Brushless DC motor for machine tool power

So I was considering getting this BLDC motor and through a 9:1 worm gear get it to power a spindle for 4000 RPM and below. Powered by Arduino and this ESC, though coding specifics are a work in progress.

Are these motors able to output the advertised (1500w) power for extended periods of time?

For the motor and controller from Aliexpress it would cost £40. A weaker 1000w motor for a SC6 is a £300 motor and £200 control board. I know the guys at Arc Euro have bills to pay but a £560 makes me think there might be some more issues with the Aliexpress sourced motor?

Having used motors of this general type in model aircraft, I would have to say no, they are not suitable for extended periods of use. Typical use would be flights of up to ten minutes or so, not all at full power. They are great little motors for what they are meant for, but not I think very suitable for powering machine tools.

John

I suspect that a worm gear wouldn't work very well either at such a high speed.

Hi Rainbows,

If you start reading from page 8 of this thread, hopefully you may get some of the answers.

Ketan at ARC.

Add to last post:

On page 8, read from thread posted by Savinay Bhari on 24 March 2015 12:58:22 onwards...

Ketan at ARC.

1500W on a 5mm shaft ?

Yer, right.

Because it delivers its power at high speed it delivers a low value of torque. I do not know what size the motor is (3" long?), allowing for an efficiency of 90 %, it means that the motor has to dissipate 10 % X 1500 = 150W. So long long will it run for before it overheats? You could try cooling it with pumped liquid hydrogen like power station alternators.

Frank

Not that motor but you MAY have some luck with a much larger motors with much lower revs / volt. It was pretty easy to find these on ebay in the past but not now by the look of it. From memory they were around down to around 500 revs per volt with reasonable power levels. In use these were generally speed reduced with a toothed belt or used to drive rather large props on model aeroplanes.

JS has missed the revs aspect of the motor power. It would be running at 20,000 rpm on 12v and the power is very probably input not output. It's possible to do some simple rough sums as spec's sometime give the winding resistance. That gets lost as heat.

They have been used for directly driving cnc machine spindles in the past but I have no idea how well this worked out. A companies was selling them on their machines.

It's also possible to find info on the web concerning winding and making your own. It's even possible to get high power from smaller laminations by stacking them on the same spindle.

Personally I would bear in mind that cables for 40 amps or so will need to be a bit bigger than the rather short lengths used in models. From memory 1/2 hp out on 12v is probably possible.

Just add that it's probably possible for a model engineer to modify a motor. It wouldn't be difficult to remake the outer casing of an outrunner to use different bearings and maybe cause it to generate some air flow to cool the stator if this sort of thing was found to be needed.

Sorry about the probables. It was cabling that gave me some pause for thought. Ideally the power supply needs to be near the motor.

John

-

Edited By Ajohnw on 16/07/2016 11:14:31

.

Just as a point of comparison ... There was an Atco lawnmower with a 1/4 hp 12v motor: the cable on that was about the size of the common lightweight jump-leads for cars.

MichaelG.

Here is an interesting conversion of a mini-lathe touse a brushless dc motor:

https://letsbuildone.wordpress.com/mini-lathe-motor-upgrade/

Mike

The motors seem to have moved on since I looked at them. If some one wanted one like this they would have to make it themselves

**LINK**

People might not realise that it's the outer casing that rotates. It's usually just supported on a single bearing at the front - unless that has been changes as well.. The casing hold the magnets and the stator carries the windings. As it uses magnets it can't be allowed to get too hot and is fundamentally well balanced.

I'm not looking further in case I get tempted but there is a whole load of them here

**LINK**

There may be some with larger diameter shafts about as well but this higher pole count is probably more suitable for spindle use as it seems to bring the revs down.

Problems, worries - well the people on here are supposed to be engineers. With a prop on the end these are generally working without any side loads and the prop will also add to cooling. The casing rotation will too. The larger helicopter motors if still about are subject to side loads as they usually drive a belt. These are around rated on the basis of the equiv cc size of an ic engine in very large sizes.

Cheapest sources of power are the 12v server power supplies that crop up on ebay. The connections have to be jumpered to make them think that they are plugged into a server. They can be very cheap even if brand new. They can be used for charging lipo's so there is info on the web about jumpering some models.

John

-

Edited By Ajohnw on 16/07/2016 12:14:59

I've read the ad several times and so far, I can't see the RPM quoted. Working backwards 4000 rpm spindle through a 9:1 worm would require 4000 x 9 = 36,000 rpm at the motor spindle - I can't see that happening, industrial routers running at 24,000 rpm have bearing issues and they are totally enclosed, built in semi-clean room conditions.

Have a look on Arc Euro Trade\s web site at the high speed spindles to see the difference. I've also looked at Zapp Automation for ideas.

.

Bob,

As I understand it [which is not very much] these motors are specified in "revs per volt" ... Which, in the case of the advertised motor would be 1600

No, it doesn't make sense! ... But here is another advert **LINK**

MichaelG.

So you're saying he wrong that it's expecting a lot to put 1500 watts (2hp) through a 5mm (0.1" ) shaft?

For small short shafts P=(D^3 x N)/38

(0.1 x 0.1 x0.1 X 20,000) /38 =0.53

In other words is that's the output power, the shaft is overloaded four times.

Even if the motor is only 25% efficient (375w output) it's still right on the limit and while that might work for a prop that's without making any allowance for the shock loads associated with shock loadings.

I think John makes a very good point - 1.5 kilowatts for that motor has as much validity in the real world a 'music power'.

Neil

Edited By Neil Wyatt on 16/07/2016 14:28:59

MichaelG,

Had a Google at LiPoly batteries and it would appear that a 6s battery, as specified as maximum in the ad, gives 22.2 volts - therefore we are expecting 1600 rpm/v x 22.2v = 35,520 rpm. Sorry, but that still sounds impractical when compared to an industrial 24,000 rpm spindle.

Putting the motor design to one side, the practicality of running a 9:1 worm at 35,500 rpm also sound impossible - maybe an epicyclic gear could cope, but I would expect that cooling and lubricating will be an issue here also.

Motors in CNC machine spindles are usually direct drive, axis drives are usually via ball-screws giving a reduction ratio of the pitch / circumference of the ball pcd and robots tend to use Harmonic gearboxes, Cyclo-drives or RV reducers all of which are driven with servo motors running at max 4000 rpm.

server supplies are a revelation, very nice recomendation. 12V 82A one for £14 caught my eye

With that existing motor it would drive it full torque at 19200RPM which becomes 2133 RPM on the spindle. Reduces power to 840W which gives a nice safety factor from the power supplies 1000w output.

One thing I noticed with the lower Kv motors was that after considering the gearing down from motor to spindle they give less torque.

The aliexpress one I linked has 0.397Nm torque at 70A, when geared down 9:1 to get 2000RPM I get 3.572Nm

The one you linked (though I did check others) gives 0.755Nm (better) but thats at 3840 RPM (12V) so a 2:1 reduction only gives 1.5Nm on the spindle at 1920 RPM.

Atleast that assuming these equations are true.

I wish that letsbuildone had a 1 year on update or something to say how well the motor coped.

Also is there a calculation to show the maximum speed for a worm gear set?

edit: oh my more replies while I was writing the post

Edited By Rainbows on 16/07/2016 14:56:04

Worm gears work by having a rubbing contact between worm and wheel, like a screw in a segment of thread if you like. Spur gears have nearly rolling contact. So a worm drive will have a lot of friction and get rather hot.

Frankly you have to ask why these motors aren't used in machines commercially. You would be much better off using a standard permanent magnet brushed motor and a standard drive unit such as a KBE one, or a little 3 phase motor and a vfd.

Short measures again Neil?

(0.2*0.2*0.2*35520)/38 = 7.48 = 5577w (1500w motor)

(0.2*0.2*0.2*19200)/38 = 4.04hp = 3000w (840w / when powered by 12v)

Edited By Rainbows on 16/07/2016 15:27:19

There is no side loading on these motors in normal use Neil. Just a prop so the shaft only takes thrust. The set ups have to be well balanced. New shafts and bearings have been available - for smash ups but I doubt if the bearings used last long.

The power figures are also overstated. Some give a most efficient rating which will be rather different to the headline figures. As amps and winding resistance are usually stated it's also possible to work out heat loss to some extent I^2R. Then there are losses in the controller and wiring. It wouldn't surprise me if the headline max amp rating was close to the fuse rating of the copper wire used for the windings or something of the order of a 10 sec rating etc.

This gives some idea how the higher power motors are sometimes used

Wot - no shaft used. Trouble is that it's still rotating on the inner shaft but doesn't bend so easily when they crash. Just add that this sort of set up could be improved by adding a more substantial bearing in a separate housing.

None of this really matters for their intended use as generally all that matters is the thrust the inevitable prop on the end gives in use. Different sizes and kv ratings are intended for use with prop's of a given diameter and pitch plus a reasonable flight time. There is bound to be info on this subject about on the net.

Worms and wheels aren't very efficient at transferring torque especially with high reduction ratios. Gears as per electric drills and angle grinders etc are better but noisier.

Personally I think Rainbows is expecting too much of them and looking at the wrong kv ratings even though these appear to offer more torque. The higher reving ones wont be easy to gear down to what are usually regarded as useful spindle speeds. For directly driving a tiny carbide milling cutter at high speed they might work out.

John

-

Edited By Ajohnw on 16/07/2016 16:33:49