[NC Cams]

At the machne tool tech center for NSK in Japan, there wasn't a press to be found in the assebmly room. ONly induction heaters and/or hot plates and chillers.

WHen you get into the thermal thing the stuff LITERALLY falls together. You don't have to have to creep up on anything as the weight of the bearing is all that affects the "press" against the shoulder.

I have a press and won't go near it when assembling bearings. THermal shrink is simpely TOO EASY.

ONce you do it a few times, there are no "slips".

As always, the reader is welcomed to follow or ignore any recommendations put forth in
these threads.

[keen]

RE post 30

YOU NEVER< NEVER< NEVER PRESS AN AC back
You heat the outer ring, chill the inner and balls and they literally DROP together.

Press them together and you might as well throw them away.

YOU will pretty much ALWAYS brinnell each of the balls as well s the raceway you manhandled the balls over.

The alleged misassembly of the bearings may NOT have been the fault of Tormach. THera are or should be chevrons ( < or >) machined or etched across the OD of the outer rings.

YOu did record the postion of the cheveron, didn't you? probably not.

If there was no cheveron, and someone simply put in generic ac/s, generic AC's will not achieve a preload unless they are of the SUL, SUM or SUH variety

YOu did take photo of them to send to tormach to prove they are/were installed backwards? Probably not.

If the chevrons were >> and wrong, they were mismarked. IF they were <> or ><, they were misinstalled. It is easy to misinstall bearings, BTDT myself. BUT, they are so sloppy it is like a saddle on a sow. I"d find it hard to believe that an experienced assembler would make that bad of a mistake - besides, if the OEM even ran the thing for 5 min, the should have heard it right away.

THings like that happen right away, not after an appreciable period of use.

Point is, one can't really blame tormach for anything if one simply knocks a set of bearings out, especially a machine tool bearing, and starts pointing fingers and waving hands on a website. Any time one points fingers, there are 3 pointing back at you.

Kluber grease is fine but pricey and not necessarily any better than Andok C over the long haul. Bridgeport used Andok C in all their milling machines and some of them ran for 20 plus years with the same Andok C grease.

Bearings get the blame for a lot when actually people cause most bearing problems.

BTW, Bob Campbell has a link on his website that has a bunch of posts listed for spindle bearing service recommendations. They are Bridgeport specific but the same applies to pretty much any machine tool spindle.

My $0.05 offered as former machine tool bearing engineer.

Hi NC Cams

Pressing bearings in is fine if the bearing is not pressed through the balls - ie if you machine sleeves etc that press only the race that is being assembled. You cannot brinell the bearings this way.

Yes i do understand the chevrons <> << >> etc - and they were correct together - but the top set were upside down. - ( I installed them correctly on reassembly).

A photo would have been good - but not proof - I noted the position carefully, told the truth, and Greg trusted me.

I did not point the finger to run Tormach down - I am very impressed with Tormach generally and Gregs response. I was only trying to identify a possible assembly issue that might need improving for the benifit of all.

My machine has run beautifully for many hours now . Free from the squawk!

[NC Cams]

"Pressing bearings in is fine if the bearing is not pressed through the balls "

I know that, you know that, but you'd be surprise dhow many "expert" technicians FUBAR bearings cuz they press them thru the balls. No pun intended.

I specificlaly refer to post 30 above which incited my rant.

Yes you/one can use a press. BTDT myself. When it comes to a spindle, however,, or any expensive machine that I"m redoing, Out come the torches and the freexer gets emptied for my bearigns.

Best case pf sjroml fot was at Samta Fe rao; sjp[/ Tje tractopm ,ptpr rebuild guy had tehis deep freexzer sitting next to his bench. He has this end plate the size of a garbabe can lid for this moster traction motor. THre was not a press withing a mile.

WHen he goes to put the outher ring in (tapered roler), he takes out this huge rose bud torch and starts heating upthe outer ring. Every so often he hits it with tempastick to check temp. WHen "showtime" came, he reaches into the freezer and takes out what I thought was a frozen pizza.

Was actually the outer ring. He Literally shakes it out of the box and it FALLS into the bore. He quickly gives it a twist to check seating and 1-15 secs later, it is tight for the next half million miles. If you know how to do thermal fits, why would you even want to use a preas??? OTher than to press OFF the old bearing.

[MichaelHenry]

Best case pf sjroml fot was at Samta Fe rao; sjp[/ Tje tractopm ,ptpr rebuild guy had tehis deep freexzer sitting next to his bench. He has this end plate the size of a garbabe can lid for this moster traction motor. THre was not a press withing a mile.

Interesting set of typos there - almost like a cat was exploring your keyboard.

What sort of hot and cold temperatures are necessary for your bearing installation method?

Mike

[NC Cams]

sorru, misplace fingers on keyboard.

IFyou freeze the bearings, you dno't have to get stuff as hot. If you get the "hot" side to about 150 or so, you have a differential of roughly 130 deg. The guy in the railroad shop, you know fjlkajls above, used a rose bud torch to heat the traction motor end bell. He used temp-a-stick to check temp. In simple terms, if it feels werm to very warm and "just toleraboel". it is probably hot enough. It is probably cold enough if it sat in freezer overnight.

The "hot is barely comfortable, the cold is "stick your tongue to it and it will stick" cold. The stuff falls together.

Sorry for the dull moment.....

[philbur]

That's OK providing you are assembling spindles made to NSK standards.

What about disassembly.

Phil

At the machne tool tech center for NSK in Japan, there wasn't a press to be found in the assebmly room. ONly induction heaters and/or hot plates and chillers.

WHen you get into the thermal thing the stuff LITERALLY falls together. You don't have to have to creep up on anything as the weight of the bearing is all that affects the "press" against the shoulder.

I have a press and won't go near it when assembling bearings. THermal shrink is simpely TOO EASY.

ONce you do it a few times, there are no "slips".

As always, the reader is welcomed to follow or ignore any recommendations put forth in
these threads.

[NC Cams]

In case you've never looked, ALL bearings comply to the same ABEC standards. Whether it be for ABEC 1, 3 5, 7 or 9. Thus they are all interchangeable. THings like raceway smoothness, raceway coaxiality & runout and other "secret" internal features ultimately determine one from the other. Even Hardinge spindle techs, at the time, couldn't figure out why the NSK bearings ran quieter and had less axial runout on their spindles than their former bearing supplier. Theoretically same bearing, noticeably different performance.

Aside from this, whether bearings came from CHina or NSK, they all react the same to kluge job versus "NSK" standards of assembly.

WHy NOT to use press with a/c's?

SImple. THe raceways are off of verticle by 15, 25, 30 deg. even deg 60 when it comes to ball screw bearings. Thus, they become more and more sensitvie to axial distortion by pressure applied to the SIDES (you know, where you put the pipe spacers againgst to press them on). Besides, depending on the press fit, metal to metal rubbing can induce scuffing or microwelding from metal to metal rub. Gall the bore or the OD while presing, you WILL distort the raceway.

COnsidering that bearing internal clearance is as low as 0.000005mm depending on size and literally 0.000000 in a preloaded AC, it don't take much press induced distortion to fubar the raceway no matter WHO makes the bearing. IT is simply too easy to "drive it home" or "get er done" and wreck the geometry of the raceway with a press.

Ok, so use some lube. As things heat up, they expand unevenly. ANY lube under the bearing can affect uneven heat transfer- why lobe the bore when youre' relying on press fit and FRICTION to keep it from spinning? Much of bearing cooling takes place by conductivne transfer via the bore and OD. Uneven heat transfer will distort the raceway. No big deal right? My engineers found that writnig on the back of the bearing with magic marker affects the heat rise and noise output of bearings. When disassembled, they found literally uneven raceway contact that coincided with the marker.

Pressing off bearings is not an issue. Reason: it is presumed tht the bearing is being replaced so who cares? Re use of the pressed off bearing becomes the question.

The scuffing of the pressed ring becomes the first concert. Second is the fact that most bearing forks or clamps (press off tools) can easily apply load to outer and inner ring concurrenlty - thus you press thru the balls which we alrady agree is a no no.

FInally, distortion. Depending on how well you can grip the ring, you often end up gripping just the very top of the inner ring or at two points rather than evenly. This is the WEAKEST part of the ring and easiest to distort. Distort the ring, you distort the raceway. Guys reuse bullets after being in and out of the chamber. The only time I reuse a bearing is if it slip fits the shaft/housing. I avoid repressing at assembly simply by thermal fitting. Period Paragraph.

In case you didn't know, thermal fit is quite common in the top spindle shops. THis is why they have induction heaters. Bridgeport/Hardinge uses only that method. THe shops I saw all used them to heat outer rings for preload set/check - no damage potential is the reason. Not a press to be found in the assembly rooms I was in - and many did not use NSK bearings..

Finally the issue is that of time. THe more bulky the bearing, the harder it is to press squarely and do so expediently. IT would take a good press and a lot of GOOD tooling (no hacksawed pipe) to press in a 8" od outer ring. ONce the guy had the housing hot at the railroad shop (lsshrxoeh above), he literally 'Shook" the ring into the housing as if he was shoveing pizza into a box.

One of my guys took the same speculative view toward press vs heat with me, until he saw me replace a 6310 (50mm bore) bearing on a lathe motor. Wasn't about to take out the armature to press that puppy. It would take a crane that I didn't have. So, I Iced down the sahft, heated the inner ring with propane torch, Literally dropped it on. Spent more time cleaning up the melted ice than screwing with the bearing.

I probably won't/can't convince you that thermal fitting is a superior way to installing bearings over anything that involves a press or god forbid a hammer. I don't care and am not about to try. As far as Im' concerned, you're more than welcome to "press baby press" I avoid it like the plague.

One can pound screws with a hatchet. IT works but the use of the screwdriver is much less primitive. Likewise is that of the thermal fitting of bearings. THis diatribe is being written for the benefit of the unconvinced lurkers.

[philbur]

Yes but does that Chinese spindle have the same geometrical precision as the bearings you are trying to shrink fit onto/into it. If not you could be pissing into the wind.

Phil

In case you've never looked, ALL bearings comply to the same ABEC standards. Whether it be for ABEC 1, 3 5, 7 or 9. Thus they are all interchangeable. THings like raceway smoothness, raceway coaxiality & runout and other "secret" internal features ultimately determine one from the other. Even Hardinge spindle techs, at the time, couldn't figure out why the NSK bearings ran quieter and had less axial runout on their spindles than their former bearing supplier. Theoretically same bearing, noticeably different performance.

Aside from this, whether bearings came from CHina or NSK, they all react the same to kluge job versus "NSK" standards of assembly.

WHy NOT to use press with a/c's?

SImple. THe raceways are off of verticle by 15, 25, 30 deg. even deg 60 when it comes to ball screw bearings. Thus, they become more and more sensitvie to axial distortion by pressure applied to the SIDES (you know, where you put the pipe spacers againgst to press them on). Besides, depending on the press fit, metal to metal rubbing can induce scuffing or microwelding from metal to metal rub. Gall the bore or the OD while presing, you WILL distort the raceway.

COnsidering that bearing internal clearance is as low as 0.000005mm depending on size and literally 0.000000 in a preloaded AC, it don't take much press induced distortion to fubar the raceway no matter WHO makes the bearing. IT is simply too easy to "drive it home" or "get er done" and wreck the geometry of the raceway with a press.

Ok, so use some lube. As things heat up, they expand unevenly. ANY lube under the bearing can affect uneven heat transfer- why lobe the bore when youre' relying on press fit and FRICTION to keep it from spinning? Much of bearing cooling takes place by conductivne transfer via the bore and OD. Uneven heat transfer will distort the raceway. No big deal right? My engineers found that writnig on the back of the bearing with magic marker affects the heat rise and noise output of bearings. When disassembled, they found literally uneven raceway contact that coincided with the marker.

Pressing off bearings is not an issue. Reason: it is presumed tht the bearing is being replaced so who cares? Re use of the pressed off bearing becomes the question.

The scuffing of the pressed ring becomes the first concert. Second is the fact that most bearing forks or clamps (press off tools) can easily apply load to outer and inner ring concurrenlty - thus you press thru the balls which we alrady agree is a no no.

FInally, distortion. Depending on how well you can grip the ring, you often end up gripping just the very top of the inner ring or at two points rather than evenly. This is the WEAKEST part of the ring and easiest to distort. Distort the ring, you distort the raceway. Guys reuse bullets after being in and out of the chamber. The only time I reuse a bearing is if it slip fits the shaft/housing. I avoid repressing at assembly simply by thermal fitting. Period Paragraph.

In case you didn't know, thermal fit is quite common in the top spindle shops. THis is why they have induction heaters. Bridgeport/Hardinge uses only that method. THe shops I saw all used them to heat outer rings for preload set/check - no damage potential is the reason. Not a press to be found in the assembly rooms I was in - and many did not use NSK bearings..

Finally the issue is that of time. THe more bulky the bearing, the harder it is to press squarely and do so expediently. IT would take a good press and a lot of GOOD tooling (no hacksawed pipe) to press in a 8" od outer ring. ONce the guy had the housing hot at the railroad shop (lsshrxoeh above), he literally 'Shook" the ring into the housing as if he was shoveing pizza into a box.

One of my guys took the same speculative view toward press vs heat with me, until he saw me replace a 6310 (50mm bore) bearing on a lathe motor. Wasn't about to take out the armature to press that puppy. It would take a crane that I didn't have. So, I Iced down the sahft, heated the inner ring with propane torch, Literally dropped it on. Spent more time cleaning up the melted ice than screwing with the bearing.

I probably won't/can't convince you that thermal fitting is a superior way to installing bearings over anything that involves a press or god forbid a hammer. I don't care and am not about to try. As far as Im' concerned, you're more than welcome to "press baby press" I avoid it like the plague.

One can pound screws with a hatchet. IT works but the use of the screwdriver is much less primitive. Likewise is that of the thermal fitting of bearings. THis diatribe is being written for the benefit of the unconvinced lurkers.

[NC Cams]

Steel is steel. Bearigns all have the same ID size and ID tolearance. The bearing makers all ahve their own recommendations for tolearnce RECOMMENDATIONS for their bearigns.

NSK, Torrington, Fafnir, SKF all have engienering manuals.which outline their recommended stamdars" for spoindle fiat and finish. I"ve yet to see catalogs or manauals for any of the "no name" chinese bearing makers.

Buy a chinese spoindle, get chinese quality. (which improves almost daily). BUy Hardinge or Haas, you get Hardinge/Haas. You get waht you pay for.

NSK tricked up bearings suck jsut as bad on a FUBAR'd spindle as anybody's. Chinese bearings don't have the quiteness/runout of the NSK's on a "piece of jewelry" spindle.

Put lipstick on a pig and you still have a prettied up pig.

{Jo;.you and I had just better agree to disagree. WE've never convinced each other in the past (4th axis food fight) - I don't expect this issue to differ..

[LeeWay]

Now I do agree that sometimes thermal is the only way to go for some things. Frozen nuts and bolts were often loosened with a smoke wrench (oxy/acet torch) in the Army.

The bearings in the lower quality X2 heads aren't A/C unless you specifically buy those as replacements. I have three different heads and spindles and sometimes reused stock bearings and sometimes new ones. I really couldn't tell any difference in fit. They all slipped into place using that tool kit. It is CNC machined true, BTW.

Another thing a novice would need to watch out for is over heating. I think a heat gun might be a far better tool to use than a torch. To freeze the spindle and heat the inner race, you have to apply heat right near the seals. Again not that much heat should be required, but too much and you have boiling grease and a blown seal.

I do appreciate all your input NC Cams. This is what makes a great Forum. Guys willing to share experience and methods and explaining why one is better than another.

In this case, your method requires even less tooling than pressing a bearing. That is always a good option. That is if you are assembling. Removal can't really be done this way.
Now I don't use a press or a bearing fork, but the toolkit I do use works nicely for the quality of bearings I do use. It does double duty as well, by pushing and pulling.

[philbur]

You miss my point as always. I am talking about the housing/shaft tolerences not the bearing tolerences. Chinese built spindles regularly fail to meet the required tolerences on the housing/shafts.!!!!!!!!!!!!

If the bearing housing or shaft does not have the correct tolerence then trying to shrink fit the bearing can be akin to pissing into the wind.

I'm not disagreeing with you I am pointing out that not all spindle assemblies are equal. Talk to your average 9x20 lathe owner about correct spindle bearings fits. He will laugh.

Phil

Steel is steel. Bearigns all have the same ID size and ID tolearance. The bearing makers all ahve their own recommendations for tolearnce RECOMMENDATIONS for their bearigns.

NSK, Torrington, Fafnir, SKF all have engienering manuals.which outline their recommended stamdars" for spoindle fiat and finish. I"ve yet to see catalogs or manauals for any of the "no name" chinese bearing makers.

Buy a chinese spoindle, get chinese quality. (which improves almost daily). BUy Hardinge or Haas, you get Hardinge/Haas. You get waht you pay for.

NSK tricked up bearings suck jsut as bad on a FUBAR'd spindle as anybody's. Chinese bearings don't have the quiteness/runout of the NSK's on a "piece of jewelry" spindle.

Put lipstick on a pig and you still have a prettied up pig.

{Jo;.you and I had just better agree to disagree. WE've never convinced each other in the past (4th axis food fight) - I don't expect this issue to differ..

[NC Cams]

Heat gun is not hot enough

LEt small bearings sit on a 100w light bulb. Larger ones on a spot light.

SLip fits are called transition fits and they pose no problems. ONce you scotch brite them, they usually slip with the slightest of warming.

Putting ABEC 7 bearings on a "abec 1" shaft is not necessarily a good idea. THe running accuracy differences can put the bearing into a bind, figuratively and literally. I wouldn't go above an ABEC 3 on a turned shaft. IN today's market, the ABEC 1's (most unmarked bearings) that are easily ABEC 1 can also readily meet most ABEC 3 specs.

THis agrees with the point that PHil makes. Appologies to him are owed as he caught something I forgot to pontificate about.

[LeeWay]

My heat gun has two temp settings. 1000 F is the highest setting.
Pretty dang hot.

[NC Cams]

We use a recycled electric skillet

It does a better job of conductively putting heat into the rings and NOT the seals than either torch or heat gun. Tis bigger than any bearing I"ll ever work with. Easily holds 10-12 little ones. Even has thermostatic gage

Paid all of $3.50 for it. Teflon lined too for the ever important no stick feature..

[keen]

Hi NC Cams - I like the idea of thermal fit in theory. In the case of the Tormach Spindle I assume assembly would be something like - (correct me if wrong.

Heat the housing/chill the bearings, and drop them in opposite ends, so hold them in place until the bearing heats up.

Heat up the housing and bearing assembly alittle more, chill the spindle, and slip it in.

Now that is the part that would worry me as the bores are quite small, the shaft may have quite a interference fit - and the coefficient of linear expansion/contraction to allow assemble the whole way in - IE not warming the spindle and causing partial assembly 'lock up', or melting out the grease etc if more heat was needed - or do you reccomend applying the grease after assembly?.

Or is there a better procedure?

[NC Cams]

IF the bearings go into opposite ends of the housing, that does make things interesting.

YOU have the sequence pretty much spot on.

WHen you get the bearings chilled and housing heated, that part is easy. Temps that i'd use for starters would be 180 deg for hsg and deep freeze the bearings. These should drop together FAST and easy.

If you have a PDF ot the cross secotion, I could give better insight on how to insert the spindle into the housing. Right now, I can't picture things clear enough

Keep in mind that this is not the preferred way to do a spindle. Should the spindle grow from heat, you end up affecting any preload you end up with.

Frankly, this sounds like an inexpensive "hobby" grade spindle. Most precision spindles FIX and preload the work end of the spindle and fix the top bearing to the spindle. THe top bearing is then left to Float axially in the housing to take care of thermal growth.

I could probably tell you how to tune the thing up, but not without pictures of a spindle in cross section.

[zephyr9900]

Frankly, this sounds like an inexpensive "hobby" grade spindle.

With respect for your experience and knowledge, NC (and I acknowledge that it is considerable), that is the only spindle under discussion. It is the only design Tormach has, and it is the only one all we Tormach owners are using. If your advice has not been specifically aimed at this spindle and its characteristics, on what have you been basing your recommendations in this Tormach-specific forum?

Best regards,

Randy

[NC Cams]

As I recall, most of the machine tool grade spindles had ground surfaces for spindle shafts and ground/honed surfaces for the housing. These were often half the tolerance for grade 5 or 7 bearings. As machined were considered "hobby grade".

In a prior post, it was pointed out the grade 5 or 7 bearings were not supplied with the machine - they were or might have been offered as an option. From personal experience with Sherline which is a precisely built HOBBY machine, the use of ABEC 5 or 7's on this sort of machine is like putting lipstick on (or in, in this case) a pig.

I can't tell you where to find the recommended specs for other bearing manufacturers bearings - I only worked for 1. I got familiar with their recommended specs. THus I can only cite chapter and verse from their hymnal. Until or unless some other expert from another company shows up to spout their mantra, I guess I"m your source.

As machined/turned spindles and housings are often selectively and/or hand fitted. Hard to do when you only have 1 set of bearings available to choose from. Bridgeport used to do this - they had 5-6 different inner and outer bearings to selectively fit from. Imagine stocking 25-36 bearing p/n's to build 1 spindle.

I"ve tried to splain things from a bearing standpoint. I don't mean to offend with regard to my comments - clearly I may have. However, your spindle design as I percieve it has some unique challenges built into it with regard to servicing it. Challenges as in built in design and machining issues that make the "simple" act of R and R of bearings a bit of a careful exercise.

LEts make it easy on everyone.

SOmeone scan a cross section or an exploded view of the spindle along with the fits (press, loose etc) of the bearing at the position. I can then make realistic suggestions rather than try to fit generic machine tool suggestions to what I consider to be a hobby grade spindle. WHen you see the specs of MACHINE TOOL GRADE spindle shafts and housings and ABEC 5 or 7 bearings used there in, you'll see why I say "hobby grade" when referring to this machine.

PM me with the drawing/scan as I'll no longer follow this thread to look for it. My acerbic natured good intentioned inputs have clearly caused too many raw nerves.

Btw,had I not ambled along, the stupid, irresponsible and ruinous act of pressing a separated AC back together would probably have gone UNCHALLENGED. I might piss you off but that stunt would have ruined ANY good set of A/C's.

No more bearing info until I see drawings. It is too hard to anticipate everything by the hunt and peck acerbic braille method that I"ve used so far and acccordingly have pissed people off with..

Then again, I guess I just don't get along well with "rormac people".

[zephyr9900]

NC, first let me apologize for omitting the smiley from my last post. It was meant in good humor. But really, to have several screens of really good bearing fitting info and then "oh, if you have a Tormach spindle never mind" was a bit of a let-down...

The following screenshot is from the publicly-available http://www.tormach.com/document_libr...View&Parts.pdf , which is the only information I have available. The upper bearing set is called out as "DT7008/DT (double tandem pair)" and the lower set is called out as "D7707C/DT (double tandem pair)". Item 73 is a spacer and item 13 is a split nut (setscrew spreads it axially to lock it in place). Conceptually it doesn't look any different than my former Sherline's spindle suspended between opposing bearings (as contrasted to my Hardinge clone lathe with duplex nose bearings and a floating rear bearing...)

Tormach's http://www.tormach.com/document_libr...gnAnalysis.pdf doesn't address the spindle bearing design philosophy at all.

I hope this helps, and thank you for your input. I did appreciate your story about the Santa Fe traction motor rebuilder (I'm enough of a railfan to have read through the obfuscation...)

the stupid, irresponsible and ruinous act of pressing a separated AC back together would probably have gone UNCHALLENGED

I read that totally different, NC. I think you were talking about reassembling a bearing, whereas keen was talking about reinserting intact bearings into a housing. Maybe I took it wrong--keen's photos certainly showed intact bearings being pressed on/into the shaft/housing...

Best regards,

Randy

[keen]

I read that totally different, NC. I think you were talking about reassembling a bearing, whereas keen was talking about reinserting intact bearings into a housing. Maybe I took it wrong--keen's photos certainly showed intact bearings being pressed on/into the shaft/housing...

Yes NC misunderstood my comment - it was a specific reply to Philburs question - (I can explain this fully if anyone is interested).

But thats OK.....
What really interests me is the direction this is now going. NC misunderstanding my comment has got him involved in what I hope will be a constructive critic of the Tormach head/spindle design. I hope Tormach/Greg is following this also as a endorsement or refinement of the current design would be good for all.

Thanks Randy for posting the drawing - I look forward to your comments NC.
I have studied machine tool spindle designs myself and can see some strengths and weakness of the current design. Before I intrude and perhaps bore you all with my thoughts, Lets see what NC has to say...