Elegent Scroll Frame Skeleton Clock

[Kim]

Thanks Minh! 

I originally went with a press fit.  Loctite to be used for final.

Yeah, I considered a press fit, but they are so hard to disassemble if needed! 

Kim

[Kim]

I’m all excited to try out this new method for making a gear cutter.  It’s described by Robert D Porter in his book, “The Clock & Watch Makers Guide to Gear Making.”  He gives a lot of good info on how to make cycloidal gears for clocks.  And frankly, this is the only place I’ve found on how to make cycloidal gears.  I’ve made gear cutters using Ivan Law’s button method, but those produce an involute gear.

So I’m all excited to try something new!

Robert’s book also describes several gadgets that he uses to make his gear cutters. I, however, am not planning on using those. I’m going to go a slightly different route.  He makes a jig that allows him to grind a very small, precise radius on the tip of a form tool.  This form tool is used to create the actual gear cutter.

I’m going to try and do the same thing, only use a rotary table to mill a precise rounded tip on a piece of W-1 tool steel for the form tool.

Here’s how I started out.  The important thing to note in this picture is that I have the rotary table tilted about 8 degrees in the Z axis.  This will provide the relief for the cutting edge.  The tricky thing about this setup is that the ‘center’ I’m finding here won’t be the center for something that is closer to the rotary table.  The center of the RT is tilted by 8 degrees!  But I’m confident that with a little trig, I can figure out how far to offset the X-axis on my mill to make up for this difference.


The next thing I did was to make a way to hold a tool blank over the center of the RT.  I decided to do this by clamping one of my lathe tool holders to the table. The trick here was to get it so that the slot was radial to the center of the table.  I set it so that the tool holding edge was 3/16” from the diameter of the RT.  This should let me create any radius up to 3/16", which is way more than I need for what I'm doing.


With all the pieces ready, I re-found the center.


Next, I clamped my tool blank in place.  I used a stack of feeler gauges to offset the edge of the tool to just about’ the radius of the tip of the tool.  Then I found the top of the tool blank in the RT. Using this info, I calculated my X-offset to get the mill spindle over the center of the RT at the location of the top of the tool blank.  Sorry, I forgot to take a picture of this important step.

I’d also marked the location where I wanted the radius of the form tool to be, which helped give me some confidence that I’d done my math right.

This picture shows my calculation page for this operation (why’d I take a picture of this rather than the real work?  Who knows!  )   It also shows my gear spreadsheet.  This spreadsheet is based on the workbook page that Robert Porter has in his book to help you calculate the radius of the form tool and all the other important things you need to know to make the form tool, the cutter, and the gear.  Lots of good detail info in that book!


I then proceeded to shape the form tool.  I cut a flat along one edge, then the radius at the specified dimension for a little over 90 degrees.  The last part is not that important as it won't be used.  The important part of the form tool is the flat edge that is facing forward and the radius size.  The line scribed on it there (parallel to the front edge) is where I marked that the center of the radius should fall.  And I think it came out in about the right place! 


After heat treating & tempering the tool, I honed it a bit and here it is, ready to go.


This took me the whole shop session.  Hopefully it’ll go much faster next time, now that I’ve figured out how to do it all!  I've got several more of these to make! 

But I’m excited that it seems to have worked out!  (at least so far  )

Kim

[cnr6400]

  Re your 8 degree trig - if you think you need a tan, use some sunscreen, then use the calculator...     

[crueby]

Fun to learn more techniques!  And to see how much Math has escaped from the little  gray cells.


You mention having to make more form cutters. Are they all to cut different  portions of the actual gear cutter, or to make different  shape gear cutters for different  gears?


 

[Kim]

Re your 8 degree trig - if you think you need a tan, use some sunscreen, then use the calculator...     

Yes, that would be a much easier way to work on my tan().   

And do you work on your ArcTan() by sitting out in the moonlight?   

Fun to learn more techniques!  And to see how much Math has escaped from the little  gray cells.

You mention having to make more form cutters. Are they all to cut different portions of the actual gear cutter, or to make different shape gear cutters for different gears?
 

The latter - different tooth shapes, different gears.   There are only two DPs used for the clock, 34 and 42, with most being 42 DP. And only two have the same tooth count.  Some of them are pretty close in tooth count, like 96 and 90.  But the tables give me different radii and angles for the tools, and I don't have the experience to know if they are 'close enough' or not. So I'm just making a separate cutter for each tooth count.  That will be five different cutters.  So five unique form tools.

All of the pinions are lantern style, so no cutters for those (which is nice!)  And all the ratchets will use the one ratchet cutter I already made.

Kim

[bent]

Bit late to the party, but catching up today.  That fusee assembly looks great, Kim!   

[Kim]

Thanks Bent! 

You mention having to make more form cutters. Are they all to cut different portions of the actual gear cutter, or to make different shape gear cutters for different gears?
 

The latter - different tooth shapes, different gears.   There are only two DPs used for the clock, 34 and 42, with most being 42 DP. And only two have the same tooth count.  Some of them are pretty close in tooth count, like 96 and 90.  But the tables give me different radii and angles for the tools, and I don't have the experience to know if they are 'close enough' or not. So I'm just making a separate cutter for each tooth count.  That will be five different cutters.  So five unique form tools.

All of the pinions are lantern style, so no cutters for those (which is nice!)  And all the ratchets will use the one ratchet cutter I already made.

Kim

After reviewing my gear cutter spreadsheet, I've decided that I don't have to make five different form tools for five gear cutters.  The only thing that differentiates the form tools is the radius of the tip.  While the spreadsheet shows different radius values for each one, some of them are only different by ~0.001" or less.  I don't think my machining skills can actually produce that accurate of a radius.  So when they are that close, I'm going to use a single form tool to generate multiple cutters.  Similarly, the main thing that differentiates the cutters is 1) the radius of the form tool used, and 2) the angle you set the form tool at when creating the gear cutter.  So I've combined gear cutters when the form tool angles differ by a fraction of a degree.  For example, with two of them I'm combining, the specified Form Tool Flank Angles are 1.023 degrees and 1.074 degrees.  I might be able to set a half a degree on the tool angle, but to think I can resolve 0.05 degrees seems a little ridiculous.

So, with my new reuse plan, for five different sized gears I will be making four unique gear cutters. And to make those four cutters, I'll only use 3 unique form tools.  That's better than 5 cutters and 5 form tools that I was planning! 

Also, while revisiting my gear spreadsheet, I realized that I'd used the wrong table for some of my gears.  In Porter's book, he has many sets of tables:
     For making gears that are driven by pinions (speed reduction gearing)
     For making gears where either the gear or pinion drives (like hour/minute movement gears)
     For making gears that drive leaf pinions (speed increasing gears - like the drive train of a clock)
     For making gears that drive lantern pinions (This is for my drive train, since all the pinions are lantern style)

And while reviewing them, I realized I should have been using the second set of tables for my hour/minute gears, not the fourth set, which I'd appropriately used for the drive train.

After fixing that, I now have a more accurate set of gear numbers in my spreadsheet. While it did make a little difference, it's likely that even using the wrong table, things would have been OK.  The differences are quite small.  But I'm glad I caught that and have it fixed BEFORE I cut all the gears!

Kim

PS   And yes, I know I'm supposed to be calling gears wheels, since I'm making a clock.  I'm working on it, but I'm not there yet.

[crueby]

Good catch!  Subtle differences  but can be important.   

[Kim]

Since things were set up for making the form tools, I went ahead and made all three of them that I will need.  One with a 0.071 tip radius, one at 0.058, and the final one at 0.050.  The last two I made on opposite ends of the same blank.  Not sure that was such a great idea – it was harder to heat treat them that way.  But I think I got it done.  I punched the tip radius on each one.  The 0.058 is upside down (on the left) in this picture.


Then I made all four gear cutter blanks.  Each one had a different ‘minimum thickness’ requirement, and I marked each one so I’d remember which was which.  The 0.142” blank will be for the great wheel, which is what I’m ostensibly making now.


To make the gear cutter, I first have to set the flank angle.  The left edge of the form tool is flat, so to set the angle, you have to position that edge of the tool at the angle you want for the edge of the cutter (what he calls the flank angle).

These flank angles are quite small.  In this case, it was 0.938 degrees.  I used my trig and determined that this would be about 16.4 thou over a 1” length.

I set the compound slide parallel to the lathe spindle, then set the tool as perpendicular to the spindle as I could. Next, I used a Dial indicator to set the compound slide 0.0164” across one inch of travel of the compound.  Once I got that, the tool SHOULD be 0.938 degrees from perpendicular, which is what I wanted.


With my setup complete, I touched off of the outer rim of the blank, shaving off a few thou to make sure it was concentric to the mandrel, then moved the cutter in the specified distance (I had to remember to DOUBLE the distance, since my DRO reads out in diameter, rather than radius!) and moved it in the specified distance from the edge of the tool.


Then I flipped the blank around on the mandrel and did the same to the other side.  The important measurement in these operations is to make sure that center part has the appropriate width (as close to centered as possible).  This value is ALSO specified when you do the gear worksheet in Porter’s book.


On the mill, I moved the RT back to the vertical position and spent the not insignificant amount of time getting it and the tail stock squared up to the X-axis and to each other. While setting up the tailstock is a right pain, it was well worth it. With the tailstock, I was able to get the RT runout on the part to be +/-1.5 thou.  That’s the best I’ve ever seen!  I think the tailstock is a win.

With that setup, I was able to cut the face of the gear cutter teeth with a 1/32” slitting saw.  I made 10 teeth, because that’s what Porter shows in his book.  The teeth are cut off center to give a 7 degree rake to the cutting face.


Then, moved the slitting saw down a good bit and adjusted the RT angle some till it looked like the right position for the back of the teeth.  And I cut all 10 of those.  (But only 3 so far in this picture )


And here’s the cutter after the heat treatment.  Oh yeah, before heat treating I also marked it. The C is for “Cycloidal” so I don’t mistake it later for an involute cutter. And the 96 is because that’s the number of tooth gear this cutter is designed for.


Next, I will actually start on the Great Wheel gear itself!

Thanks for looking in,
Kim

[samc88]

Some good maths going on there Kim, that cutter looks great!

[cnr6400]

  With that C Kim I thought you might be playing a trick on the next generation looking at that tool 50 years from now. I can imagine a great grandson saying "Look at that C mark! Great Grampa carved that cutter outa solid carbide!"   

Seriously it looks great! Good luck with the wheel cutting. 

[internal_fire]

I can imagine a great grandson saying "Look at that C mark! Great Grampa carved that cutter outa solid carbide!"   

Nah. In 50 years any kid with an ordinary phaser could do it. 

Gene

[Kim]

I can imagine a great grandson saying "Look at that C mark! Great Grampa carved that cutter outa solid carbide!"   

Nah. In 50 years any kid with an ordinary phaser could do it. 

Gene

[cnr6400]

I can imagine a great grandson saying "Look at that C mark! Great Grampa carved that cutter outa solid carbide!"   

Nah. In 50 years any kid with an ordinary phaser could do it. 

Gene

I hadn't thought of that Gene, you're probably right! 

[Kim]

While I’ve been working on the gear cutters, I’ve had the bandsaw whirring away in the background, cutting gear blanks from that big hunk of 3.25” round 303 stainless rod I got from Yarde Metals.  These are for the Maintaining Ratchet, and the Great, Center, and Third Wheels.


I took the thickest blank and faced off both sides to make it 3/16” thick for the Great Wheel. Then I sandwiched it between two scrap pieces of 3/32” aluminum sheet (I have a bunch of this aluminum sheet that I got from a dumpster at work, back when I did such things!).  My thinking was that the aluminum sheet would help provide a little more strength for the gear blank during machining.  Then I stacked them on the 5/16” spindle and brought them to the required OD of 2.897”.


Then I moved the gear blank, er, I mean wheel blank, over to the mill and started cutting teeth.  Boy, is this ever slow going!

In the past, I’ve cut brass gears.  But this stainless stuff is tougher.  After cutting 12 teeth, I looked at it (rotated the RT around to where I could inspect the teeth) and I was shocked! They looked terrible!  I was so flummoxed by the thole thing I forgot to take a picture.

I finally decided (or at least hoped) that it was just a really bad burr that got turned up by the machining.  It seemed to be pushing a big burr along the top of the teeth.  And the burr from the first cut would get pushed around by the second cut.  It looked really ugly.

I spent some time trying to remove the burrs with some small files – all with everything still mounted on the mill in the rotary table.  I couldn’t move it, or I’d lose registration and not be able to cut the remaining teeth!

Removing that burr seemed to help quite a bit, and what remained looked a lot more like some reasonable facsimile of a gear tooth.  That was a relief.

I was worried that my cutter just wasn’t up to the task.  I thought it was fairly sharp, but this isn’t brass I’m cutting, it's stainless steel.  Much less forgiving stuff to work with.

I rotated the gear back around into position and cut another dozen teeth.  This time, I decided to lower my cutting speed.  I’d originally chosen 100FPS, which is what the table says to use for 303 stainless with an HSS tool.  But I’m using a W-1 cutter.  Maybe W-1 isn’t as good as HSS.  So I brought the speed down to to ~80fps, and cut another dozen teeth.

Interestingly, these came out MUCH better than the first dozen. That seems to be a much better cutting speed for W-1.  The teeth will still need some cleanup, but they're much better than the first set (at least, before the filing to remove the ugly burr).
 
So here are my pictures.  In the photo, the top dozen teeth are the first set that originally looked so bad.  The lower dozen are the second round of teeth, cut at a slower cutting speed.


They still aren’t gorgeous, but I think they will clean up well enough.  And I think they actually look like gear teeth, so I’m feeling somewhat pleased about that.

This is just another angle of the same thing, giving you a better view of the profile.  There’s still a bit of a burr that will have to be removed, but this photo is without touching the lower dozen teeth at all.  It’s just as they came off the cutter.  They all appear to tip upward a little bit, but I believe that is the burr left by the second cut.


I did another dozen teeth before I quit for the day.  It took me about 30 minutes to do those teeth.  So another 2.5 hours to finish this wheel. (BTW, 12 teeth is 45 degrees on this wheel.)

But I was getting tired, and the last thing I wanted to do was to mess up positioning the RT when cutting a tooth!  So, I’ll start again tomorrow when I’ve got a fresh attention span 😊

I’m still hoping the gear will come out nicely (he says, crossing his fingers!)

Thanks for looking in,
Kim