Hermle Universal index table - Page 2
Close
Login to Your Account
Page 2 of 5 FirstFirst 1234 ... LastLast
Results 21 to 40 of 90
  1. #21
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    I've just pulled out the M16 DIN2080 drawbar:



    The way you get it out is interesting. It is retained with the lower threaded part (top right in the photo) that threads through the bevel gear that rotates it. To prevent the two threaded parts from rotating a 5 x 28mm pin is pressed through a radial hole. The pin is rolled from thin spring steel. To get it out you nod the table top towards the mill, then reach underneath with a 4.5mm steel pin and hammer, and knock it out. I was very happy that it was not a frozen taper pin! Then you can simply unscrew the drawbar.

    As expected, Deckel tooling fits perfectly: you just remove the drive lugs:

    I'm going to buy the correct drawbar from Hermle, the cost is low enough that it's not worth making it myself.
    Last edited by ballen; 02-14-2021 at 06:48 PM.

  2. #22
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    The drawbar (see photo in the previous post) is actually made from three pieces. In fact I did not need to remove it to change from M16 to S20x2. Here is how it is assembled:



    On the left is the part that is fixed in the table (which I did not need to remove). On the right hand side of that part the thread is "hollow" and slit in four places like a collet. In the middle is a taper-head screw (takes a 6mm hex driver) which expands the slit thread to clamp it on the right-hand part. Finally on the right-hand side is the part which gets exchanged to shift between M16 and S20x2. If you make a special tool you can remove this without taking the drawbar out of the machine as I have done. You will need a hollow tool that grips the two notches, and then a very long M6 hex driver (~150mm = 6" long) which fits through the hollow tool, to loosen and tighten the M6 screw. See drawing below.

    The helpful guy at Hermle gave me this set of instructions, translation below:



    Drawbar:
    The machine is normally fitted with an M16 drawbar thread (2). At additional cost an S20x2 fitting (3) can also be delivered.

    If an exchange is needed, carry it out a follows:

    The conical screw (1) should be loosened about 1/2 turn with an extended hex driver (4). The threaded fitting (2 or 3) can then be screwed off or on using a special wrench (5). The normal accessories include a (different) wrench which can be used to hold the drawbar fixed. After the threaded fitting has been fully screwed into place, the conical screw (1) should be tightened. (Wrenches 4 and 5 are available for purchase.)
    Last edited by ballen; 02-19-2021 at 10:19 AM.

  3. Likes michiganbuck liked this post
  4. #23
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    I got the parts from Hermle today, and installed the S20x2 drawbar. Works perfectly.



    I mounted a dial indicator on the top face of the taper and swept the cast iron "bearing surface" that supports the table. There is less than 5 microns on runout at 300mm diameter. The table is in good shape!

    I put a test bar in the taper and tested the runout. There was no motion on the dial indicator (this is 0.01mm = 0.0004" per division), so there can't be more than a micron or two of runout. Here's the video:



    By the way, I am using the drawbar locking gearing to spin the table. I don't like the way it sounds, but am not sure if that actually indicates a problem. It's a pair of straight bevel gears.
    Last edited by ballen; 02-20-2021 at 06:42 PM.

  5. Likes Mud, sigurasg, TNB liked this post
  6. #24
    Join Date
    Feb 2005
    Location
    Vancouver B.C.
    Posts
    862
    Post Thanks / Like
    Likes (Given)
    16
    Likes (Received)
    140

    Default

    Rather impressive that you can still get small parts for this machine from Hermle.

  7. Likes ballen liked this post
  8. #25
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Quote Originally Posted by Chris999 View Post
    Rather impressive that you can still get small parts for this machine from Hermle.
    Agreed. Even more remarkable: the prices are reasonable. The S20x2 fitting plus a new conical screw cost 70 Euro including shipping and 20% VAT. Perhaps the reason they still have these parts: they must use the same drawbars on some of their mills, because the instructions they provided (posted above) are for the drawbars of a milling machine.

    There are parts that Hermle no longer supply. For example I asked about this rubber protection sheet, which has some small tears:



    and they said it's no longer available. But it's easy enough to cut a new one out of 1mm NBR (Ross, that will wait until a repaint).

    Chris, do the gear teeth on your drawbar sound like my video? Or are they quiet?
    Last edited by ballen; 02-21-2021 at 04:33 AM.

  9. #26
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    OK, here's something weird. The top of the rotary spindle is 89.00mm diameter, ground and true to the rotation axis. Guess what the ID is of the standard Deckel dividing head tooling?



    Well, it doesn't look like it in this photo because of the perspective, but you got it, 89.00mm. So if the people at Hermle had just made their top a few mm higher and provided four locking holes, the Deckel accessories could have been mounted directly on the Hermle spindle. Go figure.
    Last edited by ballen; 02-23-2021 at 04:33 AM.

  10. #27
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,709
    Post Thanks / Like
    Likes (Given)
    2256
    Likes (Received)
    3101

    Default

    Bruce:
    Bet if you mounted a ring on top of that spindle held with counter sunk Allens with its lower edge machined with an angle at the correct height , the locking screws would just wedge in and pull everything
    down,nice and tight.....
    Cheers Ross

  11. Likes ballen liked this post
  12. #28
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Hey Ross!

    Quote Originally Posted by AlfaGTA View Post
    Bet if you mounted a ring on top of that spindle held with counter sunk Allens with its lower edge machined with an angle at the correct height , the locking screws would just wedge in and pull everything down,nice and tight....
    That's a great idea! If I'm following you right, the idea is to use the existing ground OD for radial registration. So my part would need to be bit under, say 88.96mm OD. It would need to have a top surface that runs true, to register the top flat underside of the Deckel accessory flanges. But that's easy, because I could machine it in place.

    Have I got it?

    By the way, I agree with what you wrote earlier, that the lower bearing surface does not look good. Curiously, the upper surface on the table is scraped for bearing and with oil pockets. Given that it's ~50 years old and still runs true, I'm probably not going to wear it out in my lifetime, cranking it by hand. I wonder if they flycut it then ground it, but left some of the flycutting marks behind as oil pockets.

    Regarding an encoder, Hermle did make a version of the table with a Heidenhain ROD 250 encoder built in (18000 lines per rotation). Trouble is that you have to give up the SK40 drawbar to get that, and I'm not ready to do that. Anyway, when I eventually take it apart for cleaning and painting, I'll see if there is a decent way for me to add an encoder above or below the ring gear. There are some nice Renishaw optical disk encoder ring/read head combinations that might fit in there very well.

    Cheers,
    Bruce

    PS: would I be better off with normal SHCS or with tapered countersunk heads?
    Last edited by ballen; 02-23-2021 at 04:26 AM.

  13. #29
    Join Date
    Aug 2002
    Location
    France
    Posts
    1,728
    Post Thanks / Like
    Likes (Given)
    330
    Likes (Received)
    787

    Default

    Is it me, or do pictures now take a while before showing in the threads (may be a few hours to one day or two) ??
    And of course I'm logged in...

  14. #30
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Quote Originally Posted by TNB View Post
    Is it me, or do pictures now take a while before showing in the threads (may be a few hours to one day or two) ??
    Tien, for me they appear instantly. But I think PM may be using some web caching services for European traffic, and those might be different in France than in Germany. That could be the cause.

  15. #31
    Join Date
    Nov 2004
    Location
    SLC, UT
    Posts
    793
    Post Thanks / Like
    Likes (Given)
    1111
    Likes (Received)
    209

    Default

    CAT40 receptacle always seemed like a good idea, but upon reflection it seems pretty small for my uses, like mounting a 6" lathe chuck.
    Find the encoder my Deckel Universal table extrremely usefull and would not want to give it up.
    Genuinely curious- what is the intended/typical use for this mounting interface on a universal table?

  16. #32
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,709
    Post Thanks / Like
    Likes (Given)
    2256
    Likes (Received)
    3101

    Default

    Bruce:
    I am not holding the table in my hands (duh) But looking at the pattern on the base face for the table, looks that it might have been "Blanchard" ground...
    Fairly coarse feed given, but the circular cross hatch is a tell tale.

    Your evaluation of my idea to mount the Deckel dividing head tooling is accurate....Might change the setup to make it easier.....Make the disc have a 40 taper shank so that the disc could be mounted via the tapered socket and drawbar....

    Steve:
    Think the idea of the 40 taper socket is that it then doubles as a dividing head fro small work...Stuff that is better held in a collet.

    Could easily do the same with a fanged mounting that could bolt directly to the table face of your Deckel tool makers table.....If you wished you could even incorporate a draw bar through
    the center hole in the table.....
    I have an 8" Bison chuck with reversible jaws (can take soft jaws as well) mounted with a flat table mounting (factory) that i use all the time for round work on my FP4NC....easier to mount than removing a table top.
    On the NC usually don't need to rotate the table and if you indicate to the part, the holding accuracy is immediately assured...
    Cheers Ross

  17. Likes Colt45, ballen liked this post
  18. #33
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Quote Originally Posted by AlfaGTA View Post
    But looking at the pattern on the base face for the table, looks that it might have been "Blanchard" ground.
    That makes sense -- it's very flat, and the "fly cutting marks" are very shallow, so they are more likely from Blanchard grinding.

    Your evaluation of my idea to mount the Deckel dividing head tooling is accurate....Might change the setup to make it easier.....Make the disc have a 40 taper shank so that the disc could be mounted via the tapered socket and drawbar....
    Main question is, would the support be more rigid if it is bolted to the spindle face? Or would it be more rigid if supported on an SK40 taper but not bolted to the spindle face? What do you think?

  19. #34
    Join Date
    Aug 2002
    Location
    France
    Posts
    1,728
    Post Thanks / Like
    Likes (Given)
    330
    Likes (Received)
    787

    Default

    Quote Originally Posted by ballen View Post
    Main question is, would the support be more rigid if it is bolted to the spindle face? Or would it be more rigid if supported on an SK40 taper but not bolted to the spindle face? What do you think?
    Bolted to the face anyday.

  20. Likes AlfaGTA liked this post
  21. #35
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Hi Tien,

    Quote Originally Posted by TNB View Post
    Bolted to the face anyday.
    Maybe, but I'm not sure. There are four M12 bolts, so the stiffness corresponds to the elastic modulus of four rods, each about 10mm in diameter. Total cross-sectional area is about 30 square mm. In comparison, the SK40 option is about 40 mm diameter at the top, cross sectional area perhaps 1200 square mm.

  22. #36
    Join Date
    Aug 2002
    Location
    France
    Posts
    1,728
    Post Thanks / Like
    Likes (Given)
    330
    Likes (Received)
    787

    Default

    May be, but the bolts are best located (read : on a wider circle) to handle the loads, than an plain central arbor.

    Plus the flat surface of the spindle nose also plays a big role in the stiffness of the assembly.
    In that regard, I wonder if a central arbor would work the same way as the four bolts : a central SK40 arbor, will work in flexion, because you'll have to design the assembly with a small gap between the flange of the equipment and the spindle nose to prevent any interference between the taper and the spindle nose/flange.
    On the contrary, in a desgin where a full contact between the spindle nose and the flange (or backplate) of the equipmement is desired, the studs will work in pure traction.
    That may make a difference.

    Think about the way a D1-x or DIN KK lathe spindle nose is designed...
    The short taper is here to locate the backplate, the cross-section of the camlock studs is ridiculous relatively to the diameter of the backplage, but it's the flat surface of the spindle nose that gives all the stiffness.

    Please note that all the considerations above are pure guts (not brain) feeling !

  23. Likes thanvg, AlfaGTA liked this post
  24. #37
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,709
    Post Thanks / Like
    Likes (Given)
    2256
    Likes (Received)
    3101

    Default

    T:
    Your example of the "D" style spindle nose is germane...But remember that both surfaces make up in your example. Both the taper and the flat face...No reason that one could not achieve the same
    dual contact using a taper and flat . The taper does not need to be full length after all and it should be noted that now milling machines are being produced that
    have dual contact spindle noses.....Contact on the face and taper.....
    Just some idle thoughts after all.....Might be an interesting exercise fro someone with an accurate cylindrical grinder.....

    Cheers Ross

  25. Likes thanvg liked this post
  26. #38
    Join Date
    Feb 2001
    Location
    Redwood City, CA USA
    Posts
    5,247
    Post Thanks / Like
    Likes (Given)
    226
    Likes (Received)
    1108

    Default

    Quote Originally Posted by TNB View Post
    Plus the flat surface of the spindle nose also plays a big role in the stiffness of the assembly.
    The short taper is here to locate the backplate, the cross-section of the camlock studs is ridiculous relatively to the diameter of the backplate, but it's the flat surface of the spindle nose that gives all the stiffness.
    Tien
    You're brain is AOK! You have it exactly right, and the situation is the same as a properly tensioned bolted joint. Like a bolt or screw, when the pin is properly tensioned, the stiffness of the assembly includes the stiffness of the clamped material, which typically huge due to larger area compared to the bolt or pin. It is only after enough force is applied to fully decompress the clamped material and make the joint start to separate, that the joint stiffness is governed only by the bolt or pin stiffness. Hence, keep your car's lugnuts and D-mount camlocks torqued.

  27. #39
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,551
    Post Thanks / Like
    Likes (Given)
    2307
    Likes (Received)
    1407

    Default

    Quote Originally Posted by rklopp View Post
    The situation is the same as a properly tensioned bolted joint. The stiffness of the assembly includes the stiffness of the clamped material, which is typically huge due to larger area compared to the bolt or pin. It is only after enough force is applied to fully decompress the clamped material and make the joint start to separate, that the joint stiffness is governed only by the bolt or pin stiffness.
    This makes sense, so I think Tien is right. In other words, if I have an extension ring bolted to the top of the spindle face, and force is applied to this, the force can

    (1) push the ring against the face,
    (2) pull the ring away from the face
    (3) move the ring tangent to the face, or
    (4) rotate the ring on the face

    For case (1) if the fit is good, the amount of motion is small because one must compress the entire area of the spindle face.

    For case (2) in order to generate any motion, one must first provide a force larger than the pre-tension of four M12 bolts. This is about 48kN for one class 8.8 bolt, so 4.8 metric tons per bolt.

    For case (3) it depends if the accessory flange overlaps the spindle or not.

    -- If the accessory flange overlaps the spindle by (say) 5mm, then the cross section resisting the force is 5mm x 89mm (diameter) =445 square mm. That would be very stiff.

    -- If the accessory flange does not overlap the spindle, then we rely on friction. The coefficient of static friction of greasy smooth steel on smooth steel is about 0.05 (see below for reference). So this will move when the force exceeds 1/20th of the clamping force. That's about 48kN x 4/20 = 9.6kN so about 1 metric ton. That's not so good.

    For case (4) the analysis is the same as case (3) without overlap. To resist the torque, drive keys are probably needed.

    Conclusion: if the accessory mounting flange overlaps the spindle by a few mm, and there are keys to prevent rotation, then the bolted-on solution will be very stiff. This is the best solution, but I'll have to make a good drawing to see if this is possible.

    If there is no overlap, then sufficient stiffness will require a feature that interferes with horizontal motion. This could be a short taper section inside the spindle, at least a few mm long, or an overlapping feature of a few mm on the outside of the spindle. As Ross says, an interesting exercise in ID or OD grinding.

    Note: this analysis explains the design for clamping the table onto the spindle. There are two M12 bolts to provide vertical clamping force, an eccentric cam mechanism operating on a split ring to ensure a snug radial fit, and keys to block rotation.

    Cheers,
    Bruce


    For the coefficient of friction of greasy smooth surfaces, I took the worst case (lowest coefficient) shown in Figure 7 of H. S. White and Dino Zei, Static Friction Tests with Various Metal Combinations and Special Lubricants, Journal of Research of the National Bureau of Standards Vol. 46, No. 4, April 1951, Research Paper 2198) https://nvlpubs.nist.gov/nistpubs/jr...n4p292_A1b.pdf

  28. #40
    Join Date
    Aug 2002
    Location
    France
    Posts
    1,728
    Post Thanks / Like
    Likes (Given)
    330
    Likes (Received)
    787

    Default

    Quote Originally Posted by AlfaGTA View Post
    T:
    Your example of the "D" style spindle nose is germane...But remember that both surfaces make up in your example. Both the taper and the flat face...No reason that one could not achieve the same
    Yes, but that kind of arrangement is much more difficult to make. Plus in Bruce's situation, the centering of the attachment would be provided by the OD of the spindle nose. So no need (nor even possibility) for the short taper.

    Quote Originally Posted by ballen View Post

    -- If the accessory flange overlaps the spindle by (say) 5mm, then the cross section resisting the force is 5mm x 89mm (diameter) =445 square mm. That would be very stiff.


    Conclusion: if the accessory mounting flange overlaps the spindle by a few mm, and there are keys to prevent rotation, then the bolted-on solution will be very stiff. This is the best solution, but I'll have to make a good drawing to see if this is possible.
    I don't think it's that simple.

    If you make your accessory flange so that it overlaps the spindle, you'll also have to make it high enough *above* the spindle nose so as to have enough meat.
    You don't want your mounting flange to loolk like a "shouldered washer" !
    But that will bring your vice (or whatever attachment) away from the spindle bearings, wich will necessarily translate into a loss of rigidity so whathever gained could be lost.
    I agree that this may be negligible, since height increase won't probably exceed something like 35mm, but anyway...

    Plus I think your calculation of the resisting force is not valid. If you consider a radial load, the resisting force of the overlaping portion of the mounting flange will decrease as you go away from the point where the load is applied.
    Say one apply a purely radial load at 12:00. The resisting force will be max at 12:00, but will decrease toward 03:00 and 09:00 where it will be zero (and it will be close to zero well before reaching those points.
    So it's not simply a matter of the overlapping portion x diameter.

    After thinking about it, I think I have a neat solution to offer.

    What I would do is get a cheapo Deckel MK4 centering vice that I could butcher without hesitation.
    I would chop the threaded part of the vice mount, down to the bottom of the bore (ending with a large flat disc). I would then bore and thread the center of the vice base, so as to accomodate a properly sized "drawbar".

    The drawbar would look like an SK40 threaded mill holder, but with the taper reduced to a short taper.
    Of course, the front end would be threaded accordingly to the threaded bore machined in the vice mouting base.

    The last feature would be a snug fit key bolted on the flat part of the vice base, that would nest in one of the spindle nose slots to prevent any rotation under hard use of the vice.

    Once everything ready, I would mount the drawbar without tightening it, mount the vice with some loctite on the threads, tigthen the vice then thighten the drawbar so as to insure a perfect contact between both the spindle nose/flat and spindle/drawbar tapers. I would let cure, then mount the locating key.

    Best of both worlds.
    The large flat would provide as much stiffness as possible, the short taper would handle any radial loads.
    The making would be absolutely easy, the correct interference between the taper and the flat beeing obtained after machining both parts separately.
    Plus no height loss under the spindle, and a very quick on/off the machine operation.

    So whatcha say 'bout that ?


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •