Where to next?

[macardoso]

I would personally turn up my nose at the 1.8kW DMM motor. A speed increase through pulleys is possible but always more difficult than speed reduction. You would not be able to run that motor on a VFD as you would need to pull the commutation data out of their 17-bit serial encoder (likely impossible). When you go above the rated speed, the motor torque will drop off a bit so you likely wouldn't have the torque you would hope for at higher speeds.

My setup is the best I have been able to come up with so far. 1:1 gearing of an Allen Bradley MPL-A330P-xxxxxx servo and an Ultra 3000 (2098-DSD-020x-xx) drive. The "xx" are characters which don't affect the performance of the motor. That is 1.8kW but a rated/maximum speed of 5000rpm. With a 1:1.5 gearing you could easily get 7500 rpm. These are available on ebay and I could help you set it up.

Think hard about if you really need a servo. They are great, very compact, and very powerful, but also very expensive. A 3600rpm inverter duty induction motor could be set up to run at 7200rpm within reason. You wouldn't get the stability in speed that a servo gives you, or any kind of rigid tapping/spindle positioning, but you could save some $$.

I would say that DMM is a great choice for the axis motors.

All of this is just my personal opinion/experience.

 

[bakrch]

I would personally turn up my nose at the 1.8kW DMM motor. A speed increase through pulleys is possible but always more difficult than speed reduction. You would not be able to run that motor on a VFD as you would need to pull the commutation data out of their 17-bit serial encoder (likely impossible). When you go above the rated speed, the motor torque will drop off a bit so you likely wouldn't have the torque you would hope for at higher speeds.

My setup is the best I have been able to come up with so far. 1:1 gearing of an Allen Bradley MPL-A330P-xxxxxx servo and an Ultra 3000 (2098-DSD-020x-xx) drive. The "xx" are characters which don't affect the performance of the motor. That is 1.8kW but a rated/maximum speed of 5000rpm. With a 1:1.5 gearing you could easily get 7500 rpm. These are available on ebay and I could help you set it up.

Think hard about if you really need a servo. They are great, very compact, and very powerful, but also very expensive. A 3600rpm inverter duty induction motor could be set up to run at 7200rpm within reason. You wouldn't get the stability in speed that a servo gives you, or any kind of rigid tapping/spindle positioning, but you could save some $$.

I would say that DMM is a great choice for the axis motors.

All of this is just my personal opinion/experience.

Good to know, many thanks!

I am fine with just the 5k rpm really, for now anyway. Maybe when I mount it all to the next machine we can shoot for more.

"need"?, heh ... noooo. This is more about building an optimal setup without getting too ridiculous. Like you, my interest is something that can later be moved to a bigger machine and perform reliably. Putting this on the PM25 would get me familiar with the control setup, and hopefully, make my inevitable mistakes cheaper.

$3k is a reasonable budget for the entire control system I think. If I go a bit over, not worried.

 

[macardoso]

"need"?, heh ... noooo. This is more about building an optimal setup without getting too ridiculous. Like you, my interest is something that can later be moved to a bigger machine and perform reliably. Putting this on the PM25 would get me familiar with the control setup, and hopefully, make my inevitable mistakes cheaper.

I get that It is a bug!

$3k is a reasonable budget for the entire control system I think. If I go a bit over, not worried.

Sounds very reasonable to me. Figure adding in these rough costs. Spending a little more to do it "right" will make it much more painless. Can certainly go cheaper here though if you end up against a budget.

• $100 - Electrical Enclosure
• $125 - Disconnect, Breakers, etc
• $200-250 - Motion Controller, Breakout Boards
• $0-200 - CNC Software
• $75 - Cable, Wire, Crimpers, etc.
• $50-100 - Bulkheads or connectors for cables
• $25 - Wire Duct

 

[bakrch]

Okay, Mike already knows this but spindle servo motor and drive acquired!

I will update when everything is assembled for a bench test!

The tapered roller bearings should be cowering in fear, because I plan to cook those first before replacing them.

 

[JimDawson]

Spindle motors: Using a small (1.8 KW) servo for a spindle motor would not be my first choice, especially when running on single phase power. Take a look at the published torque curves. Then divide that by your planned gear ratio. Ok if you are running in the lower RPM ranges, but do not make for a good high speed system. Most of the small, good quality (Baldor, Marathon) inverter duty induction motors (<5 HP) are rated at 5000 - 6000 RPM max, and run very well, and will give you 100% torque from 0 to the motor rated speed when run with a sensorless vector VFD. You want to get a 4 pole (1800 rpm) motor, they have double to torque of a 2 pole (3600 RPM) motor for a given HP. You can get purpose built high speed servo spindle motors, but the cost is more than the rest your machine. Just not practical for the home shop.

Rigid tapping is possible when using an induction motor driven by a VFD, just requires an encoder on the spindle and the controller must support rigid tapping. The only thing I have not been able to accomplish yet is to index using VFD control, but not important on most mills without tool changers.

You were posting while I was writing this, I see that you have acquired a servo motor. Your opportunity to prove me wrong, I have no first hand experience with servo spindles, only running the numbers. I'm very interested in the results, you may cause me to change my mind.

 

[macardoso]

I'm very interested in the results, you may cause me to change my mind.

We were able to find @bakrch a servo that has a rated/max speed of 5000rpm, a flat continuous torque through the entire speed range of 3.5Nm and peak torque of 9Nm up to 2500rpm which tapers down to the continuous rating at 5000rpm. Should be a good fit for a machine like the PM-25MV since you cant really use much more power than that. That's all on 240VAC 1P.

I've found that when I am roughing aluminum at the absolute limit of my machine's rigidity, I'm nowhere near maxing out that spindle (maybe 40%). I do use most of the torque for drilling at the lower speeds and roughing steel which is why I've kept the gear ratio on mine at 1:1.



I will be interested to get his thoughts on the matter once it is up and running!

Rigid tapping is possible when using an induction motor driven by a VFD, just requires an encoder on the spindle and the controller must support rigid tapping. The only thing I have not been able to accomplish yet is to index using VFD control, but not important on most mills without tool changers.

Good point on the rigid tapping. I was wrong on that!

If only induction motors weren't so big and heavy for these little machines

 

[bakrch]

We have a Summit EVS349b (3hp 4300RPM) manual knee mill at work, I use it often and know (roughly) what to expect in the future when I move the servo motor to a larger machine. Wanting more than this would just be greedy, at that point it's time to get a HAAS TM1 or Mini Mill.

 

[macardoso]

Boy that's a nice looking mill. Never seen a company selling modern manual machines at 26k lbs. Geez

 

[bakrch]

Boy that's a nice looking mill. Never seen a company selling modern manual machines at 26k lbs. Geez

Yep, that's about what our HBM-110 weighs, apparently.

We have their 16",18",22", 42" lathes, the horizontal pictured below, EVS349B and EVS-550B knee mills.

They had big shoes to fill (replaced Bridgeport, Giddings & Lewis and Monarch equipment) ... but they get the job done.

 

[bakrch]

I still haven't messed with the servo motor testing, need to get on that within the next week or else I am out of the return zone should anything go wrong.

What I have been busy with, taking my PM-25 apart to diagnose chatter while climb cutting in X.

Nothing was loose, or even close to coming loose at all ... which surprised me. All that stood out was shoddy work on the double ball nuts. Here is a short (crappy quality) video that I took.

https://imgur.com/UOkA1gd

My backlash has between .002-.003 in X and Y since the day I got it, which is explained by less-than-perfect preload. Not the greatest but have been able to make decent parts using compensation. I would have liked to take them apart, but the key screws are ground too far down.

I checked videos from Hoss from back in the day and I guess even that unbound lateral play is just normal for a Chai C7 setup.

The dovetails are still quite perfect, with incredibly smooth action throughout X and Y travel range.

Now I am going to look at the column again. It needed to be shimmed quite a bit in both directions, which had to impact the already weak rigidity. Next, I may take all of the column shims out, run a few aluminum parts to see if there is an improvement. I went ahead and put it back together, then checked tram from the head to the column ... it is still darn near perfect.

There is always the Stefan Gotteswinter epoxy method for the column shimming, but I do have machinery available to simply skim cut it square. The setup would be tricky, but it is doable.