had no obvious benefit on my files. I wonder whether soap would help on the 36 grade angle grinder disks? Might be worth a try.

With the fork slope decided I could duplicate the tines. They are now 20mm thick [0.8"] and look far more substantial. As does the 20mm thick base though it has yet to be trimmed.

Open fibrous abrasive disks have been recommended for cleaning up rough aluminum edges and surfaces. I tried the local DIY chain store but they had nothing like it. Another online search and purchase is obviously  required. 

A 44 mile ride on my touring tricycle produced some goodies to speed up smoothing the aluminium. The card, file brush [bottom] will help to clear aluminium from my files.

I spent another hour trying to level the sawn edges of the duplicated fork tines. Rubbing dry soap on the disks and files is a definite improvement.

The Loctite 'Metal' epoxy should arrive today to fix the plates firmly together. Plates simply held together with nuts and bolts tend to behave more like leaf springs under bending loads. The epoxy should make them behave more like the thicker material the laminations pretend to be. I shall have to have all my G-cramps ready to squash the plates hard together.

Click on any image for an enlargement.

*

9.10.16

2" shaft mounting Pt.46: Going down below.

*
A trial assembly with the RA wormwheel at the bottom of the Polar axis made far more sense. It also showed that I should move the altitude pivot upwards along the PA housing. This would better balance the rather top-heavy appearance. It would also push the wormwheel further away from the base plate, pier and fork. Perhaps even allowing considerably shorter fork tines.  I have sawn off the studs while still leaving a small reserve for unforeseen changes.

Am I a robot?

Rain forced me into the workshop today but at least I had the chain hoist to help me lift things into place. The arrow in the image shows the direction for a new altitude pivot point. If I removed the nearby furniture screw it could rise even further. The [much larger] pivot stud would provide far more compression than the smaller one. So there would be no sacrifice of strength. Once the new pivot point is properly established I can then measure how far I can shorten the fork tines. I am trying to maintain the full radius curve on the top of the fork tines for appearance and maximum friction.

Here I have moved the pivot further up the PA. The highest point was not the most favourable and I had cross studs to avoid.  So I moved it 3/4" down on the same line. I think the whole mounting looks much better balanced now. It is suddenly looking rather impressive [in real life] thanks to the large scale. The tape measure shows the overall height to the top of the saddle is now 40" or 100cm from the base plate.

The mounting's overall balance is now backwards or clockwise seen from this viewpoint. So I had to make up a stub axle which could hold a weight and be locked against falling by tightening the flange bearing grub screws. Adding the OTA, declination axis shaft and counterbalance weights will overcome the imbalance. Allowing for a slight bias to be overcome by the turnbuckle for fine altitude adjustment.

As the fork tines are sloping I have to be careful that the large RA wormwheel does not get any closer. Just cutting off bits of fork would bring the wormwheel nearer. I could cut a more acute angle on the base of the tines. This would tilt them over more and bring the mounting's CofG further within the fork base. While simultaneously moving the wormwheel slightly further away from the base plate and pier. I have made up a cardboard "set square" for 55° to aid rapidly setting the mounting at the correct angle.

Now I ought to make the large pivot stud to get rid of the G-cramp and allow free access to the fork. I'm thinking of boxing in the fork with front and back plates held by cross studs, along much the same lines as the bearing housings. This will help to stiffen the fork more than simply adding channel profiles to the sides. It will also allow greater freedom to fit the dual angle profiles. I used a 12mm [1/2] galvanized stud for the pivot. The nuts look a little 'understated' against this scale of mounting. Some larger washers will help. Perhaps I should step up to the 15mm stud size I used in the bearing housings.

Click on any image for an enlargement.

 *

8.10.16

2" shaft mounting Pt.45: Forked base.

*
I have been putting off the base while I sought heavier [scrap] materials. They were not forthcoming so I am going to use what I have in 150mm wide strip x 10mm [6" x 3/8"] aluminium plate.

I shall build a simple support fork for the polar axis bearing housing. Which will pivot the housing to allow fine PA altitude adjustment via a sturdy, stainless steel turnbuckle.

I carefully marked out where the 50mm [2"] shaft resided inside the casing as well as marking the position of the heavy studs. [All threads] There seemed no particular purpose in moving the pivot any further north than half way up the housing. I may change my mind over this but at present it allows room for the RA wormwheel if I should move it to the bottom of the PA shaft.

I then drilled small holes in opposite plates and supported the polar axis on two nails. The G-cramp [C-clamp] keeps it all steady while I admire the initial layout to see if I have overlooked anything important. Which I often do these days. The worm support plate will be relocated on the side of the housing to allow a control rod to be brought back to the eyepiece.

The "nail" holes will be opened out to take the largest possible threaded rod [stud or all thread] to tie the fork tines rigidly to the PA housing. This continues the compression idea to make the housings as solid and stiff as possible.  Otherwise I could easily have used captive pivot bolts pointing outwards through the side plates. This would not have provided the desired compression as the side plates would be held only by the smaller cross studs.

I try to imagine flexure modes exaggerated into catastrophic failures. Hence the through stud rather than two loose bolts. Offsetting the pivot holes between the axis shaft and large studs offered two options. I could pivot the PA housing above or below its center line. The higher position caused clearance and appearance difficulties for the fork tines. Though the PA housing could still be flipped over if desired for a much lower housing position. But then the RA wormwheel won't fit under the housing.

Oblique view showing the fork more clearly. Once the studs are finally sawn to length below the lower flange bearing there should be room for the RA wormwheel. For the moment the studs make useful handles for carrying the hefty PA housing around.  They also provide convenient propping points for mock-ups to see how things look in practice.

The fork will be fixed down to the large 10mm [3/8"] base plate via sturdy aluminium angle both inside and out. A length of scrap angle is resting against the workbench in this image all ready to be sawn up. 

Even at 10mm [3/8"] thick the fork tines look extremely flimsy. So the sides of the fork will be further reinforced against flexure with channel section aluminium carried to full height. I have propped up a short length of channel to show the basic idea. You will have to imagine the channel rising up both fork tines and having a radius applied to the tops to match the rounded fork plates. I am considering using more furniture screws to hold the channel in place.

Perhaps I should double the fork tines to 20mm [13/16"] thickness? I still have enough 6" wide strip for this job. Epoxy between the two layers would bond the sandwich into a solid mass with far greater stiffness. I don't want to follow this route until I have finally decided on RA wormwheel position as longer fork tunes consume a lot of scarce material.

Should I decide to leave the RA wormwheel at the top of the PA axis then the fork tines could be considerably shortened. This is where a bolt-together assembly is handy for prototyping different ideas.


Click on any image for an enlargement.
*

5.10.16

2" shaft mounting Pt.44: Nuts and weights.

*
The silvery finished, furniture top, cap, ferrule, sleeve, flange, connector nuts arrived. As did the five x 5kg weight disks finished in grey Hammerite. The nuts were fine and arguably rather better than the brassy ones I had been using so far. All a matter of taste though.

The weight disks were very rough on the outside edges where the minimum of fettling had occurred after casting. I saw no reason not to turn the rims to remove the roughness and sharp edges if only they would fit on my lathe. By using an inverted boring bar and expanding the 3 jaw chuck into the ~51mm bore I was able to take cuts on the outside surface of the 230mm [9"] diameter x 25mm thick disks. I am usually limited to a maximum of 180mm [7"] in the 4-jaw chuck. A steel disk, with soft packing, backed up by a live center ensured the disks did not escape from the chuck jaws. [see image above]

I ran the lathe at 45rpm, in slowest back gear, to remove the worst of the casting roughness. Then increased to 60rpm to smooth things off once there were no more shock loads. It was taking about half an hour per disk using the lathe apron's slowest feed but I still felt it worthwhile. I would never be happy seeing the heavily tapered and very rough circumferences stacked together all higgeldy-piggeldy on the declination shaft. A rub with a coarse file rounded off the previously sharp and ragged rear edges to make them look much smarter and much safer to handle.

I think you will agree that the final result was worth the extra time and effort. I have done four but will leave the last in case I don't need it to balance the 7" refractor OTA. Not bad for about 100DKK, £10, $13 each plus P&P. They will look far more uniform with a decent coat of paint.

Click on any image for an enlargement.
*

4.10.16

2" shaft mounting Pt.43: A different kind of pier.

*

In an attempt to avoid a massive, static pier, or even to have something in which to cast concrete, I have purchased some more used ventilation tubing. In 12" and 14" diameters, as shown, I also have a long tapered adapter to join the two sizes. This would allow me to have a tapered section on top of the large tube. Or, to join the two at any chosen height with a smooth taper between them. They only cost me a modest contribution to the firm's cake fund to allow me to freely select my materials.

Note how the 12" tube would suit my 10" f/8 mirror to perfection. A lining of cork or foam might be useful to reduce thermal issues.] It will need my serious mount to cope with the weight and moment. I am very tempted to put a steel OTA together for my mirror.

There were no longer 14" examples amongst the collection to allow a full height pier in that size. Hence the adapter to give me some flexibility over height. I can join the larger tube and adapter and then cut off the larger tube at the ground. Or bury it and fill it with concrete! Both tubes weigh about 23lbs or 10.5kg each.

Before you jump to conclusions and assume this [galvanized] steel tubing is far too thin for a pier I strongly beg to differ. When I was younger I built a downhill skate car. This was built from the flimsiest aluminium printing sheets. The secret to its incredible stiffness was the use of twin walls with thin, softwood laths sandwiched and glued between them. The lower hull of the car was a simple half circle [roof gutter] form built in this way. It was impossible to feel any flexure [at all] when I climbed in.

Aerospace materials are built on similar principles. Often of such thin materials that the honeycomb core can be seen imprinted through the twin skins. Flush doors are also constructed this way using flimsy paper/card honeycombs glued on edge between two plywood skins. A wooden frame provides the firm base for the hinge and lock.

I could use a single tube reinforced by simple, plywood ring formers just like an aircraft fuselage. Or I could add tensioned, threaded rods between a top and bottom plate. The rods would pass though close fitting holes in the formers. The assembly would be slid inside the tube and then the nuts tightened to stiffen the whole caboodle. The rings would stop the tube from becoming oval if subjected to bending forces. The rings could easily be locally reinforced to provide solid anchors for triangulated legs. Just like the very expensive piers found under many Astro-Physics mounts. The secret here is triangulation and a complete lack of flexibility while retaining lightness for mobility.

http://www.astro-physics.com/products/accessories/mounting_acc/eagle6.htm#pier

I have no need of a short, floating pier tube cut off at midriff nor its very light weight. I also much prefer the "anchored down to the ground" look and potential stability to be had from larger, triangulated legs. I shall use four legs rather than three and have previously illustrated the tiny tipping radius of tripods in comparison with identically spread quadruple legs before. [See image above for the glaring difference in tipping radius.] I observe mostly on lawn which gets quite soft at times. So I just need massive stability with only rather limited mobility.

I have outlined various options for my pier. I could use both tubes placed concentrically and have vertical laths between the two. Or reinforcing rings between the two for an incredibly stiff but rather heavy column. Or use one tube with the construction outlined above. It would be simplicity itself to provide a ballast water tank at the bottom of the pier for extra stability. Preferably with a drain tap to lighten the load if needed. Or, perhaps better, in a cold climate, a door provided to allow sandbags to be placed inside or removed at will. Sand is also heavier than water so will provide a lower center of gravity.

It is not a minor thing to place a 200lb mounting on top of a nearly 7' tall column without serious consideration given to stability. Another 100lbs will be added with the OTA plus its balancing counterweights.  It can be thought of as similar to placing a real human being up there on top. Is their safety guaranteed by unconditional stability in all directions? If not, then I should not risk my life [or others more innocent] by standing under that considerable load. Nor is there any point in building a massive mount if flexure is introduced by its own stand or pier.

My present pier is truly massive in terms of the weight of steel I welded together. It appears and feels rock stable in use. However, when I try to move it around on its wheels it readily wants to tip over. I spend a lot of time adjusting the screw jacks on the wheels just to keep the pier upright.

The new mounting will be at about the same height but weighs at least twice as much. It will require more stability than all that steel only pretends to provide. I will have to keep that firmly in mind. A circle of the same radius as the tripod or quadpod offers even better stability. As would bolting a quadpod down to ground anchors. The danger then is when it is released to be moved across the sloping garden. Life would be so much easier if I had a clear southern view. During the recent Mercury transit I had to drag the present mount and pier along the gravel drive just to be able to catch the afternoon sun. Never again! I badly need a permanent site for the new mounting which has reasonably clear views of the sky.

Click on any image for an enlargement. 
*



3.10.16

2" shaft mounting Pt.42: Worm support plates.

Both worm support plates are now roughed out. I used the lathe in slowest back gear to drill the 16mm holes and large circle for the axis shaft. Backed up by tailstock pressure I was able to work slowly and steadily through the plates without snatching. It takes a very serious pillar drill to manage such large drills and few non-industrial pillar drills run slow enough. The edges of the plates were allowed to rest on the cross slide to avoid unwanted rotation as they were being drilled.

Even the cheapo Millarco hole saw worked well at such very low rpm with plenty of lamp oil for cutting fluid. It never had a chance to overheat and cut through in a couple of minutes working [slowly] from both sides for neatness. The 6mm center hole was pre-drilled of course.


Image showing the 8.75" Declination wormwheel and worm above the 11.5" RA wormwheel. The saddle has been removed so I could work on the worm plate and clearance by elongating the worm fixing holes.

Some means of applying finely adjustable pressure to the worm housings would be beneficial.

This is the wrong end of the shaft because it is concave. It should be flat this end to allow a check for shaft depth in the Tollok bush through the saddle peephole. My lathe will not accept such long shafts to allow me to turn the end flat. So I will have to resort to an angle grinder.
 

Another view with the saddle in place. I have aligned both worms to show the difference in scale of both worms and wheels. Since the number of teeth [287] is the same in both cases the teeth and worm thread pitch are obviously different as the circumference of the wheels changes.

I have been considering for some time whether the channel section saddle needs reinforcement at the Tollok bush. Perhaps I should turn a miniature version of the large cylinder I used on the PA? While the circle of screws does apply pressure over quite a large area the saddle material is not thick enough in itself to be ultra stiff.
 Side view showing [from right to left] the channel profile saddle, the Tollok clamping bush, the Declination wormwheel and worm, the flange bearing and finally, the plated bearing housing.

The M8 furniture cap nuts are proving difficult to obtain in any of the local DIY outlets. I used up the entire stock from the one place which did have old stock. They have promised to order more but who knows when they will arrive and what they will cost. I could order more from eBay[UK]but the postage of the only stockist of M8 doubles the already ridiculous price of £1 [$1.30] each! For a few little nuts! Crackers!

I just found a Danish stockist with prices to match my DIY outlet. I have ordered 20 and will now be able to finish what I have planned. My thinking is that using enough 8mm [5/16"] studs will stiffen the bearing housing as if they were made from sturdy square tubes. For this I need to be able to clamp the narrower, sandwiched plates as well. By compressing the plates in all directions it should actually be stiffer than a simple tube. The internal 'ironwork' of heavy and lighter studs adds its own structural benefits. By tensioning them they should greatly increase in lateral stiffness over simple rods.

I trued, smoothed and beveled both ends of the Declination shaft with an angle grinder to avoid 'handedness' when fitting. A bevel on a shaft is essential to ensure smooth entry into the bearings and other components.

Click on any image for an enlargement.
*

2.10.16

2" shaft mounting Pt.41: Balance in all things.

*

The declination axis of a German Equatorial Mounting [GEM] acts much like a seesaw. Add a heavy OTA on one end of the Dec shaft and it must be balanced by matching counterweights. If the Dec shaft is longer on one side than the other than the counterweight required will be changed in the ratio of the difference in lengths. [Or moment arms as they are known in basic physics.] This simple arrangement allows an equally simple moment calculation. Moment being Mass [weight] x Distance [moment arm] from the pivot [or fulcrum.]  The lever is assumed to be horizontal for simplicity.

The moment should preferably expressed in similar units. Metric: Kg x cm [kg/cm] or Imperial:  Lbs x inches [lb/inches] are matching units. Kg and inches only makes life slightly more complicated if one forgets and swaps units mid-calculation. Otherwise Kg/inches is just as valid as Lb/cm. Those brought up on Newtons can use those. I wasn't and still prefer the simplicity of familiar units of length and weight.

Note how I have deliberately offset the center of the PA axis [Dec axis fulcrum] relative to the declination axis length. This helps to reduce the degree of OTA overhang beyond the nearest Dec bearing. As does placing the declination wormwheel on the far end of the Dec shaft from the OTA. 

The OTA extends beyond the saddle by about half the ring diameter + any thickness in the ring fixing boss. I'm calling that 13cm from the center of gravity of the saddle with the Tollok bush attached.

I have marked the image with the actual dimensions.  The 45cm dimension [on the right] is the distance from the PA axis to the center of the bare Declination shaft. Provided the weights are evenly arranged the center of the weight cluster can be considered as the center of gravity of all the weights. This is the length of the moment arm of all the weights for our balance calculations. 

On the other side of the declination 'seesaw' we have the OTA. Its center of gravity is roughly the middle of the main tube for our purposes. This distance is 37cm in our example.

37:45 = .82. So we can lighten the counterweights relative to the OTA's weight by multiplying by .82.  If the OTA weighs 50lbs then we can use 0.82 x 50 = 41lbs. Which is about 20kg. This is the total weight of counterweights required to balance the OTA. This assumes that all of the CWTs are of equal size and weight. A larger or heavier weight will alter the balance depending on its placement along the Dec axis. As will altering the position of multiple weights of course.

Placing the wormwheel at the saddle/OTA end of the declination shaft pushes the moment arm outwards at that end  by 45mm. This results in an almost equal balance occurring between the OTA and its vital counterweights.

By sheer coincidence weight lifters have adopted so-called 'Olympic' standards for their weight disks and bars. They have settled on 50mm as the hole size in the heavy disks but the lifting bar is not remotely that size. It was decided to reduce twist [torque] in the bar by fitting captive bearing sleeves at each end of the bar. This reduces the risk of damage to the lifter's wrists as the disks can now spin freely as they are lifted.

My lathe is limited in a maximum diameter of work-piece of about 7" [18cm] for turning. Fortunately  I now no longer have to find smaller diameter weight disks for counterweights to bore them out. I just choose from the Olympic standard disks knowing they will slide onto my 50mm declination axis without effort. I have just ordered five, plain cast iron, 5kg weight disks with a 50mm bore. With 130mm available shaft length their 25mm thickness is perfect to leave a little extra room for a collar.

The advantage of plain disks is that they look the part on a telescope mounting. Many weight lifting disks have handles and grips for easier handling. Which complexity would look rather strange. 5kg disks are relatively easy to handle compared with [say] 10kg and above. A total of 25kg should offer a slight excess of counterbalancing. A permanently set up mounting could [should] have an independent, quickly adjustable weight running along the declination housing to offer adjustable bias when heavier or lighter components are added to the OTA. For years I have been struggling with undersized mountings. This includes the Fullerscopes MkIV. Friction within the plain [bronze Oilite sleeve] bearings rises rapidly with increasing OTA weight. The ability to add extras to the OTA or mounting has been very limited.

Click on any image for an enlargement.

*