[philbur]

Picture 1) in this reference gives an understanding of the potential problem:

Ref: GMN Bearings Deep groove Radial ball bearings Technology

Phil

[handlewanker]

My reference as stated was/is the SKF bearing catalogue, perhaps not as recent as your one, probably circa 1990, but neverthe less a valid design for a high speed vertical milling spindle.

If'n I have the inclination to search for it, I will post the relevant page that I photocopied at the time and since my retirement (10 years) still have.

The illustration you posted in the MRC bearing cat on page 11 bears no resemblance to the spindle design I mentioned, and I would venture to say that anyone attempting to make a spindle to that design would be treading in deep water.

I think you, and the illustration in the SKF cat you posted exagerate the "looseness" of the radial bearing, even allowing for the clearance for a lubrication film, and if'n you have a shaft with two radial bearings on it, as I recently dismantled from a cheap junk Chinese wood turning lathe, you will find as I did that it is impossible to measure even with a .0001" dial indicator any deflection of the shaft, at either end, and this is for a very basic shaft layout in the extreme...with no preloading.

BTW, a very basic bearing layout design I made for a lathe as an apprentice has been in operation since 1962, and consists of two tapered roller bearings on a shaft with 19mm bore and spaced at 100mm apart....bearing preload adjusted at the shaft end with lock nuts.

The design was by no means ideal but functional, and if'n there had been a heat problem it would have manifested itself by variation in the preload, but this configuration is very basic indeed and is not the ideal way to have a lathe spindle bearing arrangement, but due to lack of any experience in spindle design it sufficed at the time, and still functions today.....50 years later.

The "silly" statement you refer to was an "if " factor, and that has been used in a design of another machine spindle, but the bearings were in a cartridge form and not preloaded, just spaced apart to give a stiffening effect to the spindle.
Ian.

[Chrisjh]

Guys, Guys,

Please cool down!!!

All have valid technical points and I appreciate the efforts to point out potential pitfalls with my design (which you will note, is completely unproven at this point in time). I for one, am now more aware that my simplistic view of spindle design was lacking knowledge and experience, and I am now in a much better position to come up with something better. I appreciate the references and experience provided by others. It gives me a better platform from which to make design decisions.

My original objective (and still is) was to come up with a low cost design that mere mortals could afford to attempt to make for themselves. That is one of the reasons for a short spindle design, to lower the cost of the material for the 90mm housing. The other reason for the short design was to make the depth of bored holes as small as practical, so that amateurs like me could have a chance of boring accurately without having to resort to exotic solid carbide boring bars with dampeners.

I would like to point out that my present Chinese manufactured Syil X3 Mill has a similar toothed belt driven spindle design using a simple low cost tapered roller bearing at the bottom with preload applied by a thrust nut through a thrust bearing. The top bearing is basically the same as what I proposed to use in the discussion subject.

The Syil X3 mill has given me good results over the last 10 years in terms of accuracy of the finished parts. The spindle has a maximum speed of 3400rpm and is noisy, but gives me, as a hobbyist, satisfactory results. For example, all of the bearing housings for my home made CNC Lathe were done on the X3 Mill. All I had to do to get the bearings fitted into the housings was a light hone using a brake cylinder hone for the smaller bearings and a engine cylinder hone for the large 100 and 110mm bearings for the headstock.

I am still researching and making changes to my design based on the feedback and experience shared by others. For example I am adding a lock nut to the spindle shaft as suggested. I ignorantly thought that friction alone would hold the spindle shaft in place axially.

Regards

Chrisjh

[handlewanker]

Hi Chrisjh, armchair warriors tend to get mounted on their pens with great tenacity and will go to battle armed with pen and paper regardless......I'd like to establish the basic requirements that the average person with average skills and also average equipment would have....possibly only having access to a lathe from a friend, institution etc, as opposed to the set-up many of us have the good fortune to have, and that would establish the end result.

However, there is a basic requirement that cannot be missed and that is a lathe....without a lathe you can't do it....even if'n you have the most detailed drawing in the world and even if'n you never touched a lathe....without one you just can't cut metal.

You can get by without a mill by adapting the lathe to do the milling etc.

For someone who does not have a lathe or any turning experience the other recourse would be to a plumber block type of bearing set-up using two bearing housings bolted to a base and driven by a vee belt...very basic but with clever design very adequate.
Ian.

[philbur]

Chris, In the Syil X3 spindle the top bearing is preloaded against the bottom bearings, by the pair of lock nuts. In your design there was no lock-nut to apply the preload. How you mount/preload the top bearing also depends on what bearing type you pick for the bottom end, your current choice is a mismatch.

If the X3 spindle works for you why try to design something different. Are you trying to make a better spindle, a cheaper spindle, or both. Which of these three alternatives will dictate your choice of bearings and their arrangement.

Phil

I would like to point out that my present Chinese manufactured Syil X3 Mill has a similar toothed belt driven spindle design using a simple low cost tapered roller bearing at the bottom with preload applied by a thrust nut through a thrust bearing. The top bearing is basically the same as what I proposed to use in the discussion subject.

[asuratman]

Chris,
It is best if you design with all feedback in it, let your customer choose the bearings. For different bearing, maybe with slightly change of design due to bearing dimension.

[Chrisjh]

Hi All,

First Philbur, you are correct re the X3 Spindle preload arrangement. It has been around 7 years since I dismantled and serviced my X3 Spindle and my memory was hazy. Thanks for reminding me.

Anyhow I have been busy with redesigning with all your comments considered. The feedback has been invaluable. ”Collective brainstorming” reminds me of the design review process I used to employ when working for a telecommunications company as a Project Manager.

Please find attached a revised design using 2 x 7207BEGAP bearings back to back and Nilos LSTO Steel Labyrinth Seals. Factory determined Preload is applied by compressing the 2 bearings together with the bearing retainer through the steel seals.

I found an indicative price for the individual bearings here 7207BEGAP SKF - Buy online now at WBDS your bearing and drive depot!

A local supplier quoted for each Nilos ring $9.90 AUD + GST. I’m not sure this price is for the LSTO version though.

I would appreciate comments re my proposed use of 2 x Nilos Seals. Is the inner seal really needed? There is very little chance of foreign matter being present in this area. Is it needed to keep the grease packed in the bearings in place? Deleting the inner seal would reduce the cost.

Regards
Chrisjh

[mactec54]

Chrisjh

You only need the front seal, The split ring does this have a thread, to screw on to the shaft, because, it needs to be threaded to put the preload on the centre of the bearings

[Chrisjh]

Hi Mactec54,

Thanks for your observations. I have been prompted to revisit the design again. I have been confused. I now have a better understanding.

I think that I need to mount the bearings “face to face (DF configuration)” to achieve preload by “crushing” the outer races. The text books state that” back to back” (DB configuration)” results in a “stiffer” design; but I think I will have to settle for a compromise. DF mounting arrangement negates the need for a threaded nut on the spindle shaft.

My design intent was not to have a threaded nut on the spindle shaft to minimize complexity and cost; and to maximize ease of manufacture and assembly.

The split clamping ring was added purely as insurance to prevent the spindle shaft from “falling down”. The intention for assembly was to apply a small axial crushing force between the front of the spindle shaft and the rear of the clamping ring before tightening the clamp cap screw.

I think that a DF approach would be usable by most hobbyists as I imagine most would not be placing large loads on spindles.

I will delete the rear seal.

Regards
Chrisjh

[handlewanker]

Hi Chris, I think you will need to thread the split ring 'cos if you don't the spindle will just be sitting in the inner races and under load will migrate up and down, the loading will be the same as for a preloaded bearing...IE it resists movement by applying a preloaded condition, without the split ring being tightened against the inner races the spindle is practiaclly floating in the bearing inner races.

The pre assembly would have the two A/C bearings, cups facing inwards as you have desired, assembled to the spindle and against the spindle shoulder, so allowing a threaded collar to be tightened against the inner races, and being split, have a locking screw to fix it in place.....this will positively retain the spindle in the bearing assembly.

The assembley can then be entered into the housing and the bearing retainer cap screws tightened to appliy bearing preload....."crushing" the outer races together.....the bearing retainer cap needs to be of substantial thickness and material toughness to prevent distortion as you will have at most 4 if not 5 retainer screws to apply the preload.

This is not a satisfactory condition....the ideal is to have the bearing cups facing outwards, so applying the load of the cutter against the lower bearing cup and the preload applied against the upper bearing cup with the inner races "crushed by the spindle lock collar, and the outer races merely retained (but with enough force to create a retaining preload to their mounting) by the retainer cap screws in the housing.

You will add complication by having to have a much thicker and bulkier bearing retainer cap with more screws, than if you just threaded the spindle behind the inner races and applied a locking collar.

However this is just splitting hairs as you probably will never exceed the capability of the bearing design you have decided on with the cutters that the R8 system will hold.

Before the proposed bearing design failed to perform, the R8 collet at it's maximum gripping capacity would have allowed the cutter to draw down (or push back) as it's only friction that retains the cutter in the R8 collet....well beyond the machining capability of the main machine design anyway, unless it was made with massive casting proportions.

If'n the R8 collets for cutter mounting is going to be a problem I would probably look for an Autolock chuck with R8 shank, and this with it's screwed collets is practically imoveable, even if'n it means allowing an extra 50mm or 60mm for the chuck stick out in the machine design.

I take it for granted that the spindle will be of a Nickel Chrome type steel and heat treated, iE hardened and ground to achieve maximum accuracy from runout and longevity of collet mounting, without which the spindle bearing configuration becomes unimportant.

You could just make the spindle from a low grade Nickel Chrome steel, the type known as case hardening tool steel,( E200 by Bohler in Australia), and only just get the end with the collet mounting case hardened to prevent overall distortion, and then only get the collet seating area ground......much cheaper than an overall tool steel and hardening and grinding approach, and if'n the worst comes to the worst, even making the whole spindle from mild steel and getting it case hardened on the end would not be too unsatisfactory either.
Ian.

[mactec54]

Chrisjh

It does not matter which way you mount the bearings, you still need the thread to lock the centre of the bearings together, Stay with back to back, the thread does not take much at all,The bearing supply company you are working with will have a nut that you could most likely use, a fine thread, like 1.0mm pitch will do the job

[RICHARD ZASTROW]

I'd use sealed A/C bearings, back to back, Nilos JVG type metal seal at the cutter end only.

1.) Back to back is stiffer and the recommended method by the bearing manufacturers.

2.) Sealed bearings have the correct amount of grease and will keep the grease in the bearing where it belongs.

3.) Nilos LSTO's are good seals. They are speed limited. That is why we stayed with the JVG type seals for ingress prevention.

4.) We have used a floating sealed deep groove bearing on the end opposite the cutter for years with no problem. On the business end, the spindle nose, that set up has historically had .0001" TIR or less run out. (We also employ spacers or multiple bearing sets).

5.) We always use PRECISION threaded locknuts to pre-load the bearings on the inner races and secure the outer races with a bearing retainer cap. Really not much more expensive or difficult.

"You get what you pay for or less"

Dick Z

[mchabot]

If you look closely an X2/X3 spindles you will see that they are not heat treated (file bites easily), they are made from pre-treated steel like 42cd4.

I have one spare X2 R8 spindle shaft. First I made some plan to use quality A/C bearing (S235), but a guys who build it's own E/C CNC mill including spindle gives me another advice to avoid long distance between R8 front lip and bearings:
Machine the front side of the bearing from 40mm down to 38mm (as there is some runout). Press fit a 38mm needle bearing inner race and use a needle bearing like NKI38/30. Then on the upper side a standard A/C bearing set (close from the needle bearing).
at first I think it's not a good idea because the needle bearing has some clearance, but the idea is to press-fit the needle bearing, so there is some sort of preload.
He has made a ISO30 spindle this way, from 42CD4 steel, untreated. And it's working fine. He got the needle bearing idea from his Deckel mill which use also needle bearing as front bearing (but precision matched ones to have controled preload).

[ihavenofish]

i read all this thread without looking at the OP's pdf. youd have thought his design must have been terrible and useless....

except its not.

ive designed a spindle with an ER chuck in a similar fashion. double row angular in the front, shrink fit to the shaft, and a slip fit radial deep groove bearing at the tail for stability. no lock nuts required.

we did use 3200 series bearings in the design, from nsk, which were preloaded (we use them in the same press fit manner for ball screw support bearings with good success).

tail bearing clearance is a non issue to me. i think a few tenths of movement when the tool is overhanging a few inches isnt the least of your worries with an R8 setup.

that said, i would only use a design like this as a cheap way of using existing straight spindle shafts, like ER chucks.

if youre buying something like the X2 shaft, or one from a G0704, it already has lock nut threads and the correct shoulders to use a more user friendly opposed angular contact/tapered roller arangement.

if youre machining your own shaft from scratch, theres other better and chaper ways of doing things as well.

[philbur]

To achieve correct preload your bottom bearings need to be a matched pair (read expensive) and have a preload nut. the difference between correct and incorrect preload is measured in microns. Your top bearing will still have radial float which makes it a crap match with the expensive bottom pair. Also the top bearing still requires a retaining mechanism and also needs to float axially by a couple of millimeters to allow for temperature expansion/contraction.

Chris to be frank you clearly don't know what you are doing when it comes to precision spindle design, so you either need to spend some time studying or you need to copy an existing arrangement. There are competent designs out there for every possible price point. Nice CAD drawings are not a substitute for sound engineering. Sorry if this sounds a bit tough but I feel somebody needs to say it.

Phil

[ihavenofish]

To achieve correct preload your bottom bearings need to be a matched pair (read expensive) and have a preload nut. the difference between correct and incorrect preload is measured in microns. Your top bearing will still have radial float which makes it a crap match with the expensive bottom pair. Also the top bearing still requires a retaining mechanism and also needs to float axially by a couple of millimeters to allow for temperature expansion/contraction.

Chris to be frank you clearly don't know what you are doing when it comes to precision spindle design, so you either need to spend some time studying or you need to copy an existing arrangement. There are competent designs out there for every possible price point. Nice CAD drawings are not a substitute for sound engineering. Sorry if this sounds a bit tough but I feel somebody needs to say it.

Phil

paired bearings arent needed, thats the whole point here, although they are CHEAPER than he double row setup. you can get 7007 bearings for $20 or less each, which are MORE than adequate for this spindle in precision. LOTS of people are using them on cnczone for the G0704 conversions. ive got a pair in my kx1 spindle too.

but back on topic, the double row bearing has a preset preload. thats done at the factory. nothing more to talk about on that subject as long as you buy the right bearing.

so with this bearing shrink fit on its own, the spindle its "done" in terms of basic function. the top bearing is only for radial support with long tool extensions (as you get in most cutting). the radial clearance of a deep groove bearing is small enough to be meaningless - again if you buy the right one. the clearance on a skf 6005 C2 bearing is 1 to 11 microns, and "normal" is 6-20 microns. so with a tool extension of 6", you might get 0.0005" of give (not play) worst case on inexpensive bearings - which is likely 1/4 of what you have in the ball screw nut, gibs, etc. his design also allows a fairly easy way of preloading the rear bearing with a bolster plate removing what little play there might have been. many high end spindles use deep groove radial bearings at the tail, so they are obviously more than adequate.

basically, his design is ok with the right choice of nose and tail bearing.

[philbur]

I would be interested to see a commercial machine tool spindle that uses shrink fit to achieve bearing preload and bearing retention?

Phil

[philbur]

I would also be interested to se a high end spindle that uses a single deep groove radial bearing at the tail?

Phil

[ihavenofish]

I would be interested to see a commercial machine tool spindle that uses shrink fit to achieve bearing preload and bearing retention?

Phil

shrink fitting doesnt achieve preload. the bearing is preloaded from the factory as ive said twice now.

shrink fitting is viable, we do it on ball screws. most commercial spindles WONT have this because it makes things unservicable, but it WILL most definitely work, and is a good option for an inexpensive high speed setup.

as for deep groove radial tail... kavo, jaeger, colombo, tekno, hsd...

the sieg x2 comes with a deep groove raidial bearing in the tail AND nose.

you forget hes talking about a cheap hobby "make it yourself" spindle. one that should cost less than a single skf p4 bearing you might find in a VMC spindle. its not about being the best ever, its about being cheap and functional.

[philbur]

But the proposed bearings are an unmatch pair how can they be preloaded at the factory.

Phil

shrink fitting doesnt achieve preload. the bearing is preloaded from the factory as ive said twice now.

shrink fitting is viable, we do it on ball screws. most commercial spindles WONT have this because it makes things unservicable, but it WILL most definitely work, and is a good option for an inexpensive high speed setup.

as for deep groove radial tail... kavo, jaeger, colombo, tekno, hsd...

the sieg x2 comes with a deep groove raidial bearing in the tail AND nose.

you forget hes talking about a cheap hobby "make it yourself" spindle. one that should cost less than a single skf p4 bearing you might find in a VMC spindle. its not about being the best ever, its about being cheap and functional.