spindle cross section view

[routercnc]

Some time earlier this year I was searching the internet for spindle drawings, prior to purchasing a ready made unit, just to see how they were made. Maybe I was looking in the wrong place as initially I didn't find much, but eventually I found this - might be helpful for those looking to make their own.

It's in Chinese, but the bits your can read suggest:
220v, 400Hz, 3.7Kw, 12.9A, (fairly powerful then), 24000rpm, 0.003mm spindle runout, 16mm collet (?) with 0.02mm runout (? not sure), and some figures which might be bearing preload (190+-10N and 150+-10N). Please correct me if I'm wrong!

You can see opposing pairs of angular contacts top and bottom, the motor in the centre, and what is likely to be water channels around the outside leading to the lower pair of bearings. Top cap appears to have the power supply, and the in/out water feeds.

[Swarfing]

If you look up the specs of an air die grinder tool you will see they are the same sort of layout (less the inbuilt motor and water cooling). It' just a case off finding an appropriate size collet chuck against motor with similar size spindle. Push out spindle and replace with collet chuck (collars etc). Thats what i am changing my spindle motor for. By the way the shank on a die grinder is really too short to be used like this unless the motor is mounted off the unit and belt. The chinese collet chucks can be got with a 14mm shank and 1/2" which match a lot of suitable size motors out ther (diasbled cart's/ buggy motors, DC and large spindles :-)). Pop in to your nearest mobility centre and you will see what i mean, may even sell you a secondhand motor for a few quid from there repair shops normally on site. Whilst you are there ask them for a speed controller, you change the current limiter (that's what stops the pavement road nights brigade from speeding). Maybe some ideas in there to save inventing the wheel???

[GeorgeD]

Was gonna suggest these...beaten to it
http://cgi.ebay.co.uk/STRIDE-MOBILITY-SCOOTER-REAR-SUBFRAME-MOTOR-/270614909289?cmd=ViewItem&pt=UK_Health_Beauty_Mobility_Disability_Medical_ET&hash=item3f01e7d169

http://cgi.ebay.co.uk/Mobility-Scooter-Spare-Parts-24v-Motor-Transaxle-/320559536162?cmd=ViewItem&pt=UK_Health_Beauty_Mobility_Disability_Medical_ET&hash=item4aa2d65822

[GeorgeD]

On most Aldi power tools the motors are the brushless type,thing is they're heavy but very quiet...the power tools are cheap ie the tiler,table saw,ect but the motors are good.

A bit of adapting at the spindle shaft for collet nut and you have a decent spindle ie cut the first part of shaft off and then bore out center on a lathe.

[GeorgeD]

How about a 3hp 180dc tread mill motor. :heehee:
http://cgi.ebay.co.uk/3-hp-dc-treadmill-motor-incline-motor-drive-belt-/300451332574?cmd=ViewItem&pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item45f44b81de

[Ross77]

Thanks, I was struggling to find info as well, thats why I have gone to the manufactures.


It's in Chinese, but the bits your can read suggest:
220v, 400Hz, 3.7Kw, 12.9A, (fairly powerful then), 24000rpm, 0.003mm spindle runout, 16mm collet (?) with 0.02mm runout (? not sure), and some figures which might be bearing preload (190+-10N and 150+-10N). Please correct me if I'm wrong!

I think point 2 is holding the inner races to the shaft and point 3 is the preload,


You can see opposing pairs of angular contacts top and bottom
Thats the bit thats confusing me as there are plenty of designs like that but the manufactures say dont do it. The locked bearings should be at the bussiness end and free floating bearing at the other,

http://www.skf.com/images/cat/images/9/9_0/6002_p54_1.jpg

http://www.skf.com/portal/skf/home/products?maincatalogue=1&lang=en&newlink=9_0_51b


still a little knowlege being a dangerous thing and all I'd better do some more research
I guess there are always compromises with any design and you just have to get it to suit your application

[routercnc]

Great site SKF Ross, lots to read about. Shows the lengths required to make high end machines.

Take your point about double pairs of angular contacts - increased preload when the spindle expands. Perhaps water cooled is OK, or if you look at the Chinese machine the top bearings seem to be located with either a coil spring, or are these sliding bearings? Maybe they just apply a factory preload and you get so much life out of the unit.

As a bit of light relief, after all the time and energy you are putting into this spindle, make sure you don't do this . . .:surprised: :surprised:
http://www.youtube.com/watch?v=gPPp1A1gj2M
Makes you wince. . .


And now for something completely different!
http://www.youtube.com/watch?v=yln_IGDuOCo&feature=related

[routercnc]

Interesting point about the die grinder, will free spin up to 24000 rpm, and will take all the loads. I did see one with an extended section, which might get around one of your points.

Note that I'm not looking to make one, I posted to lure Ross out with his spindle design progress. . . .

Anyone else tried to make something like this?

[Ross77]

And now for something completely different!
http://www.youtube.com/watch?v=yln_I...eature=related (http://www.youtube.com/watch?v=yln_IGDuOCo&feature=related)



Thats one brave or stupid person..how many times do yo think they ran the code before siting on the seat? still beats the hell out of Alton towers eh.....



Interesting point about the die grinder, will free spin up to 24000 rpm, and will take all the loads. I did see one with an extended section, which might get around one of your points.



Noisy and fixed speed tho so might not be good for all uses.


Note that I'm not looking to make one, I posted to lure Ross out with his spindle design progress. . . .

Mission accomplished....................................:w histling:

[Ross77]

Barry
Is that picture that you posted at full resolution? I saved it but gets a bit grainy when I zoom in, Im trying to see if the front bearings are fixed to the shaft and the spring loading is just on the rear bearings.

Cheers

[routercnc]

Hi Ross,
Sorry, that's it. I've found the original picture - scroll to post #9
http://www.cnczone.com/forums/showthread.php?t=101803

There is a model reference number in the bottom right corner of the picture, maybe search on that?

By the way, I stumbled across Steve Hilton's post from earlier this year where he was making a spindle. I notice you replied - but worth another look?

[Ross77]

Thanks Barry, this spindle thing is driving me around in circles........

Yeah I remember that, Think he already had the cartridge spindle and quality bearings, just remade the spindle for er 11 collet. cheers I will look again tho.

[coredump]

Thats the bit thats confusing me as there are plenty of designs like that but the manufactures say dont do it. The locked bearings should be at the bussiness end and free floating bearing at the other,

Deckel do the opposite: free floating (needle bearing) at the nose, locked bearings (dual andular ones) at the end. This allow very sturdy spindle with a short distance between nose and bearing (almost 0 since the needle bearing is outside the taper nose).

[Jonathan]

Deckel do the opposite: free floating (needle bearing) at the nose, locked bearings (dual andular ones) at the end. This allow very sturdy spindle with a short distance between nose and bearing (almost 0 since the needle bearing is outside the taper nose).

I might be being pedantic, but surely as the spindle heats up the shaft will expand and push the cutter down slightly further. Clearly won't happen if the fixed bearing is near the chuck. Negligible?

[coredump]

Sure if you are milling in the micron ball park. Anyway once you spindle reach thermal equilibrium you don't care anymore about this.

Edit: to be precise, FP1 have two needle bearing outside the taper and directly behind them, a trust bearing. On the other side of the spindle, a single trust bearing.

[Ross77]

Sure if you are milling in the micron ball park. Anyway once you spindle reach thermal equilibrium you don't care anymore about this.



Fair point and and seems to be the mailn difficulty with building high speed precision spindles is getting the correct clearances at operating temp.

You did only mention axial expansion but the real problem (for a diy spindle) I think would be the radial expansion and the resulting change in bearing clearance

if its preloaded when cold it can cause brinelling problems if not used regularly and if the preload increases during the warm up period then there is the risk of overheating and to much preload binding and associated damage.

Therefore I think needle bearings would be difficult to use if mounted in the front. Btw Did you mean needle bearings or roller? as I dont think needle are suitable for high load and cant have a taper added to change the preload. Are they?

Do you have a diagram and design use for that spindle?

[Jonathan]

Surely you want angular contact bearings? I've got a spare pair FAG 7206 bearings... ABEC7 grade (so intended for spindles)! If you're interested...

You need to work it out such that the spindle temperature remains in a certain range, so etc.

[Ross77]

Cheers, yeah I do I was just trying to clarify what coredump was suggesting with the needle bearings.

I already have a matched pair of Barden precision bearings 109HDL (45mm int.) for the nose, if I can make the spindle accurattly then most of an ER 32 collet will be in side the bearing anyway. I do need some bearings for the rear but they will need to be floating or have end float, allowing them to slide on the shaft or housing just adds more maching headaches.

I intend to do a practice run first so might be intested in the bearings.what size are they,contact angle , clearance etc.

cheers

[Jonathan]

They are FAG B7206C.T.P4S.UL
The info is all on the FAG site if I recall correctly.

[coredump]

Hi,

You can find the cross view of FP1 spindle here: http://www.usinages.com/manuels-documentations-machines-outils-disponibles-f82/fraiseuse-deckel-fp1-t3738.html page 55 of the documentation (sorry in french).

Still in french, if you look at this thread: http://www.usinages.com/conception-fabrication-f22/fabrication-une-fraiseuse-granite-epoxy-t15107.html the guy build a spindle very close from Deckel spindle (and he got one FP3, no wonder why).
His spindle is made from 42CD4 (no post treatment), BT40 spindle nose (done on an asian lathe), Needle bearing in front, runing on a bearing ring (NKI bearing), double angular bearing (normal SKF) on top.
The trick to have low radial clearance is to use the fact that needle bearing race is thin, so a very tigh adjustement make the race growing, or shrinking. In fact in his case the adjustement is a little bit on the high side. He explain me that Deckel use a wide range of needle, spaced in micron, to found the correct adjustement when building or refitting a spindle (he used to work a lot with Deckel in his professional life). There is also some growable needle bearing (INA), but more exotic.
Btw, all his epoxy granit mill has been done with pure DIY spirit: no big machines (not even his FP3, which he doesn't have at first), no exotic component. I have seen it IRL and it's pretty impressive.

[Ross77]

Hi Coredump
Thanks for the link thats one hell of a machine and looks like a really good site but unfortunatly I cant speak french and google translate dosnt seem work very well !!!

I couldnt see a diagram of his spindle. There was a pdf someone else posted but that used angular contact bearings not needle or did I miss an explanation or drawing?

I also noted that his design was for 5000prm max which would be fine for rollers or needle bearings, I'm looking for 18000rpm so I think that design is out for me.