machines which I have made, am making, or intend to make, and some other stuff. If you find this site interesting, please leave a comment. I read every comment and respond to most.

Tag: gear cutting

More Gearing Up, and more to come.

This was 4 days ago.
Today. Re the gears on the right, bottom row…. one was machined from bar stock, the other was cast, had the outer ring removed, and a new ring soldered on, then the teeth were cut. Can you pick which is which? And all of the round gears have spent 3 hours in the gemstone tumbler to remove sharp edges. The tumbling has reduced the surface oxidation on the large gear castings, but some more time required to totally remove it.
CNC milling the spokes in one of the intermediate gears. Neat job, but the internal corners with fillets are not really kosher. This gear is barely visible in the finished model.

And the gears with ratchets attached need 2 mates. I tried to make them yesterday, but we had a wild day with thunderstorms and high winds, and my machines were playing up. I might get back to them today. (the ratchets can be seen in the header photo).

In country Victoria we are out of level 3 lockdown! Hooray! Still can’t see family from Melbourne, and minor restrictions on visiting local friends, and need to wear masks when out and about. But things are on the improve. I doubt that we have seen the last of the virus however.

Cutting the Gear

Today Swen and I started cutting the gear.  Here is the setup.


The gear was centered on the rotary table with an aluminium bush, which fitted the outside of an ER40 collet chuck.  The chuck had an M3 shaft which fitted neatly into the spindle of the rotary table.  3 bolts secured the gear blank to the T slots on the rotary table.


This photo was taken after the setup was broken down.  Showing the M3 taper ER 40 collet holder, which I used to centralise the gear blank on the rotary table.

I started the mill conservatively at 200 rpm and a 0.5mm deep cut, but gradually increased the RPM to 450, and the depth of cut to 2mm.   Later adjusted to 300rpm, 1mm cut.

The CNC table performed flawlessly, with rapid advances between the 360/77º degree cuts (about 4.6º each tooth).  The feed rate was controlled by manually winding the X axis feed on the mill.


We initially used water soluble cutting fluid, but changed later to raw Tap Magic, which seemed to work better.

As you can hear in the following short video, the cutter teeth are slightly off centre, but working well.  Didn’t quite finish the gear teeth in this session.  Some sparks later on, indicated that a cutter sharpening was required before the finishing run.  That will happen tomorrow.

The smoke is evaporation of cutting oil.

The setup will be left undisturbed by removing the gear cutter for sharpening.

I changed my mind about sharpening the cutter in the middle of the job, and continued cutting.

Instead, I lowered the spindle rpm, and the feed-rate.  The sparks stopped.  Maybe I was just pushing too hard, or maybe there was a hard spot in the metal.  Anyway, I finished the cuts.


The finished gear cut.  Are those teeth looking a bit skinny at the peaks?


Showing the setup from the operator’s view.  The CNC table worked brilliantly.




Removing the burrs with wire brush and file.


And testing the fit with a trial run in the headstock.  Thanks Swen, for helping (actually directing) the trial run.  The new gear on the left.

I made a video of the gear being rotated through 360mm, perfectly, but for some reason it will not upload.  (did upload eventually.. see below).    It looks perfect, with a tiny amount of backlash.   Full installation in a day or two.  I was quite surprised that making the gear to the specifications worked so well.


Me, testing the backlash.

And afterwards, sharpening the cutter on the Quorn T&C cutter grinder which I made a few years ago.  An amazingly versatile tool.



Just in case I need to make another.

The lathe headstock will be properly reassembled in a couple of days.  But I am finally feeling a bit confident about this job.




CNC rotary table and preparing a gear blank

Hi readers.   Sorry for the long break.  Since my return from UK I have been severely jet lagged, then very busy, and not much time in the workshop.

The jet lag going westwards to from Oz to UK was minimal, but after the homeward trip it took 2 weeks to start feeling normal again.  It is a 22 hour flight, plus 2 hour stop over in Singapore.   I do not remember ever having such marked jet lag before, and  not much was done during those initial 2 weeks.

When I did venture back into the workshop, I discovered that my CNC mill was malfunctioning.  The Y axis has been a bit unpredictable for quite a while.  I found a broken wire and fixed it, but the problem returned.  After a previous electronic failure in the Z axis, my CNC expert advisor, Stuart,  suggested that I should  replace the electronics in a major upgrade.  The mill is a solid industrial machine, mechanically in sound condition, and is worth spending some time and money on.

It is a 1997 model, and the memory in the CNC motherboard is a whopping 7k!  I was able to get a fair bit done with the 7k, and the situation was improved by linking an external PC, and using V-Carve Pro.  But there was a limitation in that the mill is a 2.5 axis machine.  Not that I want to use 3 or 4 axes very often, but the lure of improving the mill is irresistible.

So I am in the process of ordering 3 new servo motors.  They will be AC single phase servos, rather than 3 phase motors.  I have installed one of these in my small Boxford lathe as a spindle motor, and it has proved to be reliable, compact, powerful and inexpensive (well, fairly inexpensive, comparatively speaking).  They have been ordered from China.  Cost-wise, the three axis motors will be much less expensive than one of the existing 3 phase servos.  On top of that I will need a breakout board, ESS smooth stepper to link to a computer, and various switches, wiring, power supplies etc.

I will document the steps of the rebuild.

But the item that I was getting to, was hooking up my rotary table to CNC.  I had expected to pick up a new gear for my big lathe on my return from the UK, to replace the one with the broken tooth.  I was pretty annoyed to learn that the gear maker had not done the job, and worse still he had not notified me that it had not been done.  Since he never answers the telephone, I drove to the factory, expecting to pick up the new gear, as arranged and promised, to be met with apologies and excuses.  Long story, I have decided to make the gear myself.

It has 77 teeth, an unusual number for a gear, which means that it has to be made, not purchased off the shelf.  I have a dividing plate with 77 teeth, but I could see plenty of potential for making mistakes using that, so I elected to finish the CNC conversion of the rotary table which I had started last year.  The mechanical aspects had been finished.  All that was required were the electronic hookups.   Fortunately for me, I have a friend who is an expert at these.


In the center is the rotary table, an 8″ Vertex.  The stepper motor is a NEMA 36.   The intervening aluminium block is the coupler.  The controlling program is Mach 3.  Originally I intended to use an Arduino, but it seemed more complicated and less robust than this setup, which involved using the breakout board of the CNC lathe (right), and a new Gecko driver. (see next pictures)



Looks complicated and messy.  Much better with the doors closed.  The rotary table Geckodrive is the one on the left.  The 2 on the right are for the lathe.  The black white and green wires 8,9,10, are from the breakout board.  The black and red 1 and 2 are from the 48v power supply, and the stepper motor power is from Windings A and B, 3,4,5,6, in the thick white wire.

I confess that I have little understanding of the wiring.  Stuart had it hooked up in under an hour.  A bit longer configuring and tweaking Mach 3, and it was working.  The extra Geckodrive, and some wires were the only extra components required to make the electronic connections.

I shot a video of it working, with giving a commentary.  But it is so bad that I will reshoot it, and add it to this post in a day or 2.   Sorry.  Not done yet.   But I have been busy preparing the blank for cutting a new gear.

I decided to retain the hub of the gear and to add on a new ring which will be machined, and then new teeth cut into it.

Firstly I had some 25mm steel plate water jetted approximately to size.  I chose water jetting in preference to laser cutting or oxy-acetylene cutting to avoid any inadvertent heat hardening.

I also had the original gear water jetted to remove the outer 25mm, including the teeth, because it had originally been heat treated hardened, and I did not fancy machining that on my other lathe and maybe breaking more teeth!

It was not cheap.  But a nice finish, which machined easily.  So the hub and the blank ring were machined with a 0.1mm gap, and glued together with Loctite 620.  Then Scotch pins were inserted.   Since my CNC mill is out of action, I reverted to calculating X and Y co-ordinates, using FS Pro.  See screen shot below.



My CNC mill is out of action, so I reverted to doing some XY calculations on the manual mill with DRO, using FS Pro.   Screen shot above.


And in the above shot, I have drilled and threaded some M6 holes and Loctited in some M6 grub screws.

Then machined it to size,


The 6mm holes are the Scotch pins.  The 10mm holes are to attach the assembly to the CNC rotary table for cutting the teeth.

So, this post might be a bit ramshackle and disorganised.   A bit like my workshop at present, and possibly my brain.  My GP has started me on blood pressure medication, so I will blame that.

Watch this space for cutting the gear, soon.