Buying first machine. Questions about Speedio vs. Okuma. - Page 7
Close
Login to Your Account
Page 7 of 14 FirstFirst ... 56789 ... LastLast
Results 121 to 140 of 267
  1. #121
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by SVFeingold View Post
    The 15 arcsecond came from this Haas service document for the UMC 750. It's dated 2019, so I assume it applies to the current gen, and further to the UMC-500.

    How much of an impact will coolant temp have? AFAIK the Haas doesn't have any kind of temperature control (cooled rails/bearings/screws/casting). It would be a straightforward (and pretty fun) project to add a chiller to keep the coolant at a given temperature. I may still have a line to get a recirculating chiller for cheap.
    UMC-750 Applications Troubleshooting Guide

    Some more ^^^ detailed stuff from the OLD 750.

    That's not the new cyclonic drive + scales … (service bulletin you link).

    From video grabs on screen from the last IMTS Mark Terry Berry and NYCCNC and other's I've seen the accuracy stated (on screen (on the control) as +/- 20 arc seconds … I.e. 40 arc second spread , I believe the encoders are read down to 20 arc seconds positioning.

    Accuracy 40 arc seconds , repeatability 20 arc second (small angle single moves) ~ Possibly to two sigma and maybe to one sigma 15 to ten arc seconds (like what you are using @SVFeingold.) but without any verification from HAAS or current users , it's kinda hard to map out. Maybe HAAS can make a test part for you like @Finegrain is saying but there are ways to skew great results. Or maybe a rotational plot would not go amiss. But HAAS doesn't really do that.

    I can dig about time willing to piece that together (on screen stated accuracy).

    I would hope the machine could be better than that but the cyclonic drive is a different beasty all together but much more bomb proof / neat idea. (originally they drove the knuckle trunnion off one of their bigger 4th axis rotaries.). So the cyclonic drive should be a real improvement.

    As you know all errors are additive so your normal XYZ errors have to be added to your rotational errors. The combined rotational error of both axes have a compounding / cumulative effect. (you seem to be working off that premise).

    The videos that empower has linked to that he has shot sure looks like the machine moves really nicely (much more so than one would expect).

    Repeatability tests and various sequences of rotations and their inverse back to a DTI and test bar might be handy.

    Seemingly the backlash may be less with the cyclonic drives, hence lower backlash compensation for each rotary move ?

    ~ That's a very DMG Mori-ish adaptaion you made to your UMC 500 table there …

  2. Likes empwoer liked this post
  3. #122
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    Quote Originally Posted by cameraman View Post
    I would hope the machine could be better than that but the cyclonic drive is a different beasty all together but much more bomb proof / neat idea. (originally they drove the knuckle trunnion off one of their bigger 4th axis rotaries.). So the cyclonic drive should be a real improvement.
    With any luck maybe they'll update it in a couple years with a more accurate drive that's a drop-in replacement...

  4. #123
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by SVFeingold View Post
    With any luck maybe they'll update it in a couple years with a more accurate drive that's a drop-in replacement...

    With HAAS they Dooooo make constant improvements, some you see and some are more just by "Stealth".

    So yup that has been known to happen.

    It still concerns me that their control is still limited on bandwidth and IF one day they really took that side of things seriously then there would much much less limitation on the hardware.

    Still … The UMC 500 and UMC 750 reboot Seems to move quite a bit more nicely in sim-5 axis "Mode" than the old 750.

  5. Likes empwoer liked this post
  6. #124
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    Quote Originally Posted by cameraman View Post
    The videos that empower has linked to that he has shot sure looks like the machine moves really nicely (much more so than one would expect).
    Which videos by the way?

  7. #125
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    617
    Post Thanks / Like
    Likes (Given)
    1016
    Likes (Received)
    299

    Default

    Quote Originally Posted by SVFeingold View Post
    Which videos by the way?
    YouTube

  8. #126
    Join Date
    May 2018
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    133
    Post Thanks / Like
    Likes (Given)
    106
    Likes (Received)
    79

    Default

    Can the tool probe be rotated 90 deg or its out of travel?

  9. #127
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by SVFeingold View Post
    Which videos by the way?
    Ohhhh Kaaaaaaaay you had to ask didn't you ;-)

    YouTube


    ^^^ This is the cycloidal drive offered by Nabtesco (if it doesn't link then go to HAAS UMC 500 details page).

    States "Minimal backlash 1 arc minute+ - (60 arc seconds).

    umc_1000_acc_repeatability.jpg

    ^^^ Screen grab from NYCNC at IMTS 2018 with Mark T. (you can click on the image to make it full size.).



    From this video 12minutes and 5 seconds in.

    __________________________________________________ __________________________________________________ ____


    UMC 500 Nabtesco inset vid on HAAS page indicate 1 arc minute / 60 arc second backlash.


    HAAS UMC 1000 shows / states +/- 40 arc seconds accuracy, …, +/- 20 arc seconds repeatability.

    _______________________________

    It's better to think in terms of positional uncertainty rather than accuracy.

    Secondly there are combinations of errors for rotation so the "Laws" of central tendency apply.


    sigmas-.jpg

    ^^^Standard deviation One sigma, two sigma, three sigma and then applied to central tendency and positional uncertainty.

    So 68% of the time it could be that the HAAS UMC cycloidal drive can position to within +/- 10 arc seconds.

    I'm clinging to the imaginary idea that the HAAS scales read down to 20 arc seconds.

    The cycloidal drive is amazing for low starting torque hence better sensitivity of movement / lower inertia and hence BETTER MOVES :-)

    However I wonder about the runout or if there is a separate bearing ? It can run without a slewing bearing.

    Also I imagine the rotational calibration has more in common with pitch error compensation than conventional rotary drives. Interesting how those assemblies gyrate against multiple pins. lower accuracy but longer life as the wear is more evenly distributed.

    Better sensitivity of movement and lowered backlash should allow for better corrective probing strategies especially those worked into a more automated scenario.


    Overall a lot of interesting design trade offs that are biased towards better moves / arcs in cut, and longer life of the drive over and above raw precision and accuracy. ~ That's kind of where HAAS's market really is. They know their market and customers well.

  10. Likes SVFeingold liked this post
  11. #128
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    So quick update, I spent about an hour and a half on the phone with Selway this morning asking tons of questions.

    A few key points:

    • It sounds like the rotary axes probably do need to home after a power loss, but they're looking into it.
    • CAMplete or NCSimul are both $10-$15k. Damn
    • The 70A breaker rating is very conservative, it used to be 100A for the same hardware. Sounds like they have people running 50A with zero issues.
    • The 1k TSC sounds like it's just a high pressure pump, provided you got the TSC-ready option. Interestingly you can only select through-tool air blast if you also select the 1k TSC option. It won't let you select it online with only the TSC-ready option. That doesn't make a ton of sense so they're looking into it.
    • The chip conveyor may need to be installed at the factory, as if it's anything like the UMC750 the machine has to be lifted in order to install it in the field. That would be a bummer but they are double-checking.
    • We should be able to go from "truck pulling up" to "garage door close" in ~1 hr. That will be good to minimize time neighbors have to ask questions. The tool carousel can then be re-installed once inside.
    • Might be able to get a ~50% break on the sales tax due to a CA incentive for R&D outfits...


    All in all I got a good feeling from them. They were upfront about what they did/didn't know and seem willing to find answers to all my questions and to help figure out any issues I may have getting up and running. So far this has been true of Yamazen and Gosiger as well, so it seems like there are no bad options here on the support front. I still dream of owning a Speedio one day though. And maybe an Okuma. And hell why not a Mori too.

  12. Likes empwoer liked this post
  13. #129
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    617
    Post Thanks / Like
    Likes (Given)
    1016
    Likes (Received)
    299

    Default

    Quote Originally Posted by SVFeingold View Post
    So quick update, I spent about an hour and a half on the phone with Selway this morning asking tons of questions.

    A few key points:

    • It sounds like the rotary axes probably do need to home after a power loss, but they're looking into it.
    • CAMplete or NCSimul are both $10-$15k. Damn
    • The 70A breaker rating is very conservative, it used to be 100A for the same hardware. Sounds like they have people running 50A with zero issues.
    • The 1k TSC sounds like it's just a high pressure pump, provided you got the TSC-ready option. Interestingly you can only select through-tool air blast if you also select the 1k TSC option. It won't let you select it online with only the TSC-ready option. That doesn't make a ton of sense so they're looking into it.
    • The chip conveyor may need to be installed at the factory, as if it's anything like the UMC750 the machine has to be lifted in order to install it in the field. That would be a bummer but they are double-checking.
    • We should be able to go from "truck pulling up" to "garage door close" in ~1 hr. That will be good to minimize time neighbors have to ask questions. The tool carousel can then be re-installed once inside.
    • Might be able to get a ~50% break on the sales tax due to a CA incentive for R&D outfits...


    All in all I got a good feeling from them. They were upfront about what they did/didn't know and seem willing to find answers to all my questions and to help figure out any issues I may have getting up and running. So far this has been true of Yamazen and Gosiger as well, so it seems like there are no bad options here on the support front. I still dream of owning a Speedio one day though. And maybe an Okuma. And hell why not a Mori too.
    do yourself a favor and stay away from dmg/mori!

    sounds like you've got your mind made up on the 500, i think you'll like it. other than my gripe about the processing speed limitation (80 lines look-ahead and 1000blocks per second limit) i'm damn impressed with this little machine so far. i'd definitely get a chip conveyor.

    simulation is pricey, but saves hours/days of fucking around with a program, verifying clearances on tools, spindle/table etc. worth every penny!

  14. Likes cameraman liked this post
  15. #130
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    Quote Originally Posted by empwoer View Post
    do yourself a favor and stay away from dmg/mori!

    sounds like you've got your mind made up on the 500, i think you'll like it. other than my gripe about the processing speed limitation (80 lines look-ahead and 1000blocks per second limit) i'm damn impressed with this little machine so far. i'd definitely get a chip conveyor.

    simulation is pricey, but saves hours/days of fucking around with a program, verifying clearances on tools, spindle/table etc. worth every penny!
    Is that 80-block look ahead the limit on the UMC500SS as well? I kinda thought the HSM option would push it out more. I suppose I can lean more heavily on the CAM to smooth out toolpaths.

    I really, really do want CAMplete or something similar. You have a preference? They did mention it comes with a post, which I imagine will be better than whatever I get with HSMWorks if I even get one for free w/ HSM. For CAM itself I've got HSMWorks, and am going to evaluate Powermill. Hypermill seems very nice but expensive as hell.

    Now to pull the VF-2SSYT model into CAD and see what I'd be giving up in rotary space and gaining for table space.

  16. #131
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    617
    Post Thanks / Like
    Likes (Given)
    1016
    Likes (Received)
    299

    Default

    Quote Originally Posted by SVFeingold View Post
    Is that 80-block look ahead the limit on the UMC500SS as well? I kinda thought the HSM option would push it out more. I suppose I can lean more heavily on the CAM to smooth out toolpaths.

    I really, really do want CAMplete or something similar. You have a preference? They did mention it comes with a post, which I imagine will be better than whatever I get with HSMWorks if I even get one for free w/ HSM. For CAM itself I've got HSMWorks, and am going to evaluate Powermill. Hypermill seems very nice but expensive as hell.

    Now to pull the VF-2SSYT model into CAD and see what I'd be giving up in rotary space and gaining for table space.
    yes, that applies to all machines. no HSM is even less lines.
    we have camplete, and i do like it a lot! depending on what cam system you go with, it can either integrate into it so transferring the program to complete is a snap (mastercam does this very well) or you need to export/import your part, fixture, stock and toolpath data separately.
    you'd be surprised at price between powermill/hypermill, do yourself a favor and quote both. hell, quote mastercam and others too. get demos from them all, make them work for you and earn your sale.

  17. #132
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    Quote Originally Posted by empwoer View Post
    hell, quote mastercam and others too
    I intend to stay as far away from Mastercam as I can.

    I went ahead and pulled in the VF-2 to do some comparison. Honestly the difference in envelope isn't that great. I mean yes, by volume the VF-2 is nearly double, but it's really not much bigger in XY. VF-2 is a pretty small table.

    The difference with a trunnion is pretty massive though. The pics below show a comparison of the working envelopes - the smaller transparent box is the UMC. Also shows the distance between a dovetail vise mounted on a subplate and the top of the Z envelope for both.

    Front view, comparison of working envelopes.
    umc-vf-2-work-envelope-comparison-front.jpg

    Side view
    umc-vf-2-work-envelope-comparison-side.jpg

    VF-2 with TR160, 5th Axis subplate, and a dovetail vise.
    vf-2-trunion-dovetail.jpg

    UMC500 with same.
    umc-subplate-dovetail.jpg

    Here's that giant subplate from my earlier post in comparison to the VF-2 table. Honestly there's not much in it, if I can set up a good macro to re-probe a known reference every time I flip the C for simpler 3-axis parts. I hope it's rare enough that I can deal.

    giant-subplate-vf-2.jpg

    The UMC seems to make more sense at this point. Plus the casting looks far more substantial even though I'm sure there is an implicit rigidity hit going from 3 -> 5 axes. Anybody know what kind of ways Haas uses? Are they ball or roller? Preload? Questions I'll ask Selway as well.

  18. #133
    Join Date
    Sep 2018
    Country
    UNITED STATES
    State/Province
    California
    Posts
    617
    Post Thanks / Like
    Likes (Given)
    1016
    Likes (Received)
    299

    Default

    Quote Originally Posted by SVFeingold View Post
    I intend to stay as far away from Mastercam as I can.

    I went ahead and pulled in the VF-2 to do some comparison. Honestly the difference in envelope isn't that great. I mean yes, by volume the VF-2 is nearly double, but it's really not much bigger in XY. VF-2 is a pretty small table.

    The difference with a trunnion is pretty massive though. The pics below show a comparison of the working envelopes - the smaller transparent box is the UMC. Also shows the distance between a dovetail vise mounted on a subplate and the top of the Z envelope for both.

    Front view, comparison of working envelopes.
    umc-vf-2-work-envelope-comparison-front.jpg

    Side view
    umc-vf-2-work-envelope-comparison-side.jpg

    VF-2 with TR160, 5th Axis subplate, and a dovetail vise.
    vf-2-trunion-dovetail.jpg

    UMC500 with same.
    umc-subplate-dovetail.jpg

    Here's that giant subplate from my earlier post in comparison to the VF-2 table. Honestly there's not much in it, if I can set up a good macro to re-probe a known reference every time I flip the C for simpler 3-axis parts. I hope it's rare enough that I can deal.

    giant-subplate-vf-2.jpg

    The UMC seems to make more sense at this point. Plus the casting looks far more substantial even though I'm sure there is an implicit rigidity hit going from 3 -> 5 axes. Anybody know what kind of ways Haas uses? Are they ball or roller? Preload? Questions I'll ask Selway as well.
    haha, dont blame you on mastercam, its far from my favorite program to use, but it is powerful.

  19. #134
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by SVFeingold View Post
    I suspected as much, but I don't mind buying nice tools when needed, plus the few extra I'll destroy while learning to run them. A good face mill and a set of boring heads should be enough to hit the tolerances I actually need, right? I can't increase my budget by 100k, so if I do end up being convinced to go full 5-axis the Haas is pretty much the only choice as far as I can see. In both cost and size. Even on the used market, what else is there that doesn't have a dramatically smaller envelope? Other choices for 3x mentioned above...Haas, Brother, Okuma (which is a big stretch).

    I will say this, it's easier for me to relax my requirements then stretch the budget for a machine that can meet them all. "Requirements" is too strong a word. More like "wish list." If I can only accept 20% of the potential work instead of 30%, with the benefit of more freedom for my own work, so be it.

    On a related note, just for shits and giggles, I drew up a dummy part and subplate just to see what I could fit in the UMC500. This is a 15" x 27" x 3" part. Provided the subplate is tall enough to clear the probe and trunnion mount, it'll fit with a 180° index (and maybe re-indicate) required to reach the entire part, and just enough room to fit a 1/2 end mill around the perimeter. Clearly a pain in the ass bumping this close to the limits, having to order cut-to-size stock, making sure I don't exceed the torque limits machining the ends, etc. but it's nice to know it's possible the one time a year I need it. Without re-probing after the index the stated 15° arcsecond accuracy gives a positioning error up to +/- 25um at the furthest extent of that part from the C-axis centerline, and a perpendicularity error (relative to platter surface) of +/- 5um on a bore that deep due to B-axis error. I can live with that.

    Speaking of re-indicating after an index, what about a couple of ring gauges or something bolted to the subplate or platter to use as a known reference? Anyone do anything like that?

    Attachment 281574

    Attachment 281575

    Attachment 281576

    Attachment 281577

    Now someone tell me why this is a horrible idea.
    Coming back to this again …^^^ and "sub plate " design and other requirements...


    I find it inescapable that the DMG Mori CMX 70 U would fit your set of application requirements better than the HAAS UMC 500. (at least on paper). especially your sub plate design practically IS the table from the CMX 70 U.

    But it's 4+1 so chamfering using the side of the "end mill" can be accomplished for complex edges lieing on the same plane.

    No sim 5 on the CMX 70 though, but scales all axes (XYZ and B and C) and rotational accuracy +/- 8 arc seconds... Good thermal control , cooled spindle etc. better / good roller bearings on the B and C axes etc.

    Base has mineral casting bottom,

    Heindenhain control (if you want it or Siemens).

    weighs 14,000 lbs though... but is 89" to the parapet of the machine excluding spindle height fully extended ?

    Machine could probably be had for less than $200K.


    buuuuuuuuuuuut , you might not want to do that for a whole host of reasons...

  20. #135
    Join Date
    Aug 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    140
    Post Thanks / Like
    Likes (Given)
    137
    Likes (Received)
    52

    Default

    Quote Originally Posted by cameraman View Post
    I find it inescapable that the DMG Mori CMX 70 U would fit your set of application requirements better than the HAAS UMC 500.
    All but one! The price... If we define "less than $200k" as 190k then after adding workholding, power, phase converter, compressor, stock management, etc. We're at like 235k vs. a hair above $170 for the 500SS. That's a pretty big gap, and I believe probing is an extra option on the DMG. CAM and verification add a happy $15k to either machine. I just don't see justifying it but maybe I'm looking at it wrong. Seems like I'd have to get >$65k in extra work that I couldn't have gotten on the UMC due to the lack of capability. It really starts to transform from a machine it'd be nice to get paid work for to one I have to get paid work for.

    But I guess it wouldn't hurt to get a quote. I do like the big table.

    EDIT: Wow, a DMU50 would fit. A man can dream.

  21. Likes cameraman liked this post
  22. #136
    Join Date
    Jun 2016
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    16

    Default

    Quote Originally Posted by SVFeingold View Post
    All but one! The price... If we define "less than $200k" as 190k then after adding workholding, power, phase converter, compressor, stock management, etc. We're at like 235k vs. a hair above $170 for the 500SS. That's a pretty big gap, and I believe probing is an extra option on the DMG. CAM and verification add a happy $15k to either machine. I just don't see justifying it but maybe I'm looking at it wrong. Seems like I'd have to get >$65k in extra work that I couldn't have gotten on the UMC due to the lack of capability. It really starts to transform from a machine it'd be nice to get paid work for to one I have to get paid work for.

    But I guess it wouldn't hurt to get a quote. I do like the big table.

    EDIT: Wow, a DMU50 would fit. A man can dream.
    Word to the wise, I highly doubt that DMG/Mori would even remotely accept your floor in a garage as being "good enough" to support any of their specs. We have a DMU60 linear drive machine and were having issues with repeat-ability and the like. Repeat-ability was tested using their built in probing cycle of the tooling ball included with the machine. Our floor was poured fresh 10" reinforced, quality floor. They blamed the machines issues on the floor and essentially wiped their hands of it. If you look at the requirements in their books for installing a machine (they didn't mention it to us upfront), you cannot use a forklift under them, they must be craned. We had a tech here waiting for the truck to show up to supervise the rigging, sales guy came out, and it wasn't until the truck showed up (after the had been here for 3 hours) that they ask, where's the crane? So we had to expedite a crane up to our facility. Skates were deemed OK but the only way to get the machine off the skates was to crane it down from above. Using a toe jack/enerpac would void the warranty because it could tweak the sensitive, mineral-filled casting by putting more load on two corners and not distributed evenly. They also use the cheesiest aluminum feet I have ever seen that can break if you tighten them a tick too much, we have had to replace a few. So now we had to get a boom lift in capable of picking up such a machine. It was a mess. Just double check, and if they tell you it's OK, make sure every single step from pickup at their facility to being in place on your floor is laid out on paper and signed off on by them ahead of time.

  23. Likes cameraman, SVFeingold, bryan_machine liked this post
  24. #137
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by SVFeingold View Post
    All but one! The price... If we define "less than $200k" as 190k then after adding workholding, power, phase converter, compressor, stock management, etc. We're at like 235k vs. a hair above $170 for the 500SS. That's a pretty big gap, and I believe probing is an extra option on the DMG. CAM and verification add a happy $15k to either machine. I just don't see justifying it but maybe I'm looking at it wrong. Seems like I'd have to get >$65k in extra work that I couldn't have gotten on the UMC due to the lack of capability. It really starts to transform from a machine it'd be nice to get paid work for to one I have to get paid work for.

    But I guess it wouldn't hurt to get a quote. I do like the big table.

    EDIT: Wow, a DMU50 would fit. A man can dream.

    The new DMU 50 3rd gen seems to be good / pretty good / really good for some... quite spendy to spendacious compared to the 2nd gen.

    I believe there are some DMU 50 2nd gen machines still available "New" (not sure if "support" is waning for them.) ~ Interestingly a lot of the Taiwanese "Universals" are kind of a knock-off of the DMU 50 2nd gen, Like the GF (AG Charmiles entry level universal) but has a narrow ram / saddle assembly in the x dimension. The broader truck spacing of the new UMC 750 seems promising, I assume the UMC 500 has a similar idea for it's Ram/ saddle assembly ? .

    Personally I think the HAAS UMC 750 NEW REBOOTED and the new UMC 500 seems more exciting somehow than the older DMU 50 2nd gen. Including the price. The Old DMU 50 2nd never really got my blood going really , can't explain why ?

    @SVFeingold as you have probably figured out for orthogonal smaller parts that actually fit on your table; first cuts / plane top surface of part op 10 (that might then be flipped in the fixture Op 20 ) means you have theoretically a top surface and a bottom surface that is parallel almost to tenths. + any features you may have profiled using the side/ flank of a tool. <--- ALL without commanding a single rotation. So for a lot of parts that's where the principal "Accuracy comes from" and then the rotations for "hitting" the other 4 sides may not be that critical in terms of how they should reference each other. Like others have said if you have good probing routines you can close the gap on rotational errors to a reasonable degree. For funky and interesting parts that possess curious angles and geometries then you have to go much more slowly and hands on. More "Artful", but that can be fun and rewarding in of itself. But definitely steps away from the idea of "Automation" and 5 axis being an enhanced form of automation. I hear for singular boring operations the HAAS-es like even the old UMC 750 are pretty sound.

    I wondered if also on the newer UMC 500 and rebooted UMC 750 that the tool setter was set to the front of the machine to compensate for Y axis "Droop" when fully extended. I.e. HAAS tries to guarantee that part geometry is never violated, so if things go slightly awry you can have a second "Bite at the cherry" and have second finish pass to sneak down onto a final surface. Similarly with thermal expansion in Z of the spindle head it seems the HAAS control backs the spindle head away from the part to sort of "Overcompensate" ; even high end machines that use a "Thermo friendly " concept like Okuma and MAZAK will still have drifts of the order of 8 micron. If you get super nerdy in you garage you can map out almost every thermal drift and situation (given time). Like what @Finegrain was saying you'll probably be making the same part five or ten times at least to really get it and get it dialed in. + tool wear.

  25. #138
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    3,307
    Post Thanks / Like
    Likes (Given)
    3791
    Likes (Received)
    941

    Default

    Quote Originally Posted by 70olds View Post
    Word to the wise, I highly doubt that DMG/Mori would even remotely accept your floor in a garage as being "good enough" to support any of their specs. We have a DMU60 linear drive machine and were having issues with repeat-ability and the like. Repeat-ability was tested using their built in probing cycle of the tooling ball included with the machine. Our floor was poured fresh 10" reinforced, quality floor. They blamed the machines issues on the floor and essentially wiped their hands of it. If you look at the requirements in their books for installing a machine (they didn't mention it to us upfront), you cannot use a forklift under them, they must be craned. We had a tech here waiting for the truck to show up to supervise the rigging, sales guy came out, and it wasn't until the truck showed up (after the had been here for 3 hours) that they ask, where's the crane? So we had to expedite a crane up to our facility. Skates were deemed OK but the only way to get the machine off the skates was to crane it down from above. Using a toe jack/enerpac would void the warranty because it could tweak the sensitive, mineral-filled casting by putting more load on two corners and not distributed evenly. They also use the cheesiest aluminum feet I have ever seen that can break if you tighten them a tick too much, we have had to replace a few. So now we had to get a boom lift in capable of picking up such a machine. It was a mess. Just double check, and if they tell you it's OK, make sure every single step from pickup at their facility to being in place on your floor is laid out on paper and signed off on by them ahead of time.
    That's a good "Share" ^^^ .

    I have to admit I'm wondering about garage floor "denial" with SVFeingold. I think the UMC 500 is between 10,000 and 11,000 lbs ? (will check) so he might be OK ?

    Even the CMX 70 U is 14,000 lbs.

    @70olds sorry DMU60 didn't repeat worth a damn by the sounds of it... (for a linear machine that's super disappointing.).

    Related ~ Was wondering with the new UMC 500 and NEW Reboot 750 does it have the ability to crane parts in overhead ?

    Obviously in a garage crane-ability is limited ;-)

  26. #139
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    5,178
    Post Thanks / Like
    Likes (Given)
    931
    Likes (Received)
    2810

    Default

    One could always go with one of the many Makino models where the base casting is strong and rigid enough that they use a 3 point leveling system yet still maintain extremely high accuracy and productivity.

  27. Likes cameraman liked this post
  28. #140
    Join Date
    Jun 2016
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    40
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    16

    Default

    Quote Originally Posted by cameraman View Post
    That's a good "Share" ^^^ .

    I have to admit I'm wondering about garage floor "denial" with SVFeingold. I think the UMC 500 is between 10,000 and 11,000 lbs ? (will check) so he might be OK ?

    Even the CMX 70 U is 14,000 lbs.

    @70olds sorry DMU60 didn't repeat worth a damn by the sounds of it... (for a linear machine that's super disappointing.).

    Related ~ Was wondering with the new UMC 500 and NEW Reboot 750 does it have the ability to crane parts in overhead ?

    Obviously in a garage crane-ability is limited ;-)
    I don't want to say it was terrible. It may have been on par with machines that are spec'd out as "lesser" machines, but it wasn't what it was supposed to be, and that was followed with so many excuses as to why which was very disappointing. It was more the handling of the situation than the actual situation that was a problem for us.


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
  •