PDA

View Full Version : Seals for toolpost grinder



TexasTurnado
06-27-2008, 12:19 PM
Hello from a new member. I have been reading posts for some time now, but had not joined until yesterday.

I'm giving consideration to building my own toolpost grinder and have acquired some ultra precision bearings for the project. They are rated for up to 50,000 rpm with oil lubrication and are the angular contact type with light preload (DUL type)

Does anyone here know what kind of seals are used for the highspeed bearings on toolpost grinders? Mfgrs like Themac and Dumore must be using seals to keep the oil in the spindles of these machines. Given the high rates that these turn (up to 30,000 + rpm) it would seem that regular rubber seals would not last long and might even melt. Any comments?

TexasTurnado

pcarpenter
06-27-2008, 12:36 PM
I don't know the answer for sure, but I would suggest that "regular rubber" seals are not made from latex sap from a tree anyway:D

I would also suggest that if you are running them hot enough to melt the seals there is a preload problem. I have read that preloads on high speed spindles are often on the gentle side to allow for growth. Really high speed bearings are usually oil lubricated as grease is a problem at high speed. That often makes them lossy too so they may not be sealed at the bearing, but rather at an outside housing somewhere.

Are you really going to turn them that fast? I would guess not since many stones will blow up at speeds much slower than your bearings are rated
to.

Edit-- welcome aboard. I would guess from your screen name you have a South Bend "Turnado"? If you are new to machining, do read up on how awful toolpost grinders are for lathes and the precautions you need to observe.

Paul

SGW
06-27-2008, 12:47 PM
I doubt you need or want much oil in the spindle. All ball bearings need is a thin film of oil. The Quorn spindle has no oil seals at all. It uses "labyrinth" seals to keep the grit out, but that's it.

Get the book about building the Quorn. It has a good design for a high-speed spindle, and some good discussion of design considerations. The Quorn's spindle design isn't anything radical or unproven; I have an old New Departure bearing handbook that shows a spindle design very similar to the Quorn's.

Forrest Addy
06-27-2008, 01:17 PM
High speed spindle are usually sealed (a loose definition to be sure) with flingers and a simple labyrinth clearance. Grit has to enter against centrifugal force fo a spinning flinger and oil has to exit against the labyrinth. Simple really once you see a sectional drawing.

http://en.wikipedia.org/wiki/Labyrinth_seal

A flinger is a simple rotatiing disk intended to throw contaminant from a clearance space between shaft and bearing housing.

TexasTurnado
06-27-2008, 03:02 PM
I don't know the answer for sure, but I would suggest that "regular rubber" seals are not made from latex sap from a tree anyway:D

I would also suggest that if you are running them hot enough to melt the seals there is a preload problem. I have read that preloads on high speed spindles are often on the gentle side to allow for growth. Really high speed bearings are usually oil lubricated as grease is a problem at high speed. That often makes them lossy too so they may not be sealed at the bearing, but rather at an outside housing somewhere.

Are you really going to turn them that fast? I would guess not since many stones will blow up at speeds much slower than your bearings are rated
to.

Edit-- welcome aboard. I would guess from your screen name you have a South Bend "Turnado"? If you are new to machining, do read up on how awful toolpost grinders are for lathes and the precautions you need to observe.

Paul


Thanks for the welcome.

I think "regular" seals are made from nitrile or viton. I did some snooping on the web and found some that were rated for 13,000 max. No mention of life at that speed tho. I doubt I will want to turn the spindle at 50,000, but one of my desired uses is to grind small id's using the stones mounted on 1/8 and 1/4 shanks using a collet to hold them. These seem to work fairly well in a Dremel at 20,000.

As for the contamination, most of the apps I have in mind will be the id type as mentioned above, but I'm sure there will be times when I will want to do the od type grinding and will have to protect the rest of the lathe as well as I can whenever grinding.

SGW: Thanks for the suggestion regarding the Quorn. I will look into that design.


Forrest Addy: Good point. I was thinking along these lines - ball bearings are available with "mechanical seals" as well as rubber seals.

On the SKF site, I found some "sealing washers" listed (part no Z004 is representative) which seem to show promise. A few of these side-by-side on the shaft would appear to result in a labyrinth seal.

I will try to post a picture of it but am not yet familiar with photobucket.

TexasTurnado

TexasTurnado
06-27-2008, 04:12 PM
Here's the image:
http://i288.photobucket.com/albums/ll168/TexasTurnado/SKFSealWasher20mm-1.jpg

lazlo
06-27-2008, 08:06 PM
Hi Guys,

High speed spindles use open bearings for three reasons:

High speed bearings need oil mist lubrication

You're preloading the bearing races on an angular contact pair, so a seal would interfere with the preload offset

Seals rub against the inner ball race, and create a lot of friction. So sealed bearings have a lot higher rolling resistance, and heat generated, than a shielded j or open bearing.


Here's a Dumore 7X-250 grinding spindle that I have. It uses a pair of opposing angular contact bearings (Barden 203HDL's), one at each end (a lot like the Quorn spindle), and the spindle cartridge itself is the preload cylinder. Like Forrest alluded, the bearings are "sealed" by the labyrinth seal on the endcaps.

The Dumore spindles use an oil mist system that's dirt simple: there's a wool wick in the center chamber, and centrifugal force flings a fine mist outwards towards the open bearings. It's described nicely in US Patent 21888251 and 2326161. By the way, the Haas VMC uses a similar oil mister: a drip oil system that lands on a felt pad that's immediately above the spindle bearings.

http://i164.photobucket.com/albums/u15/rtgeorge_album/Dumore7X-250.jpg

TexasTurnado
06-29-2008, 03:47 PM
QUOTE=lazlo]Hi Guys,

High speed spindles use open bearings for three reasons:

High speed bearings need oil mist lubrication

You're preloading the bearing races on an angular contact pair, so a seal would interfere with the preload offset

Seals rub against the inner ball race, and create a lot of friction. So sealed bearings have a lot higher rolling resistance, and heat generated, than a shielded j or open bearing.


Here's a Dumore 7X-250 grinding spindle that I have. It uses a pair of opposing angular contact bearings (Barden 203HDL's), one at each end (a lot like the Quorn spindle), and the spindle cartridge itself is the preload cylinder. Like Forrest alluded, the bearings are "sealed" by the labyrinth seal on the endcaps.

The Dumore spindles use an oil mist system that's dirt simple: there's a wool wick in the center chamber, and centrifugal force flings a fine mist outwards towards the open bearings. It's described nicely in US Patent 21888251 and 2326161. By the way, the Haas VMC uses a similar oil mister: a drip oil system that lands on a felt pad that's immediately above the spindle bearings.[/QUOTE]


Wow, that's just the kind of picture I was wanting to see! Sorry I didn't reply sooner, but have been out of town a couple of days.

The oil misting system sounds very interesting. Can you tell me if the shaft is smooth where it passes thru the felt pad or does it have ridges of something to make the mist?

Not having seen this kind of detail before, I was planning to use a pair of angular contact bearings face-t0-face at the grinding end and anchor the spindle there. Then I was planning to use another pair of a/c bearings at the belt driven end, but these would be fit with a sliding fit on the spindle so any thermal expansion/contraction forces would not be transmitted to the bearings.

The type of a/c bearing I have are furnished face ground so that mounting them face-to-face applies the correct preload (for these DUL types, about 5-6 lbs).

Has anyone seen high speed spindles designed this way? Or should I save one set of bearings and just use a single pair as on this Dumore spindle and the Quorn design?

For lubrication, I was planning to put just enough oil in to cover the bottom of the outer race. That way, when the spindle was turning the balls would run thru the oil at the bottom and transfer it all over the bearing(s). I would expect that after the spindle had run a short time at high speed, the oil would be distributed all over the bearings and the inside of the spindle support tube so the lubrication would be closer to a mist situation than actually running in an oil bath.

If I use the felt misting system design as in the Dumore, is there enough mist to get to the outer bearing of the pair, or will this bearing tend to be starved for lube? Any comments will be appreciated.

TexasTurnado

TexasTurnado
06-30-2008, 10:42 PM
Lazlo - may I bend your ear for a few questions regarding the Dumore spindle you have?

If so, can you tell me how the bearings are held in place? From the pic, it looks like there is some sort of special washer next to the bearings that has several fingers around its circumference - is this correct? Do these fingers contact the shiny face on the end cap when it is tightened in place? I can't tell from the pic if these are there to hold the bearings in or if they are used as a locking device to keep the end cap from coming loose. Or am I off base entirely, and the "fingers' I see are really slots in the bearing retaining nut?

You listed a couple of patent numbers for the misting system - do you know if these are available on-line? Or do I need to order them from the patent office? You mentioned the felt was in the center cavity - is this just the space between the two bearings, or is there a separate cavity for it?

Thanks for your time. :)

TexasTurnado

TexasTurnado
07-11-2008, 11:12 PM
Thanks for all the comments guys. I learned/found a lot of info I didn't previously have. Thanks especially to Lazlo for the patent no of the oiling mechanism Dumore uses. I found the second patent online, but am still looking for the first.

If anyone has a good suggestion for a motor to use, I'd like to hear it. I recall one of the previous toolpost grinder articles used the motor from a router, but the router I have (Porter Cable) seems to have quite a bit of vibration for this application.

Thanks again!:D

lazlo
07-11-2008, 11:15 PM
Sorry Turnado, I didn't see your last post with the questions about the Dumore bearings.

I'll have to find the spindle again, but IIRC there's a wave washer on the drive pulley end, and the business end is pre-loaded directly against the bearing cap.

I haven't been into the oil slinger yet -- the patent you found is the only diagram I've seen.

oldtiffie
07-12-2008, 02:27 AM
You can buy pairs of angular contact bearings that when mounted "face-to-face" ("end-to-end") as a "matched pair" - at the "grinding wheel end" - provide the designed pre-load when clamped together. That provides both radial and axial bearing load as well as restricting end-float. The bearing at the other end (usually the"pulley" end) can be a normal ball or even a roller bearing. That allows the spindle to "float" with the float being mainly between the two bearing sets but moved the pulley but not the grinding-wheel relative to the pre-loaded bearing/s set.

This arrangement is very effective and requires a lot less precision machining (and risk of error) and adjustment ("Belleville" and other spring washers and fine-threaded holes and nuts etc.). This sort of adjustment requires axial pre-load and end-play measured in the order of 0.0005", where-as the "pre-loaded pair and free end-float" as suggested cancels out most of the problems.

I agree with Forrest Addy as regards sealing and lubrication.

TexasTurnado
07-12-2008, 10:59 AM
You can buy pairs of angular contact bearings that when mounted "face-to-face" ("end-to-end") as a "matched pair" - at the "grinding wheel end" - provide the designed pre-load when clamped together. That provides both radial and axial bearing load as well as restricting end-float. The bearing at the other end (usually the"pulley" end) can be a normal ball or even a roller bearing. That allows the spindle to "float" with the float being mainly between the two bearing sets but moved the pulley but not the grinding-wheel relative to the pre-loaded bearing/s set.

This arrangement is very effective and requires a lot less precision machining (and risk of error) and adjustment ("Belleville" and other spring washers and fine-threaded holes and nuts etc.). This sort of adjustment requires axial pre-load and end-play measured in the order of 0.0005", where-as the "pre-loaded pair and free end-float" as suggested cancels out most of the problems.

I agree with Forrest Addy as regards sealing and lubrication.

The pair of preground a/c bearings you describe is what I have (2MM9104WI) and I found them particularly interesting in that the factory has permanently marked the high spot on both the inner and outer rings. They recommend the high spots be aligned (inner-to-inner, outer-to-outer) during installation for best results. Have you seen this before? Considering the extremely high accuracy and low runout of these bearings in the first place, I found it interesting they would go to the trouble of marking the high spots.

In use, I assume mounting them this way causes the preload to remain more constant and prevents the bearings from working against each other in the radial direction during rotation - and thus results in less temperature rise. Or is there some other reason I haven't thought of?

I considered using a roller bearing, but cannot find any rated for 20,000 rpm or higher (these a/c ball bearings are rated for a max of 50,000 with oil). There are precision single bearings (like the 2MM204K) but I have not been able to find any at less than big$$. Since I was able to purchase two pairs of the a/c bearings very reasonable, at this point I am still planning to use them at both ends, but float the outer race of the driven end to allow for self alignment

Another factory recommendation for the precision a/c bearings I found interesting was that they recommend fitting them line-to-line on both inner and outer races. I guess I am just used to ordinary deep groove bb's where one or the other race is usually an interference fit to its bore/shaft. Thinking about it though, I presume this line-to-line fit is used to prevent the factory precision fit of the balls and races not to be disturbed by the mounting. When a standard bb is pressed in, the internal clearances are reduced.

lazlo
07-12-2008, 11:18 AM
They recommend the high spots be aligned (inner-to-inner, outer-to-outer) during installation for best results. Have you seen this before?

All precision bearings (ABEC-5 or higher) have the high-spots marked. Barden's have a little 'V' laser-etched into the high spot. Timken's have a light red circle stamped on the high spot. OEM spindles are usually ground with the bearings installed, so it's critical that you mark where the high spot on the old bearings were installed, and install the new bearings with the high spot in the same place.

If you're building a spindle from scratch, I'd find the spot of maximum TIR on the spindle, and install the high spots on the bearing opposite that position.


I assume mounting them this way causes the preload to remain more constant and prevents the bearings from working against each other in the radial direction during rotation - and thus results in less temperature rise.

Correct -- the point of the wave washer pre-load is to allow for thermal expansion.

TexasTurnado
07-12-2008, 11:51 AM
If you're building a spindle from scratch, I'd find the spot of maximum TIR on the spindle, and install the high spots on the bearing opposite that position.

Good point! I will keep that in mind while turning the spindle.

TexasTurnado
07-12-2008, 12:10 PM
Would anyone here care to comment on the advantages/disadvantages of the spring loaded belt tension system used by Dumore and others vs the rigid mounting with eccentric cam method Themac uses?

Themac claims their method reduces vibration compared to the Dumore by preventing the motor from moving. Their literature claims the motor on the spring loaded design is allowed to moved back and forth as the belt revolves while their rigid design doesn't. Advertising BS or fact?

lazlo
07-12-2008, 12:24 PM
When I was calculating the belt drive for the Servo/Poly-V retrofit for my Bridgeport head, I skimmed past an entire chapter in the Gates engineering handbook on spring idlers. It's definitely not marketing BS -- there are definite pro's and con's to idlers versus static tensioning, and it depends a lot on the belt type, diameter, horsepower load, and RPM.

Themac is a very fine toolpost grinder, some say even better than the Dumore. I think ultimately the engineers at Dumore ended up with a different solution.

If you really want to think about using a spring idler, you can download the free Gates DesignPro software, set it up for the belts you're going to use on your toolpost grinder, and it will tell you the various options you have.

For my Bridgeport retrofit, I used Poly-V sheaves that were big enough that I don't need a lot of static tension, so a spring idler would be overkill.

By the way, Poly-V belts are just perfect for a toolpost grinder (high speed, high efficiency, low vibration) if you're going to have to turn a new sheave anyway...

TexasTurnado
07-12-2008, 02:07 PM
If you really want to think about using a spring idler, you can download the free Gates DesignPro software, set it up for the belts you're going to use on your toolpost grinder, and it will tell you the various options you have.

For my Bridgeport retrofit, I used Poly-V sheaves that were big enough that I don't need a lot of static tension, so a spring idler would be overkill.

By the way, Poly-V belts are just perfect for a toolpost grinder (high speed, high efficiency, low vibration) if you're going to have to turn a new sheave anyway...

Thanks for that info. I was aware of the Poly-v belts but would not have guessed they would be capable of the surface speeds needed for a grinder.

lazlo
07-12-2008, 02:36 PM
Actually, high surface speeds are where Poly-V's excel. They lose traction at low RPM, so they're not as ideal for low speed applications.

TexasTurnado
07-12-2008, 04:30 PM
Actually, high surface speeds are where Poly-V's excel. They lose traction at low RPM, so they're not as ideal for low speed applications.

I have been off reading what I could find on the Gates site. I didn't find mention of Poly-v, but they do offer what they call a Micro-v belt. The rating graphs they show end at 20,000 rpm and in another place they say the surface speed should be less than 6500 fps. Using sheave dias of 1 and 2 inches (assuming 10,000 rpm motor) comes out to about 5500 fps if I did the calcs correctly.

These are industrial rated belts, so they may be giving these limits based on continuous service. Needless to say, I won't be even approaching that kind of use, so this may be a great drive belt for my grinder. Monday I will contact them directly and see what they recommend.

I did not see any offer of design software - can you provide a link?

Back in the early 90's I redesigned the sorry drive system the RF-30 comes with using 3VX belts and design info from Phelps Dodge. I really like the 3VX belts and have used them on most of my projects since they where applicable.:)

lazlo
07-12-2008, 04:44 PM
Yeah, one of the maddening things about Poly-V belts is that each manufacturer calls them something different, so it's a pain to Google...
Machinery's Handbook calls them "V-Ribbed" belts.

The Micro-V's are Gate's Poly-V belts, with the standard 40 included angle. They also sell Polyflex JB belts, which are proprietary V-belts with 60 included angle. The Gates Micro-V J-series sheaves/belts are the most common for machinery applications. I just bought several of the Gates Micro-V belts and sheaves with the TaperLoc system.

I used the Gates PolyFlex cogged belts on my RF-30, based on a HSM magazine article, and they were a dramatic improvement -- even better than the PowerTwist belts.

You can download the Gates DesignFlex Pro here. It's free, but you have to register (no spam):

http://www.gates.com/designflex/index.cfm?location_id=809

TexasTurnado
07-12-2008, 07:46 PM
Thanks for the link. I dled it and installed, but it won't let me design anything above 12,000 rpm and the min sheave size is 1.94 in.:confused: So I will still be calling them Monday.

Do you have any suggestions for a suitable motor?

Machtool
07-12-2008, 09:10 PM
Texas

You might take a look at these GMN labyrinth seals. Very common on High speed European equipment. Very easy to mount, and they do a good job.
http://www.gmnbt.com/labyrinth.htm (http://www.gmnbt.com/labyrinth.htm)

Regards Phil.

lazlo
07-12-2008, 09:37 PM
You might take a look at these GMN labyrinth seals. Very common on High speed European equipment. Very easy to mount, and they do a good job.

Phil, unless I'm reading that page incorrectly, it looks like that's GMN's standard non-contact (labyrithine) seal that you can get on normal radial bearings? You can get just about any standard radial bearing with labyrithine seals, they're the bearings with the -LLB or -VV suffix.

If you look at GMN's angular contact bearings, seals are not an option. They have the weird -X2Z extra-wide angular contact bearing with two shields, but they're not available in duplex (matched) pairs:

http://www.gmnbt.com/spindlebearings_selection.htm

Machtool
07-12-2008, 11:00 PM
G’day Lazlo.

Maybe my link took you to another place?

When Tex mentioned sealing his spindle, these would be my favourite preference.
http://www.gmnbt.com/images/dichtungen_labyrinth3.jpg----------- http://www.gmnbt.com/images/d_a2_1.gif

http://www.gmnbt.com/images/dichtungen_labyrinth3.jpg (http://www.gmnbt.com/images/dichtungen_labyrinth3.jpg)
http://www.gmnbt.com/images/d_a2_1.gif (http://www.gmnbt.com/images/d_a2_1.gif)

A totally separate component to the bearing, and fairly easy to design in. Rather than all that fun trepanning and turning trying to make your own labyrinth. Preparation is simple. Just carry then on the spindle, say the 20mm journal for his 2MM9104WI (7004’s). Then just bore a location journal in the front face of his face cover / retaining journal. Just be careful of concentricity,

They have a 20 x 30 x 10, that you could squeeze in the front of the spindle nose. They are totally free running – non contact. That size above has an axial displacement allowance of +/- 0.38mm. So they don’t bind up with thermal growth etc. The L type are the plain ones. The M type are for heavier contamination. Those have a groove on the OD which you can port for drainage.

I first found these years ago. GMN also make integral spindle motor grinding spindles through to 105k r.p.m I have a customer with 4 of them on a Studer CNC grinder.

You have G.M.N in the States. (Houston T.X). I have to import these. I’m guessing but small ones like that aught to be about 30 bucks.

Just another option to those SKF washer things he was looking at.

Regards Phil.

TexasTurnado
07-12-2008, 11:11 PM
Texas

You might take a look at these GMN labyrinth seals. Very common on High speed European equipment. Very easy to mount, and they do a good job.
http://www.gmnbt.com/labyrinth.htm (http://www.gmnbt.com/labyrinth.htm)

Regards Phil.

Good day to you and thanks for the post. I took a look at the catalog online and they look interesting. In your last post you mentioned a price of $30. Is that in US dollars and for each or a pair?

Also, it appears they are meant to be pressed in/out - is that correct?

lazlo
07-12-2008, 11:43 PM
A totally separate component to the bearing, and fairly easy to design in. Rather than all that fun trepanning and turning trying to make your own labyrinth. Preparation is simple. Just carry then on the spindle, say the 20mm journal for his 2MM9104WI (7004s).

Oh, that is cool! Thanks for that Phil!

Machtool
07-12-2008, 11:53 PM
Tex.

Hello. That price would be a guess, and a wild one, but based on each. No real way for me to tell from this far away. Theres an agent here in Melbourne, but they never have stock, so Ill pay $30 - $40 for the seal, but another $125 for a rocket DHL out of Germany. Our dollar and yours are about the same at the moment, with in 6% anyway.

Recommend fit for those seals is Housing K7, shaft h6. That would be a housing bore +6 to -15 micron. Shaft 0 -13um. Thats going to be controlled by your bearing seat size. I tend to try and keep towards the larger end of the tolerance on the O.D Thats aluminium any way, its fairly forgiving. For some thing that needs to be assembled or taken apart, I might blow that out a bit and use a few drops of the bearing seat Loctite -641

There dead easy to install, especially if you can design in a shoulder to drive them onto. It pays to turn up an assembly sleeve to push them on. Pin punches and hammers are risky. Dont ask how I know.

Looking at the distributors page for GMN, they list themselves as the contact for Texas. So you can go direct and not through a bearing company. A phone call on Monday morning should come up with a price to see if they are feasible. 800-323-5725. Id be interested to know what they what for them on your side of the Pacific if you do get a price.

Regards Phil.

lazlo
07-13-2008, 12:42 AM
Phil, these seals seem to have strange dimensions. A 7202 AC is 15 x 35 x 11mm, but the closest seal is 37 mm OD? Is the seal supposed to be recessed by 2mm?

http://www.gmnbt.com/labyrinth_Mseries.htm

By the way EB Atmus (online distributor) carries GMN, but I can't find these seals on their web page:

http://www.ebatmus.com/index.html

lazlo
07-13-2008, 12:57 AM
Thanks for the link. I dled it and installed, but it won't let me design anything above 12,000 rpm and the min sheave size is 1.94 in.:confused:

I haven't had a chance to play with DesignFlex today, but the Micro-V engineering guide says:

"One major advantage of Gates Polyflex JB and Micro-V belts is the ability to operate smoothly and efficiently at high belt and shaft speeds. For example, shaft speeds of 30,000 rpm and higher have been achieved."

I've got a set of precision grinding heads I bought off Ebay, and they're belted with Browning Poly-V belts: 2.5" on the grinding head, and a 4.5" drive belt - 6-VJ.

Machtool
07-13-2008, 01:07 AM
Lazlo.

I’ve noticed that, They don’t seem to follow Din 70 or 72 series dimensions. I guess they are far more slender in section. That’s probably a good thing, in that you can use different diameters to the bearing bores to locate them and not be fighting fits for the seal against the bearing requirements.

Typical arrangement on a spindle carrying 7202’s Assuming you wanted to use the 15mm spindle size to carry both the bearing and the seal. What you would normally do is have your 35mm male journal protruding off the back face of the retainer plate / face cover. The one facing the housing. That’s the journal that pushes against your bearing outers, locking them into the housing.

At the front of that plate, facing out towards the wheel. You would bore a dia26mm x 8mm deep bore. Your going to have a 15mm & clearance hole right thru to bring the spindle out.

Quite often when there is a thread for a lock nut, they will drop the next journal down a size. More often than not your bearings will be against a shoulder or a nut, so you pick a larger one and run it on top of the shoulder journal.

The German parent site has far better Pics of different arrangements. Take a look at this.

http://www.gmn.de/front_content.php?idcat=52&idart=155&lang=2 (http://www.gmn.de/front_content.php?idcat=52&idart=155&lang=2)

If that doesn’t work, follow your nose from http://www.gmn.de (http://www.gmn.de/)

Regards Phil.

TexasTurnado
07-13-2008, 11:31 AM
Tex.

Hello. That price would be a guess, and a wild one, but based on each. No real way for me to tell from this far away. There’s an agent here in Melbourne, but they never have stock, so I’ll pay $30 - $40 for the seal, but another $125 for a rocket DHL out of Germany. Our dollar and your’s are about the same at the moment, with in 6% anyway.

Recommend fit for those seals is Housing K7, shaft h6. That would be a housing bore +6 to -15 micron. Shaft 0 -13um. That’s going to be controlled by your bearing seat size. I tend to try and keep towards the larger end of the tolerance on the O.D That’s aluminium any way, its fairly forgiving. For some thing that needs to be assembled or taken apart, I might blow that out a bit and use a few drops of the bearing seat Loctite -641

There dead easy to install, especially if you can design in a shoulder to drive them onto. It pays to turn up an assembly sleeve to push them on. Pin punches and hammers are risky. Don’t ask how I know.

Looking at the distributors page for GMN, they list themselves as the contact for Texas. So you can go direct and not through a bearing company. A phone call on Monday morning should come up with a price to see if they are feasible. 800-323-5725. I’d be interested to know what they what for them on your side of the Pacific if you do get a price.

Regards Phil.

I have the same reaction as Lazlo: these are pretty cool seals. Thanks again for pointing them out. I will definitely call tomorrow for pricing and availability - probably for one with a 15mm bore. The bearings I have are supposed to be mounted with a line-to-line fit, so trying to mount the seal on the shaft would not be an interference fit for retention.

Do you know how they make these? Aluminum expands faster than steel when heated, but do you think it would expand enough to allow the inner ring to be inserted (even if cooled)? Or do you suppose the aluminum ring is made large enough to clear and then "crimped" onto the steel inner ring? Aluminum will spring back some to increase the clearance. A very interesting design.:cool:

TexasTurnado
07-13-2008, 01:44 PM
Lazlo: check this out:


http://cgi.ebay.com/FAFNIR-2MM203WI-CR-DUM-FREE-SHIPPING_W0QQitemZ330238787958QQihZ014QQcategoryZ1 11618QQrdZ1QQssPageNameZWD1VQQcmdZViewItemQQ_trksi dZp1638Q2em118Q2el1247

These are DUM's, but the way they are used in your spindle, I don't think it will matter.

lazlo
07-13-2008, 02:50 PM
Lazlo: check this out:

These are DUM's, but the way they are used in your spindle, I don't think it will matter.

Thanks -- they're the guys I bought the SKF Explorer (ABEC-5) AC's from. $99 is out of my price range for an old beat-up Dumore spindle. I made them a best offer, but they're not willing to bargain much...

By the way, I played with DesignFlex last night, and I'm getting a max 10,000 RPM with the Micro-V belts (3:1 ratio on a 3450 RPM motor) and a max of 20,000 RPM with the Polyflex JB belts (60 Poly-V's).

Are you trying to hit reach 39,000 RPM the Dumore and Themac can do (for tiny little wheels)?

TexasTurnado
07-13-2008, 03:29 PM
By the way, I played with DesignFlex last night, and I'm getting a max 10,000 RPM with the Micro-V belts (3:1 ratio on a 3450 RPM motor) and a max of 20,000 RPM with the Polyflex JB belts (60 Poly-V's).

Are you trying to hit reach 39,000 RPM the Dumore and Themac can do (for tiny little wheels)?

When a played with it (DesignFlex:D ), the problem I ran into was not being able to set the sheave dia to less than 1.94 in (which I suspect is their smallest standard size). Turning at 10k, this dia gives a belt speed of ~5k, push the speed to 12k, and the belt speed is at the limit. If I can figure out a way to reduce the sheave dia, then I think it will show about what I calculated before.

If I use these belts, then I will probably have to use the spring loaded type of belt tension. Their shortest standard belt is 18in, so the c-c distance will be on the order of 6.5 in.

BTW, I have heard of people turning these belts inside-out and using them as flat belts. I have never been a fan of flat belts, but apparently this has been the best approach in the past. With the appearance of the multi rib belts, that may no longer be the case. When I get a chance, I will take a look at the Browning units you mentioned. And wasn't it Goodyear that first brought out the poly-v design?

As for my speed design goal, I am aiming at more like 20k for the small stones with a 1/8 shank like the Dremel uses. The bearings I have though, are rated to 50k, so it should be possible to get to 40k if the assembly is balanced well enough. Which begs another question - how does one go about balancing a spindle like this? Do they make balancing machines on the same principle as tire balancers that will tell you how much weight to add (or take off) at what angle??:confused:

lazlo
07-13-2008, 03:48 PM
The PolyFlex JB sheaves go down to 1.17". The highest I've been able to get with any Gates belt is 18,000 RPM with the PolyFlex JB series. Notice I was able to get the center distance down to 4.6 inches:

http://i164.photobucket.com/albums/u15/rtgeorge_album/PolyflexJB.gif

By the way, I just took a look at the Gates Micro-V (TaperLoc) sheaves I bought last week from EB Atmus, and the sheaves have a max speed of 17,440 RPM.

Machtool
07-13-2008, 08:53 PM
Tex


Do you know how they make these?
I have no idea. The hard copy of the catalog I have, that only says “ The high accuracy of inner and outer rings labyrinth profile is achieved by a special process”. That all they tell you about how they are made.

I have a couple of M 72-90-10 I keep here for a repeat job. That’s a fairly common size I find on a few different manufacturers BT40 spindles. I’ve had a real good look at those this morning. The M type has the slots in the OD profile, so you can see into it. The vees in there seem way too deep to achieve any sort of heat shrinking assembly. The inner vees must protrude into the aluminium outer at least 1 mm per side.

You could be correct that they are swaged on. I also think that it is possible the outer is rolled on as a strip and some kind of high tech welding joins the lap. Possibly ultrasonic or laser. But I cant see any evidence of a join on the one I’ve got on my desk.


so trying to mount the seal on the shaft would not be an interference fit for retention.
I’m figuring your shaft size for those 2MM9104’s will be smack on nominal to a tenth or two under. All my catalogs for shaft and housing fits for ABEC 7 bearings of that size give me a tolerance of 0 to -0.004um That’s tighter than the h6 spec (0 to -13um) for the seal, but it falls right in the top end. They will run nicely on any decent fit for the bearing.

You have to remember they are totally non contact, so excepting inertia, They take close to zero energy to turn. So the fit doesn’t have to be that heavy.

Regards Phil.

TexasTurnado
07-13-2008, 10:29 PM
You have to remember they are totally non contact, so excepting inertia, They take close to zero energy to turn. So the fit doesnt have to be that heavy.

Good point. As long as the outer half is retained in place, the inner will be restrained. However, if it drifted over to make contact, I suspect that would would not be good.

In my case, I may put a spacer washer between the inner ring of the bearing and the seal, then put the retaining nut on the outside face of the seal. The other side of the second bearing inner ring will be against a shoulder on the spindle. This should give a rather compact design and minimize overhang. Since the inner half of the seal is steel, this should work, shouldn't it?


You could be correct that they are swaged on. I also think that it is possible the outer is rolled on as a strip and some kind of high tech welding joins the lap. Possibly ultrasonic or laser. But I cant see any evidence of a join on the one Ive got on my desk.

They claim the process is high tech and fits exactly, so it could be something like coating the mating surface of the inner ring with some material to the thickness of the gap, and then plating or metal spraying the outer ring in place. After the seal is finished, they dissolve or melt out the material in the gap. Such a process might explain the cost of these units.

TexasTurnado
07-14-2008, 12:55 AM
The PolyFlex JB sheaves go down to 1.17". The highest I've been able to get with any Gates belt is 18,000 RPM with the PolyFlex JB series. Notice I was able to get the center distance down to 4.6 inches:

By the way, I just took a look at the Gates Micro-V (TaperLoc) sheaves I bought last week from EB Atmus, and the sheaves have a max speed of 17,440 RPM.

lazlo:

I had a chance to look at DFPro again, and you are right about the Polyflex JB designs. And I was right about the micro-v design problems:)

Their smallest sheave is 1.94 in (probably because they are taper lock) and their shortest standard belt is 18 in. OTOH, the PF-JB sheaves go down to .67 (but with a small bore) and the available belt lengths are much shorter. The software came up with a recommendation of 2.36/1.17 for a speed of 20.17k and 2.48/1.24 for a speed of 20k and found belts to go down to a 3 in c-c distance.

I suspect the 17.4k limit you see for the sheaves you have is because of their dia - 1.94 in at 17.4k is a rim surface speed of more than 8800 fpm = this may be the strength limit of the metal when safety factors are accounted for. Larger dias would be even higher, of course. What size(s) did you purchase?

Considering the list prices of their sheaves and the fact they have to be bored to size anyway, I will probably make my own from pre-hard 4140. I would think that would give me plenty of safety factor.:)

lazlo
07-14-2008, 03:28 PM
I suspect the 17.4k limit you see for the sheaves you have is because of their dia - 1.94 in at 17.4k is a rim surface speed of more than 8800 fpm = this may be the strength limit of the metal when safety factors are accounted for.

I know, that's pretty amazing when you think about it. The speed of sound is 65,106 fpm, so that sheave is spinning at 1/8th the speed of sound :D

One limit to the Gates' sheaves are that they're made from sintered steel. The Browning Poly-V's I have are cast iron, but don't have a Max RPM stamped on them. Cast iron sheaves from Gates are nearly twice as expensive as the sintered steel.


Larger dias would be even higher, of course. What size(s) did you purchase?

In my typical obsession/compulsive fashion, I decided to convert my tool and cutter grinder to Poly-V's, and I'm going to use Poly-V's for the servo gear hobber project I'm working on.
So I'm using 5.3" driving, 4.0 driven (1.325:1) on the Bridgeport servo retrofit, and 4.5" driving, 2.24" driven (2:1) on the grinder workhead.


Considering the list prices of their sheaves and the fact they have to be bored to size anyway

They actually sell machinable TaperLoc's for cheaper than you can buy the raw materials. I bought several.

TexasTurnado
07-14-2008, 03:38 PM
They actually sell machinable TaperLoc's for cheaper than you can buy the raw materials. I bought several.

Who does? and for how much? I could use a TaperLoc on the motor, but the 1.xx dia of the spindle sheave is too small for a TaperLoc is it not?

TexasTurnado
07-14-2008, 03:58 PM
Looking at the distributors page for GMN, they list themselves as the contact for Texas. So you can go direct and not through a bearing company. A phone call on Monday morning should come up with a price to see if they are feasible. 800-323-5725. Id be interested to know what they what for them on your side of the Pacific if you do get a price.

Regards Phil.

Phil, you were right on the money. When I called Houston, she said about $30 each, but I would have to go thru a distributor. When I called A&W Bearing here in Dallas the quoted low 30's for either 15 or 20 mm sizes.

I believe I will try my own approach to sealing it first - that's more than I paid for the 2 pairs of bearings I have:eek: Then, if I'm losing too much oil, I may reconsider. But I still thank you for pointing them out!

lazlo
07-14-2008, 04:14 PM
Who does? and for how much?

Gates sells them (machinable TaperLocs). They range in price from $7 for the smaller sizes up to $30 for the giant sizes. I bought a 1 11/16" machinable TaperLoc for my lathe spindle (long story) and it weighs around 8 lbs.

By the way, props to Wierdscience for turning me on to TaperLoc bushings. They're very cool, and much more compact than the standard QD tapered bushings.

Edit: I just checked my EB Atmus invoice, and the Gate machinable TaperLocs are part numbers beginning with 3020. So the 1 11/16" machinable TaperLoc I bought is Gates part number: 3020 1.11/16NKS. It was $30.11, and like I said -- it's massive: weighs around 8 lbs. One of the little ones I bought is a 1/2" machinable TaperLoc -- that's 3020.1/2NKS, and it was $6.97

Good luck!

Robert

TexasTurnado
07-14-2008, 06:04 PM
Gates sells them (machinable TaperLocs). They range in price from $7 for the smaller sizes up to $30 for the giant sizes. I bought a 1 11/16" machinable TaperLoc for my lathe spindle (long story) and it weighs around 8 lbs.

By the way, props to Wierdscience for turning me on to TaperLoc bushings. They're very cool, and much more compact than the standard QD tapered bushings.

Edit: I just checked my EB Atmus invoice, and the Gate machinable TaperLocs are part numbers beginning with 3020. So the 1 11/16" machinable TaperLoc I bought is Gates part number: 3020 1.11/16NKS. It was $30.11, and like I said -- it's massive: weighs around 8 lbs. One of the little ones I bought is a 1/2" machinable TaperLoc -- that's 3020.1/2NKS, and it was $6.97

Good luck!

Robert

TaperLoc... aren't those the ones that mount with two set screws that are located in the taper with half the threads on the bushing and half on the sheave? I designed some of that type into my RF-30 drive. I bought the bushings and machined my own step sheaves from aluminum. Getting the threads in the sheaves was an interesting exercise.:)

The only problem I had was on one sheave the two setscrews did not give me the ability to adjust the sheave for absolutely no wobble at the rim like the QD's do with their three screws. Maybe I didn't get the taper exactly right on that one, but I used the same angle as on the others that gave no problems. Do the larger sizes use three (or more) set screws?

Machtool
07-14-2008, 09:27 PM
Tex

Thanks for the feed back on that price. Seemingly we are paying about the same for them. Which is unusual, moderately rare stuff like that often attracts a premium.


Then, if I'm losing too much oil, I may reconsider

Am I going to do your head in by mentioning this, and do they still burn heretics at the stake here?

Dont discount grease in this application. I sat down and crunched some numbers last night, before you mentioned oil. I ran out of time to write up a reply.

As I understand it youre looking to run up to 20k rpm? Thats a scary number, but your bearings are tiny. Are you familiar with Speed Mean factor calculations? Or the figures shown as Dn in all the Bearing & Lubrication catalogs. That just a factor or unit used to show how relatively fast a bearing is turning.

Its the easiest engineering calc on earth. Its just the average diameter of your bearing I.D / O.D multiplied by the design speed.

In your case 20 + 42 / 2 = 31 x 20,000 rpm = 620,000 Dn

A speed mean of only 620k is well in the zone of grease lubrication.

Heres a list of Barden specified grease cross references. The last figures are the Limit speed

GREASES
G-2 Exxon Beacon 325 65F to +250F 400,000
G-4 Nye Rheolube 757 SSG 40F to +200F 650,000
G-12 Chevron SRI-2 20F to +300F 400,000
G-18 Nye Rheotemp 500 50F to +350F 500,000
G-33 Mobil #28 80F to +350F 400,000
G-42 Nye Rheolube 350-SBG-2 30F to +250F 650,000
G-46 Kluber Isoflex NBU-15 40F to +250F 750,000
G-74 Exxon Unirex N3 40F to +300F .650,000
G-75 Arcanol L-75 60F to +250F 1,200,000


5 of those greases exceed your speed rating, Theres also a newer Arcanol L210 which is up in the million plus range.

Im a Kluber guy, The Isolflex NBU 15, Id use that on any thing in the 500,000 800,000 Dn range. Past that Id jump up to LDS 18 Special A or the newer Kluber Speed 22-72BF. Thats pushing grease lubrication up to the 1.4 million Dn numbers.

The other thing I note is the Fafnir spec for your bearing speed under grease lube. It has a rating of around 38,000 r.p.m Keep in mind those figures are for one only bearing - lightly spring preloaded. Other factors come in to play when you rigidly pre-load them, increase pre-load or start stacking them in pairs. But youre along way off getting close at 20k.

Just because Dumore were using oil back in the 60s, doesnt mean you have to in 2008.

Ill go and hide now.

Regards Phil.

lazlo
07-14-2008, 09:49 PM
I’m a Kluber guy, The Isolflex NBU 15

Yeah, folks are nuts about Isoflex here too, but a 50 gram (1.8 oz) tube will run you $32, which is more than the bearing costs ;)


Just because Dumore were using oil back in the 60’s, doesn’t mean you have to in 2008.

Good point Phil -- modern synthetic barium/lithium greases are way better than the stuff available when Dumore designed these spindles. But the Dumore's go up to 38,500 RPM, so that's really pushing it for grease...

TexasTurnado
07-14-2008, 10:01 PM
Tex

Thanks for the feed back on that price. Seemingly we are paying about the same for them. Which is unusual, moderately rare stuff like that often attracts a premium.



Am I going to do your head in by mentioning this, and do they still burn heretics at the stake here?

Don’t discount grease in this application. I sat down and crunched some numbers last night, before you mentioned oil. I ran out of time to write up a reply.

As I understand it you’re looking to run up to 20k rpm? That’s a scary number, but your bearings are tiny. Are you familiar with Speed Mean factor calculations? Or the figures shown as “Dn” in all the Bearing & Lubrication catalogs. That just a factor or unit used to show how relatively fast a bearing is turning.

It’s the easiest engineering calc on earth. Its just the average diameter of your bearing I.D / O.D multiplied by the design speed.

In your case 20 + 42 / 2 = 31 x 20,000 rpm = 620,000 Dn

A speed mean of only 620k is well in the zone of grease lubrication.

Here’s a list of Barden specified grease cross reference’s. The last figures are the Limit speed

GREASES
G-2 Exxon Beacon 325 –65F to +250F 400,000
G-4 Nye Rheolube 757 SSG –40F to +200F 650,000
G-12 Chevron SRI-2 –20F to +300F 400,000
G-18 Nye Rheotemp 500 –50F to +350F 500,000
G-33 Mobil #28 –80F to +350F 400,000
G-42 Nye Rheolube 350-SBG-2 –30F to +250F 650,000
G-46 Kluber Isoflex NBU-15 –40F to +250F 750,000
G-74 Exxon Unirex N3 –40F to +300F .650,000
G-75 Arcanol L-75 –60F to +250F 1,200,000


5 of those greases exceed your speed rating, There’s also a newer Arcanol L210 which is up in the million plus range.

I’m a Kluber guy, The Isolflex NBU 15, I’d use that on any thing in the 500,000 – 800,000 Dn range. Past that I’d jump up to LDS 18 Special A or the newer Kluber Speed 22-72BF. That’s pushing grease lubrication up to the 1.4 million Dn numbers.

The other thing I note is the Fafnir spec for your bearing speed under grease lube. It has a rating of around 38,000 r.p.m Keep in mind those figures are for one only bearing - lightly spring preloaded. Other factors come in to play when you rigidly pre-load them, increase pre-load or start stacking them in pairs. But you’re along way off getting close at 20k.

Just because Dumore were using oil back in the 60’s, doesn’t mean you have to in 2008.

I’ll go and hide now.

Regards Phil.



No need to hide:) I try to have an open mind about things like this and remain open to all options. I was surfing the Timken site today and noticed they have a grease called "ultra high speed spindle grease" and apparently sell only greases and not oil. Interesting, eh? The bearing mfgr does not sell oil..

So the thought of using grease has already passed thru my dark matter. One problem I see with grease is frequency of re-application. With oil, it's no real problem. Properly designed, one can just drop in in and it will get to where it should be. But with grease, it needs to be put in place.

So my question is: if one applies the correct amount of grease (whatever that is) how long could a spindle running at 20k be reasonably expected to run before it needs to be re-greased?

Taking the spindle apart to re-grease is a pain at the least, and a nightmare in some cases. What are your thoughts on this and do you have some real world cases to share?

As to mounting, I plan to use what Fafnir (Timken) calls DB or back-to-back mounting with the races rigidly clamped together on both ends of the spindle. One end pair would be mechanically restrained while the other end outer races would be allowed axial movement.

For the DUL's I have, this is supposed to result in a light preload of a nominal 10 lbs. Is there an advantage to using a bellville or wave washer to hold the races together? I would think the spindle would be more stable with rigid mounting than some sort of spring loading - no?

lazlo
07-14-2008, 10:07 PM
I was surfing the Timken site today and noticed they have a grease called "ultra high speed spindle grease" and apparently sell only greases and not oil. Interesting, eh? The bearing mfgr does not sell oil..

I wouldn't read anything into that. Timken doesn't make grease -- I'd bet a cold beer that it's re-badged Shell Alvania, which is the most commonly used OEM grease pack. Same deal with Barden, which started selling Barden-brand bearing grease at the same time Timken did.

They just haven't gotten around to rebranding Mobil Velocite and call it Timken spindle oil :)


Taking the spindle apart to re-grease is a pain at the least, and a nightmare in some cases. What are your thoughts on this and do you have some real world cases to share?

Angular contact bearings are open. Dunk the bearing in clean mineral spirits, and let the factory grease pack dissolve out. You're going to need to do that to run oil anyway.


Is there an advantage to using a bellville or wave washer to hold the races together? I would think the spindle would be more stable with rigid mounting than some sort of spring loading - no?

Belville washers are strictly to provide for thermal expansion of the spindle. A grinding spindle gets hot when it's running at 20,000 RPM :) If you don't include some kind of dynamic preload, the preload will increase a lot as the spindle heats up...

TexasTurnado
07-14-2008, 10:26 PM
Belville washers are strictly to provide for thermal expansion of the spindle. A grinding spindle gets hot when it's running at 20,000 RPM :) If you don't include some kind of dynamic preload, the preload will increase a lot as the spindle heats up...

Is this true when the bearings are mounted race to race in duplex configuration? If I need to use preload on the duplex pair, how much preload should be used for these DUL's I have? The same 10lbs they are supposed to have at room temp or something else? And what about the friction of the race on the shaft to allow it to move axially?:confused:

BTW, do you have a 800 no for Timken support? I couldn't find one on their site and ended up sending an email on their "contact us" option.

lazlo
07-14-2008, 10:47 PM
Is this true when the bearings are mounted race to race in duplex configuration? If I need to use preload on the duplex pair, how much preload should be used for these DUL's I have? The same 10lbs they are supposed to have at room temp or something else? And what about the friction of the race on the shaft to allow it to move axially?:confused:

See, now you're wishing you took me up on that offer to send you the Quorn grinding spindle articles :)

PM sent.

Machtool
07-15-2008, 05:46 AM
Tex.


As to mounting, I plan to use what Fafnir (Timken) calls DB or back-to-back mounting with the races rigidly clamped together on both ends of the spindle. One end pair would be mechanically restrained while the other end outer races would be allowed axial movement.

Im thinking your heading for difficulties right there, by making a 4 bearing arrangement. Thats basically a DBB set, even though they are spaced apart. That figure for speed in the catalog. You can probably deduct %25 from that in that in a light preload arrangement.

You mentioned running small mounted point stones. Going for Duplexed Back to Back sets is way over kill. That picture of Lazlos spindle, with one bearing at each end. That would be the most common arrangement for a light load, high speed spindle like this. Additional heat generation / expansion will make that a real tricky spindle to get right. Sure it can be done, but your chewing up a lot of engineering.

If youre trying to picture that, its basically like the arrangement of a back to back arrangement where the bearings are touching each other, but your adding space between them. Via shoulders on the shaft, housing or true spacers. Getting them apart stops cross pollination heat of each bearing feeding temp into the other. Thats a whole merry go round. Things warm up and expand. Differential expansion of the shaft to the housing do all sorts of grief with unseating preload.

Im away tomorrow & Im real short on minutes right now. But I can find you an example. Id be securing the front bearing, and spring preloading the rear if you wanted to get real sexy. Not with a Bellville or wave washer, but with a spring pack.

Lazlo.


Same deal with Barden, which started selling Barden-brand bearing grease at the same time Timken did.

That list Grease cross reference. I didnt make that up. That was in one of the on-line Barden engineering catalogs I found once. I saved a copy. Whats above is cut & pasted. Its since disappeared, or I cant find it online again. But its the only place Ive ever seen where they declare what the actual grease is that they are selling by the G number.


50 gram (1.8 oz) tube will run you $32
I see a big EBay future for you, buying 1kg tins and decanting it into 10 & 20CC syringes, then flogging it on EBay for blokes like Tex. He would be lucky to need half a teaspoon of the goop.

If you every need a tiny little bit, hit me up. It wouldnt be any trouble at all to put some in a post pack.

Cheers Phil.

TexasTurnado
07-15-2008, 06:34 PM
I’m thinking your heading for difficulties right there, by making a 4 bearing arrangement. That’s basically a DBB set, even though they are spaced apart. That figure for speed in the catalog. You can probably deduct %25 from that in that in a light preload arrangement.

In itself, that should not be a problem since they are rated for 50k max.


You mentioned running small mounted point stones. Going for Duplexed Back to Back sets is way over kill.

Damn, now you've done it! I knew sooner or later someone would bring that up.:D Over-design must be in my blood.:)

Actually, at first blush, the back-to-back mounting appeared to be a simple solution to the preload problem. Just let the mfgr grind them for the proper preload and be done with it.... Now I find that's not necessarily the case.

But if one can't mount them this way, then why does the industry make these expensive little pairs with DUL, DUM, and DUH ratings?:confused: Can these bearings be used as advertised at lower speeds, say 1k to 5k rpm?

I don't plan on running just the small little stones, but that seemed to be tallest pole in the tent as far as the design is concerned, The larger wheels have to be run at a slower speed, which should not be a problem if the spindle will do 20k.

And there's another side to it: I like to do something different and/or new and explore new territory. I had not seen a spindle design in the amateur publications like this, so I chose the back-to-back approach.

But here I am venturing into unknown waters, at least for me, so I it looks like I need to rethink the whole design. Perhaps I will end up making two spindles: one for lower speed external wheels and one for high speed internal grinding. I don't want to destroy two perfectly good pairs of bearings in one attempt - particularly if I can just possibly destroy one set at a time having fun:D


If you every need a tiny little bit, hit me up. It wouldn’t be any trouble at all to put some in a post pack.

Does this apply to me also? I will be happy to pay for the quantity I need -which I'm sure is not anything like 50g. I remember getting a 10g tube of silver loaded heat sink compound and it was a lifetime supply for computer cpu's.

lazlo
07-15-2008, 06:50 PM
But if one can't mount them this way, then why does the industry make these expensive little pairs with DUL, DUM, and DUH ratings?:confused:

DUL, DUM, and DUH specify the amount of preload: low, medium, and high, respectively. As preload increases, stiffness increases, heat increases, and maxiumum RPM decreases.

A toolpost grinder would want/need light preload (DUL).

Note that this is completely orthogonal to how you duplex the bearings.


I don't plan on running just the small little stones, but that seemed to be tallest pole in the tent as far as the design is concerned, The larger wheels have to be run at a slower speed, which should not be a problem if the spindle will do 20k.

For a 2 1/2 - 3" wheel, you only need 6600 RPM for the optimal sfpm.

TexasTurnado
07-15-2008, 07:05 PM
A toolpost grinder would want/need light preload (DUL).

That's why I chose the DUL's for this project.


Note that this is completely orthogonal to how you duplex the bearings.

I'm not sure what you mean by this. As I understand it, a universal DUL pair is ground so you can mount them race-to-race in any of the possible combinations and the preload will be the same DUL level. If they cannot be mounted this way, what is the point of face grinding in the first place? Aren't we talking just tenths of a thou difference between the three preload levels?

TexasTurnado
07-15-2008, 07:07 PM
What are the tradeoffs of wave washers vs belleville washers for preloading?

lazlo
07-15-2008, 07:40 PM
Note that this is completely orthogonal to how you duplex the bearings.

I'm not sure what you mean by this. As I understand it, a universal DUL pair is ground so you can mount them race-to-race in any of the possible combinations and the preload will be the same DUL level?

We're going in circles :)

Phil mentioned that duplexed Back-to-Back AC's was overkill for a toolpost grinder (I agree), to which you responded by asking about DUL, DUM, and DUH, which is the factory preload, and is orthogonal to how the AC's are duplexed.



You mentioned running small mounted point stones. Going for Duplexed Back to Back sets is way over kill.

Actually, at first blush, the back-to-back mounting appeared to be a simple solution to the preload problem. Just let the mfgr grind them for the proper preload and be done with it.... Now I find that's not necessarily the case.

But if one can't mount them this way, then why does the industry make these expensive little pairs with DUL, DUM, and DUH ratings?

Any AC (light, medium, or heavy preload) can be tandem, back-to-back or face-to-face duplexed. Tandem duplexing increases the thrust capacity in one direction but doesn't provide for thrust in the opposite direction. Back-to-back and face-to face duplexing provides bi-directional thrust capability. Back-to-back arrangement is the most common configuration.

lazlo
07-15-2008, 07:45 PM
By the way, Universal AC bearings mean that the preload was ground individually, and not in a matched pair. Not as precise as a duplex AC pair, but still much better axial play than non preloaded bearings.

TexasTurnado
07-15-2008, 10:44 PM
We're going in circles:)

I know my head is:) I'm going to have to go off and think about all of this for a while. Thanks for all your input!

BTW, when did Timken acquire MPB? I was just surfing their site and saw a super precision catalog for MPB bearings.

TexasTurnado
07-23-2008, 10:10 AM
lazlo: When you purchase from EB Atmus, do you have a local source there in Austin or do you order online?

Thanks

lazlo
07-23-2008, 11:09 PM
Sorry, didn't see this earlier. EB Atmus is an industrial supplier based out of Springfield, MA. The only reason I used them is because they sell most industrial distributors (SKF, NSK, INA, Timken, Gates, Browning, GMN,...) at about a 20% discount.

However, I just had a bad experience with them on the large order of Poly-V sheaves I bought for my Bridgeport servo retrofit: they drop-ship all their orders from the OEM, and you end up paying for multiple UPS shipments, so the shipping costs kill the discount.

To make matters worse, they accidentally dropped the big 24mm Gates Poly-V drive sheave off my order (I was charged for it, and it's clearly documented on the order confirmation), so I got 50 lbs of sheaves, but not the one I needed most. :mad:

Now that MSC has their 25% off sale every two weeks (including now, until Friday), they're a better deal if they have your item, even with sales tax. If MSC doesn't stock the item, you're probably better off with McMaster, who actually stocks the items (doesn't drop ship them), and has cheap shipping.

TexasTurnado
07-23-2008, 11:24 PM
Sorry, didn't see this earlier. EB Atmus is an industrial supplier based out of Springfield, MA. The only reason I used them is because they sell most industrial distributors (SKF, NSK, INA, Timken, Gates, Browning, GMN,...) at about a 20% discount.

However, I just had a bad experience with them on the large order of Poly-V sheaves I bought for my Bridgeport servo retrofit: they drop-ship all their orders from the OEM, and you end up paying for multiple UPS shipments, so the shipping costs kill the discount.

To make matters worse, they accidentally dropped the big 24mm Gates Poly-V drive sheave off my order (I was charged for it, and it's clearly documented on the order confirmation), so I got 50 lbs of sheaves, but not the one I needed most. :mad:

Now that MSC has their 25% off sale every two weeks (including now, until Friday), they're a better deal if they have your item, even with sales tax. If MSC doesn't stock the item, you're probably better off with McMaster, who actually stocks the items (doesn't drop ship them), and has cheap shipping.

Sorry to hear they screwed up your order like that - I know how irritating it is to not get all your order. I never heard back from Paul in Australia about the grease, and so far EB Atmus is the only place I have found that offers to sell just one syringe. The local suppliers here do not have it and they say the factory has a minimum order of 6 pcs. I will try McMaster tomorrow.

MSC doesn't have it in the online list, so I suspect they would want a minimum order of 6 also (if they can get it), but it won't hurt to call and ask. How do you get the 25% discount? I have seen discussion here about the Enco codes, but I missed the code for MSC if it was posted. I need to order some other items from them anyway, so it would be nice if they can supply all the items.

lazlo
07-24-2008, 12:01 AM
Save up to 25% off orders of 5 items or more PLUS Free Ground Shipping on orders over $250!

Enter WEBM25FF in the promotional code box when ordering online or mention the code when ordering via phone to receive the discount. Promotional price will be reflected in the Shopping Cart.

Machtool
07-24-2008, 02:47 AM
I never heard back from Paul in Australia about the grease
Sorry this thread dropped off my radar as well. Last I heard, your head was spinning and you were going off to take a break.

The offer still stands. Drop me your contact details to my Email. machtool at bigpond.net.au I keep all of them here in 1 kg tins, its not a problem to decant a syringe. I waste more than that.

Regards Phil.

P.S Whos Paul? *L*

TexasTurnado
07-24-2008, 11:06 AM
Sorry this thread dropped off my radar as well. Last I heard, your head was spinning and you were going off to take a break.

The offer still stands. Drop me your contact details to my Email. machtool at bigpond.net.au I keep all of them here in 1 kg tins, its not a problem to decant a syringe. I waste more than that.

Regards Phil.

P.S Whos Paul? *L*

Oops. My mistake for not looking back at the thread - sorry Phil:o I should know by now not to trust my memory....

I will send you an email shortly.

John

TexasTurnado
07-24-2008, 12:13 PM
Oops. My mistake for not looking back at the thread - sorry Phil:o I should know by now not to trust my memory....

I will send you an email shortly.

John

Please check your PM.