2" shaft mounting Pt.37: Joining the axes-


 The Tollok clamping bush screwed fast to the declination axis plate. Access for tightening and removal of the bush requires the removal of one of the declination housing plates. Most easily achieved by removing one of the narrower plates. This saves having to remove the 6 screws which clamp the housing together. Merely slackening off the furniture nuts and the nuts on the large studs will allow one of the narrow plates to be slipped free.

The 7" cylinder fits snugly over bush show above. This leaves the grub screws on the flange bearing accessible. Removing the heavy shafts seriously reduces the weight of the bearing housings.

This image shows the inside of the bearing housing plate.

I have yet to fit any through studs to hold the cylinder firmly to the opposite plate of the bearing housing.

Note the "spy" hole to confirm the shaft is fully home in the Tollok bush. Once the bearing housing is fitted together it is impossible to see the end of the shaft.
Now I have fitted six 8mm [5/16"] studs to help to hold the bearing housing together.

This image shows the inside view.
The mounting all but assembled except for the wormwheels, worms and supporting base. 

The weight is now well beyond my capacity to lift the entire mount.

The mounting broken down into fairly manageable parts.

The long declination shaft is quite a lift in itself.

I shall have to weigh all the components to see how heavy the mounting is at this stage.

Dec Hse + PA shaft....62.3
Bare dec shaft.............28.2
PA hse.........................28.2
RA wm-wheel..............8.6
Worms + bush..............6.0
Dec wm-wheel.............5.2
2 x worm plates............5.0

Click on any image for an enlargement.


2" shaft mounting Pt.36: Holding it all together.

I was struggling with ideas for an attractive and efficient means of holding the axis housings together. Then I found some oversized 8mm [5/16"] hex socket furniture nuts at a DIY superstore. Some 8mm studding [all thread] provided the means to join opposing sides.

Marking suitable points to site the threaded rods suggested a position to pass close to the main studs. The smaller studs would then be denied the ability to move sideways if the housing plates should manage to slide. Not that it would be likely with lengthways compression applied by the large studs. The smaller studs sandwich the side plates between the wider ones. Great care must be taken if I should decide to add more smaller studs at intervals to increase the clamping pressure. Clear room must be left for the large cylinder. Any bolts or studs to hold the housing to the cylinder must be over 6" long.

Each new fixing will increase lateral resistance to plate movement by improving the clamping pressure. Note the scale of the heavy 2" axis shaft and the large and smaller studs in the internal image above.

The images show the general idea. With earlier images taken without side plates to show the "innards."

The final image [left] shows the declination axis standing on end with the side plates firmly clamped in place. Endways movement of all four plates is fully constrained by the sturdy bearing flanges against the pressure applied by the large studs and nuts. The plates cannot move inwards, towards the axis, because they all rest against the large studs along their entire length. Being under tension the sturdy main studs are incredibly stiff and so highly resistant to bending.

I will still need to use long, through bolts [or studs] to join the declination axis housing to the large cylinder. I must avoid the single connecting plate carrying all the [heavy] loads via [only] the resistance of the four smaller [8mm] studs.

The exact location of these long fixing screws is another problem to be solved. The housing plate nearest the polar axis can easily be drilled to match the Tollok bush. Thus providing another level of reinforcement at the joint between the axes. Unfortunately, the screws for the Tollok bush cannot pass right through the declination axis shaft. Which all leaves very little clearance for additional fasteners within the 7" diameter footprint of the cylinder and the stud-filled declination bearing housing.

Nothing must interfere with the mating surfaces of the cylinder and the housing plate. Even the position of this joint along the length of the declination housing needs careful consideration. The declination housing ought to be offset to allow just enough clearance for the large wormwheel and its  worm. Any extra offset must be balanced by even heavier counterweights.

Click on any image for an enlargement.


2" shaft mounting Pt.35: Cutting 10mm [3/8"] thick alumminium.


I haven't yet decided if butt joints are sufficient if I use long cross screws to clamp the opposing plates together. I'd rather not spoil the clean lines of the plates by cladding the corners with angle profiles.

I could make up some more oak worktop strips to go between the studs to give the alloy plates some internal reinforcement. Though given the nature of the task I could simply produce some internal squares out of 3/4" birch plywood. These would slide over the studs and axis shaft to provide some extra resistance to any local distortion caused by compression loads from the cross screws.

These 10mm [3/8"] plates are probably sturdy enough that they are unlikely to bow. Though it is important to maintain the stiffness of the box structure to avoid flexure. Most commercial mountings use castings for this purpose.

Before rushing off for new saw blades I did some homework online. It seems I need a coarser toothed blade with hooked teeth. Along with kerosene, paraffin or petroleum as a cutting fluid. A 13 mile cycle ride to the shops produced some new specialist aluminium blades from DeWALT. DC2163. DC stands for deep cut.

The new blades worked wonders with a small dab of lamp oil applied with a brush after every 2cm or 3/4" throughout the cut. Much cleaner cutting without jamming or snatching and obviously much quicker than yesterday. I managed the first full length cut of 42cm, 16.5" cut in only 1/4 of an hour. The next at much higher speed in only 5 minutes! Experience showed that less pressure and medium speed worked better than low speed and heavy pressure. A setting of half the available oscillation on the Bosch seemed best.

After smoothing and filing the new cut edges I clamped up all four sides to check alignment. It seems my original try-square was off by a tiny fraction. I found another square and nulled it by reversing the stock after the first sharp pencil mark on a square plate. Both perpendicular lines matched perfectly. Further checking of each transverse cut on the wider strips proved the angle to be slightly out on only two ends. This produced tapered overlaps of the narrower cladding strips because the wider strips were being slightly tilted towards the diagonal. The 6" width is slightly oversized when all four plates are pressed tightly together. More angle grinding to narrow the overlap.

Half way through smoothing and squaring the saw cuts my angle grinder ground slowly to a halt. That meant more expense to buy a new one. [Cheapest DeWALT for about £40/$50] My third angle grinder in a couple of decades and I really don't use them that much. Not to mention the cost of purchase doubling just to buy some flap disks and a plastic backing pad and a few coarse paper disks just in case. They didn't have any mixed packs of sanding disks. The real profit is obviously in the sanding disks and accessories.

Several hours of angle grinding later I had four more plates to complete the PA bearing housing. 40 grit cuts fastest but tends to clog with lumps of aluminium. I tried running steel against the disk and that helped clear some of the built up junk. The flap wheel 80 grade just seemed to polish more than cut.

Finally I could prop it all up to admire the semi-finished appearance. It looks too top heavy! The RA wormwheel pushes the declination housing too far above the top PA bearing. The cylinder depth is not a matter of choice as it is only the length of the vital clamping bush. It looks as if I really ought to move the RA wormwheel down to the bottom of the PA to reduce the overhang above the top bearing. This is not so easily achieved at 55N PA altitude angle compared with lower latitudes. It will probably require an offset fork base to allow enough clearance for the 11.5" wormwheel. I could reverse the top bearing to bring the inner race extension inboard. This would alter the top-heavy appearance rather than the stiffness of the arrangement. Access to the bearing's grub screws would need consideration. 

An image rotated to show the 55 PA altitude.  The PA does not look quite so understated now. Nor does the RA wheel appear so heavily cantilevered. A substantial base enclosing the PA housing, to allow altitude adjustment, will help to regain the visual balance.

Several hours of angle grinding later I had four more plates to complete the PA bearing housing. 40 grit cuts fastest but tends to clog with lumps of aluminium. I tried running steel against the disk and that helped clear some of the built up junk. The flap wheel 80 grade just seemed to polish more than cut.

A made up image with the PA tilted at 55° and the RA wormwheel removed. The inner bearing race cannot be easily rotated to reduce the visual overhang. A small ball is deliberately fixed on the outer race to stop rotation of the spherical outer race. This ensures the lubrication groove matches the placing of the  grease nipple on the cast flange housing. Important for a heavily loaded bearing but not for a lightly loaded one which will rotate only about once a day. 

Click any image for an enlargement.


2" shaft mounting Pt.34: All metal bearing housings.


All that lovely 6" wide 10mm strip is drawing me back to the scrap yard like a siren. If only it were an inch wider! I drew a blank at the second, local scrap yard. It is entirely by chance that one can find really decent stuff like this. Most of it is corroded, drilled all over, scraped or bent to hell or is quickly gone. Some yards used to keep the good stuff aside to sell to willing customers at above the daily scrap metal rate.

Most [virgin] metal stockholders in Denmark will not deal with private customers. Even if they did, then their delivery charges can make a mockery of scrap yard, cash prices.

The siren call of the 6" alloy strips was answered. I bought 6 meters of 15cm x 10mm x  [20' x 6" x 3/8"] alloy strip [plate] in excellent condition for about £40 [$50US.] For 24kg or 50lbs I thought this price was very fair indeed. I should have easily enough to clad both bearing housings and lots more left for the base. It is smarter and stiffer than oak strips and completely immune to damp! It will even take a thread.

The metal clad housing certainly looks far better than wood. At least to my eyes. My desire for 7" wide strip proved to be a hangover from the wooden housings. 6" is actually a smidgen too wide where the plates rest naturally against the enclosed studs. [all threads]

I need to trim two plates if they are going to be sandwiched between two 'outers.' That won't be a lot of fun because it takes a quarter of an hour just to saw across one 6" width with the electric jigsaw. [US: sticksaw?] That was with frequent stops to lightly oil the blade and the cutting line. Otherwise the teeth just build up metal and won't cut. Perhaps I should examine other jigsaw blades intended for light alloys. I just used the last remaining 'metal' cutting blade in my limited collection. I mostly cut wood with the jigsaw.

There was lots of noisy abrading with a coarse disk on the angle grinder to get this far. The ends needed to be smoothed, squared and trued with a try square after the sawing. Since the plates define the squareness of the flange bearings the ends must be absolutely square.

The declination housing plates are 42cm or 16.5" long between the flange bearings. Slightly longer than the polar axis housing at 14". The declination wormwheel has to be added and room left for the counterweights. I don't need to add load spreading plates now to protect the previous wood panels from being crushed. Though I do need a worm support plate for both axes.

Click any image for an enlargement.


2" shaft mounting Pt.33: PA top plate and mounting mock-up.

 As I was only able to find 10mm x 150mm [3/8" x 6"] alloy strip I decided to use a piece to make a Declination axis adapter.

There was a bit of marking and drilling to do for the 10 screws which hold the clamping bush on the end of the Polar Axis. Plus 4 more holes for the bush extractor screws and a shaft depth spotting hole in the middle.

If i can find some heavier plate I shall swap over to that. Greater thickness would allow the bush holding screws to sit in counter-bored holes to allow them to be made flush with the surface of the top plate. Having them standing proud just means I need to rout a circle in the oak housing to clear them.

The Polar assembly is getting seriously heavy now. I was just able to lift it upright and lay it down on the bench again for the photographs but that was about it. The cylinder, wormwheel and top plate need to be removed to carry the basic bearing housing around with any remaining reserves of strength.

A slightly different view. The Polar Axis will be tilted up and held firmly at 55°. Not lying down as shown here. 

The plan is to run long screws vertically through the cylinder and the top plate to add extra strength and stiffness. Even longer screws will be passed through the top plate and right through the Declination housing. You can se where I practiced drawing clearance lines on the PA "box" before proceeding.

Here I have very gingerly placed the very heavy declination axis shaft, the saddle and its flange bearings on the top plate. You'll have to imagine the bearing housing covering the four studs between the bearings exactly like the PA housing. Unless I go with aluminium cladding...

For scale the saddle is 60cm or 2' long.  The PA housing is 7" square x 14" tall. The Dec shaft is 80cm, 32" long. The polar axle 24" or 60cm. The wormwheel is 11.5" Ø. I hope that symbol for diameter shows up on other computers. I just borrowed the capital Danish Ø. Google Groups makes a mess of the three non-standard Danish letters: Æ, Ø & Å.

Click on any image for an enlargement.