spindle motor VFD

[PeterT]

Is anyone on the forum running one of these (typically Chinese) ER collet spindle / vfd combos like pic?
I've been checking out websites & YouTube vids, just curious about a few things as the feedback seems to vary. BTW, I don't have a cnc router/mill & not really contemplating a build. I'm kind of curious about the motor for other purposes.

 

[DPittman]

Not me but now I am curious about them also. Would those be any good for a tool post grinder?

 

[gerritv]

I am using one (3kw) on my cnc mill/router: Will spin at 1000 rpm but not a lot a torque there. I use it between 8000 and 24000. Water cooled so no fan noise. Very quiet until you start cutting steel

Gerrit

 

[Mcgyver]

Hi Gerrit, have you measured the runout and whats the vibration like; super smooth or vibrates like crazy?

 

[gerritv]

I'll check the runout again, at the spindle taper (inside) and with a 6mm end mill inserted. Last time I did this it was less than 0.0001 but I have better gauges now
Noise-wise, very quiet. There is of course some noise, more at lower rpm. Heck, Ill just do a video on Sunday.

Cutting, the cuts are great in delrin, aluminum and steel. I still need to add a second ball nut to the X axis, adding one to Y made a noticeable difference. They are Thomson 0505's from 1980's, belleville washers between the 2 bearings.

Gerrit

 

[Mcgyver]

sounds really good. I've been modifying a BCA mill/jig borer, part of which has included a new T slotted Z axis so that I can swap out different accouterments on the Z axis. I've already made a sinker EDM head for it, thinking of a slotting head and a new spindle is just waiting for me to finish balancing iot. I thought adding a high speed spindle would be a nice addition.....may end up splurging for one with that endorsement.

 

[PeterT]

Sorry for ignoring thread, Gerrit & I have been chatting offline a bit. I've generally been curious about the bearings in these things as they seem to vary all over the map by supplier despite looking generically the same. Some are 2 banks of 2 conventional bearings, usually unspecified grade, sometimes deep groove sometimes not. Sometimes the motor end is set to allow thermal float with a spring plate or spring washers for temperature compensation. Other spindles have lower bearing count and/or what appears to be low grade bearings. Others are matched angular contact. Some air cooled others water cooled.

Clough42 has a good video series on a CNC spindle/motor/VFD retrofit to Asian mill for CNC work. I learned a lot about setup & electrics. I think the particular spindle he chose is a step up from most of the typical offerings but I'm not really up to speed on these things yet.
https://www.youtube.com/c/Clough42/search?query=ATC Spindle

 

[PeterT]

This is an interesting video. He disassembles his Chinese spindle to replace with a high end, expensive euro spindle. At ~2:00 he discovers ‘regular’ bearings for the business end, not even deep groove. Curious what the purpose of the V scribe is. Maybe someone can suggest a practical reason. Hopefully not trying to pretend they are matched set or AC bearings. Naww... they would never do that! LoL

 

[PeterT]

I've been modifying a BCA mill/jig borer, part of which has included a new T slotted Z axis so that I can swap out different accouterments on the Z axis. I've already made a sinker EDM head for it, thinking of a slotting head and a new spindle is just waiting for me to finish balancing iot. I thought adding a high speed spindle would be a nice addition.....may end up splurging for one with that endorsement.

Cool stuff. Lets see some pics!

 

[Mcgyver]

Cool stuff. Lets see some pics!

ok.... this'll be long...lots of photos but since you started the thread I take it as permission to go OT

this is a BCA jig borer, sort of the British version of the Boley mill

New Z axis

Spindle

The rod at the bottom will hopefully support an overarm for hobbing operations. As the Z tilts, the plan is to hob on it via an electronic coupling between the spindle and work axis

All that's left to do is balance it (oh, and grind the spindle ID).....this got interrupted a few months ago as I fell down a rabbit hole recondition a Schaublin 70. The beauty of 3D printing .....I just print out new holders/arms to fit whatever it is that needs balancing. I'd finished printing them when I had to put it aside for the Schaublin

The EDM

The controller/power supply is from Robert Langlois series in HSM ages ago. I bought it from the estate auction of the late Chris Bell - he was a guy knew how to make stuff!

Designed my own head that relies a lot on 3d printed parts

 

[PeterT]

McGyver, wow, neat stuff. So much to absorb here. Lets start with these two pics
- is the shaft OD being ground on a dedicated OD grinding machine or is the grinding wheel /spindle somehow mounted on a lathe cross slide?
- re the ID grinding, can you tell me more about it. Is it a dedicated TPG or spindle cartridge from another machine in a tool post mount clamp? What about the motor & speed control? Those arbors look quite stout, are they shop made or part of the kit?

ps by dedicated OD grinding machine, I'm talking about these machines

 

[Mcgyver]

Thanks. The OD grinding is on a tool grinder with a motorized work head - lets it do double duty as a light cylindrical grinder. I did a ground up scraping job on this machine so its really nice - its not difficult to get things straight and round so a digital tenths mic reads the same end to end. The ID grinding is a Wolf tool post grinder. Decent quality, made in UK although I gather a Indian firm as bought the brand. the TPG is pretty much stock, but I did balance the motor rotor. One day I'd love to have a proper cylindrical grinder with the internal grining spindle, but simple don't have the space in the current abode

 

[gerritv]

Wow. I bought Chris' Universal Pillar Tool at the auction a couple of years ago. Beautifully made. Missed out on the die filer, should have bid a bit longer :-(

 

[gerritv]

I'll check the runout again, at the spindle taper (inside) and with a 6mm end mill inserted. Last time I did this it was less than 0.0001 but I have better gauges now
Noise-wise, very quiet. There is of course some noise, more at lower rpm. Heck, Ill just do a video on Sunday.

Cutting, the cuts are great in delrin, aluminum and steel. I still need to add a second ball nut to the X axis, adding one to Y made a noticeable difference. They are Thomson 0505's from 1980's, belleville washers between the 2 bearings.

Gerrit

Runout at spindle itself, inside the taper: 0.003mm
Runout on 8mm end mill shank, regular ER20 collet: 0.051mm I didn't clean it off first, this is more runout than I am used to.
Runout on .125" carbide end mill in AA ER20 collet: 0.020mm

Sounds at 5k, 8k, 12k, 15k and 24k rpm. The VFD is noisy. Just a hint of vibration when I touch the spindle mount.

This is the machine now with its shrouds in place:

Gerrit

 

[PeterT]

Runout at spindle itself, inside the taper: 0.003mm
Runout on 8mm end mill shank, regular ER20 collet: 0.051mm I didn't clean it off first, this is more runout than I am used to.
Runout on .125" carbide end mill in AA ER20 collet: 0.020mm
Gerrit

Well the spindle runout is encouraging. I'm always leery of chucking dowels & measuring because now the dowel, the collet, collet nut & seating torque all enter the picture. But I guess that what the work piece would see so important to know. Did you have water flowing during this run up video? Some of the bench run YouTubes I've seen are rather impressive - spindle just laying on a workbench & winding up to 25K & it just sits there, not buzzing with appreciable vibration. I think there is something to the 3 phase motor type too for smoothness, but above my pay grade. And there seem to be a lot of VFD tweaks that make it run this way or that especially if not set right. Clogh42 did a really good rundown of chokes & box installment. There is actually a lot of electrical work to these things from what I can tell. But they sure look useful.

 

[gerritv]

No, no water running. I find, as did Stefan G., that it takes a bit of effort to get these warmed up. When actually cutting I do run water. It is just a pond pump, nothing fancy.
And yes, the collet, collet chuck and tool itself all add to the runout possibilities. For tiny cutters I will have to pay more attention.

I'll be writing up a spindle warm-up routine in the next little while. Running it for a while does seem to change the sound, I assume as the bearings warm up a bit. maybe 4 minutes or so at various RPM's.

Gerrit

 

[PeterT]

Gerrit sorry if I already asked you - did you get any parts/schematic documentation included with your spindle package that indicates the bearing types being used? or is it one of those 'take it apart & discover' when the time comes?

 

[gerritv]

Presently the bearings are a mystery. When I did the runout measure the other day there was no slop so that is a positive indication
At the time of my order the ad did not specify bearings but the same ad now says 2 7005P4DT and 2 7003P4DT bearings are installed. (cnc spindle motor kit 2.2 kw 110v/220v/380v water cooled spindle+ YL/HY Inverter + pump +80mmbracket+1SET ER20 for CNC|Machine Tool Spindle| - AliExpress)

I hope it will be a while before I need to disassemble the thing.

Gerrit

 

[PeterT]

seems to indicate towards angular contact but I'm not really up to speed on these bearing spec codes

https://www.robotdigg.com/product/1204/Spindle-motor-bearing-7002-or-7005-P4

 

[Tom Kitta]

The state runout is 0.01mm so over 2 tenth is still acceptable as per stated tolerance.

Given that collets are certainly not better then 0.01mm I tend to agree... If you got better collets you should go for better tollerance on the spindle.

Also I know Chinese know this is mostly used for hobby work - so price is more of a ruling factor vs. uber tight tolerances.

I expect standard taper roller bearings - nothing fancy.

Almost all hobby CNC machines use rolled ball screws. Ground screws are available from aliexpress and cheapest ones are at least 3x the price of rolled stuff. Next level is 10x or so the price (or more).

That spindle would cost 10x more if stated runout was at most 0.001mm. I certainly would not pay 10x more - there is no way to squize 0.001 accuracy from a home for 99% of home users.