24,000 6-32 holes to make, how fast can I go? - Page 3
Close
Login to Your Account
Page 3 of 5 FirstFirst 12345 LastLast
Results 41 to 60 of 87
  1. #41
    Join Date
    Oct 2014
    Country
    CANADA
    State/Province
    Ontario
    Posts
    1,433
    Post Thanks / Like
    Likes (Given)
    975
    Likes (Received)
    1184

    Default

    I definitely want to see video once this part is dialled in!

  2. Likes Volitan liked this post
  3. #42
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    838
    Post Thanks / Like
    Likes (Given)
    271
    Likes (Received)
    385

    Default

    Quote Originally Posted by Mtndew View Post
    They are pretty insane when tapping. I'd be surprised if the average rpm isn't the programmed rpm or damn close to it.
    I tap 8k holes every 2 months or so in 6061, #4-40 threads. And it doesn't mess around. There isn't a machine that compares to a Brother when tapping even our brand new Okuma mills don't even come close to the tapping efficiency.
    So I am gathering from what everyone is saying! I would love to see a video rigid tapping at 8000 rpm Maybe I'll seach around youtube for one. My hiney hole is puckered when I tap at 2000 rpm

    Maybe some day I'll have a newer Okuma but likely never a Brother, just doesn't suit the type of work I do.

  4. #43
    Join Date
    Dec 2013
    Country
    UNITED STATES
    State/Province
    California
    Posts
    848
    Post Thanks / Like
    Likes (Given)
    255
    Likes (Received)
    495

    Default

    Quote Originally Posted by Hardplates View Post
    So I am gathering from what everyone is saying! I would love to see a video rigid tapping at 8000 rpm Maybe I'll seach around youtube for one.

    YouTube

    This was an older machine tapping at 8k. Brother had a 20b that was BT15 that tapped at 10k. Accel-Decel on Speedio is even faster. EagleMike, your High Torque Speedios reverse from 10k to 10k in about 0.2 seconds. Make sure those collet nuts are torqued properly!

  5. Likes PegroProX440 liked this post
  6. #44
    Join Date
    May 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    838
    Post Thanks / Like
    Likes (Given)
    271
    Likes (Received)
    385

    Default

    Quote Originally Posted by BROTHERFRANK View Post
    YouTube

    This was an older machine tapping at 8k. Brother had a 20b that was BT15 that tapped at 10k. Accel-Decel on Speedio is even faster. EagleMike, your High Torque Speedios reverse from 10k to 10k in about 0.2 seconds. Make sure those collet nuts are torqued properly!
    That's impressive

  7. #45
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    3,938
    Post Thanks / Like
    Likes (Given)
    2296
    Likes (Received)
    1549

    Default

    I'd up coolant strength to 10% and start tapping at 5k and see how it goes from there.
    Fastest i ever did on a Robodrill was 4k rpm and it was like a riveting machine!

  8. Likes Volitan liked this post
  9. #46
    Join Date
    Aug 2014
    Location
    Indiana
    Posts
    147
    Post Thanks / Like
    Likes (Given)
    108
    Likes (Received)
    256

    Default

    I've never seen one in person, but they look pretty quick.


  10. #47
    Join Date
    Apr 2014
    Country
    UNITED STATES
    State/Province
    California
    Posts
    873
    Post Thanks / Like
    Likes (Given)
    1116
    Likes (Received)
    577

    Default

    Quote Originally Posted by BROTHERFRANK View Post
    YouTube

    This was an older machine tapping at 8k. Brother had a 20b that was BT15 that tapped at 10k. Accel-Decel on Speedio is even faster. EagleMike, your High Torque Speedios reverse from 10k to 10k in about 0.2 seconds. Make sure those collet nuts are torqued properly!
    Oh yeah, you might not remember, but the first job I ran on the 450 with the H/T I have a collet nut come loose. They just stop so fast.
    I tap at 4k-5k all the time, 4mm and 5mm threads, form tapping. 6mm I run at 4500, IIRC. Even 8mm I tap at 3000. I'd go faster, but I'm chicken. I've been using the 229 series (MA Ford) 3 flute carbide drills, and they are the bomb. Size and finish is great.
    Stuff like this is why I'm a Speedio fan.

  11. Likes 2outof3, hanermo, Pete Deal liked this post
  12. #48
    Join Date
    Dec 2013
    Country
    UNITED STATES
    State/Province
    California
    Posts
    848
    Post Thanks / Like
    Likes (Given)
    255
    Likes (Received)
    495

    Default

    Quote Originally Posted by Mtndew View Post
    Wait, so I should be using G77 instead of G84 for my thousands of #4-40 tapped holes?
    I haven't had an issue so far after >30k tapped holes, is G77 faster?
    The G84 works ok. One of the nice things with the G77 is using the I and J for the thread. No rounding off like with a 18 tpi. or trying to figure metric feed... More accurate and convenient. Also you can change the S and not have to recalculate the Feed it locks in with the I and J. Brother has been doing Synchronized tapping since 1985 and it is different from rigid tapping. Not just for speed but for accuracy. Helps tap life in tougher materials because the tap isn't being pushed on or pulled on by inaccuracy. We used to tap bars of soap at the Westec show to demonstrate this. No pulling out threads in the soft material.

  13. Likes eaglemike, hanermo, TeachMePlease, Mtndew liked this post
  14. #49
    Join Date
    Sep 2009
    Location
    barcelona, spain
    Posts
    2,458
    Post Thanks / Like
    Likes (Given)
    552
    Likes (Received)
    1462

    Default How to tap really fast - 24.000 holes

    Depending, like always, but ..

    I suggest this is probably an excellent app for a custom-ground precision drill-tap tool.
    The tool might cost 400$, one, but might save 20-50% in total cycle time per hole, and might be longer lasting.
    Next 3 tools maybe 80$ each, set of 4.

    Your local (national, international) custom toolmaker will probably be able to suggest the right way, coatings, etc.
    The brother dealer will know or should know a few shops who make custom tools.

    For 24k holes, how many broken taps do you expect and how much scrap//costs from failed parts with broken taps ?

    It is non-trivial to choose what is the "best" or "fastest" speed achievable, to make more money.

    If a scrap part costs 200$, vs 1000$, or 30$ in material/work/piece, it might for example be much better to tap at 3000 rpm with a custom tool, and zero failures per 8.000 holes, swapping the tool 3 times, depending.

    I suggest a single hole takes about 2 secs and the tap about 2 secs, normal tools, given your specs (6-32, brother, mild steel, shallow, through hole).
    A custom tool might do both drill/tap in 2 secs, typical parameters.
    Saving 48k seconds per 24k holes.
    13 hours.

    And perhaps 1.2 secs if you run it "fast".
    Saving 48k + 19,2 secs. 67 k seconds.
    18.6 hours.

    The extra savings of 5 hours runtime at max speed of 1.2 secs for the op are probably about 5x-10x harder to get right and reliable.
    Inclusions in material and random hard spots may affect this.

    How expensive is it to stop/clear/restart a piece ?
    In time, materials, tooling, work done, opportunity cost, etc.

    Your original Q. was how fast can it be done, and I suggest these numbers (1.2 secs) as ballpark estimates for best-possible.

    But my real answer is more towards 2 secs for the total op, commercial realities, and optimising the whole process and not just the drill/tap.


    I also expect that better coding of the program can often save a lot of time, per op (hole, tap).
    Endless ways to optimise, potentially.

    It might be better to do all holes at once, if they are arrays, next to each other, for example.
    This would avoid tool changes, and keep the spindle spinning, avoiding spindle up/down times.
    These can be 2 secs for tool changes and 1 sec for spindle up/down cycles.

    Somewhere I recall iirc about 1.5 secs for new brother tool changes and maybe similar for spindle up/down.
    If the brother spindle is a servo, don´t know, it might spin up effectively instantly.
    A new ac brushless servo spins to 3000 rpm in 20 msec, 0.02 secs.

    If the code does a move, and only then does a spindle speed, perhaps the spindle can spin up while the move is being done.
    This will depend on the control, and on the programming.

    Often, such things are possible but not used and not in standard cam posts.
    So "better" programming might save 2-4 secs per hole themselves, without any tool/tap/drill stuff.
    Likewise, getting close to the hole, can save time in axis movements vertically.
    Endless options.

    Just use a custom macro.

  15. Likes Volitan liked this post
  16. #50
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    5,191
    Post Thanks / Like
    Likes (Given)
    933
    Likes (Received)
    2816

    Default

    Just for giggles, I made up a quickie program to tap 10 holes, one inch apart and 32 TPI to Z-.25" with an R plane of .1". Programmed at 2K rpm took 14.7 seconds. Bumped speed to 4K and time dropped to 12.6. Bumped speed to 6k and time stayed the same. So for this machine (Mori TV30) the spindle does not get up to commanded speed before having to stop and reverse with this depth (.35 total Z move or 11 turns) at somewhere between 2K and 4K.

  17. Likes aarongough, Volitan, Rick Finsta liked this post
  18. #51
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    4,945
    Post Thanks / Like
    Likes (Given)
    886
    Likes (Received)
    2071

    Default

    Quote Originally Posted by BROTHERFRANK View Post
    The G84 works ok. One of the nice things with the G77 is using the I and J for the thread. No rounding off like with a 18 tpi. or trying to figure metric feed... More accurate and convenient. Also you can change the S and not have to recalculate the Feed it locks in with the I and J. Brother has been doing Synchronized tapping since 1985 and it is different from rigid tapping. Not just for speed but for accuracy. Helps tap life in tougher materials because the tap isn't being pushed on or pulled on by inaccuracy. We used to tap bars of soap at the Westec show to demonstrate this. No pulling out threads in the soft material.
    What's the difference between rigid and synchronized?

  19. #52
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    North Carolina
    Posts
    12
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    7

    Default

    Quote Originally Posted by Cole2534 View Post
    What's the difference between rigid and synchronized?
    This may not be the full picture, but G77 (sync) derives the feed from the spindle speed and the given metric pitch or tpi (I and J). This means any rounding/etc. is happening in the machine and at the last possible moment, which increases speed and accuracy. The feed is synchronized with the spindle speed as it accelerates and decelerates. G84 (rigid) by contrast requires a speed and a feed that together produce a given thread. Any inaccuracies in this calculation will put additional strain on the tap or create less accurate threads.

    An additional benefit of G77 is the L return speed, which can be programmed to be faster than the cutting speed since you're just retracting.

    So G77, specify speed and pitch, feed is synchronized during spindle accel/decel. G84, specify speed and feed, pitch is a result of those two.

  20. #53
    Join Date
    Sep 2009
    Country
    UNITED STATES
    State/Province
    New York
    Posts
    227
    Post Thanks / Like
    Likes (Given)
    79
    Likes (Received)
    69

    Default

    I like the DRAP idea. But I have never used one. I would add that tool life may be extended using a form tap or roll tap. A;so, it will reduce the amount of chips in the machine since a roll tap does not cut. Tool material selection is an important consideration when form tapping.

  21. Likes Volitan liked this post
  22. #54
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    89
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    8

    Default

    Quote Originally Posted by mutiny View Post
    This may not be the full picture, but G77 (sync) derives the feed from the spindle speed and the given metric pitch or tpi (I and J). This means any rounding/etc. is happening in the machine and at the last possible moment, which increases speed and accuracy. The feed is synchronized with the spindle speed as it accelerates and decelerates. G84 (rigid) by contrast requires a speed and a feed that together produce a given thread. Any inaccuracies in this calculation will put additional strain on the tap or create less accurate threads.

    An additional benefit of G77 is the L return speed, which can be programmed to be faster than the cutting speed since you're just retracting.

    So G77, specify speed and pitch, feed is synchronized during spindle accel/decel. G84, specify speed and feed, pitch is a result of those two.

    This doesn't make much sense. Because in order for g84 to work, the feed must be synced to spindle speed as well.

  23. #55
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    North Carolina
    Posts
    12
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    7

    Default

    Quote Originally Posted by sirAIG View Post
    This doesn't make much sense. Because in order for g84 to work, the feed must be synced to spindle speed as well.
    You're right, of course. Some synchronization must still be happening with G84, but presumably letting the control derive the feed entirely from the desired pitch improves accuracy and speed. Perhaps BROTHERFRANK will be able to chime in and clarify a bit.

  24. #56
    Join Date
    Dec 2013
    Country
    UNITED STATES
    State/Province
    California
    Posts
    848
    Post Thanks / Like
    Likes (Given)
    255
    Likes (Received)
    495

    Default

    Brother's Synchronized method synchronizes the spindle position to the Z position. You have the responsiveness and control of the Brother spindle, 10 to 17 hp with a relatively light 2-3 pound collet chuck, synchronizing to the heavier less responsive (slightly ) Z Axis. Z servo is typically 2-3 hp, working through a coupling and ball screw trying to control the mass of the spindle head casting etc... As far as I know, no other builder is doing it this way. It takes a higher quality spindle motor plus Brother's proprietary firmware and fine tuned Accel-Decel curves.

  25. Likes TeachMePlease, Volitan, hanermo, vegard liked this post
  26. #57
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    5,191
    Post Thanks / Like
    Likes (Given)
    933
    Likes (Received)
    2816

    Default

    Rigid tapping is the same as synchronized tapping. Just different terms for the same function. In rigid tapping on Mitsu controls the F address is expressed as pitch. For Fanuc controls F can express feed in IPM in all cases and if the machine builder configures the control for IPR function then F can be IPM or IPR depending on what mode, G94 or G95, is in effect.

    Here is a clip from the Fanuc 16/18 connection manual (functions) describing rigid tapping....

    The rotation of spindle and feed of drilling axis are controlled so that they
    are always synchronous each other in the rigid tapping cycle. Namely, in
    other than rigid tapping, control for speed only is performed. In the rigid
    tapping however, position control is also performed during the rotation
    of spindle, that is, the rotation of spindle and feed of drilling axis are
    controlled as linear interpolation of two axes.

  27. #58
    Join Date
    Jan 2019
    Country
    UNITED STATES
    State/Province
    Pennsylvania
    Posts
    89
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    8

    Default

    So other then fancy wording and arguably better spindles/servos/synconization, brother is doing nothing the others aren't from synchronizing the spindle to the z axis ball screw. At least it has yet to been explained differently.

    Sent from my SM-G955U using Tapatalk

  28. #59
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    5,191
    Post Thanks / Like
    Likes (Given)
    933
    Likes (Received)
    2816

    Default

    Brother has been doing rigid tapping longer than any other control I know of. As far as synchronization goes, Fanuc and Mitsubishi have lots of parameters used to "tune" the rigid tap function. Machine builders get to set those parameters to suit how they want rigid tapping to work to work with their machine. Some may do a better job than others at it.
    Last edited by Vancbiker; 03-08-2020 at 10:46 AM. Reason: clarified some wording

  29. #60
    Join Date
    Sep 2010
    Location
    Oklahoma City, OK
    Posts
    4,945
    Post Thanks / Like
    Likes (Given)
    886
    Likes (Received)
    2071

    Default

    Quote Originally Posted by sirAIG View Post
    So other then fancy wording and arguably better spindles/servos/synconization, brother is doing nothing the others aren't from synchronizing the spindle to the z axis ball screw. At least it has yet to been explained differently.

    Sent from my SM-G955U using Tapatalk
    That's what I got out of it. Lol. Brother may do it better, but the concepts remain consistent.


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
  •