Discussion on Small CNC Mill Spindle and Axis Motor Performance

All the "real" mills ive seen/used have an induction motor just like Jim has suggested. Not the 15kRPM+ spindles, but everything under that. Modern VFDs are really blurring the line between servos and traditional dymb spindle drives.

The problem i have with induction motor/vfd setups is simply space and weight. Unless you can beg/borrow/steal a really high performance spindle-specific induction motor you're stuck with fairly standard motors. And standard induction motors, up to about 10hp or so, arent as power-dense as PM servos or BLDCs.

Thats the reason Ive selected a PM motor for a spindle. Weight and size advantage means my Z axis isnt trying to lift 100lbs of motor and I can fit a standard PDB indtead of some kludge lever thing.

If i had more room, or if the motor was a smaller percentage of spindle/head assembly weight I'd probably use an inductio motor, vfd, and spindle-mounted encoder like Jim described
 
I managed to get the spindle motor and belt system temporarily mounted on the mill. There is definitely some fine tuning to be done. Due to the orientation of the pulleys, the high gear belt works perfectly, but the low gear belt doesn't get tight enough for my liking. This is due to the larger pulley being on the motor for high gear and on the spindle for low gear; hence the geometry of the belts are different. This was an oversight on my part and i'll have to re machine a new lever for the low gear. One thing that needs to be considered in a design like this, is that each tensioner needs some over travel to compensate for belt stretch. The powerdraw bar will come later - I need to solidify the belt system before I move onto other things.

I've mounted all the spindle motor electronics in a separate enclosure to mitigate any noise issues and so far everything is running smoothly.

PXL_20210415_023844781.jpg


PXL_20210427_031251243.jpg

I haven't been able to run any programs to put the new motor through its paces, but I couldn't resist doing a few test cuts at 8000 rpm. I have to say, what a crazy difference it makes over the stock 2500 rpm motor. 3/8" 2 flute, 0.5" DOC, 0.100" WOC at 70ipm puts a massive grin on my face.
 
tkalxx said:
I managed to get the spindle motor and belt system temporarily mounted on the mill. There is definitely some fine tuning to be done. Due to the orientation of the pulleys, the high gear belt works perfectly, but the low gear belt doesn't get tight enough for my liking. This is due to the larger pulley being on the motor for high gear and on the spindle for low gear; hence the geometry of the belts are different. This was an oversight on my part and i'll have to re machine a new lever for the low gear. One thing that needs to be considered in a design like this, is that each tensioner needs some over travel to compensate for belt stretch. The powerdraw bar will come later - I need to solidify the belt system before I move onto other things.

I've mounted all the spindle motor electronics in a separate enclosure to mitigate any noise issues and so far everything is running smoothly.

View attachment 364073


View attachment 364074

I haven't been able to run any programs to put the new motor through its paces, but I couldn't resist doing a few test cuts at 8000 rpm. I have to say, what a crazy difference it makes over the stock 2500 rpm motor. 3/8" 2 flute, 0.5" DOC, 0.100" WOC at 70ipm puts a massive grin on my face.
Click to expand...
Looks amazing! You have to share some video of the belt change and maybe some cutting performance. Awesome work!
 
Speaking of motors, I might have stumbled across a couple of candidates... $100/ea, $15 for the cables.

It appears the Coply can accept sin/cosine encoder signals, so no butching the Siemens stuff.

Check out the specs on the dataplate.

20210427_152337.jpg20210427_152433.jpg
 
spumco said:
Speaking of motors, I might have stumbled across a couple of candidates... $100/ea, $15 for the cables.

It appears the Coply can accept sin/cosine encoder signals, so no butching the Siemens stuff.

Check out the specs on the dataplate.

View attachment 364090View attachment 364091
Click to expand...
Very nice. I had a few Siemens motors but they all had resolvers. Sold them for a good chunk of change
 
I'm pretty excited to be able to get 8kRPM with a 1:1 spindle now. And no extra spindle encoder/index signal needed - just using the drive output.
 
tkalxx said:
I haven't been able to run any programs to put the new motor through its paces, but I couldn't resist doing a few test cuts at 8000 rpm. I have to say, what a crazy difference it makes over the stock 2500 rpm motor. 3/8" 2 flute, 0.5" DOC, 0.100" WOC at 70ipm puts a massive grin on my face.
Click to expand...
that is really slick and very impressive. You could even do something like that manually for a non-CNC machine :)
 
mattthemuppet2 said:
that is really slick and very impressive. You could even do something like that manually for a non-CNC machine :)
Click to expand...

Good thinking. A couple of over-center levers (I'm thinking toggle clamp-type things) to select each ratio. And pull the second one to lock the spindle during a drawbar tool change.
 
I replaced my spindle bearings this weekend with SKF 32005X/Q and 32007X/Q tapered roller bearings. The larger of the two bearings has a reference speed of 8500 and a maximum of 10,000. Using Kluber isoflex nbu-15 grease, I'm unable to run the spindle at 8000rpm comfortably. During the break in period, temps shot up very quickly within the first 5 minutes running at 8000rpm (upwards of 95C). 7000rpm seemed OK - I ran it for 45mins and temps leveled out around 80-85C. I tried playing around with the pre-load and did not see any different results.

Throughout this thread it's been mentioned that 10,000rpm would be the desired top end for our hobby cnc's. Unfortunately, I can confidently say you wouldn't be able to run at those speeds with the top of the line SKF tapered bearings. This might push me down the road of investigating alternative bearings or lubrication methods. On the plus side, I'm measuring between .0001"-.00015" of runout on the R8 taper with these new bearings.

Does anyone know how tormach pulls off 10,000rpm? As far as I'm aware, they don't even use an oil drip feed to the spindle bearings. Maybe angular contacts would improve top speed figures, but you risk increasing total runout. Another option is a new spindle cartridge rated to higher rpms as Spumco has mentioned, but I'm not really interested in heading down that path.
 
You must log in or register to reply here.
Back
Top