The Rumors are True New HAAS UMC 500 ! - Page 2
Close
Login to Your Account
Page 2 of 17 FirstFirst 123412 ... LastLast
Results 21 to 40 of 335
  1. #21
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by fmari --MariTool- View Post
    Thank you. Since then we have added a Studer CNC grinder. Even more accurate and has 60,000 rpm id grinding capability.
    @maritoolusa on Instagram: “Daily Grind - Day 28 #mth2019 Big shout out to the great guys at Integrated Machinery Systems @integratedmachinerysystems . They did a…”

    Even is "form" or angle is more accurate on a spindle taper it does not have any effect on concentricity due to elastic deformation of the spindle. Unless of coarse angle if off so much that spindle cannot deform enough.
    That's a brilliant point about elastic deformation of the spindle.

    Maaaaaan ohhh man that ID grinding spindle is massive, beyond a beast a total monster on the "God machine" being the Studer grinder... Unreal / amazing. That should last at least 20 years right ?

    Do you guys ever take on custom work that is not 100% tool related / other application areas that are quite similar or is your production line and focus iron clad/ immutable ?


  2. #22
    Join Date
    Nov 2005
    Country
    UNITED STATES
    State/Province
    Illinois
    Posts
    1,978
    Post Thanks / Like
    Likes (Given)
    209
    Likes (Received)
    1418

    Default

    Thank you cameraman !!
    Ya machine should give a good 20 years of life. Hopefully. Should pay for itself in a year or two. Sorry we don't take any work unless a tool holder. Any new work or custom is alreay in 2020 timeframe. 2 more machines coming in, already have them booked. Anyway back to the UMC500.

  3. Likes cameraman, doug925 liked this post
  4. #23
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    Quote Originally Posted by cameraman View Post
    Yup, going through the "Book of Face" for HAAS seems they may have announced it on July 12th. Or be on their HAAS Automation site earlier hidden in plain sight.

    Good point about the resale value / low/lower risk with HAAS purchases.

    The UMC 750 had quite complex alignment and installation procedures as the machine being larger volume is more dependent on it's foundation and final installation. The UMC 500 seems to have a more compact casting and "floor" plan, maybe easier to rig and re-rig. Up sticks and move.
    I don't recall anything out of the ordinary when they did ours. I don't think our foundation was anything special either, but...?

    I think it's a little crappy that you used to get this with the UMC750
    oldprobe.jpg
    which was the really nice Renishaw

    Now you get this
    probekit.jpg
    DOn't know if it is relavent or not, but they show doing the mzrp(?) with the new style on center, or roughly center. Not sure how this would check accuracy??

  5. Likes cameraman liked this post
  6. #24
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Orange Vise View Post
    I think it's been on the website for about a month.

    I was talking to somebody about ROI calculations for a new shop, and mentioned that the TL-1 I got back in 2012 was sold in 2015 for 97 cents on the dollar. That led to a discussion about UMCs, which is when we noticed the 500. These things are going to sell like hotcakes.
    I "Pinged" Haas on FB and they said UMC -500 was officially launched July 1st in the USA and Canada , surprised there was not more 'Buzz" about it , but maybe there was ?

  7. #25
    Join Date
    Aug 2010
    Location
    Minnesota
    Posts
    1,984
    Post Thanks / Like
    Likes (Given)
    150
    Likes (Received)
    1464

    Default

    Quote Originally Posted by cameraman View Post
    I "Pinged" Haas on FB and they said UMC -500 was officially launched July 1st in the USA and Canada , surprised there was not more 'Buzz" about it , but maybe there was ?
    If you can't build enough to meet demand it better not to make buzz... why pay to advertise something that sells itself. Tesla has never advertised, don't need to, can't build enough to meet demand as is.

  8. Likes Garwood, empwoer liked this post
  9. #26
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Mike1974 View Post
    I don't recall anything out of the ordinary when they did ours. I don't think our foundation was anything special either, but...?

    I think it's a little crappy that you used to get this with the UMC750
    oldprobe.jpg
    which was the really nice Renishaw

    Now you get this
    probekit.jpg
    DOn't know if it is relavent or not, but they show doing the mzrp(?) with the new style on center, or roughly center. Not sure how this would check accuracy??


    It's ^^^ This Leveling and alignment procedure that concerns me... (At least for the UMC 750 ).

    [I v'e designed and built / implemented long "self" leveling trackways for precision moving camera towers and precision camera positioning equipment that weigh several thousand pounds over uneven terrain and can be a real art to level things out properly with many feet to get super precise smooth safe level and even load distribution.].

    In the case of the UMC 750 seems to require skill and patience to level and align properly and seems that it would require several iterations to get everything to converge nicely in a stable way (as it should). I don't know if the techs they send out for that are all really "Down with that"/ top notch ? Seems an average foundation could or would settle in or shift a bit ? Wonder how many people that bought second hand UMC 750's got them installed really as they should be ?

    Wondering if the more compact UMC 500 is naturally less ponderous and reliant on it's foundation / skill of installation tech ? [Probably all of a muchness maybe ?.]

    __________________________________________________ __________________________________________________ ___

    So (@Mike1974) is that New MZRP gizmo is inert ? No probing electronics ? Just a mini test bar ? Is one supposed to use cigarette paper and jog the machine to the ball ? Is that for the tool setter ? No comprendo ? Surely they assume you have Renishaw probe ?

  10. #27
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by kustomizingkid View Post
    If you can't build enough to meet demand it better not to make buzz... why pay to advertise something that sells itself. Tesla has never advertised, don't need to, can't build enough to meet demand as is.
    Exactly that's my "read" / 'vibe"...

  11. #28
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    Quote Originally Posted by cameraman View Post


    It's ^^^ This Leveling and alignment procedure that concerns me... (At least for the UMC 750 ).

    [I v'e designed and built / implemented long "self" leveling trackways for precision moving camera towers and precision camera positioning equipment that weigh several thousand pounds over uneven terrain and can be a real art to level things out properly with many feet to get super precise smooth safe level and even load distribution.].

    In the case of the UMC 750 seems to require skill and patience and seems that it would require several iterations to get everything to converge nicely in a stable way. I don't know if the techs they send out for that are all really "Down with that"/ top notch ? Seems an average foundation could or would settle in or shift a bit ? Wonder how many people that bought second hand UMC 750's got them installed really as they should be ?

    Wondering if the more compact UMC 500 is naturally less ponderous and reliant on it's foundation / skill of installation tech ? [Probably all of a muchness maybe ?.]

    __________________________________________________ __________________________________________________ ___

    So that New MZRP gizmo is inert ? No probing electronics ? Just a mini test bar ? Is one supposed to use cigarette paper and jog the machine to the ball ? Is that for the tool setter ? No comprendo ? Surely they assume you have Renishaw probe ?
    No I should have clarified I was just talking about the attachment with the ball. It used to be a nice 'kit' in a case with the calibration master (tool length and diameter to set tool probe with), and the mag base with the ball offset at 45deg to run your 5 axis calibration for center of rotation. Now it is just a stick with a ball on it that you screw to the table with T nuts. You still need the work/spindle probe with both setups. I think probing is 'mandatory' on the 5 axis, it is not an option you can choose not to buy (I think).

  12. Likes cameraman liked this post
  13. #29
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Mike1974 View Post
    No I should have clarified I was just talking about the attachment with the ball. It used to be a nice 'kit' in a case with the calibration master (tool length and diameter to set tool probe with), and the mag base with the ball offset at 45deg to run your 5 axis calibration for center of rotation. Now it is just a stick with a ball on it that you screw to the table with T nuts. You still need the work/spindle probe with both setups. I think probing is 'mandatory' on the 5 axis, it is not an option you can choose not to buy (I think).
    Ok that's good to know (thanks for the clarification) , I DID laugh out loud when I went from your first picture and then flicked to the second picture … Literally a ball on a janky stick. (Passing on savings to the customer lol).

    That ball with 45degree angle + mag base looks / looked pretty sweet / solid.

  14. #30
    Join Date
    Nov 2002
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    3,242
    Post Thanks / Like
    Likes (Given)
    1818
    Likes (Received)
    800

    Default

    Quote Originally Posted by g-coder05 View Post
    The little bastard would be perfect to free up our 750's and 1000's so figured we would pre-order five of them. I get the contract back from HFO Shenzhen and China has not only tacked on the 25% tariff they added the old 18%VAT on top of it as well.


    Might want to wait on that PO.

  15. #31
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    Quote Originally Posted by cameraman View Post
    Ok that's good to know (thanks for the clarification) , I DID laugh out loud when I went from your first picture and then flicked to the second picture … Literally a ball on a janky stick. (Passing on savings to the customer lol).

    That ball with 45degree angle + mag base looks / looked pretty sweet / solid.
    Admittedly I have not looked to try and find a video, but I am confused how the new style even works?

    The 45deg one had you putting it in the corner of the table with the ball pointing to the center, then hand jog probe down to just over the top (approximate) center of the ball then start the calibration macro/routine. This would take measurements of the ball initially, then rotate the table around to various positions and re-probe the ball and at the end record the numbers from which you could adjust your center of rotation in the parameters. IIRC the first 4 moves of the program just indexed the C axis (platter) every 45 degrees, then it would move C and B together and repeat. Not sure how you could do that with a straight ball...

  16. #32
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Mike1974 View Post
    Admittedly I have not looked to try and find a video, but I am confused how the new style even works?

    The 45deg one had you putting it in the corner of the table with the ball pointing to the center, then hand jog probe down to just over the top (approximate) center of the ball then start the calibration macro/routine. This would take measurements of the ball initially, then rotate the table around to various positions and re-probe the ball and at the end record the numbers from which you could adjust your center of rotation in the parameters. IIRC the first 4 moves of the program just indexed the C axis (platter) every 45 degrees, then it would move C and B together and repeat. Not sure how you could do that with a straight ball...

    Apologies for density here. [Feel free to skip over this if this is not your "Thang". ].


    I did watch a few of the older videos for the older MZRP …

    Couldn't find the newer one being used.

    As far as I understand (mathematically) the procedure I'm not sure the 45 degree angle adds anything in it's current method for finding the intersection point of both rotary axes. Or at least define the position of each axis independently. [Theoretically they don't have to intersect perfectly for things to still be mapped out and controlled properly. That's a thick and difficult discussion for a different time and place perhaps. ].

    The probe (like a blind man with a stick) and the on control probing cycle just finds the center of the ball and assumes that the ball's center is fixed in space relative to the table. It's pulling the XYZ coordinates of the center of the ball. The table rotates to the next position records the coordinates, rotates to the next position and records the coordinates etc. [You know all of that better than I do.]. However that doesn't rely on the fact that the crooked neck points 45 degrees to the center. I think the positions opposite each other can be used to compute a line that crosses the center area of the table and a second opposing set of positions/ XYZ coordinates (from opposite sides of the table) can be used to construct a second line (in 3d space) so that the first line intersects the second line (in plan in the XY plane but not necessarily intersect perfectly in 3 space) a third line can be used in true three dimensional space or more, to fudge together the center points for rotation and redundant data can be used to good effect. Technically there are more advanced techniques like least squares adjustments that would allow one to compute a more likely center point ("adjusted'/ weighted correctly) than a simple "Average" or a minimum number of XYZ positions for the calibration ball being collected. It seems the B axis calibration would be dependent on the C axis calibration like what you are saying, to switch into opposing positions.

    There are other types of calibration carried out at the factory as commanded rotation can be wildly different form actual position (on most 5 axis machines). Sometimes on other machines the more complex kinematic calibration is used to refine that.

    Maybe there's something I'm not visualizing about the Janky ball stick vs. 45 degree crook'ed neck (nicer gizmo) It could be that the 45 degree angle with the second ball could have been a belt and braces approach where the second larger ball could be probed also so you get additional projected lines. (Maybe that's something they used as a temporary "crutch" in early testing and then gave it to customers (just in case) ? We use similar mathematical techniques for mapping our radial distortion on precision lenses (with autocollimators from two different angles) and rotate the lenses along its axis (very precisely) but an angular approach is not strictly necessary especially if you have good XYZ coordinates and can map various transformations properly. Maybe HAAS abandoned the need for aditional constructed lines in 3d space from the points mapped from the two balls vaguely pointing to the intersection point of the two axes (maybe) forming "conic" or slightly twisted teepee type projection (when the C axis is rotated.). The lines in 3d space wouldn't have to converge like a perfect conic projection (to or from an single point in 3d space), but just make a projected conic / slightly twisted truncated polygonal cone (like the sticks (as lines in 3d space) that make up a teepee / wigwam ) where the tarp at the top of the tent makes a concentric polygon shape (If you get what I mean). That data can tell you how flat your rotaions are and where the center point of rotation is likely to be.

    Sorry that's a bit wordy as I can't imagine exactly the "mathematical" musing of what HAAS may have chosen to do... originally versus what they have abandoned and now do instead. I think the 45 degree crooked necked double ball arrangement may be vestigial now ( as mathematically and computationally they skin that cat differently now and don't need a second auxiliary method as a sanity check to confirm everything has gone well with the primary method.).

    Was a nice piece of kit it seems. Everybody likes nicely fashioned / nicely engineered things.

  17. #33
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    Maybe the 45deg ball adds nothing? I can't say for sure now as it has been a few years. It was a nice kit however, and convenient with the mag base. Might be off base here, but I would assume the more distance you have between checks (ie the ball mounted in the corner) would give more accuracy, but that is a WAG. With the new ball config they show it very near the center of the table, but it might specify in the cal routine to place it *here* with some approximate dims?

    Rotary-Axis Calibration Tool

    click on the details tab then scroll to the bottom for pics

  18. Likes cameraman liked this post
  19. #34
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    264
    Post Thanks / Like
    Likes (Given)
    312
    Likes (Received)
    110

    Default

    Quote Originally Posted by Mike1974 View Post
    Maybe the 45deg ball adds nothing? I can't say for sure now as it has been a few years. It was a nice kit however, and convenient with the mag base. Might be off base here, but I would assume the more distance you have between checks (ie the ball mounted in the corner) would give more accuracy, but that is a WAG. With the new ball config they show it very near the center of the table, but it might specify in the cal routine to place it *here* with some approximate dims?

    Rotary-Axis Calibration Tool

    click on the details tab then scroll to the bottom for pics
    45 deg ball DOES add something.
    think about it, when it tilts the axis, if you have a straight shaft, the probe will run into the shaft when probing one of the sides of the ball. if the shaft is angled, you'll essentially have relief for the probe to find the center of the ball.

  20. Likes cameraman liked this post
  21. #35
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Mike1974 View Post
    Maybe the 45deg ball adds nothing? I can't say for sure now as it has been a few years. It was a nice kit however, and convenient with the mag base. Might be off base here, but I would assume the more distance you have between checks (ie the ball mounted in the corner) would give more accuracy, but that is a WAG. With the new ball config they show it very near the center of the table, but it might specify in the cal routine to place it *here* with some approximate dims?

    Rotary-Axis Calibration Tool

    click on the details tab then scroll to the bottom for pics
    That's interesting... Thanks for that ^^^ Not such a terrible piece of kit perhaps , I've seen worse and nice that it breaks down to different heights.

    technically IF they are just trying to find centerlines / center points then having the ball be closer to the center of that table might be more accurate (for them) as the positional errors at the edge of table will be larger* (and that is a combined error that needs to be filtered out).

    Having a tooling ball at the edge of the table would be good for mapping out positional errors/ rotational errors.

    Maybe they chose to separate out or minimize positional errors combined with the need to find the centerlines.

    If you find your centers for rotation first then mapping out angular and positional errors (i.e. amplified at the edges of the table) would be easier later. Again mathematically there are combined solutions for that with taking many more points around the table but I guess HAAS has elected not to do that keep everything as simple a possible.

    Also you have to take into account where the 'Sweet spot' for the machine is along it's linear X, Y and Z travels... If some of these measurements are made where the Y axis is fully extended (on that type of layout) and similarly the Z axis is full travel and the whole ram style is positioned at the furthest corner of the X travel then then your XYZ coordinates are not going to be so great as compared to the well balanced "sweet spot" of the machine. It's a shame that machines don't have inbuilt laser interferometers (maybe one day). Never the less much can be compensated / mapped out mathematically using reversal methods especially for rotary axes.

    __________________________________________________ ___________________________________________

    * Just in general having large rotary trunnions or universals (of the order of a meter) are extremely difficult to make to be super accurate unless you are going for equipment that is 20X the price. So I'm hoping HAAS has managed to make the UMC-500 be proportionately more accurate than the UMC 750 per se as there is much less heavy 'Floppy" metal to move around/ inertia> Maybe Overall 'Tighter" more precise and repeatable machine.

  22. Likes Mike1974 liked this post
  23. #36
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    @cameraman post #35

    Actually the way the UMC750 (no exp with the 500 or ss models) is, it can sometimes (alot IMO) live near the max XY travels. The table rotates I think 30 to the left (looking in machine) to 110 to the right (meaning the platter is on the left side at 90deg), hopefully I explained that well enough (it makes sense in my head anyways!)... That means when even when doing work on the left side of table/platter at 0 deg tilt/rot, the X axis is already at about half it's max travel. Y is less extreme, but still even at the back of table the Y is already out maybe 2-4" (?). The reason, from what I could see was X was designed like that so you can get to the work when tilt is at 90 deg'ish, not sure on why the Y axis was designed like that, other than it goes home to touch tools on the toolsetter probe.

  24. Likes cameraman liked this post
  25. #37
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    3,430
    Post Thanks / Like
    Likes (Given)
    1480
    Likes (Received)
    1621

    Default

    As a side note of sorts. The table on the UMC750 is large enough to fit 2 Kurt D688's side by side AND clear all rotation angles fwiw. Just wanted to share since we used it for alot of 3 axis work too since it was the only machine we had. I doubt the 500 table is that large. I downloaded a model at work just to have a look see, I will try and remember to inspect it for size, or throw a vise model on there and see how it fits.

  26. #38
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by Mike1974 View Post
    As a side note of sorts. The table on the UMC750 is large enough to fit 2 Kurt D688's side by side AND clear all rotation angles fwiw. Just wanted to share since we used it for alot of 3 axis work too since it was the only machine we had. I doubt the 500 table is that large. I downloaded a model at work just to have a look see, I will try and remember to inspect it for size, or throw a vise model on there and see how it fits.
    Ohhh for sure that's one of the amazing things (I think about) the HAAS UMC 750 and especially the newer UMC 1000 is this 40" of travel (That's huge) (it's like a 40x20 vertical than can spin sh*t around ) so in the older true sprit of a "Universal" that longer flatter work piece can be machined in a 3 axis way if you want (or mild tilts for more sophisticated work along vertical walls for large flat work pieces).

    I think what's interesting with the UMC 500 's table is that it's 400 mm diameter and yet has bags of room around it and the X travel is still 610 mm (that's pretty impressive) they don't say what the maximum collision circle is. Just as non sensical comparison the Hermle C250 (their entry model) has 600 mm ~ in X but theoretically a 700 mm ~ish collision circle. But longer than that you bump into the shoulders of the trunnion.

    But the 400mm table certainly takes things in a different direction for them HAAS/ (Captain Obvious ) and stirs things up a bit and treads on the toes of other MTBS with their offerings 400mm to 250 mm diameter tabled 5 axis verticals (kind of thing).

  27. #39
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,721
    Post Thanks / Like
    Likes (Given)
    3235
    Likes (Received)
    718

    Default

    Quote Originally Posted by empwoer View Post
    45 deg ball DOES add something.
    think about it, when it tilts the axis, if you have a straight shaft, the probe will run into the shaft when probing one of the sides of the ball. if the shaft is angled, you'll essentially have relief for the probe to find the center of the ball.
    So are you saying straight ball an' stick thing won't work ?

    Dang... Tool steel is not that bendy.

  28. #40
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    264
    Post Thanks / Like
    Likes (Given)
    312
    Likes (Received)
    110

    Default

    Quote Originally Posted by cameraman View Post
    So are you saying straight ball an' stick thing won't work ?

    Dang... Tool steel is not that bendy.
    i dont see how...
    think about it, imagine a straight shaft with a ball on top, then rotate your tilt axis 90*, you need to probe the tool from 4 sides, you can get to 3 sides, 4th will have the shaft in the way.


Tags for this Thread

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
  •