Announcement

Collapse
No announcement yet.

Bridgeport Spindle Preload

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • sauer38h
    replied
    <font face="Verdana, Arial" size="2">Originally posted by wierdscience:
    Since the greater the contact angle equals the greater radial load capacity </font>
    Huh? Where are you measuring contact angle from? Zero contact angle means the balls are running in the center of the raceways, and radial capacity is maximized. They can't take any axial load and stay in that configuration, of course.


    [This message has been edited by sauer38h (edited 02-04-2006).]

    Leave a comment:


  • wierdscience
    replied
    On second thought,since your bearings are"standard"that should indicate standard flush ground races,which would provide the minimum contact angle of 15*.Since the greater the contact angle equals the greater radial load capacity it is possible that the factory bearings were custom grinds.If your old bearings were factory spec grinds,then they won't be flush ground.You may be able to measure any amount of offset grind by simply using a mic to measure the width of the inner and outer races on the old bearings.That dimension would yield the correct preload setting for the new bearings.Basically it would tell you how much difference you would need to machine into the spacer set.I'm betting the inner races of those bearings(old ones) are narrower than the outer races.If thats the case your inner spacer will have to be ground shorter by twice that amount to provide the preload you need for the flush ground bearings you have.

    Leave a comment:


  • wierdscience
    replied
    Take the thing apart and check the spacers with a tenths reading mic,they should be both THE SAME LENGTH.Do this before going any further.If one or both spacers is spun,has a ding or is just dirty you will not get the proper preload no matter what you use for bearings.

    After doing that if you still have up and down motion you will have to measure than and correct the grind on the spacers.The easiest way to accurately measure the endplay is to bring the spindle nose into contact with the back jaw of your vise and clamp the nose in it.Position a dial indicator to read off the end of the quill housing and use the quill handle to move the quill up and down.The reading you get from this is what must be removed from the inner spacer ONLY.Once that amount is removed,reassemble the quill and make a test cut,at that point your problems should be over.The spindle should have a slight drag,this will disappear once the machine is up to operating temp because the spindle WILL expand in length as it reaches temp.

    The temps should never climb above 150-160*f,if this occurs you must remove(by lapping)material from the outer spacer only.

    Am I to assume that you paid $380 for off the shelf bearings when the exact replacements are availible here for $303.00,click on part #52.

    http://www.alternativemachinetool.co...9290f9e08e554b

    Leave a comment:


  • sauer38h
    replied
    <font face="Verdana, Arial" size="2">Originally posted by Peter S:
    Question for bearing gurus:
    ** Out of interest, I am not sure how the pre-load is provided - are the spacers identical length (ie ground together) with the preload set by a difference in height in the bearing races - or are the spacers different lengths?
    </font>
    The usual way to do it is to buy a duplex pair and clamp them together in a suitable housing. The preload will then be whatever was ordered, if the bearing manufacturer did his job right. Most commonly the outer races are clamped together, and the inner races will then press against each other, take up the play, and provide the designed preload.

    However you get the same effect if both raceways are separated by spacers of identical length. The outer races get clamped down, and the inner races are then squeezed by exactly the same amount that they would be if they were in direct contact.

    The tolerances involved are very tight. Think of it as a spring preload with a very stiff spring - the elastic constant of the steel. So standard practice is to buy pairs of bearings from a competent manufacturer, and grind the spacers together in one setup so they come out the same length.

    Note that spacer separation is not something done casually - it is an integral part of the spindle design. Spindles made with duplex bearings may have a duplex pair at one end and a single bearing at the other, or may have duplex pairs at each end, or may have a single pair with long spacers positioning the halves of the pair at opposite ends of the spindle. Those are the simple arrangements, I've seen others too. As usual in design, there is no "best", the choice is a balance of tradeoffs in a specific application.



    [This message has been edited by sauer38h (edited 02-03-2006).]

    Leave a comment:


  • aboard_epsilon
    replied
    Just had a think about this.
    tread carefully .........once you grind them spacers , they will be useless if you ever have to go back to the genuin type bearings.
    Think when you send the spindle to bridgeport or a reconitioner, they re-use your spacers.

    all the best.mark

    Leave a comment:


  • JCD
    replied
    I don’t know about the rest of the world but, my Bridgeport J head spindle calls for Bridgeport part number (code number) 11190238. I talked with a Bridgeport service technician yesterday, he stated that, from the drawing of the spindle, the lower two bearings are: JM207K-PRB-DB/A/2729. I talked with a customer service representative from Bridgeport today; he confirmed the number of the bearings. The bearings I removed from my spindle are marked MRC 207 S D1 ABEC X. According the information published by MRC the bearings are Special 8 15/32 Dia. ball ABEC 7 with MRC exc. race modifications. According to information published by the bearing manufacturers, the MM207K bearings I installed in the spindle, as replacements are the same.

    From the numbers above, all of the bearings are CONRAD type bearings. They look like Conrad bearings. When I measure the outer race width with a dial caliper, it is symmetrical on both sides, i.e. the same on both the front and the back of the bearing. When I measure the width of the inter races, it is also the same on both sides of the bearing. I am not measuring the thickness of the races, which is the same as the thickness of the bearing. The width of the race I am measuring is developed by the differences between the outside diameter and the inside diameters of each race.

    By reason of the above, I am assuming the correct bearings for my spindle are Conrad bearings.

    After talking with engineers employed by two different bearing manufacturers that have, or do, supply the spindle bearings to Bridgeport for this application, I find that one of the special features of the bearings used in the application is pre set preload. This is accomplished during the manufacturing process by adjusting the relative positioning of the ball grooves to the face of the bearing races so that when the bearings are “at restâ€‌, or not installed, there is an offset between the faces of the races. Then the bearings are installed between the correct spacers in the machine the preload is established. In my spindle the correct spacers between the outer and inner racer are of the same length thaw will establish the correct preload for my spindle.

    Now, as I did not purchase the “special bearingsâ€‌ from Bridgeport, and have made an error, and purchased $380.00 dollars worth of Timken MM207K “standardâ€‌ ABEC 9 bearings, ABEC 9 was substituted for ABEC 7 at the same price because of availability, from the local bearing supply house, I am still faced with the task of adjusting the preload of these bearings to the proper specifications so that I will obtain the designed spindle performance. As these are “standard bearingsâ€‌ no preload is “built inâ€‌ to the pair.

    If you can give me some indication as to what preload may give a 20 to 30 degree temperature rise in this application, as noted above. I would sure be grateful. Or even someone could help me understand what a significant amount of preload may be for the application. I just don’t know if .001â€‌ is significant or if .005â€‌ or .0001â€‌ is significant. From the bearing manufacturers as I stated above there specifications are given in pounds of force. I understand why. But that doesn’t solve my problem either.

    HELP!

    P.S. I am assuming I have a preload problem, because I have .005â€‌ vertical movement, .0015â€‌ horizontal movement, .0006â€‌ eccentricity, and when spindle is loaded sideways, by hand, it rattles, the spindle chatters when machining, and when I measure the clearance between the outer bearing spacer and the outer bearing races, in the assembles spindle, a .008â€‌ gap is present.

    Leave a comment:


  • wierdscience
    replied
    <font face="Verdana, Arial" size="2">Originally posted by aboard_epsilon:
    quote"You can remove material from either spacer,but that usually isn't required."


    what if you could not afford a set of bearings ..would removing material off the spacers take up the wear in the bearings and allow perhaps a few more months/years out of them tired bearings.

    all the best.mark

    [This message has been edited by aboard_epsilon (edited 02-02-2006).]

    [This message has been edited by aboard_epsilon (edited 02-02-2006).]
    </font>
    Sure,why not,got nothing to lose.The spindle in question you would remove material from the inner spacer.Not very much thou,we're only talking .0001" or less.Lapping would be the method I would use.

    Leave a comment:


  • aboard_epsilon
    replied
    quote"You can remove material from either spacer,but that usually isn't required."


    what if you could not afford a set of bearings ..would removing material off the spacers take up the wear in the bearings and allow perhaps a few more months/years out of them tired bearings.

    all the best.mark

    [This message has been edited by aboard_epsilon (edited 02-02-2006).]

    [This message has been edited by aboard_epsilon (edited 02-02-2006).]

    Leave a comment:


  • Peter S
    replied
    Darn it, pressed the wrong button...

    [This message has been edited by Peter S (edited 02-02-2006).]

    Leave a comment:


  • wierdscience
    replied
    <font face="Verdana, Arial" size="2">Originally posted by Peter S:
    JCD,


    The pre-load on these spindles is not "adjustable".
    It is not a Bridgeport secret, because you don't need to know it (unless you have lost the factory spacers, even then I am not sure if that matters**).

    When you tighten the nut on the spindle, the bearings and spacers pull up and the preload is automatically set. It doesn't matter what you do, you can't alter this preload.

    To change the preload, you would have to remove or add material to one of the spacers which goes between the paired bearings.

    Question for bearing gurus:
    ** Out of interest, I am not sure how the pre-load is provided - are the spacers identical length (ie ground together) with the preload set by a difference in height in the bearing races - or are the spacers different lengths?
    </font>
    That is exactly how it works,you can tighten that nut down until the threads give and not change a thing.That's why when those bearings wear you throw them away and replace them.

    All the b-port spindles I have worked on have had both spacers exactly the same length.

    You can remove material from either sapcer,but that usually isn't required.

    Leave a comment:


  • Peter S
    replied
    JCD,

    I am guessing anything further is going to muddy the waters???

    I am not sure if you have the correct angular contact bearings, nor if you have installed them in the correct configuration (ie back-to-back).

    However, there is one thing I am not sure you are clear about:

    The pre-load on these spindles is not "adjustable".
    It is not a Bridgeport secret, because you don't need to know it (unless you have lost the factory spacers, even then I am not sure if that matters**).
    Edit note - it is not a BP secret because the pre-load is built into the bearings. There is no secret to give away...

    When you tighten the nut on the spindle, the bearings and spacers pull up and the preload is automatically set. It doesn't matter what you do, you can't alter this preload.

    To change the preload, you would have to remove or add material to one of the spacers which goes between the paired bearings. Edit note: Do not even think of doing this!

    Question for bearing gurus:
    ** Out of interest, I am not sure how the pre-load is provided - are the spacers identical length (ie ground together) with the preload set by a difference in height in the bearing races - or are the spacers different lengths?

    Edit note - I think I just figured out the answer to my own question. The preload must be in the bearings themselves, because they are sometimes (often?) used in back-to-back pairs without any spacers......

    [This message has been edited by Peter S (edited 02-02-2006).]

    Leave a comment:


  • Mark Hockett
    replied
    JCD,
    I dont have your email anymore. Send me an email, I might be able to help you with the bearing problem. You can also call me at 360-914-6026.
    Mark

    Leave a comment:


  • Forrest Addy
    replied
    It's true that "conrad" or radial ball bearings have a thrust capacity roughly double their radial capacity and they can be preloaded in pairs to eliminate end float as is done with a wavey washer spring in electric motors. Doing so results in a situation where the ball path describes a cone angle but one too shallow for the precision axial restraint needed in a machine tool spindle.

    One can calculate this cone angle from the ball diameter and half the assembled axial clearance. I took a rough estimate from a 207 radial ball bearing out my collection and got a 1.3 degree cone angle. The cone angle reccommended for Bridgeport angular contact spindle bearings is 15 degrees.

    However axial positioning of a preloaded radial pair is poor in comparison to that of an angular contact pair having otherwise like characteristics. Plot axial deflection per unit of load of the radial pair Vs the angular contact pair and the axial deflection of the one will be several times that of the other. A preloaded radial pair will be downright "spongy" compared to an angular contact pair.

    The spindle bearings commonly found in turret mills built on the Bridgeport pattern are indeed duplex pairs arreanged back to back with a pair of spacers separateing them by an inch or so to enhance rigidity.

    The terms "angular contact" "duplex" and "DB" or "DU" should have loomed large in the discourse between JCD and his bearing supplier. Unfortunately it did not - or - did it? We'll know for sure after JCD reads this and looks futher.

    [This message has been edited by Forrest Addy (edited 02-02-2006).]

    Leave a comment:


  • sauer38h
    replied
    <font face="Verdana, Arial" size="2">Originally posted by Forrest Addy:
    They are made with built-in internal clearance and so have no intrinsic control over axial motion. They are commonly used as "float" bearings to support the driven end of the spindle -indeed there is a conrad or double row flow bearing at the spline end of your spindle. It's not uncommon for a conrad bearing to have 0.0008" assenbled radial clearance and 0.012 axial motion. Thus they cannot be used in the business end to the spindle.</font>
    That's not actually so, if the luxury of pairs of bearings is available. Conrad bearings with even greater clearances start to have the axial load capacity of angular contact bearings, and the excess play can be taken out by loading them axially - which is what the "duplex" arrangement is all about. But so as to avoid muddying the waters, if the Bridgeport spindle doesn't use duplex pairs, I'll say no more about them.

    [This message has been edited by sauer38h (edited 02-02-2006).]

    Leave a comment:


  • Forrest Addy
    replied
    JCD emailed me with an appeal to discuss his his problem. I thought he was being well advised here so I only followed the discussion. However, since he asked me to weigh in:

    My Email text to JCD follows:

    A wide variety of periodic phenomina have been refered to as "chatter". Here's is the procedure for determining whether an angular contact bearing pair has the correct preload.

    Mount a mag base on the extended quill.near the retainer nut. Mount a 0.0001" dial test indicator to the mag base so the contact is registering motion of the spindle collet taper. Rotete the spindle by hand to see if the collet taper runs within 0.0003". If it's OK accept it and move on.

    Move the indicator and install your largest clunkiest end mill or shell mill adaptor. Restore the indicator to register the OD of the spindle. Draw a circle around the contact point with a felt tip pen for repeatable positioning. Shake the spindle by hand strongly attempting to force the spindle sideways. You will see a few tenths of deflection but when you relax the spindle should go back to zero (if the circle is centered under the contact point). Move the indicator so it registers on the spindle end and repeat, shaking vertically. If the indicator registers any free motion at all there is something seriously wrong.

    Remove the mag base and indicator. Stick a small meat thermometer to the lower end of the quill with modeling clay. Set the spindle to high speed and run the spindle full out for a half hour. The spindle should warm up about 10 to 20 degrees F as measured by the thermometer. This is similar to the final check in the most sensitive and accurate machine tool spindles: the temperature control is disabled and the machine run at high speed for some time while the observer looks for a range of temperature rise. If the spindle does not warm up that's in indication of insufficient preload. If the spindle warms up over 20 degrees at thermal equalibrium the preload may be a bit too snug. Over 30 degrees definitely too snug.

    Use a screwdriver as a stethoscope to listen to the bearings: you should hear a soft whir. If you hear any tickings or grumbling you have a problem with dirty assembly.

    As for the chatter you describe: (you don't mention material, cutter details, machine settings, spindle extension etc) it may be an artifact of spindle endplay or it may be the behavior of a properly set preload. The old bearings may have loosened up enough to damp out this phenomenon whereas the new because they are rigid may not. The same phenomena may have different interpretations depending on context.

    Spindle work is tricky. It's not like changing out wheel bearings because you have to be so careful of cleanliness, parts handling, deburring withhout affecting geometery and dimension, installing the bearings so the burnish marks align and all the rest. I always shudder when people have a "chatter" problem who blithly blame it on "sloppy" spindle bearings. There are so many more quite simple causes for the problem. Few seem willing to conduct a few simple diagnostic tests before tearing into the spindle. On the strength of an assumption or what "some guy" told them, a naife will take apart a precision machine tool spindle (often using rough methods) and only then ask questions and get concerned about bearing Classes. of fit, suitable sources etc.

    I'm not suggesting the inexperienced can't do good spindle work. I am suggesting they often bite off more than they can chew thanks to little or no research, blithe assunptions, giving credence to the advice of someone who means well but is ignorant, inadequate tools and equipment, and lack of the excrutiating care necessary for spindle work.

    The circularity of a spindle axis is about 20 times more accurate than that necessary in an asle bearing and 10 times more accurate in an induction motor bearing. The class of workmanship and the cleanliness and ove- the-top care employed in spindle work has to reflect this accuracy.

    End of text. Adding JCD write: "The spindle bearings are Conrad type (single row radial) and after doing a considerable amount of research, including conformation from Bridgeport, Conrad type bearings are the correct bearings for this spindle." Bridgeport spindle bearings are NOT conrad type but an angualar contact matched pair.

    If you installed conrad (actually "radial) bearings (this term refers to non-filling groove ball bearings whose inner race is eccentrically displaced for filling: conrad - concentric radial) there is your problem. If you have indeed installed radial bearings it will be immediarely apparent in the axial shake test. You'll see anywhere from 0.005 to 0.020" end play when there should be none.

    Radial (conrad) bearings can be made to high class numbers and their accuracy and circularity can be held to as close a tolerence as the finest of angular contact bearings. They are made with built-in internal clearance and so have no intrinsic control over axial motion. They are commonly used as "float" bearings to support the driven end of the spindle -indeed there is a raidal or double row flow bearing at the spline end of your spindle. It's not uncommon for a radial bearing to have 0.0008" assenbled radial clearance and 0.012 axial motion. Thus they cannot be used in the business end of the spindle. I cannot imagine anyone knowledgeable at Bridgeport using "conrad" is connection with spindle bearings.

    I strongly suggest you look up "angular contact bearings" and clarify some concepts.

    http://www.ntnamerica.com/Knowledge/...ng_AngCont.htm

    Look at the ourter race of your old bearings. If one side of the outer race is much thinner that the other (look at the cross sections in the link) that's a sure identifier for an angular contact bearing.

    The identifying number for angular contact bearings suitable for spindles is a long affair with many designators. Each designator has to be de-coded to determine if you have a correct cross match with the original spindle bearings.

    Many bearing manufacturers offer off-the-shelf precision angular contact bearings suitable for immediate installation in a turret mill spindle. There's no point to paying a $150 premium for the manufacturer to run the same set of bearings in and out of his inventory if you could purchase from the local bearing supply house and get quicker service.

    JCD, without looking at the whole bearing and its number its hard to say what you have but when you said "conrad" it set off all kinds of alarm bells. If have indeed installed plain radial ball bearing in your BP spindle you've been led to a bad conclusion and your bearing supplier needs to have his a$$ kicked - very hard. There is no remedy but to obtain the correct bearings and install them.

    Edited to insert the word "radial" as appriate. It kept slipping away from me last night.

    [This message has been edited by Forrest Addy (edited 02-02-2006).]

    Leave a comment:

Working...
X