Author Topic: Brushless DC Motor Spindle by c raynerd (Read 49028 times)

raynerd · « **on:** March 08, 2012, 06:14:15 PM »

Hi Guys
I`m making a spindle using a brushless DC motor and I`m using the idea that someone has come up with on mycncuk.

You basically take a brushless DC motor with a 8mm shaft. Purchase an 8mm straight shank er collet chuck, build up a bearing housing on the front of the motor and and replace the shaft with the chuck.

Something like this: http://www.mycncuk.com/forums/attachment.php?attachmentid=4786&d=1286643523

I`m sure many many of you on here understand these much better than me, but this evening I took apart the outrunner motor and now have an idea on how it works. Basically, the outside of the motor "case" itself actually spins and this is locked to the shaft, which therefore forces the shaft around. In this picture:
http://www.mycncuk.com/forums/attachment.php?attachmentid=4609&d=1316538477

...you can see how the ally bearing housing is screwed to the motor winding section. It is then the blank looking piece to the top right, filled with the magnets that spins and by being connected to the shaft, causes the shaft to rotate.

The bearing house as per the original idea shows the ally house bored to a deep shoulder, a washer, then bellvelle washers and then finally a second washer with the entire housing being screwed to the four mounting points of the motor. Although these means that the belleville washers will squeeze together when clamped to the motor and put tension against the washers against the motor and bearing house, it doesn`t put any tension or load on the spindle! The spindle can freely move up and down through the washer and is clamped with the grub screw at the bottom end.

In my opinion, would I not be better boring a piece of ally straight through to act as a guide to centre the washers. Inserting the washer, belleville washers and then last washer. Screwing the housing to the motor and then .... pushing the er chuck hard against the top bearing, consequently compacting the washers and then tightening the top grub screw to hold this tension on the bearings AND spindle.

Any help appreciated!

Chris

raynerd · « **Reply #1 on:** March 09, 2012, 02:42:26 AM »

I`ve just re-read last nights late woffle I wrote above and I think perhaps I need to simplify my question.

When making a spindle, should there be tension on just the bearings in the housing and the spindle just runs between these loaded bearings OR should the spindle actually be clamped up and be the thing that is actually putting the tension on the bearings?

I have two option in my head.

1. Is to make a bearing housing that is just a little bit too short with a blind hole into which the bearings and washers slide. When the housing is clamped up against the motor body, the outer bearing which will be pretruding just a small amount will compresses the beveled washers and puts tension on the bearings inside the housing.

2. The other option is to have a straight through bore in the housing. Bolt it upto the motor and slide in my bearings, beveled washers and last bearing. The bearing will then be protruding my a tiny amount on collet side. I`ll then push the chuck hard up against the bearings, compressing them a little and then tighten the grub screw to hold it in place. The bearings will then be loaded and the chuck will be holding this tension.

I`m not clear which I am aiming for...if either!!

Chris

John Rudd · « **Reply #2 on:** March 09, 2012, 03:37:12 AM »

I'm a little surprised to see such a large motor with what would seemingly be having a low kv used as a spindle driver...

It is usual to find the smaller motors having higher kv ratings than the bigger ones...Still if it does the job why not....and a novel idea of attaching the shaft to the front..

David Jupp · « **Reply #3 on:** March 09, 2012, 03:43:05 AM »

I wouldn't trust a grub screw to put any pre-load on the bearings. Definitely need something more positive than that.

Your plan 1 for compressing the belleville washers sounds good. Careful control of dimensions (and/or shims) would allow you to choose the pre-load level.

John Hill · « **Reply #4 on:** March 09, 2012, 04:37:51 AM »

I have some old tape drive capstans (1" mag tape) and they each have a knurled ring threaded on the shaft to put end load on the two ordinary-looking ball bearings.

HS93 · « **Reply #5 on:** March 09, 2012, 08:37:25 AM »

normally there is a circlip on what is the prop end and then a grub screw in the back Of the outer ring I use them a lot in boats and have had them apart quite a lot , I wonderd about a pre load but after looking at the way they arrived with a small gap between the circlip and the washer in front of the bearing I have gone with that what happens in a plane use is the the prop would push the shaft back and take up any slack this in turn just moves the out magnet ring back. I have been using some of the new and bigger motors and just built some mounts with cooling for a mate replaning the 5mm shaft with a 5 mm prop shaft so no couplings, we are also using one of the scorpion motors that are 450 watts through a belt drive that is a right beast.
be aware that these motors require cooling they are normally cooled by a whacking great prop or in the boats case a heat sink and water they tend to get the heat welling up after turn off and concentrating on the front cover getting very warm we have a pump to keep cooling the motor after a run for a few min this has worked well, but we are pushing it a bit.

https://youtube.com/watch?v=e5-YmmElIfk

the boat is 7kg
cheap uk supplier http://www.giantcod.co.uk/brushless-motor-c-25.html
Peter

HS93 · « **Reply #6 on:** March 09, 2012, 08:51:35 AM »

have you thought of a bearing and a thrust bearing.
Peter

raynerd · « **Reply #7 on:** March 09, 2012, 10:15:43 AM »

I appreciate all your advice.

Can I just clarify, does there just need to be tension applied to the bearings or does the spindle need to applying the tension on the bearings itself?

I`m not clear...

Chris

David Jupp · « **Reply #8 on:** March 09, 2012, 10:48:29 AM »

There needs to be pre-load in you want to eliminate end float of the spindle. There are alternative ways to achieve this which can put spindle in tension or compression. The strain may be localised rather than applied to the bulk of the spindle. You will likely want to consider the effects of thermal expansion on the arrangement for giving the pre-load.

Noitoen · « **Reply #9 on:** March 10, 2012, 02:18:52 PM »

They sell pairs of matched angular contact bearings. The "stiffness" will be right next to the spindle with no expansion problems.

tonybraz · « **Reply #10 on:** March 10, 2012, 04:30:32 PM »

Hi craynerd, I use a brushless motor on my cnc mill. It is a direct drive on to ER16 milling chuck. I run this on a 80 amp radio control speed controller with a servo tester to adjust the speed. The chuck is supported by a large bearing that fits on a 20 mm diameter on the chuck shaft . I have fitted a watt meter in the supply to the motor to monitor loading. I find this set quite successful if you need any more information let me know.

Tony

HS93 · « **Reply #11 on:** March 10, 2012, 05:30:19 PM »

Have you got any pictures Please ,and what kv / make was the motor

thanks

peter

tonybraz · « **Reply #12 on:** March 10, 2012, 05:40:54 PM »

Hi all the motor is 600kv and is powered by a Pro Peak 20 amp power supply running at 13.5 volts here are some photos

Tony

HS93 · « **Reply #13 on:** March 10, 2012, 07:11:05 PM »

thanks for that
peter

raynerd · « **Reply #14 on:** March 12, 2012, 08:29:37 PM »

It is late and time for sleep but I`m buzzing after just putting my brushless motor spindle together! Worked a dream once I had all the bits ready!

Cheap phone camera shot for now, but more over the next few days:

Thanks for the advice asked for in the original message before I had made a start

Chris

HS93 · « **Reply #15 on:** March 12, 2012, 08:51:14 PM »

how did you do the thrust in the end?

Peter

raynerd · « **Reply #16 on:** March 12, 2012, 09:03:59 PM »

Compacted it up to a thrust bearing and the beveled washers provide the tension on the bearings. Over the next few days, I`ll fix it up properly and post some better pics! Looks very solid..hope it works as well as it looks like it should!

John Rudd · « **Reply #17 on:** March 13, 2012, 04:28:10 AM »

Nice job Chris...

How are you going to control the motor speed and what you going to power it with?

raynerd · « **Reply #18 on:** March 13, 2012, 08:02:29 PM »

Hi guys,
Got it up and running tonight, not mounted on the intended machine, but spinning.

ESC 3phase driver is Sky...something. 100A
Speed controller is a servo tester - not explored yet but I presume the control is linear!? I intend to get rid of this and use PIC in conjunction with EMC to control the spindle once up and working.
Power supply is a cheapo variable supply upto 13v, 8.5A power supply.

It spins damn fast and with seemingly enough torque... the proof will be in the pudding!

Runout looks good

but more on that when it is fully mounted and not vibrating in the vice resting on the bench!

Warning... sound is stupidly high which is nuts because it is less than half noise of a dremel on full, the camera just captures it for some reason.

I still need to strip it, clean and take pictures. Has anyone any suggestions as to a secure way to grip the shaft in bell at the end. It is currently held in by a grub screw but since this is going to be taking some of the cutting force, it needs more...any thoughts?

HS93 · « **Reply #19 on:** March 13, 2012, 08:34:31 PM »

when you pull it apart centre pop the mark made by the grub screw and drill you for the grub screw to drop in to , that's how we are doing the boat shafts as they are turning a 50mm brass prop that may come out of the water at times and go back in on full power,so snapping action .
you could always drill a second grub screw on the other side as well.

peter

raynerd · « **Reply #20 on:** March 14, 2012, 06:30:16 AM »

Cheers Peter.

I`m going to put my hands up here and say I`m useless! My grub screws always slip.

I`m guessing here, but I think the grub screw on the original shaft is M4 ish. So would you drill 4mm so the grub goes into the shaft or just shallow mark so that it is essentially sitting on a flat? I`ve filed flats before and my grub screw always ends up slipping and I don`t understand why since in theory, it can`t!!

Chris

Bluechip · « **Reply #21 on:** March 14, 2012, 07:06:51 AM »

Chris

What's on the end of your grub-screws ?? A point ??

The cup/knurled cup/dog point hold a lot better.
Particularly the latter if you do a small hole to accept the end.

You can , if you have enough room, stick a second screw on top of the first, as a 'lock-grub-screw'.
ie. two short instead of one long one.

If you are going to put two in different places, the second goes at 90 deg.
For some reason shaft retaining collars / gears etc. on machines are always like that. Why, I don't know.
IIRC there was some explanation given in the ME postbag section decades ago.

Why grub-screws anyway? If it's permanent, poke a taper pin in the wretched thing, or even a roll-pin.
Taper for preference.

BC

John Rudd · « **Reply #22 on:** March 14, 2012, 08:51:27 AM »

Thought I'd chip in with a couple of things......

Is your power supply capable of being current limited? Brushless motors can draw a fair amount...

I'd stick with the single grub screw...If you stall the motor (potentially this could happen) you will either burn out the speed controller if the power supply isnt limited or may damage the motor windings...

It makes sense to have a 'weak link' so to speak..

If I'm talking rubbish just ignore me....

raynerd · « **Reply #23 on:** May 04, 2012, 04:34:12 AM »

Hi guys, just updating this thread as I`ve not had much time to do so recently.

I stripped and rebuilt the spindle and while doing so, I took some pictures:

Here are all the bits. You`ll notice the surface marks where the bearings were skipping on the previous built due to not enough tension and slight misalignment. I hope this is now stopped!?
You can see the thust bearing on the shaft up need the head of the chuck.

The bearings are separated by a number of washers which will compress to apply tension to the bearings.

Bottom half of the motor is screwed up:

And the bell slid on the shaft with a grub screw tightening this up. There is now a flat milled onto the shaft:

It is working well so far. I`d love to know how it would perform as a tool holder grinding spindle on the lathe!

According the my hall effect sensor and Mach3 - I can get 10,100rpm max speed.

Chris

raynerd · « **Reply #24 on:** January 16, 2013, 02:55:36 AM »

Just to bump this back to the top as I`m hijacking Erics CNC build thread with comments on the spindle.

As John has mentioned in his post above, I am having issues with the power supply. It works fine for wood and plastic, but starts stalling on other harder materials. I`ve cut brass and ally, but the cuts are so small it is almost impossible unless the material is thin.

I was told the ATX probably wouldn`t do.

Swarfing - I know you said to tie two ATX supplies together, problem is, I already have 2 tied together to run my cnc machine steppers so that would be 4 in total! Rather than going to the trouble of trying to find two more, I think one of these cheap Chinese made 12v supplies will be OK. What sort of current will this spindle be trying to pull? As John said, I`m guessing if I allow it to pull too much current it will burn itself out!

Chris

News:

Author Topic: Brushless DC Motor Spindle by c raynerd (Read 49028 times)