Strutt Epicyclic Train Clock (maybe?)

[kvom]

Yes it's a counterweight, and the arm assembly is complete.

[kvom]

I  booked a 3-night trip (Tues-Fri) to Las Vegas 3 weeks from now, and I decided to extend it a day and fly to San Diego.  From there I'll rent a car and drive up to Carlsbad to visit the Craftsmanship Museum.  Aside from all the great models there (some made by people I know), there is also the Strutt Clock.  Some clear color photos ought to answer the questions I'm puzzling over.  I'll stay over Friday night to avoid a red-eye return flight.

CNC mill build is proceeding without much trouble. I have all the axes mounted, and the next step is to get everything trammed square.

[kvom]

I think I've figured out how the fusee and the great wheel interact.  I think this arrangement must be common on such clocks as Kim's fusee has a similar ratchet disk attached to the big end.

The great wheel has an identical ratchet with the same orientation.  It engages the steel "click".  The click is thicker than the ratchet so that when the wheel is located close to the fusee the click also engages the fusee's ratchet.  So when the fusee is unwinding it's locked by the click and the wheel turns at the same time.  But when winding the clock the fusee turns and the wheel doesn't. The click is proud of the wheel ratchet by only 1/64", so the fitting of these parts is pretty delicate.  I'd think the click could be made thicker (?) 

The click is tensioned against the ratchets by a click spring mounted on the wheel, and not shown in the screen shot.

The wheel engages a lantern pinion attached to the center arbor and drives the minute hand.

[Kim]

That's a very interesting mechanism, Kvom.  Looking forward to seeing this come together to see how it all works.

It seems like having one click on two ratchets could be a little squirly. As you wind the fusee, the click will lift up before it falls past the next tooth. And during this time, it would seem that it would also release the second ratchet and allow the great wheel to move backwards.  But I'm sure there's something I'm missing there...

On the Wilding clock I'm making, it has a 'maintaining ratchet' that would serve a similar purpose to that second ratchet you show on your great wheel.  The maintaining ratchet is much larger than the one you have, and it has a separate click from the fusee click. 

Looking forward to seeing your progress!
Kim

[kvom]

Because of the unusual gear train, the hands cannot be fixed to their arbors.  So each hand has a "collet" that attaches it to the arbor with the end of the collet riding in a groove machined into the arbor.  There's enough pressure from the collet to keep the hands from rotating on their own but to allow moving the hands manually to set the time.  Attachment shows the minute hand collet.

On the CNC mill front, I decided to not use 220V power for the spindle motor and will return the spindle motor and inverter (VFD) in exchange for 110V.  One reason is that I have two 220V outlets that limit where I can position the mill, and the cord/plug will be heavy.  I'd need to attach a short cord with a male plug at the mill and use my existing extension cord that I built for the powder coat oven.  A second reason is that in addition to the spindle power cord, the mill needs one 110V cord for the power supply,  two 110V plugs for voltage converters that plug into the Mach3 board and provide power for the coolant pump and the board itself.  And the computer needs a cord.  So 5 in all for the running system.  Easy with a power strip, and I have a lot of 110V outlets to choose from.

[kvom]

I am on the verge of getting my CNC Mill "kit" ready to make parts.  I'm getting a fixture plate and modular vise from Saunders Machine Works (tracking for Saturday), and just got delivery of a pair of .045" 2-flute carbide endmills that I'll use to cut the teeth of the gears.  Feed and speed calculator says 20K RPM at 7 ipm in brass at .06 depth.  That's full depth but I'll cut that in half and ramp. 

Building the mill has been pretty slow.  Limited shop time including travel, some missing parts, and a bad control board slowed me down.  And I also had an out-of-range precision on the Y-axis that I though was backlash but turned out to be my not tightening a nut that secures the ball screw to a bearing.   

[kvom]

It occurred to me that it could be possible to machine a fusee on a CNC mill with a 4th axis.  While there's no single command that would follow the curve, you could divide the job into multiple g-code moves per groove.  With a ball nose mill whose diameter is the width of the groove, the g-code move would be a coordinated move in X (width of the groove) and A (360 degrees) and Z (depth increment).  Supposedly there will be an available 4th-axis at some point for my mill.

[uuu]

Well, I frequently cut a hole using a helical toolpath - using a G02 code, to perform a circular move in XY, with a target Z depth so the tool plunges down as it cuts the circle

So it ought to be possible to do a curve in X and Z while the A axis has a target rotation.

Wilf

[crueby]

If you can define the tapered spiral path, using a keyway style cutter, then you would not need the extra axis? Cut it with the fusee axis vertical. The cutter would cut a little extra as it moved upwards (or downwards), but it would be an even effect, and would taper the sides of the channel, which is probably a good thing for a fusee.

[kvom]

After some months of building my mini CNC mill, I was finally at a point to attempt a prototype gear.  This is the planet wheel, 68 teeth, 2.217" OD.

This is not the final version as the brass is thinner than specified, but I was happy to get a decent result.  Endmill is .045" diameter, 2-flute, flute-length .125", .125" diameter shaft.  I used very conservative feeds: 7ipm with .014" depth of cut, 19500 RPM.

I placed a piece of MIC6 (e.g. flat and smooth) aluminum in the vise, and covered it with blue painters tape.  Then spread super glue onto the tape to hold down the brass.  Left it to cure a while before machining.  Afterwards soak in acetone to release the part.

[Dave Otto]

Looks good!
What are you thoughts on the SMW Mod Vise?

Dave

[crueby]

That came out great!

[kvom]

Looks good!
What are you thoughts on the SMW Mod Vise?

Dave

This is the Gen3 hobby edition.  Using the stock jaws the clamping surface is narrow and clamping pressure is a lot less than a regular screw vise.  But for parts like this they're good, and they're well made.   For something much heavier I'd want to clamp to the plate using the 1/2-20 threaded holes.  The fixture plate is made for this mill.

I have a set of 4 Mitee-Bite clamps, but they use 1/2-13 screws.  I think making a plaate with 1/2-13 holes for the clamps and 1/4-20 clearance holes for attaching to the fixture plate could work pretty well.

[Kim]

That gear looks great!  You should feel pretty pleased with that I'd think! 
Kim

[kvom]

Next time I'll run a chamfer around the spokes.

I should probably machine a depthing tool before going much further.