[limbo]

Hi,
Has someone tried before to use a brushless model airplane motor as a semi- HF spindle? They are much more powerful than a dremel and run at high very RPMs. I'd expect that they maybe give low vibration because of the small rotor diameter. Possibly also quieter during rapids (no constant brush-noise).

Could this be done? I'm cutting ply and balsa, so 2HP is overkill for me. My 130W "dremel" has power to spare. Low vibration and noise are my first interest. I'd have to buy one to try it, but maybe someone here knows if an unloaded brushless motor vibrates more or less than a comparable brushed one...

And, what about bearings? I'm thinking of an aluminium shaft coupler between motor and bit (both1/8'). Probably its best to have an extra bearing as close the bit as possible. Would a shielded abec9 skate bearing do, or are they just not up to high RPMs?

Greetings,
John

[MrBean]

I too have been wondering about using this type of motor. Once my machine is finished I'm sure my neighbours won't be best pleased with a >2000-Watt router screaming for hours on end.

Was thinking along the lines of the homebrewed LRK motors.
http://www.aerodesign.de/peter/2001/...index_eng.html

Seem to have good torque for the size, not sure if RPM would be high enough tho'.

Could even be a fun CNC project. Imagine trying to cut those laminates by hand.

Anyone think something like this could be worth using?

[limbo]

The outrunners are known for higher torgue at lower RPM. I'm not sure what is more important for routing ply. I run my dremel at setting 4 (~20000rpm?). What about inside runners, they have small rotordiameter and run fast. A smaller rotor might be better, as every gram unbalance will have a smaller arm of force, I think. Is this correct? Maybe this is why HF spindles are long and thin? The nice thing is that brushless motors can be extremely small, even speed500 equivalents spit out 200W or so. Maybe I"ll try something like in the drawing.

A few questions remain:
1) brushless motors run very stepper-like at low RPM, giving more vibration than a brushed one. Does this change around at high speed? Any Electroflyers?
2) What would be best, 1 skate bearing with fixed coupler or two with flexible coupler?
3) Do skate bearings explode at 30-40 kRPM?
All input welcome.
greetings,
John

[Swede]

I like the way you guys think... getting the most power in the smallest size.

Two things which would concern me would be the duty cycle of the motor (it'll get hot VERY fast) and a suitable power supply. Correct me if I'm wrong, but don't these motors draw 20 to 40 amps at 12-18VDC or so? I suppose you could use one of the cheaper 12V/30A supplies.

A continuous duty high-speed spindle motor in the 1000Watt range, such as the HF type seen routing and drilling printed circuit boards and such, often requires water cooling. I'd bet you could make up a water jacket with a small pump to handle the cooling chores.

Fascinating topic.

Swede

[limbo]

Swede,

Thanks for your input. I don't know about the temp effects. When first posting this I was thinking of the hacker C40 in the pic. Its ment for model cars so there is no big prop or airflow to cool it. Good tip on the water cooling. It should not be too difficult to twist some copper tubing in between the cooling fins or so.
You're right about the low voltage and high current. I am happy to have a regulated 0-20V/40A supply lying around... So no big problems here, but I've found some new ones that will probably kill the project:
Most likely the smoothness in a true HF spindle comes from the fact that they use pure sinewave controllers. A model airplane brushless controller gives trapezium waveforms, causing the "stepping" at low rpm. Maybe at high rpm the rotormass will smooth things out, but I doubt it. Also, my dremel really shreaked yesterday when cutting only 6mm ply, putting my feet back on the ground concerning the necessary rigidity and quality of bearings.

greetings,
John

[Hobbiest]

Check out this link at Dave Kush's site, http://www.buildyouridea.com/cnc/pcbmill/spindle.html. And here is the link for the specs for the different ABEC numbers, http://www.buildyouridea.com/cnc/hblb/abec_specs.gif. Abec 7 bearings are usually good to about 35k rpm. And that is under vertical load.

[Cold Fusion]

I am very deep into RC cars and deal with brushless motors all the time. Feel free to send me a over a PM about any of them.


I have a C50 in my truck right now and a Lehner 5300 in our team truck.


There are alot better motors out there then the c40. First that comes to mind is a Hacker B50 8s. It is cheap, easy to find on ebay, and has some crazy torque.


Limbo: Would it be possible to talk to you on the phone about all of this? I could answer all your motor questions and you could tell me alittle bit more about your fine router.

[Cold Fusion]

http://cgi.ebay.com/ws/eBayISAPI.dll...&category=1200

Hacker B50 8s.

[Alan T.]

Aways a "However" in the bunch.

Good idea guys but I would be concerned about the effects of lateral forces on the bushings or bearings. These motors are not really designed with cutting forces in mind. You would I suspect, very quickly develop problems unless you kept the cuts very, very light and a slow feedrate.

Alan T.

[Hobbiest]

If you run a seperate spindle, and use step pulleys, then you not only save the bushings, but can have multiple speeds.

[limbo]

Yesterday I got the chance to do a test! A fellow flyer came over with his Mega Acn brushless motor (speed 400 size) and Yeti controller. Not bad, Not bad at all... We tried running it full throttle with the case held to the table to amplify the vibration-induced noise. Just a faint high pitched whistle and thats it. Try that with a regular speed400 or 600.... So the rotor appears to be nicely balanced and the stepping effect I expected is totally absent at high RPM. Of course, this size motor is too small to provide enough torgue to cut 6mm ply, but it gives an idea.

I too am worried about the lateral forces on the motor bearings. Maybe if you make the Alu housing in the drawing a bit longer most of the force will be at the skate bearing. I like the pulley solutions although I'd prefer the more compact inline setup. But I think Swede hit the nail right on the head with his duty cycle/ temperature comment. The motor got warm, even though it was running unloaded! These things are supposed to have super efficiency? I'm in the dark here, anybody knows what happened?

John

[Shiraz]

Probably the eddy currents kicking in, warming up the stator...if it gets too hot, the high powered Neodymium /Iron/Boron magnets used in the rotors of these motors could get fried too, at about 80 Celsius...one would need to use Samarium/Cobalt magnets that are more resistant to heat... I also understand that the trapezoidal drive also leaves marks in what is being processed, so I guess high speeds don't particularly help (assuming the machining was done at high speed..)

[ViperTX]

If you're talking about the spindles where the rotor is the spindle, then what I remember is that most of the motor controllers for those motors monitor the temperature of either the bearings and or the shell that contains the bearings and magnets. Sometimes they use either the cutting fluid (through the rotor / bit), air, or a separate resovoir to cool the motor, otherwise the motor mounting is used to conduct the heat away from the motor. Alot of these also use air to keep contaminants from entering the bearings.

[Shiraz]

I don't think it would be easy to arrange for cooling when using an airplane motor except by using heatsink ridges on the rotor (assuming it is an outrunner)...and although theoretically there are no sparks, commutation being electronic, sending a fluid through the motor may be a problem, as it may deteriorate the insulation of the magnet wire windings and otherwise foul up the insides, bearings etc...the motor and electronics would further have to be sealed to prevent wood chips and sawdust from clogging the inside, and so ridges on the rotor may not help dissipate heat either, unless it somehow is able to communicate with the outside...fluid conducting pipes running through the stator is going to make winding the stator very complicated, given that it is difficult to even wind sufficient turns given the amount of spacing available at each "tooth" of the stator...

[signIT]

Limbo,
I like your drawing it's a nice idea and very easy for diy:ers, you you can make your own cheap spindles using Mabuchi 12V DC motors and a metal tube equipped with two skate bearings, Yahoo! this is it, it would be so cheap that one could afford to make a whole bunch of complete spindles.

Thank You!

Leffe

[limbo]

Its nice to see that this thread came to life again. However, with some new insights, I doubt that a speed600 with skate bearings would make a better spindle than a cheap dremel clone. Originally I was thinking about a brushless inside runner but, as Swede said, watercooling seems mandatory. The combined cost of a big brushless motor plus speed controller is similar to that of a good quality router. I bought a kress fm6990 in Germany for 150 euro. It made for a whole new machine. Quiet as a hairdryer and high torque at lower RPMs. The dremel had to run at 20000rpm while cutting ply, the kress can run at ~10000 for equal cutting speed. The sound when cutting has changed from a high pitched shreak to a soft low hum. I'm convinced that the mass of the router and the quality of the bearings were the biggest factor in these improvements.

John

[ViperTX]

limbo,

You are correct the mass of the rotor and the type of bearing make the most difference. Generally to prevent "cogging" the cheaper routers must run the motor a bit faster...generally what I've encountered is the higher the HP the smoother they run and that is all due to the mass of the rotor.

[arvidb]

Why not add a flywheel to the spindle to smooth things out?

Arvid

[ViperTX]

Why not add a flywheel to the spindle to smooth things out?

Arvid

You certainly could...it works for automobile engines...it's an additional load and you have to contain it somewhere.

[arvidb]

It only adds a load during acceleration... so only for a few seconds. Once it's spinning it does not inflict any extra load to the motor. Well except maybe some air drag...

Arvid