I have been looking everywhere for a milling spindle to use on a future project. My requirements are:

1/ Support about 2.5 to 3 hp
2/ Have a 30 NMTB taper
3/ Run to 8000rpm
4/ Have class 7 class bearings

Setco have something that fits the bill (Sentry (http://www.setcousa.com/files/pdf/S-0005-1a-Sentry.pdf) ) but it is $2980 dollars. This is just too much to justify.

Would anybody be interested in collaborating in the design and manufacture of a general purpose spindle and share in the development costs. I was thinking of something along the lines of:

1/ Support up to 3hp

2/ Have the option of a R8 or 30 NMTB taper

3/ Support class 7 class bearings or lower class ones for people who want a reduced cost.

4/ Support a powered draw bar (in the future)

5/ Designed for belt drive

Anyone interested?

the problems with standards is that there are so many to choose from ! my friend and I are toying with the idea of making a spindle and we were thinking SK40 taper, definitely with a pneumatic gripper drawbar (anyone know what typical forces for these are?)

could you expand on the bearing reuirements ? what are the specs for class 7 bearings ?

belt drive sounds OK, maybe with two wheels for two speed ranges (with 3000rpm motor we would have 1:1 and 1:2 for max 6000rpm)

Frankly and with all due respect for your creativity, your specs are inadequate to spec out the bearings for such a spindle.

As soon as you talk 8000 rpm, your putting yourself into a tough area (engineering wise) so as to come up with a bearing that will support the loads and the speeds, not to mention lube and preload and fit them.

You can easily transmit 3hp thru a shaft with little concern over the bearings - until you start to apply combined radial, axial or combined radial loads to the bearings. Thus, you start having to pay more cash for speed and more for radial capacilty as size and speed go up.

Small size bearings are competitively priced (IE: 6203's) as everybody makes them like aspirin and there are jillions to choose from. Bigger bearings and/or ABEC 7's are scarcer and not so widely made thus they command and get a higher prices. Classic law of supply and demand.

Re: the cost factor, check out the cost of ABEC 7 bearings regardless of size. Depending on the size and configuration (duplex, triplex, quad, roller,etc) you can easily tie up $1-2K in bearings alone.

For grins, check out the price of a set of 7207CTYDUMP4's (duplex, abec 7, 207 size) for the nose and 6206VVP4 would probably work as a guestimate at the other end of a mill spindle.

Compared these to the $$$ for regular 6207's and 6206's and you'll see why spindles are so pricey.

THen figure what it will cost to machine, heat treat and OD and ID grind the shaft afterwards. Don't forget the housing maching and ID grinding (to tolerances of 0.0001 or better).

Unless you have the machinery (OD, ID grinders, lathes, mills, etc in house already) that is capable of doing it, you'll be talking a small fortune to have it done by a job shop.

Yes, it would be nice to find an ABEC 7 equipped spindle that would do what you want for much less than $2900. Chances are, houweve, that it have already been done if it were financially feasibile/possible. Or one would be on the market from Taiwan or the ROC. In today's price sensitive market, people would be beating a path to your door for one - again if it were possible.

Re bearing specs: get a hold of a bearing catalog from any precision machine tool bearing supplier. Most show the specs for ABEC 7 (same as ISO P4). NSK's catalog #E124 has them on page`A8 thur A10. Spindle and housing machining specs are found in same catalog on pages A15 thru A17. Same catalog shows other stuff you'll need to know/apply if you still intend to make a spindle on your own.

I wish you well in your endeavors....

When you see the numbers you realy should be holding, you'll see why the $2900 price for a precision spindle from stock is really pretty cheap....

anyone know of sources for hobby-level spindles with standard tapers used in cnc machines ?

I found an MK3 spindle at littlemachineshop, but it's a bit small:
http://www.littlemachineshop.com/products/product_view.php?ProductID=1508

ISO/BT40 size or similar would be nice. preferably one that fits a pneutic gripper.

Since it's apparently really hard to make the spindle DIY I'm thinking that fitting the bearings and the holders for the bearings + spindle box can be done DIY.

NC, Thanks for the frank reply. I understand this is far from a trivial exercise. I have being doing a bunch of research and have read most of the bearing manufactures literature. I knew going into this that the bearings cost a fortune.

I have a CNC mill, a Myford super7, a hardinge TFB and a Dunmore toolpost grinder. A long time ago I built a grinding spindle just using a Myford and toolpost grinder. It certainly was an exercise and with a certain amount of luck actually turned out pretty good (around 0.00015” run out, that I could measure) and was capable of grinding at 15,000rpm using a 3/4hp motor. However, it took a long time to make and I wouldn’t fancy my chances at being able to replicate this feat with any certainty of success.

To get the cost down, I am thinking I will have to get a bit creative. I have looked into the availability of existing R8 spindles (minus housing and bearings). I have decided to buy the spindle from a Grizzly G3102 small Bridgeport type mill. This claims to have ABEC7 bearings as standard so hopefully the tolerances on the shaft should be OK. They are asking $160 for it so it’s worth the risk. I don’t like their housing or bearing setup so will design my own. There are some local precision grinding shops I am going to talk to about doing the ID grinding on the bearing seats. For my first attempt I may fit ABEC5 bearings as they will be good enough for my initial requirements. Based on the spindle design I think it should be ok for about 1 1/2hp under typical hobby type milling conditions. The important aspect for me is rpm capabilities of at least of 7000rpm as I plan to use it with small end mills mainly.

Andy, I plan on fitting a power drawbar similar to Power Drawbar (http://cgi.ebay.com/AUSSIE-air-power-drawbar-R8-NT30-CNC-Bridgeport-mill_W0QQitemZ7573827743QQcategoryZ12584QQcmdZViewItem)

I will probably use either R8 tooling or Tormach tooling holders (TTS (http://www.tormach.com/Product_TTS.html)) I use on my
CNC mill.

The spindle is backordered so I probably won’t get it for 5 or 6 weeks. While I waitI will start drawing up some design ideas.

There are several philosophies to consider:

Make a precisiong spindle, use precision bearings, assemble it and grind the ID to get the axis of the spindle to run tru to the axis established by the bearings.

or

Make semi precision spindle, use semi precision bearings, selectively assemble (move them around until you get axis to run true) then "lock it down" and grind the ID.

Make whatever for a spindle, use whatever for bearings, grind after selective fit but don't be surprised if you end up with something that is oversize and doesn't holed tolerance.

Whenever you talk high speed, you want the cutter running true. Otherwise, finish, chatter and size control become concurrent problems. Thus, precision machining and bearings become more critical as speeds mount. Like we tell our racing customers, "Speed costs money - how fast do you want to go and how long do you want to go that fast???"

I bought an import mill and it ran well as is. Since I worked for a bearings supplier, I figured I"d "tune it up" with some surplus ABEC7's that were being tossed.

Result: Disaster

Reason: the housing and spindle were selectively fit (cheap labor enables you to assemble/reassemble sloppy stuff until you get it to work much cheaper than buying expensive bearings and holding tite tolerances when grinding).

Yes, you can buy the spindle as a "core" and then via grinding, plating and selective fitting, you can blueprint the part to do what you want. However, it will cost time and/or money. Do you have either/both to spare????

andy, did you see this: http://www.littlemachineshop.com/products/product_view.php?ProductID=1407

an R8 spindle for the mini mill.

can anyone tell me where I could look for bearing that would be suitable for use with this spindle? I found a seller on ebay, VXB Bearings I think it was, that has a very large amount of auctions for various bearings, so I was looking for a relatively low cost alternative.

Low cost and machine tool grade (ABEC 7 or ISO P4) bearings are an oxymoron.

IF you want to do high speed grinding, ABEC 7 bearings make a huge difference with regard to size control and finish. How so???

A straight drill will drill a precise hole but a bent on will wobble and cut a funky hole.

Generally speaking ABEC7's have HALF the runouts an a fraction of the tolerance of ABEC 1's or 3's (typical "electric motor" quality bearings). They are priced accordingly UNLESS somebody is blowing them out to get their money back because they were stuck with them.

From my experience in the bearing industry, any quality bearing supplier will gladly provide an inspection dwg that "certifies" the fact that you'd be getting what you pay for.

If someone says, "trust me they are ABEC 7's" , don't unless they'll provide certs. See prior post on how to selectively use/fit lower spec bearings and spindle for low net spindle runout.

In short, you'll pay for it in time or labor but high running accuracy costs.....

i am just in the process of building a small benchtop cnc mill. nothing spectacular, it is primarily a school project (taking mech engineering technology right now), but i am using good quality parts such as ballscrews and precision ground shafting.

My requirements for a spindle of such type would be:

NMTB #40 40mm taper
up to 3,000 max RPM
down to 275 min RPM
belt driven by NEMA motor @ 1,750 motor rpm
belt is a (much wider than automotive type) serpentine belt style.
capable of at least 4" Z travel
precision to 0.001 to 0001 better.
Uses ABEC 7's.

Never seen any reasonable cost 2 to 5HP motors, not even 3's in NEMA enclosures. Anyone have good Baldor part #'s?

I wont have a tool changer.

I thought about CAT 45 but holders take up too much room and I already have a whole pile of NMTB 40 taper holders.

would like it to support FANUC 6 series spindle drives & servos as I've got them already

don't really need the power drawbar but the downfeed

I'd love some help in this respective too myself.

private comments can go to gbowne1@yahoo.com if you want.

I am personally looking into making a NMTB #30 spindle for up to about 5000rpm or so(more if I can get the bearings) for my next mill project.

These tool holders use a simiar drawbar and similar power drawbar like what you see on bridgeports right?
I have one of those half built that I would like to complete and would probably work well on this type of setup.
I was origionally looking at a retention nob tool holder setup, but that requires high pressure on the spindle bearings and a high pressure large air cylinder and all.

Jon

Yeah, those require the 5/8" threaded drawbar like you see on bridgeports. I think I remember seeing bridgeport-type mills w/either 30 taper or 40 taper, but I don't like the design of the head.

I've got so many NMTB #40 40mm taper tooling (most of them Weldon brand)
that it is relatively economically sound for me to use the NMTB #40 taper. I thought about using the CAT45 taper tooling which I have only a few of. (Came with some other NMTB tooling I got at an auction) but it's not what I like. Though there is the possiblity I could change the spindle out later, and add an ATC.

I'd love to find a decent sized belt and NEMA frame motor to use. Anyone got any suggestions? V-Belts don't appeal to me, even though I've seen them cheap.

Anyhow, there's always a possibility out there somewhere.

I wish I could find a moog spindle, I have boatloads of the tooling. Got it cheap on ebay. I wanted it because there were a batch of the double angle collets included.

Ok, I abandoned the Grizzly spindle idea, just unknown quality and not ideal for my proposed use.

I did manage to track down a local shop that has the grinding capabilities required to match the tolerances for ABEC7 bearings. They seem open to a bit of a sideline project. The plan is for me to do as much of the work as possible and they will do the final grinding ops. Only after I get a basic design together will I know if it is financially feasible.

So the plans are to draw up a design for small NMTB30 taper spindle. I want to use a single sealed/greased duplex angular contact bearing set at the front and a rollerbearing at the rear. Its not going to be industrial quality for sure, but I hope it will do for a small hobby type mill up to 1.5hp. For my first attempt I was planning on skipping the hardening of the spindle to simplify things. I know this wouldn't fly in a commerical application but I am pretty easy on things and its not like the spindle is going to be used 7 hours a day. I guess I could plate the internal taper to increase the wear resistance.

So next steps are to get some specs together so I can talk to a bunch of bearing manufacturers. If anyone knows of any suppliers that have decent prices on ABEC7 bearings let me know.

On a side note, I stumbled upon some air spindles that look like they would be useful if you need really high rpms (>25,000rpm).

http://www.airturbinetools.com/html/600Series/650XJS.html

SETCO has great spindles. I'd use one right now if i could afford it. Someday maybe. LoL. It'd be nice to have one of their spindle blueprints to make one from, then given that a person could make something quite similar.

Makes me wounder if they could make a belt driven lathe spindle in 2 or 3MT or 5C (or a 3J/3AT version), or have a threaded chuck mounting or a D style.

I'm going to choose LEESON for my motor. They've got a nice web page with details. I have the page link if anyone is interested to the data.

Well, there's a start.

Greg

Mispost Ignore this.

Thats looking good. what taper is that?

Do you have everything correctly drawn and to spec?

The face grooves on the spindle and for the end caps, are those necessary and all, what are they for?


Jon

OK, so here is the first of what I am sure will be many design iterations for my spindle.

http://www.cnczone.com/gallery/data/500/Spindle1_2.JPG

http://www.cnczone.com/gallery/data/500/Spindle1_1.JPG

First, I will point out that the design doesn't show any threads, cap screws, detailed bearings or the belleville washers used to clamp the holder.

I managed to pick up 8 sets of Fafnir Super Precision duplex bearing pairs for a steal from a guy getting out of the bearing business. Specs are 30mm ID, 15 degree pressure angle, light preload, pregreased. Having these has somewhat constrained the design. I originally wanted to fit 35mm bearings at the front and a 30mm bearing at the rear.

The current design has a matched pair of bearings at the front with precision ground spacers to increase the stiffness. The bearing configuration is DB (/\). There is no way I can make to spacers to the required precision so I will have to purchase them. Given the price I paid for the 30mm bearings it may be cheaper to just double them up in DT-DB configuration (//\\).

The only problem with going to a 30mm front bearing is that there isn't as much material on the spindle near the mounting taper. Plus it forces me to have more projection of the spindle from the front bearing than I would like. In the design shown is a rough approximation of a NMTB 30 taper. I may stay with this or go to a smaller custom taper. I still haven't decided yet.

At the rear I have a 25mm bearing. I can think of two possible solutions to account for spindle expansion. Either a deep groove bearing where the outside race is a sliding fit in the housing bore. Alternatively I could use something like a barden floating displacement bearing which has the inner and outer races clamped to the spindle and housing respectively (the bearing allows the inner and outer races to displace laterally).

The 3 step pulley shown is somewhat of a mock-up (no grooves or key way). I will add the detail for this later. At this point I am considering using a poly-v belt.

I would be grateful to hear any comments on my design. I am not an engineer by profession so I have a lot to learn.

Jfettig, the face grooves are labyrinth seals. The bearings I am using are only partially sealed so they need extra shielding. Even if they had full shields I would want some other protection against coolant getting in. Super precision bearings have zero tolerance for contamination.

Everything is drawn to scale, although no final tolerances have been added. I need to select the rear bearing before I can finalize the basic dimensions.

As mentioned above, the taper is approximately NMTB 30. I haven't made my mind up on the final taper yet. Really a 30 taper is a little too big given my bearing size. I may just make a custom taper and machine the holders myself. However, it would be nice to have something standard.

Also forgot to mention, this spindle is designed to be run from a 1000w servo (4500rpm). The gearing will provide 1-1, 1-2 and 2-1 for a top speed of 9000 rpm for use with small cutters. I want to use it as an auxillary high speed spindle for my Tormach and a further down the line on a small precision CNC mill I am designing (a little bigger than an X3).

Seems the section on the spindle shaft itself is awfully thin at your front bearing locations. I don't think you have enough backup mass there to really do justice to your super precision bearings.

I just have ugly visions of stress failure starting at the corner of the bearing seat when I envision the mass of the spindle nose and tooling whipping along at 8k and having cutting forces added to the mix. I suspect just a few chatter events would put fini to it.


Tiger

Thanks, you confirmed exactly what I was thinking. Its a real pity the bearings I got weren't 35 mm but for a quarter of the price I couldn't pass them up. I am going to modify the design to use a smaller retention knob and probably will go with a smaller taper than NTMB 30. I think I could get away with NMTB 30 if I had a manual draw bar. However, I have long term visions of implementing an automatic tool changer using this spindle design.

Your "top" single bearing needs to float BUT you also want it preloaded for noise reduction and smooth operation.

An eletric motor preload spring (IE: bellville washer) from McMaster car will suffice nicely. Shoot for 1%-2% of the radial capacity for preload amount.

If you make the housing and spindle of materials with similar thermal expansion characterisics, the preload will stay relatively constant.

If you buy preloaded bearings, YOU MUST MAKE THE SPACERS EXACTLY THE SAME WIDTH TO THE MICRON so as to keep the preload the same as the OEM generated values. They also must be PERFECTLY SQUARE to keep the preload even and constant.

You have to take speed into consideration when you establish preload. Reason: the rings grow as speed increases and ball preload is affected. Contact the bearing manufacturer with the bearing p/n's and preload info (L, M or H) and ask them if the preload is OK for the anticipated speed. Ditto that for shaft and housing fits for same reasons.

Bearing stiffness (as in lb/in deflection) is not affected by the spacers HOWEVER the wider stance does improve overturing moment resistance.

IF your bearings are DU( ) where ( ) denotes preload, you can make up a quad set mounted // \\ for a STIFF bearing. For high speeds, you're gonna want to stay with L preload.

If you want/need preload adjusted, KAF Mfg in Stamford Ct can do it for you - they do bearing preloading for a number of bearing and machine tool OEM's.

The basic design looks good, and is similar to some of our Hitiachi-Seiki spindles at work. The blue and yellow peices that captivate your drawbar should be hardened for better life. If not they may deform when a tool isn't in all the way when clamped. You don't want to have to hammer it out if(when) that happens (chair) Also make sure you have a way to retain the yellow piece from moving.

Dale

looks good. Q. is that A/C arrangement with spacers better than straight BB? seems like the spacers are a complication - there must be something gained but not sure what. I often see diagrams showing the bearings back to back which addresses thermal expansion and the need for accurate spacers

what about seals? I suppose the top could be a sealed unit, but don't you need a seal at the A/C end?

the A/C endcap shown in purple, I had thought of it being threaded so that I could control the force on the out races - ie via torque. yours has a capbolted into place - other than being made very accurately is the any control on how much force is used in the squeezing the outer races together?

thanks

Thanks for the feedback dpuch, I will be hardening the gripper assembly for sure.

NC, I wasn't aware of the requirement to preload the rear bearing, however, it makes sense. I will add provision for a spring washer. Like I said, I understand making spacers is out of my capabilities due to the reasons you state. My bearings are DUL so I think I may just double up the bearings at the front and remove the spacers. I think spacers will cost more than my bearings.

Mcgyver, a lot of spindles I have looked at use spacers to increase the overturning moment (I wrongly quoted stiffness before as pointed out by NC). I know they use spacers on the small Setco spindles for two bearing arrangements. For four bearings I don't think they do.

On high speed spindles they don't normally use contact seals and rely entirely on labyrinth seals, sometimes in conjunction with compressed air purge. From what I have read the labyrinth seals work pretty well by a combination of convoluted path for the contaminant to get in and pressure drops. You can't really see the details, but I will post a close up when I have finished dimensioning them.

Regarding the front bearing cover, duplex pairs of precision bearings are designed to be clamped firmly together (they give torque specs for bearing cover bolts). The inner and outer races are ground in such a way to apply the correct amount of preload once clamped together. There is no provision for adjustment typically. To change the preload you get a different set of bearings (or have them ground).

Last night, I made a decision to go with my own taper. A 30 taper was just too big and was compromising the design. So I will use custom holders which are 1.125" reference diameter (30 tapers are 1.25") with the same taper (3.5" in 12"). I have modelled a holder and adjusted the spindle, retention knob and gripper to match. I am much happier with the cross section of the spindle now. Making the holders will be a good test of my Hardinge CNC conversion when it is complete. I will post an updated design in a few days once I have added some more detail.

So here is the second iteration of my design

http://www.cnczone.com/gallery/data/500/Spindle2_1.JPG

http://www.cnczone.com/gallery/data/500/Spindle2_2.JPG

http://www.cnczone.com/gallery/data/500/medium/Spindle2_3.JPG

As I mentioned above, I decided on a custom sized taper. This has allowed me to beef up the spindle and reduce the overhang. I have also decided to double up the front bearings, as spacers would cost more than I paid for the bearings. The housing has been increased in diameter from 3" to just under 3.5". At the rear I have added two belleville washers to give some preload to the rear bearing. Once I determine the rear bearing I will calculate the preload required etc.

The next step is to finalize the calculations for the pulley, belleville washers for the draw bar and rear bearing specs.

Looks like a winner :)

Much better proportions on the taper cross section, and a nice mass distribution through the three front zones.


Tiger

Only comments would be to watch fits and tolerance stacks....

especially on the top bearing with respect to concentricity with bottom so that it slides properly and does not create a moment in the lower bearing.....

With that much bearing stiffness in the lower bearing, concentricity is going to be VERY critical to prevent moment loading....

You could remove the balls from the two inner bearings (remove cage and slide balls to one side) and voila, 2 inner spacers and two outer spacers.

Carel

NOT really, Carel.

The inner bearing is ground with a load applied to obtain an offset with the matching outer. THis takes into consideration the raceway offset, the bearing flex and other tolearance stacks.

Making spacers is not that big of a deal IF you have access to a blanchard or decent surface grinder. Machine the rings, preferrably harden them and then grind them as a matched assembly (inner and outer set). As long as they are same width (inner and outer for that set), the preload is contained in the bearings.

Tubing or billet will work fine as would the inner and outer rings for a set of regular deep groove ball bearings with the balls removed. I would NOT tear apart some M/T grade A/C's to make spacers. If you do use a generic bearing to make spacers, you'll still need to surface grind the rings to make them parallel and equal width.

The Quad set of bearings will give TREMENDOUS capacity (radial and axial). Theoretical failure potential is higher - more moving parts = higher probability of failure but OH the capacity....

BTW, to take apart an A/C bearings, the inner ring will literally fall out as long as you orient it properly and heat and support the outer ring ONLY. A propane torch works fine for heating it....

NcCams,

We keep crossing swords over different objects. I don't like the stacked bearing design, because if you view an axle from a flexible point of view, you will see that the outer bearings of the stack take the load. The inners will float. The OP stated in post #18 that he had no way of getting spacers. If I look in my catalog i see that spindle bearings can be used in X and O configurations. That's a claim for axial finish. Your story applies also to the 4-bearing design as I only took out the balls of the middle bearings. Taking a bearing apart requires no violence. In the past I needed non available bearings. Assembled them out of two. Won the championship.

Carel

Sometimes knowing something can't be done, Keeps you from doing it! ;)

Prior to my prior post, I looked to see what kind of loads might be involved - I don't recall seeing any.

I also recall that the original design had a spacer included. It wasn't until later that the issue of not wanting to build spacers but to use bearings to fill the hole became an option. I wasn't following the thread on a word by word basis, I did chime in when I felt my technical experience would be of benefit.

Quad bearings mounted in tandem provide additional axial and radial capacity. Hence, it is a quite viable way to add capacity to a spindle.

The reason for suggesting the use of regular bearings for spacers was simple - I hate to see someone tear up a set of ABEC 7's just to make up some spacers. No more, no less.

Rather than "pop out the cage" as I believe was suggested, I pointed out an industry accepted way to disassemble the bearing if the user wanted to make spacers.

I have no qualms with folks being "creative" with regard to their use of bearings. However, since this is a DIY site with a lot of folks who merely read and do, I feel it is more appropriate to provide the technically accepted method for doing something. This provides the highest probability of success.

People can, will and do ignore well intended advice - usually at their own peril. And, yes, I too have taken short cuts with regard to bearings. Some have worked and others were unmitigated disasters (as outlined in a different spindle thread re: ABEC 7's).

To clarify, yes you can disassemble bearings to make spacers - I'd merely used generic ones instead of ABEC 7's - just my style and attitude towards the use of expensive bearings versus generic ones to serve as space fillers.

The user can also install quads for a very robust cutting end of the spindle - the user's call.

I don't see the exchange we've had a as a crossing of swords but rather merely a cross polination of ideas that can create something synergistically better than the ideas taken individually.

Well stated!

My 2cents is that if you don't have the money or the know-how to do to industry standards, don't let that stop you. Some of the best inventions were thrown together just to see what would happen. Some without regard for longevity some with.

And as I am sure you can put something together under perfect conditions and have it fail miserably, you can also put something together under the worst conditions and have it exceed your wildest dreams.

It’s been my experience that if you "overbuild something" you better do it right (as NC Cams points out) or you will have some severe headaches. Or you build it a little looser and make it work as best you can and be highly rewarded.

NC Cam (correct me if i am wrong) builds Race car Camshafts and used to be in the bearing industry, so he speaks from experience, Take head of his advice especially if your using all the high end stuff.

On the other hand, a friend of mind asked if I knew why I opened my own business? I said yes "I wanted to make more money and be my own boss" He said "NO". He had me hooked so I said, "Well then why did I?" He said, "Because you didn't know it was impossible!"

While obviously not true, you get the point! I would have never attempted it knowing the pitfalls ahead of time. Although as I become more successful I will regret it less and less.

Here are some interesting points:

1. a bearing has a "rated" life span.
Deviating from operating parameters shorts this time period, usually depending on how for you deviated.

2. A looser bearing is rated for a specific amount of run out.
If you don't get the class "7" bearings in Perfect they will Most likely blow up.
If you don't get the looser class bearing in perfect they won't last as long but they will work and you may even get them to run truer than spec (this is what has happened to the spindles I have made, sure the bearing life might not get the 10k hour listed but they still run fine for me after 3years of intermittent use)

3. If you buy the expensive stuff and don't follow the bearing wizards advice don't expect to get you moneys worth! But On the other hand if you buy cheaper bearings for you experiment and you suck wind, well you’re only out a little time and money.

Just my 2cents.

Generally speaking, mechanical devices follow a rule I call "the perverse nature of inanimate objects". That is, stuff that should work often doesn't and for some reason stuff that shouldn't, does - only God knows why. If you know what IS needed to make it work, it is often easier to understand why it doesn't or how to get by if you don't have what you need.

Engineers are theoretically taught how to solve problems - some are better than others and some are more creative than others. We tend to follow a logical path dictated by rhyme, reason and technical principles. Thus, when folks ask for information, I try to provide the best info that will provide for the most reasonable degree of success.

The points made by miljnr are absolutely correct (with some teeny tiny opinion differences regarding semantics but correct enough even for my overly sensitive tastes). You simply CAN'T expect to have good results if you make bad compromises (2 wrongs don't make a right kind of deal).

The lowly Chevy small block engine that was designed for 190hp in 1955 is, in today's trim, capable of nearly 800 in full NASCAR trim. How? Surely NOT "abec 1 parts and lots of luck" but truly "abec 9's or better parts" plus LOTS AND LOTS of TLC and research.

I learned bearings as it was a job. I did the cam thing because I wanted to learn how to do it. The ego part kicked in when I thought I could offer a better service that wasn't being offered by the name cam companies. wanted to sell the engineering service but couldn't do it without a grinder to make them. Bought a grinder and learned to use it to sell engineering services. Left handed logic but it's how I got into the busines.

It takes a LOT of work, thought and knowledge to make simple parts do extravagant things - you create luck when preparation coincides with an opportunity.

To the budding DIY'er, the worst thing the veterans could do would be to give them the impression that you simply "bolt this stuff together" and "if it was machined wrong, Loctite will fix it".

I feel the veterans should teach them the GOOD and proper technical skill (not wives tales or tricks)- they'll pick up the bad ones (and learn the wives tales) on their own in due time :cheers: ....

There are several philosophies to consider:

Make a precisiong spindle, use precision bearings, assemble it and grind the ID to get the axis of the spindle to run tru to the axis established by the bearings.

or

Make semi precision spindle, use semi precision bearings, selectively assemble (move them around until you get axis to run true) then "lock it down" and grind the ID.

Make whatever for a spindle, use whatever for bearings, grind after selective fit but don't be surprised if you end up with something that is oversize and doesn't holed tolerance.

Whenever you talk high speed, you want the cutter running true. Otherwise, finish, chatter and size control become concurrent problems. Thus, precision machining and bearings become more critical as speeds mount. Like we tell our racing customers, "Speed costs money - how fast do you want to go and how long do you want to go that fast???"

I bought an import mill and it ran well as is. Since I worked for a bearings supplier, I figured I"d "tune it up" with some surplus ABEC7's that were being tossed.

Result: Disaster

Reason: the housing and spindle were selectively fit (cheap labor enables you to assemble/reassemble sloppy stuff until you get it to work much cheaper than buying expensive bearings and holding tite tolerances when grinding).

Yes, you can buy the spindle as a "core" and then via grinding, plating and selective fitting, you can blueprint the part to do what you want. However, it will cost time and/or money. Do you have either/both to spare????

These are probably too large but I got them "cheap".
Was thinking of using them for a lathe but got something wrong, the bearings in the lathe was imperial and the bearings I got was metric, ended up buying a used lathe instead of rebuilding the old one.

You seem to know what you´re talking about, is this too cheap?

http://www.cnczone.com/classifieds/showproduct.php?product=844&sort=2&cat=6&page=13

Take the part number and price it by asking for price and delivery for that same part number from any bearing supplier....

The seller-to- user pricing structure usually does NOT accurately reflect the true pricing of a part.

Depending on the situation and one's "connections", the prices are either quite reasonable (a deal) or totally irrational (rip-off)....

I don't have a clue anymore about bearings prices - been out of the sales end of the business too long.

Oh well, guess I´ll just have to wait and see if anyone bites! :rolleyes:

Thanks anyway.

I was contacted by a former employee of one of the big spindle manufacturers who has offered to help me out on the project. He has pointed me in the direction of a couple of useful links regarding toolchangers and power drawbars

http://www.transconn.com

http://www.ame.com/line.cfm?id=spin-ott

I am kicking around the idea of offering a kit for the spindle with the more complicated components complete e.g. complete bearing set with matched ground arbor and housing. I don't really want to get into spindle manufacture (better to go to the big guys for ready built units). However, I think a kit may make it a more achievable project for serious home builders. Anyway, I want to get a prototype up and running before I think about this too much.

Be VERY carefull your anticipation of the skill level of your potential/target customer base!!!!

It is one thing to offer an assembled product but another to offer a spindle kit and especially to expect a DIY'er who may be clueless regarding the assembly of a precisions spindle to ultimately assemble it properly let alone accurately.

I thought I knew a lot about aircraft bearing assembly practices (I did). Anybody who's ever assembled a race car quick change rear axle and have it live and NOT whine, does. Anyway, I thought I was pretty sharp until I went to Japan and was taught how to assemble a M/T spindle by the guys who do if for a living. What a practiced ART and SCIENCE.

The term "press" as in press fitting a beraings is NOT in their vocabulary. They were impressed with my skill level but it STILL WASN"T HIGH ENOUGH!!! M/T spindles literally FALL toghther WHEN you have the fit & alignment tolerances met and use thermal fit methods. If you have to force anything, you did it WRONG.

Based on the "but this is DIY and it don't have to be that accurate for our needs" comments I"ve gotten any number of times in rely to the procedures I"ve presented on this and other DIY forums, I'd encourage doing the prototype or an assembly. BUT, I'd think LONG and HARD about marketing a DIY product, especiall a M/T grade spindle.

Did that sort of thing (created kit parts to mix and match) in a former career with hot rod parts for a famous aftermarket hot rod part supplier. Found that even the so-called experts assembled the stuff wrong and blew it up shortly thereafter.

Sadly, this was In spite of the inclusion of explicit instructions that were successfully tested by our secretaries (who were generally clueless regarding assembling said parts). Imagine how some expert felt when he learned that he couldn't/didn't do it right and a "clueless secretary" could assemble it properly the very first time.

This loss of faith in the average DIY'er is the main reason why my cam business is "by professional referral only", why we DON'T/WON"t SELL RETAIL and more importantly why we DON"t have a sign on the door.....

Just my opinion which you are free to ignore.....

I agree with that assesment. My company sells front ends for motorycycles to the manufacturers. And you wouldn't believe some of the calls We get! And these are guys that put together 70-80k motorcycles for a living!

We supply instructions! I guess most people don't read, or think they don't have too. So don't think instructions will stop the phone calls.

That being said.

I am still doing what I do, haven't let the mentaly challenged stop me. \

So don't let our real world negatives stop you, just remember them as your selling and plan ahead!

Some time ago I tough of building a spindle using a collet extender, some bearings and retainers, here are the drawings maybe they give you some ideas :)

http://peu.net/mods/spindle1
http://peu.net/mods/spindle2
http://peu.net/mods/spindle3
http://peu.net/mods/spindle4

Hope it helps


Pablo

I've been reading this thread as kind of an educational excursion. I think I understand a lot of what has been said, but perhaps one of you could quantify the differances between a very good spindle including one with the best bearings, and a home-made spindle with something less in the way of bearings.
For simplicity assume the same machine and solid mounting, and let's eliminate the high speed requirement.
From the standpoint of operating the mill how will I know a good spindle from a bad one.
Maybe along those same lines someone could explain the differances between a Taiwan spindle and a Bridgeport, both in construction and performance.

Many Thanks,
Jerry

From my experience, the biggest improvement you will see is in run-out and reduced tool deflection under cutting loads. The end result is that high quality spindles typically result in better surface finishes and improved accuracy. Hardinge super precision lathes are a good example. They have very precise bearings and the turned finish you can get from them is almost like a mirror.

However, to complicate matters, higher quality spindles also typically have better bearing configurations. This results in more rigidity than cheap Asian import type setups.

Here is the 3rd iteration

http://www.cnczone.com/gallery/data/500/medium/Spindle3_2.JPG

http://www.cnczone.com/gallery/data/500/medium/Spindle3_1.JPG

http://www.cnczone.com/gallery/data/500/medium/Spindle3_3.JPG

The headstock is from the small mill design I am working on. There will be a precision spacer to avoid clamping the spindle housing too much and crushing the bearings.

I was looking at the previous design and thinking if there was anyway to simplify it when it struck me why not use a TTS style holder setup. This eliminates the need for complicated grippers and the associated problems. Also construction is easier because internal hardening of the arbor can be avoided. The other nice thing about this setup is that if I skip any kind of drive dogs, no spindle orientation is required on the toolchanger design. I was planning on using the TTS holders as is but they are just a bit too big for this spindle. So I will make some slightly smaller holders with the necessary retention grooves for the toolchanger.

Now I am considering making my carousel toolchanger design adaptable to my Tormach PCNC1100. This will require some slight changes to the holders (shrink a ring on for the V grove and chamfer the end of the shank to allow easy entry into the collet). The only really tricky bit will be coming up with an automatic drawbar. This would have to be the type that screws/unscrews the drawbar. Anyway something to think about.

As you can see, I decided to go to just 2 pulleys as the overhang was too much given the size of the poly-v belt required. Hutchinson has a nice online calculator for belt size

http://www.hutchinsontransmission.com/transmission/index3.html?langue=EN

I need to source the rear bearing and start work on the detailed drawings. I suspect there will be a lot of work as the devils in the details. All the dimensions require firming up and required tolerances need working out. I probably won't get this done for a while as I am busy setting up the Tormach mill and working on the lathe conversion.

What do you think your rough cost of materials will be?

As an aside; I was intending to build a spindle for a new milling head but became so intimidated by reading this thread that I opted for buying a spindle on Ebay.

Jerry

As an aside; I was intending to build a spindle for a new milling head but became so intimidated by reading this thread that I opted for buying a spindle on Ebay.

Congrats on your new spindle - but cutting your own is not that hard or expensive. I documented my work on this forum when I did mine. You can see I used common and cheap tools to do the work with the help of the good folk from the Zone. I ended up spending about $150 for a spindle that has virtually no runout and can turn at 30K RPM all day. It took a little over a month to do including time waiting for parts shipment and learning curve. It was my first lathe job so experience is nice but not required. The spindle just turned a year old last month with no troubles to report.

I'd set reasonable goals, define the requirements and freeze the project. Don't try to add more stuff or make it the be all, end all of spindles. You'll be surprised how much better even a modest design is over what you can buy ready-made.

JR

I would guess less than $600. However, this is only possible because I picked up a batch of ABEC 7 bearings really cheap. If you had to buy the bearings full price I would estimate around $1100. If I manage to pull it off, you would have to pay over $5000 for a spindle approximately the same size/number of bearings with a power drawbar. It may not be as good but for around a 10th of the cost I'd be very happy. Plus this is more of a learning/fun exercise than anything else. I have already learned a lot and I am sure will learn a lot more before its done.

Congrats on your new spindle - but cutting your own is not that hard or expensive. I documented my work on this forum when I did mine. You can see I used common and cheap tools to do the work with the help of the good folk from the Zone. I ended up spending about $150 for a spindle that has virtually no runout and can turn at 30K RPM all day. It took a little over a month to do including time waiting for parts shipment and learning curve. It was my first lathe job so experience is nice but not required. The spindle just turned a year old last month with no troubles to report.

I'd set reasonable goals, define the requirements and freeze the project. Don't try to add more stuff or make it the be all, end all of spindles. You'll be surprised how much better even a modest design is over what you can buy ready-made.

JR

Where can I find that documentation?
Thanks,
Jerry

I would guess less than $600. However, this is only possible because I picked up a batch of ABEC 7 bearings really cheap. If you had to buy the bearings full price I would estimate around $1100. If I manage to pull it off, you would have to pay over $5000 for a spindle approximately the same size/number of bearings with a power drawbar. It may not be as good but for around a 10th of the cost I'd be very happy. Plus this is more of a learning/fun exercise than anything else. I have already learned a lot and I am sure will learn a lot more before its done.

Yes, that was the conclusion I was comming to, the bearings would cost as much as the rest of my little machine.
I will build at least one spindle, maybe more, just to try using the info I've gotten from this thread.
Thanks,
Jerry

If I'm not too far off topic let's talk about home machined tool holders.
The spindle I bought has an integral kwik switch end.
Since I can't afford $70 to $120 a holder I'll have to make my own.
So I guess what is obvious is that I need to turn and mill the holders and then make a lathe boss to hold them while I bore and perhaps grind the female ends. It needs to be very repeatable and with as little runout as possible.
I've been able in the past to make a C-5 collet chuck which holds work repeatably with less than .0005" even with cheap collets.
Another way to go would be to bore first and then machine the tapers with the work piece on a mandrel.
I'm open to all help and ideas, so please jump in.
Tanks,
Jerry

Attention spindle building shoppers, please take a look at these bearings now selling on eBay. I bought similar ones for my spindle. Barden are some of the best bearings you can buy and these are just perfect for an ER-25 based spindle:

http://cgi.ebay.com/Barden-precision-machine-tool-bearings-105H-No-Reserve_W0QQitemZ7601017183QQcategoryZ25284QQrdZ1QQcmdZViewItem

http://cgi.ebay.com/Barden-Bearing-ABEC-7-P-N-105T6_W0QQitemZ7600478815QQcategoryZ25284QQrdZ1QQcmdZViewItem

http://cgi.ebay.com/Barden-precision-machine-tool-bearings-203H-No-reserve_W0QQitemZ7601027348QQcategoryZ25284QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

Good luck.

Jerry, for your reference, here's full thread: http://www.cnczone.com/forums/showthread.php?t=7004

JR

Im curious why you choose to make your own taper if you going with Tormach Tooling, Why dont you just go with a R8 spindle like the one for the mini mill or something?

I know that kind of foils the making your own spindle idea, but it might be a better idea.

Jon

As I said above, I can't use the standard TTS holders, a R8 collet or 3MT collet is too big for my 30mm ID bearings.

I looked into using an existing arbor but given the tolerances involved I am 95% sure it would end in failure. If you want a precision spindle you've got to make it specifically for the bearings you have chosen. I wish it was otherwise but thats just the way it is.

this is interesting. the power drawbar allows a toolchanger to change the tools?

I just dont see the reason to make a whole spindle, collet and all the holders just because you have a few bearings. The only problem with tts is that they dont have the groove for grabbing the tools.

Another problem with this style tooling is that you need to at least taper the end so it can self center if its not exactly perfect.

The springs on the drawbar will allow a air cylinder or something like that to push the drawbar down and pop the collet or tool free.

Also, the retention mechanism for retention nob style tool holder really isnt too difficult. I built one of those and set it up for my R8 spindle. I didnt have an air cylinder but used a lever. I wish I would have taken some pictures of it before I left for college again.

The whole concept works much like an air coupler, uses balls and tapered walls and whatnot to hold it.


Jon

I have been looking everywhere for a milling spindle to use on a future project. My requirements are:

1/ Support about 2.5 to 3 hp
2/ Have a 30 NMTB taper
3/ Run to 8000rpm
4/ Have class 7 class bearings

Setco have something that fits the bill (Sentry (http://www.setcousa.com/files/pdf/S-0005-1a-Sentry.pdf) ) but it is $2980 dollars. This is just too much to justify.


I think that about wraps it up. Nothing motivates a diy'er as well as visions of economy and the old "who says it has to cost that much?" ;)

Tiger

Jerry, for your reference, here's full thread: http://www.cnczone.com/forums/showthread.php?t=7004

JR

Thanks for the thread. It's big but I'll read it all.
Jerry

I just don't see the reason to make a whole spindle, collet and all the holders just because you have a few bearings

Well those few bearings would cost around $900 new and I paid $200 so that is one reason. My spindle contact put me off using grippers as he says commercial spindle producers have had a lot of problems with them, especially the ball bearing type. They nearly all use the finger type grippers now. These are hard to make and cost around $550 for ISO25 size to buy from Berg. Anyway making a few holders isn't a big deal once you are tooled up for the job. ISO 25 ER Collet chuck holders are around $150 each. If I can make my own that will save a lot. The single setscrew type holders won't cut it at 10,000 due to imbalance. I was advised that even the Tormach TTS collet holders would need balancing.

At the end of the day, this is as much an exercise to see what is possible as anything else. I am doing it for fun so I don't need to justify it to anyone but myself. This is a hobby for me so If I learn something along the way, that is good enough for me.

this is interesting. the power drawbar allows a toolchanger to change the tools?

Yes that is the idea. The model doesn't show the belleville washers used to create tension on the collet. I will use a pneumatic actuator similar to the following to press on the drawbar to release the collet and holder.

http://www.automation4less.com/store/proddetail.asp?prod=NCDQ8A250%2D050

he says commercial spindle producers have had a lot of problems with them, especially the ball bearing type. They nearly all use the finger type grippers now.

Every machine I have worked with in the past 10 years has used the ball style grippers and they have been around much longer than that. Now they might not be the best method to use (not an engineer so would hazard a guess) but they certainly work just fine. The main negatives for this style gripper that I see is that the pull stud better be made of some sturdy stuff because with only tangent contact on the balls you will develop dents fairly quickly with cheap Non-hardened material.

Like you say I am sure they work once setup correctly but I think he was referring to that fact they are a bit tricky to design to work 100% reliably and not wear out prematurely.

pstockley,
You have done a lot of nice work here!
W. Smith

If I didn't want to make an automatic tool changer I would definitely consider the ER collet route. At the moment I don't really have much at risk financially so I will go this route as a learning exercise. The TTS style setup seems to work OK so I am not worried about that.

If the design works out OK I could easily adapt it for bigger bearings and maybe a R8 or a 30 taper once I have some more experience. If someone wanted just a high-speed spindle with manual draw bar even this current design could easily be changed to fit ER25 collets.

I just got a workholding catalog from Schunk, and all of their devices use ball retention. Not quite the same thing, since they are holding chucks, but it's a data point.

I will use a pneumatic actuator similar to the following to press on the drawbar to release the collet and holder.

http://www.automation4less.com/store/proddetail.asp?prod=NCDQ8A250%2D050


You might want to check out Surplus Center for those.

Hi Group,

Does anyone have experience with the Littlemachineshop MT3 spindle ?

I'm considering buying their spindle box assembly and taking out all the gears and putting a timing belt on top.

the spindle box is at
http://www.littlemachineshop.com/products/product_view.php?ProductID=1906
and is only $85

specifically, if anyone has measured runout on this spindle I'd be very interested in hearing about the results.

Has there been any updates? I have just re-read this post from start to finish and all is looking REALLY PROMISING !!

I am tied up on my Hardinge lathe conversion at the moment. Generally I go into the workshop a lot less during summer so I don't expect to finish this until next year.

Some things have progressed regarding my spindle design. Firstly, I picked up some 40mm ID Fafnir super precision bearings off eBay for $54. So I plan on going back to a BT30 taper design. However, I first plan on building a ER25 spindle using my 30mm bearings. This will be easier to build especially since I already have an ER25 closer nut. I plan on using this smaller spindle as an auxiliary unit on my Tormach. The aim is to learn from this design and then process to the more complicated automatic toolchange version.

andy55...stick with the R8 spindle from the same company.....yes you can strip out all the speed change gears...

does anyone have any idea how the AIR Spindles work ? why are they (if infact they are) self centering friction free????

andy55...stick with the R8 spindle from the same company.....yes you can strip out all the speed change gears...


Hi, do you have any info/experience on the R8 compared to the MT3 ?
I am in Europe and seems that MT3 tooling has much better availability.

In regards to pstockley's spindle,
have any decisions been made as to the top bearing? As far as bearings go, I'm using quad NSK angular bearings, but haven't found any good info on what to use for the bearing nearest the pulleys. I remember seeing in an NSK brochure the use of cylindrical bearings, but I've also seen ball bearings somewhere else. What things are important in the choice of this bearing? The angular bearings I have are ABEC 7, so does that mean they would require an ABEC 7 radial bearing? I would also assume that I would want the radial bearing to be able to acheive the same speeds as the angular ones, if not greater. I looked through this thread, and this seems to be the one part not covered.

Thanks,
Brooks

Top bearing to be used is a function of who designed and supplied it.

If NSK did the design, they'd probably spec out a PA( ) where ( ) would involve a non-published semi standard version of a ABEC 3, 5 or 7 to make it suitable with the lower ultra stiff QUAD. How they did this was a trade secret of the machine tool dept and they didn't explain it too much.

To use off the shelf, get ABEC 7 and hold end to end mounting surface concentricity as close as possible to dead nuts 0.0000 TIR.

The top will have to "float" a bit as the bottom will tend to establish axial and radial postioning under thermal growth. Probably the easiest way to do this would be an ABEC 7 (ISO P4) cylindrical - pricey but high capacity and accuracy.

An ABEC 7 ball would work but you'd have to set it up to slip in the housing and light thermal shrink on the shaft - probably w/C3 clearance. You'll get better runout with open bearing over sealed or shielded. I'd run external seal because bearing seals are not "sealed" - they have vents to relieve pressure, hence they can let liquid stuff in when they cool.

Hope this answers the question...

most guys are saying that they are using ABEC 7 bearings but as this is the class of bearing can anyone give part Numbers??

NSK 7205CTDULPA7
And thanks to NC Cams, I know what they all mean-
05- bore(25mm)
C- 15 degree contact angle
T- Retainer
DU- Duplex (2 matched bearings)
L- Light preload
PA7- Special machine tool version of Abec-7

I'll be using 4 of them in my spindle design, which will be very similar to pstockley's. Mine will have an ER20 collet as opposed to his. Aside from that, I don't think there will be many other differences. As to what orientation for the quad bearings, I still don't know enough to write down which I will do. Research, research, research...FUN!

-Brooks

Well it's official.. I have just been to the boys in the machine shop and they are only to happy to machine what ever it is I want :banana: there is only one draw back....they all drink scotch!!! so it looks like 2x 1125ml bottles is what I need to buy

Oh did I say one draw back ... well its more like two they need specs and with spec come drawing I'm guessing

:withstupi unfortunateley unlike the reference [left ] Rather than being with stupid -I'm stupid, when it comes to drawings etc... so has any one done anything that allows for tool change?

I did like the graphics post early on in the tread but I have not seen anything with TOL.s and diamentions are there any out there....

ALso has anyone given any thought to extending the spindle up several inches past what would normally be considered the top, and fixing the motor so that when the z-axis moves up and down the stepper/servo does not need to lift the motor weight.
The spindle could have splines so the pulley was free to ride up and down (much like a drill press)

please all flames welcome

Hi.. First, I'm sorry about my english.

I'm attaching some drawings and one assembly SAT file.

I took some ideas from the first draws of this tread and I searched some.

To make this POWER DRAWBAR, I'm trying to use the cat 40 tool holder, I read that it is the most comun and cheaper tool holder.

I want to make an Autmatic Tool Changer, spindle speed of 18.000 rpm for a (strong & fast) Wood Router.

For what I read, the easy way to clamp the tool holder is using a Drawbar preloaded. I'm using a Coil Spring "Extra Heavy Load (green one on the draw )" It has 2167 Newtons (221 kgf - 487 lbs), But it isn't enough power clanmp, no even closer. But Maybe It is enough for wood.

Also I'm using 2 angular Ball berings (couple) - precission ABEC 7.


I found a Drawbar, Model: BERG SEK 2-48 , from (http://www.transconn.com/ ) It has a Tool Clamping Force: 18,000 N (4030 lbs), Spindle Speed: 20,000 RPM but it isn't for cat 40, It is for a Tool Size: HSK A 63.

I read that Cat 40 / 50 + , They aren't good for fast spindle speeds 8k and up
, so you have to use HSK tool holder's model.
Or you can use the same Drawbar for cat 40, but using a new model of tool holder, which is call BIG-PLUS (http://www.bigkaiser.com/).

Well, I'm really an amateur, bigginer :) about this, so I will appreciate very much some ideas, before I can hurt my self.

V flange Female Drawbar Dimentions aren't ok. I can't find the real dimentions.
Some help with the V flange dimentions

Re: bearing sizes in ABEC 7.

Sure can.

What is the generic size you chose for the application???

What type, ball or roller???

No guarantees about capacity or viability - that's the responsibility of the spindle designer to figure out proper size based upon load/speed requirements.

For what I read, the easy way to clamp the tool holder is using a Drawbar preloaded. I'm using a Coil Spring "Extra Heavy Load (green one on the draw )" It has 2167 Newtons (221 kgf - 487 lbs), But it isn't enough power clanmp, no even closer. But Maybe It is enough for wood.

I hear stuff like this allot!

Power used is a function of programing and fixturing, not of what material your cutting! If you have an easier to cut material (like wood) then you tend to feed faster so the load on the machines CAN be the same as cutting steel. Atlhough the load may be smoother with less hammering from the cutter the side loads are the same.

Personaly I run aluminum harder than steel just because my tooling is more geared for aluminum. And I don't do enough steel products to warrent buying the bling type steel tooling. And thus the loads cant be as high do to cutter strength.

So don't make the mistake of saying it will work for wood becuase its weaker. It would be better to say it will work because you are nursing/babying the machine to make it work.

The angular contact ball bearings Abec7, from www.McMaster.com

For Shaft Dia: 35mm
OD: 62 mm
Dynamic load: 3,507 lbs
Speed:22,000 RPM

I think you are right. I`m going for more power clamping

Does anyone know where one could procure some abec radial or cylindrical bearings, without "eBay" being in the web address? I'd love to stick with NSK, but I really don't have anything holding me to it, unless using 2 bearings by different makers is a no-no. I would also like to know what 7205CTDULPA7's cost new.

Thanks,
Brooks

Hi. Type your zip code here:

http://www.npa.nsk.com/public/enu/info_1049.asp

for a local NSK distributor. If that fails, these guys might be able to help you: http://bearingsdirect.com/products/

For what it's worth, I bought 4 new, matched pairs (8 total), ABEC-7, angular contact, ceramic ball bearings on eBay for less than a single one would go for on a typical distributor. Yes, you'll need to wait and eBay it's certainly not a reliable supplier if you have an emergency. But if you can live with a little delay and some weeks of searching, it might save you some cash.

JR

That's why I'm asking. I've been watching eBay, and that's when I swooped in and grabbed the 4 super precision NSK's for about $100 total. I just don't think some abec-7's with a 25mm bore and speeds to match my angular bearings (~30k RPM grease, ~45k RPM oil) will come up on eBay any time soon, if at all. Probably once I see the price for the bearing, I'll just go back to watching eBay.

Thanks for the link,
Brooks

I started building a spindle also and after adding up all the part dollars I was like, wow. Then this baby came around and I snaged it. It is a CAT40 Bridgeport TorqueCut spindle. As you can see they use four balls to retain the stud.

http://img50.imageshack.us/img50/8748/10005764vr.jpg
http://img321.imageshack.us/img321/4474/10005775iv.jpg
http://img513.imageshack.us/img513/9067/10005794nj.jpg
http://img513.imageshack.us/img513/2706/10005800kr.jpg

I see a lot of talk about bearings and configurations, but what are you planning on using for a spindle motor and how are you going to connect it (direct drive, gears, belt drive).

Thanks,

Johnny

As far as spindle motor goes, I've seen a lot of people use treadmill motors. They're 1.5-2.5 hp, and decent RPM. The only thing I don't get about them is the fact that some motors are treadmill duty. I searched, but did not find a definitive answer. I myself found a 2.5 hp continuous duty motor, which should work well.

For driving the spindle, Littlemachineshop.com uses gears in their mill head, I've also seen timing belts used. Search the galleries for MikeAber. If I remember right he has a lot of good stuff. As for direct drive, I've seen that a lot on engraving spindles.

Hope that helps,
Brooks

The surplus DC variable speed tread mill motors I acquired were designed/intended to only run in one direction. The pulleys were locked on with a nut as opposed to a keyway'd shaft.

Bidirectional rotation may be an issue as the nut may work loose, espeially under high vibration.

The motors I have are just DC brush motors that don't care what direction they run and are keywayed to the pully. And rated at 2.5hp continuous (which are really stout motors according to my freind in the fittness repair industry).

I think the ones you picked up are maybe lighter duty (regardless of rating) or made by one of the lower end manufacturers.

Some food for thought about horsepower ratings:

HP= (torque x RPM)/5252

Where HP = horsepower and torque is expressed in ft-lbs.

Thus, the rated HP of the motor is real sensitive to the RPM at which the HP was calculated. Moreover, different windings of motors with the same package size will have vastly different torque curves - check out brushed DC motor specs and you'll see for yourself.

Using simple but not necessarily valid numbers for illustrative purposes:

HP1 = 4 x 3450

vs

HP2 = 8 x 1750

Although HP1 = HP2 when you do the math, the torque available at a lower rpm (if that is critical) is going to be a LOT more with HP2 than HP1.

When the motor ultimately "sees" the load, it is torque at that RPM that counts, not some single HP value that determines if it will cut the mustard.

If you're going to run a motor at variable speeds, you REALLY need to know the torque curve of the motor as a function of speed and current draw.

Reason: when you do throttle it down to obtain the spindle speed you need, you still need TORQUE so that the motor has the ability to absorb the load and NOT fall off in speed under load. The current flow at that voltage determines the power that will be developed.

Simply put, the voltage generates the speed whereas the amperage flow determines the torque output.

Yet in the above example, if you run HP2 at 1:1 gearing and the other at 2:1 gearing, you should end up with essentially the same NET power at the work end of the spindle.

Point is this: simply saying the motor is rated 2.5hp continuous and is "treadmill" rated really doesn't tell you what you really need to know with regard to sizing/selecting the motor - especially one that will probably be run at throttled speeds. It takes a bit more math and science than that to make absolutely sure.

This link may give more insight:

Motor torque info/defininition
http://www.merkle-korff.com/formulas.asp#con

pstockley,
You PM inbox is full.
I tried to send you a message to run my ISO-25 spindle by you but I couldn't.

Later,
Bill

Its been awhile since this has been updated. Has there been any progress?
-Adam