The Rumors are True New HAAS UMC 500 ! - Page 5
Close
Login to Your Account
Page 5 of 8 FirstFirst ... 34567 ... LastLast
Results 81 to 100 of 146
  1. #81
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by empwoer View Post
    LOL, you think they build F1 parts on their own machines? lets be real here...

    a friend of mine was an AE for Okuma, and programmed quite a few parts for the haas F1 team. none of them were cut on a haas machine.

    it has nothing to do with the capability of technology. they have access to the same technology as any other MTB out there. they choose not to put it on their machines to keep the cost down for users that dont need that level of tech.
    If you go the HAAS website …

    They have a friendly little logo motif that tells you that it's a machine used by the real F1 team ... /

    Haas Motorsports | Haas F1 Team | Stewart-Haas Racing

    ^^^ Read all about it !

    It even haas pretty pictures of the 100 or so parts they make on their HAAS machines for their formula 1 cars.

    I'm guessing a Formula 1 car has at least 4000 parts ?

    I used to live on the neighboring hill to Gordon Murray in France [Of McLaren fame](me ~ rickety old farm house barns and woods ~ Murray blinged out casual estate with an infinity pool) ... . (tiny little town/ village in the country SW France. Some funny stories when he transitioned / did some design work for Mercedes (save for over a beer) ).

    So McLaren for a number of years has had a strong / special partnership with MAZAK, beautiful shop. So I figure If MAZAK is good enough for Gordon Murray and McLaren then it's probably good enough for me lol. BUT I still wonder about what components are also made on non MAZAk machines (probably more grinding/ lapping related).

    I'm not an Automotive anorak/ wonk / gear head / petrol head, but Gordon went on to do some really interesting design projects like a flat-pack truck for developing countries... Like IKEA except its f*cking truck!

    Gordon Murray's Latest Vehicle Is a Flat-Pack Kit Truck for Africa

    Gordon Murray designed the OX flat-pack truck for the developing world - Roadshow


    gordon_m_flatpack_truck.jpg

    ^^^ Click on picture it enlarges.

    Called the OX ^^^ (Targeted for Africa).

    __________________________________________________ ____________________________________________

    * Malaren Honda + MAZAK McLaren-Honda extends exclusive machine tool partnership with Mazak

  2. #82
    Join Date
    Aug 2005
    Location
    CT
    Posts
    7,708
    Post Thanks / Like
    Likes (Given)
    305
    Likes (Received)
    1811

    Default

    Unfortunately the modern world is marching unabated to higher toleranced complex parts that have to be manufactured very efficiently.

    OK, so as an uneducated blacksmith with 50% Haas machines ... I am asking this question:

    What kind of complex parts does this world is marching towards unabated which requires arc second tolerances?

    I mean seriously guys!
    Arc seconds? Really?

  3. Likes Cycle1000, doug925 liked this post
  4. #83
    Join Date
    Apr 2018
    Country
    UNITED KINGDOM
    Posts
    1,532
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    721

    Default

    Quote Originally Posted by g-coder05 View Post
    ... China has not only tacked on the 25% tariff they added the old 18%VAT on top of it as well.
    They always do, and have for several years but it's 17%. You can get a 14% credit against that tho so it's really 3%. Of course you need to have a tax bill to credit it against, but that's like the investment tax credits in the US.

    Doesn't help with the 25% tho .... maybe you can get Haas to tranship through Vietnam and stick the label on there. That's what China does

  5. #84
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by SeymourDumore View Post
    OK, so as an uneducated blacksmith with 50% Haas machines ... I am asking this question:

    What kind of complex parts does this world is marching towards unabated which requires arc second tolerances?

    I mean seriously guys!
    Arc seconds? Really?
    Well 5 to 7 arc seconds repeatability that's more typical.

    In my case optical … and Digital imaging systems that are themselves of a metrological grade + robot "Movements" and very precise linear travels for instruments that are themselves rotated around their principal optical axes and have to maintain high precisions and accuracies themselves in multiple orientations.

    The main problem is that errors add. You can't lose your rotational errors in a cloud of thermal errors they are all additive.

    In 5 axis there are at least 21 errors that are additive, accumulate to mess up your parts, tolerances and orientations.

    There are a number folks on this forum that have bought platforms and machines that are of 1 to 3 arc second positioning accuracy + serious automation options.

    If you do the math trying to hold a seven to ten micron part accuracy for form tolerances and 6 degrees of freedom and multiple positioning + thermal compensation / spindle growth.

    When that is all mapped the rotational errors have the largest effect, when that's mapped out to a volumetric accuracy that become quite large (your 2 micron linear positioning accuracy becomes an 8 micron volumetric uncertainty).

    Even on the highest end Okuma large 5 axis bridge style a calibrated volumetric accuracy of 25 micron is considered good.




    Leadwell ^^^ Look at these parts and why they bought this machine---> tricky aerospace part does not have to be hand scraped. + other capabilities and issues and this is not a top drawer machine. "The quadrants... quadrants... ? yeah quadrants !.."

    The point of 5 axis is to have something that is as accurate as 3 axis, if your five axis system produces results that are less accurate than doing it all your self on a 3 axis machine (multiple set ups and specially designed fixtures) then what's the point ? Takes a hell of lot to attain the equivalent accuracy (with a certain degree of automation) 'cuz the machine moves about in 5 or 6 Degrees of freedom , that "Base" and baseline from a good 3 axis machine (even referenced to the spindle ) is pulled out from under you like a carpet. If want to keep to the "two tenths" mantra that is challenging for 5 axis precision parts.

    [For example the MAZAK integrex main spindle has positioning down to 0.0001° and the B axis mill head can position to 2 arc seconds + 1 micron tool repeatability + 3.5 micron circularity in XZ and YZ planes … Functional machine parts on a five axis platform need that level of positioning certainty... Otherwise it's no good or the level of automation is severely reduced, back to grinding + hand finishing. ].


    Sim 5 axis demands accurate coordinated movements other wise it's a total mess. Only recently 5 axis machines have become accurate enough to supersede 3 axis contouring with 5 axis positioning, so that barrel tools and lens shaped profiles can be used for much more efficient machining of 5 axis parts including Blisks, impellers and screw impellers ---> read better accuracies. better performance and lower cycle time.

    A lot of todays and tomorrow's parts are less orthogonal and more "organic" and have more diagonal members and slanted planes... Part of faster, lighter , stronger and even smaller aerospace structural parts require higher tolerances than 40 years ago.

    With UMC-500 it's a smaller machine than the 750 and newer UMC 1000 . With a large work envelope like the UMC 1000 I can totally understand not to expect high rotational accuracies , that is a real challenge for a machine, but the UMC 500 is smaller work envelope and proportionately smaller parts can certainly have higher accuracies and tighter tolerances associated with them. Hopefully the UMC 500 is proportionately more accurate. But... bandwidth on the HAAS control may prevent that until a later time. HAAS know their market very well and in sense created their own market with the UMC 750, so I think they have gone in the bombproof / indestructible direction rather than more accurate direction as for right now the control might not allow that. [I actually have confidence that HAAS (HAAS factory) has the skills to build a geometrically more accurate 5 axis machine (if they were tasked to do so), it's more about design/ market/ business strategy.].

    __________________________________________________ __________________________________________________ _______


    @SeymourDunmore cool that you are a blacksmith , my best friend is a Farrier, he's pretty skilled with the iron also, does some nice work / tools.

  6. Likes Mstcnc liked this post
  7. #85
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    156
    Post Thanks / Like
    Likes (Given)
    203
    Likes (Received)
    77

    Default

    Quote Originally Posted by cameraman View Post
    If you go the HAAS website …

    They have a friendly little logo motif that tells you that it's a machine used by the real F1 team ... /

    Haas Motorsports | Haas F1 Team | Stewart-Haas Racing

    ^^^ Read all about it !

    It even haas pretty pictures of the 100 or so parts they make on their HAAS machines for their formula 1 cars.

    I'm guessing a Formula 1 car has at least 4000 parts ?

    I used to live on the neighboring hill to Gordon Murray in France [Of McLaren fame](me ~ rickety old farm house barns and woods ~ Murray blinged out casual estate with an infinity pool) ... . (tiny little town/ village in the country SW France. Some funny stories when he transitioned / did some design work for Mercedes (save for over a beer) ).

    So McLaren for a number of years has had a strong / special partnership with MAZAK, beautiful shop. So I figure If MAZAK is good enough for Gordon Murray and McLaren then it's probably good enough for me lol. BUT I still wonder about what components are also made on non MAZAk machines (probably more grinding/ lapping related).

    I'm not an Automotive anorak/ wonk / gear head / petrol head, but Gordon went on to do some really interesting design projects like a flat-pack truck for developing countries... Like IKEA except its f*cking truck!

    Gordon Murray's Latest Vehicle Is a Flat-Pack Kit Truck for Africa

    Gordon Murray designed the OX flat-pack truck for the developing world - Roadshow


    gordon_m_flatpack_truck.jpg

    ^^^ Click on picture it enlarges.

    Called the OX ^^^ (Targeted for Africa).

    __________________________________________________ ____________________________________________

    * Malaren Honda + MAZAK McLaren-Honda extends exclusive machine tool partnership with Mazak
    because pretty pictures mean its all real right?

    i'm sure they make SOME parts on haas machines. but anything that requires real precision/speed/efficiency is done on 3rd party machines. dont believe ALL the marketing hype bud!

    again, this isnt to trash haas, just trying to keep it real

  8. Likes cameraman liked this post
  9. #86
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by empwoer View Post
    because pretty pictures mean its all real right?

    i'm sure they make SOME parts on haas machines. but anything that requires real precision/speed/efficiency is done on 3rd party machines. dont believe ALL the marketing hype bud!

    again, this isnt to trash haas, just trying to keep it real
    For sure ! Note traces of mild sarcasm.

    For sure in the USA the normal "Mode" (at least on the interwebs) is to take everything to the max and either be a rabid fanboy or be a total burn to the ground hater, so a more complex constellation of strengths and weaknesses tend to fall on deaf ears.

    Pulling focus on what IS can be helpful.

    With 5 axis machines it can be someone buys their first 5 axis machine to gain experience but then later realize it's not accurate enough, then they buy a second one and that can be a "near miss" ( a near miss is still a miss / awkward ) and then buy a third one that really nails what is needed and what they want. Sometimes it's better to buy the one that nails it first time and carefully learn the platform totally inside and out. Saves more time (in the long run) if time is a factor and budgets allow for that.

  10. Likes empwoer liked this post
  11. #87
    Join Date
    Aug 2005
    Location
    CT
    Posts
    7,708
    Post Thanks / Like
    Likes (Given)
    305
    Likes (Received)
    1811

    Default

    Quote Originally Posted by cameraman View Post
    Well 5 to 7 arc seconds repeatability that's more typical.

    In my case optical … and ...
    ... And in my case 50+% Aerospace...
    Motor, air frame, structure, aeronautics support, ground support, assembly or whatever. Living and working in CT means aerospace is something you encounter no matter what.
    Not one, and I mean NOT ONCE have I seen arc-second tolerancing on ANYTHING!!!

    Nearly impossible corners, radiuses, depths, lengths, thicknesses, gaps, mismatches, surface finishes .... Yupp, I've seen those!
    Arc-seconds though ...
    OTOH, I have had a few RFQ-s on my desk for Aerospace parts where all other suppliers took a "Relax It" or "Fly a Kite" attitude on a +/- 1 degree positional tolerance.
    Rest assured, some of those suppliers either had on hand, or could afford something a bit better than what Haas is offering.
    Quite a bit better actually!

    ( for the record, I've declined on most of those )

  12. #88
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by SeymourDumore View Post
    ... And in my case 50+% Aerospace...
    Motor, air frame, structure, aeronautics support, ground support, assembly or whatever. Living and working in CT means aerospace is something you encounter no matter what.
    Not one, and I mean NOT ONCE have I seen arc-second tolerancing on ANYTHING!!!

    Nearly impossible corners, radiuses, depths, lengths, thicknesses, gaps, mismatches, surface finishes .... Yupp, I've seen those!
    Arc-seconds though ...
    OTOH, I have had a few RFQ-s on my desk for Aerospace parts where all other suppliers took a "Relax It" or "Fly a Kite" attitude on a +/- 1 degree positional tolerance.
    Rest assured, some of those suppliers either had on hand, or could afford something a bit better than what Haas is offering.
    Quite a bit better actually!

    ( for the record, I've declined on most of those )
    In my twenties I worked as a conservator at the national air and space museum and Garber facility air craft restoration facility , and had to do condition reports for the entire collection in storage, … That was a real mother of a task and risk assessment of the collection. Months and moths spent in very hot warehouses + active conservation duties. So I have pretty good handle on the range of parts, structures and components, from wing spars to power plants of the past century of aviation, even worked on the SR-71 we had at Dulles at the time... So there IS a very wide range of tolerances that are required. Air frames to avionics / instruments. But nowadays designs are getting more advanced geometrically.

    If for example the UMC -500 has decent sensitivity of comandable movement with out too much slop or stiction / stored residual movement (like a spring) then much can be fixed by hand/ dti (not even probing).

    Machine a 150 mm cube and test it with a cylinder square on a granite inspection plate.

    I know that would be meaningless to most aerospace easier/ more practically tolerance parts. But other kinds of components require accurately refenced multiple features... recently the fuel tube fiascos for F-35.

    I know there are better tests parts.
    Last edited by cameraman; 08-14-2019 at 12:38 PM.

  13. #89
    Join Date
    Jan 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    672
    Post Thanks / Like
    Likes (Given)
    102
    Likes (Received)
    351

    Default

    So what's this talk about "bandwidth" limit on the rotary? Are you saying this limits XYZ positional accuracy while using TCPC?

  14. #90
    Join Date
    Jan 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    672
    Post Thanks / Like
    Likes (Given)
    102
    Likes (Received)
    351

    Default

    Quote Originally Posted by SeymourDumore View Post
    NOT ONCE have I seen arc-second tolerancing on ANYTHING!!!
    I don't think his point was necessarily arc-second tolerance features on parts, but meant that a less accurate rotary translates to less volumetric accuracy.

    20 arc-seconds of error on a 10" radius is just about 1 thou

    If you're doing small parts, then I guess it doesn't matter so much.

  15. Likes cameraman, empwoer liked this post
  16. #91
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by thesidetalker View Post
    So what's this talk about "bandwidth" limit on the rotary? Are you saying this limits XYZ positional accuracy while using TCPC?
    Nahhh

    Right now they use lower resolution encoder and don't have linear scales (different channels) anyway / separate channels. Right now the new UMC 1000 +/- 40 arc seconds positioning accuracy. 80 arc second spread i.e. 1 minute 20 arc seconds positioning uncertainty with respect to the machine. That's 0.0222° of arc. ≈ 1/45th of a degree of arc.

    But if they went down to a ten arc second encoder ring that could be an issue. They probably aren't gonna do that for at least another 15 or 20 years. lol-ish

    Absolute encoder would be a better way to go IMO or even a hybrid absolute-incremental. But those are much more expensive for HAAS (perhaps). That way they can slew and servo and lose counts and then pick up true finer position at each end of the acc/decc curves/ moves.

    With the HAAS VM machines if you want linear scales on the machine HAAS advises that that can only be used for prismatic parts and not be used for 3d contouring. In other words if you want their 'Mold" machine and want to do 3d contouring they advise that you don't get scales for the machine.

    As empower and others have noted and other Fusion 360 related topics and videos made by HAAS in partnership with Autodesk it's pretty easy to get that HAAS control to choke... I.e. slower cutting speed as the control can't keep up with the number of points it has to munch on. Solution being to optimize/ strip out linear segments etc. and convert to arc segment definitions and other (as empwoer mentioned). Whereas a Hiedenhain or Fanuc can blast through that.

    The mantra with 5 axis machines is linear scales all axes, rotary scales all axes (but to meaningful precision) and spindle chiller / thermal management of some kind + decent geometry/build.

    But some entry level models manage to skip one rotary encoder and a spindle chiller but still have really accurate positioning and thermal management but such a machine is designed to be limited in how fast it can move (like a Hermle C250).

    Coffee time but maybe later I dig up bandwidth numbers for encoders.

  17. Likes empwoer liked this post
  18. #92
    Join Date
    Jan 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    672
    Post Thanks / Like
    Likes (Given)
    102
    Likes (Received)
    351

    Default

    Ah, right 3d contouring. That's what it was.

    In my opinion, trying to hold tighter accuracy where the (haas) control is literally choking on code, I think you're probably a fairly high feedrate, no? (or unnecessarily small lines/arcs)

    If the control could process an unlimited amount of code, I think at those feedrates, the rest of the Haas machine (mechanically) wouldn't hold the accuracy you're looking for anyway. So isn't it a wash?

    Just like trying to interpolate a bore at a high feedrate. Your feature is under/over size and out of round.

    Probably better off finding a good balance between acceptable code accuracy and feedrate. Control should handle enough code just fine for the accuracy it is capable of mechanically.



    Just my opinion...

  19. Likes empwoer liked this post
  20. #93
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by thesidetalker View Post
    I don't think his point was necessarily arc-second tolerance features on parts, but meant that a less accurate rotary translates to less volumetric accuracy.

    20 arc-seconds of error on a 10" radius is just about 1 thou

    If you're doing small parts, then I guess it doesn't matter so much.
    This is the point with a large table and a large (mill) Head on a universal type machine it's quite probable that you have to fixture the part at least 6" to 8" off the table. If your part is a few inches thick and 10" to 15" long (and maybe 6" to 8" wide) then those rotational errors "swing" in two planes and the end points of your part could be 12" away from the centers of rotation.

    conceivably (as mentioned earlier you can probe etc. and implement some sort of corrections).

    Smaller precision parts can be more tricky, we need parallel prismatic parts for fitting precision linear bearings too (in different orientations ) so for a 100 mm part 20 arc second error creates a 10 micron error (0.01 mm) on the part in just one rotated plane. I.e. +/- 0.0002" or a 0.0004" spread. (that's before any questions about how straight can a given machine cut (with best tooling)).

    The other thing to bear in mind when the Japanese MTB's quote positional accuracies and repeatability's the repeatability value may seem really ultra tight like two or thee arc seconds, but that usually a small angle rotation less than 30 degrees. That's not a series of multiple rotations (with accumulated positional uncertainties . Also the value they quote in the brochures etc. are usually unidirectional values too. So they don't show bi-directional values (lower tolerance positional accuracies) or the full rotational plot. The Germans tend to be a lot more conservative about that and typically their test sheets and test parts indicate higher accuracies and reparability's than published just so they can stay within broad limits. Also the nuclear non-proliferation treaty + higher accuracies and export restrictions. *

    I guess with the UMC 500 and 1000 the rotary encoders help correct any slop in the mechanism when the machine start to get a bit knackered / lose, after years of having the snot pummeled out of it.

    __________________________________________________ _________________________________________


    * I don't build atomic munitions, (I know people who do), but questions of arc second-ish repeatabilitys on a machine ? ; it's my understanding that the high precision angular cuts is what makes a miniaturized atomic warhead work / be feasible. Not that - that many "nukes" are made on a HAAS … or are they ???? ;-)

  21. Likes empwoer liked this post
  22. #94
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by thesidetalker View Post
    Ah, right 3d contouring. That's what it was.

    In my opinion, trying to hold tighter accuracy where the (haas) control is literally choking on code, I think you're probably a fairly high feedrate, no? (or unnecessarily small lines/arcs)

    If the control could process an unlimited amount of code, I think at those feedrates, the rest of the Haas machine (mechanically) wouldn't hold the accuracy you're looking for anyway. So isn't it a wash?

    Just like trying to interpolate a bore at a high feedrate. Your feature is under/over size and out of round.

    Probably better off finding a good balance between acceptable code accuracy and feedrate. Control should handle enough code just fine for the accuracy it is capable of mechanically.



    Just my opinion...
    I agree...

    On the VM series they have fine pitch ball screws and maximum feed rate (cutting) is 500 ipm.

    The computer scientist half of me kinda says optimize point distribution and have an efficient and compact representation that is accurate NOT throw the machine a bazzilion points needlessly I'm a fan of optimization schemes like NURBS or super nurbs like on the Okuma control (but I think I'm in the minority there). So at least correct settings / g codes / G187 settings.



    ^^^ This caused a bit of minor "Fisticuffs" on the CAD CAM wing of the forum lol.

    __________________________________________________ _________________________________________

    A Makino F3 or F5 has no problem reading scales real time for accurate movement for mold work and that's on a Fanuc control, but they have other optimization routines from their SGI (super geometric intelligence layer). DMG mori CMX 110V has no problem with scales all axes etc. Same with MAZAK of course (very fast control) same with OKUMA (if you are lucky enough to have an Okuma with scales) or even M-460V 5ax scales all linear axes.

  23. #95
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    The significant appeal of the UMC 500 (the one I "Built" on line worked out to be about $160K with 15Krpm spindle HSK 63 A and other options ) undoubtedly it is the price point that is very attractive + HAAS (service + documentation and support and community. If that's working for you. )

    Most more serious 5 axis machines (so called "Entry" level (rolling eyes)/ platforms are between $260K to $360K (easily) [Makino does not build n entry level machine.], and if you have to buy appropriate support and auxiliary equipment and tooling for that machine, tool setters , better tool holders balancers shrink fit , even better 'Air" and so on it can start to add up a bit and nudge the "Idea" of 5 axis easily over the $400K line + specially engineered foundation in some cases.

    Depending on how much you can put down as a down payment on a machine The HAAS UMC-500 costs as much (in totality) as down payment on the other 5 axis machines. At least that way if you are a sole proprietor / one man machine shop (running around "spinning plates" ) the with UMC-500 you are not necessarily chained to your machines (for the foreseeable future) to make payments.

    In other words you can go on holiday with the family, sit in deck chair with knotted handkerchief on your head, pint of Guinness in hand and not feel you are smoking hundred dollar bills every hour while you are trying to relax on the beach. So your "vacation" does not become $15K-$20K loss.

    So I got to admit depending your budget capability and how much debt you want to get into the HAAS UMC-500 seems like a pretty "Chill" option if it meets your technical requirements / business model. Gives you a bit more freedom and future flexibility.

    You get what you pay for but will be interesting to find out where the "value proposition" really lies with the UMC-500.

    __________________________________________________ ___________________________

    * There are other machines that lurk in the $180k to $220k range from other MTBs Like the MAZAK VC 500A 5ax. "Kentucky built", there are also various Taiwanese machines with large-ish trunnions on a c frame style machine that are similarly priced.

  24. Likes empwoer liked this post
  25. #96
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    2,857
    Post Thanks / Like
    Likes (Given)
    1410
    Likes (Received)
    967

    Default

    Quote Originally Posted by cameraman View Post

    Leadwell ^^^ Look at these parts and why they bought this machine---> tricky aerospace part does not have to be hand scraped. + other capabilities and issues and this is not a top drawer machine. "The quadrants... quadrants... ? yeah quadrants !.."
    Interesting that they are still making machines with box way Z.
    I bought a V25H back in 1999, which was a 10kspindle X660 machine, but it was rails in XYZ.
    Fast and solid, but they still made some larger machines with box Z (back then).
    Roll forwards to now, and I'd have thought everyone was roller rails - but obviously not.
    Matsuura used to say that the best way to damp a spindle was to have a box Z, which is why all their VMCs were this way (with wide linear rail XY) back in the '90's/2000's.

  26. #97
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default UMC-500 video / EMO action (will be)



    Stumbled across this ^^^ It's on Vimeo (not youtube)…

    Interesting ergonomics there... Nice surface finishes and some interesting moves :-)

    __________________________________________________ _________________________________


    *

    ^^^ just in case it doesn't load right. (adobe flash plugin that Microsoft doesn't like.).

  27. #98
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,504
    Post Thanks / Like
    Likes (Given)
    3100
    Likes (Received)
    673

    Default

    Quote Originally Posted by barbter View Post
    Interesting that they are still making machines with box way Z.
    I bought a V25H back in 1999, which was a 10kspindle X660 machine, but it was rails in XYZ.
    Fast and solid, but they still made some larger machines with box Z (back then).
    Roll forwards to now, and I'd have thought everyone was roller rails - but obviously not.
    Matsuura used to say that the best way to damp a spindle was to have a box Z, which is why all their VMCs were this way (with wide linear rail XY) back in the '90's/2000's.
    https://us.dmgmori.com/resource/blob...0-pdf-data.pdf

    DMG mori did something very interesting with the NVX 5100 / NVX series 2nd generation.

    Originally it was boxed ways all axes and then with the 2nd Gen. they made the Z axis have a whole bunch of linear (rolling element) rails + trucks, and beefed up the column connection to the bed.

    That way you still have a rigid machine but have very responsive contouring (in Z) for mold work.

    On page 8 of the PDF discusses pressure distribution of the boxed ways dynamically … interesting.

    The 2019 prices for this 2nd machine are absolutely obscene... It's on a different horizon all together.

    __________________________________________________ __

    * No affiliation.

    Methods do some nice looking re-badged / commissioned Boxed way machines (for a good price), haven't seen one or know anyone that runs one yet.

  28. Likes barbter liked this post
  29. #99
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    7,427
    Post Thanks / Like
    Likes (Given)
    366
    Likes (Received)
    6216

    Default

    Quote Originally Posted by thesidetalker View Post
    So what's this talk about "bandwidth" limit on the rotary? Are you saying this limits XYZ positional accuracy while using TCPC?
    It's a max speed limit thing not an accuracy thing.
    You can only pump these pulses out so fast and read them so fast across copper wires and 30 foot runs.
    You are spot on in that is not arc seconds but position.
    Part size not really the matter. A 1/4 inch part with fine features with it located 18 inches from the rotary center is gonna be hard.
    Center of rotation and where you are from it.
    5,6 8 rotary axis suck. 4's are better behaved.
    People now live in giga hertz word . If I blow one pixel on your HDTV every 10 seconds you will never know. If I blow one count in your machine every ten seconds you will.
    Bob

  30. #100
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    2,857
    Post Thanks / Like
    Likes (Given)
    1410
    Likes (Received)
    967

    Default

    Quote Originally Posted by cameraman View Post

    Methods do some nice looking re-badged / commissioned Boxed way machines (for a good price), haven't seen one or know anyone that runs one yet.
    They look like a re-badged Feeler machine from Taiwan?


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
  •