How to measure the radius of a boring bar - Page 2
Close
Login to Your Account
Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 42
  1. #21
    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 Finegrain View Post
    Quite a challenge to measure ANYTHING +/-.001mm,..... then push on the tool with your hand and watch the indicator move . Now re-evaluate the utility of making your +/-.001mm measurement.
    You want to measure/ qualify your boring setup to 0.0000393"
    I would guess that even attempting to get within 0.01mm prior to using it in the machine, will be tough.
    Best of luck!

  2. #22
    Join Date
    Apr 2013
    Country
    UNITED STATES
    State/Province
    West Virginia
    Posts
    59
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    31

    Default

    Quote Originally Posted by G00 Proto View Post
    Take a picture of the setup using your iPhone. Then email it yourself. Import the picture into Solidworks (or any cad cam system). Then draw a centerline down the center of the boring head, and a point at the tangency of the cutter radius. Then set your CAD measuring increments to .001 mm (actually to be ISO, you need to set the increments to .0001 mm). Then simply pull a measurement from centerline to the point. That times two will give you a measurement to the nearest .001 mm.
    You’re never going to get anywhere near .001mm with a photograph. Even if you had the camera in a copy stand capable of that level of accuracy the actual camera isn’t that close to parallel to the body of the phone. Parallax at just 30mm away from the object being photographed will push you outside the requested spec.

  3. #23
    Join Date
    Nov 2013
    Country
    UNITED STATES
    State/Province
    Utah
    Posts
    4,374
    Post Thanks / Like
    Likes (Given)
    1170
    Likes (Received)
    2399

    Default

    Quote Originally Posted by IrritableBadger View Post
    You’re never going to get anywhere near .001mm with a photograph. Even if you had the camera in a copy stand capable of that level of accuracy the actual camera isn’t that close to parallel to the body of the phone. Parallax at just 30mm away from the object being photographed will push you outside the requested spec.
    He was joking dude.

  4. Likes G00 Proto liked this post
  5. #24
    Join Date
    May 2013
    Location
    wi usa
    Posts
    193
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    60

    Default

    Good question, I also kinda wondered that over the years.

    We have fancy automatic tool measuring machine at work. It cost more then my house.

    We still go small and adjust to size. The material, machine differences, setups, speeds and feeds can push hole sizes all over the map. Never had one come off of pre-setter and do perfect hole yet.

    I was taught to test bore on small side and adjust final size from there.


    In my home shop for race engines, I just made up perm set boring heads to my common bores sizes.
    They were all set by going small and adjust to final size.

    I always then try to finished honed bores to +-.0002 or better.

  6. #25
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,719
    Post Thanks / Like
    Likes (Given)
    3231
    Likes (Received)
    717

    Default

    Quote Originally Posted by litlerob1 View Post
    He was joking dude.
    I thought it was pretty funny...

    I'll have to re-read all these posts.

    I know a few old school handy high-precision "Peeps" use analogue microscopes over the tip of a tool mounted on an open machine (like a lathe). Where you can easily see the curvature of the radius of the tip of the tool relative to cross hairs, and with a graticule "fit" the radius.

    I also know from various physics experiments ( old school) that you can have horizontally travelling microscopes (mononcular) on a precision straight track/ stage + scale. Maybe with DTI and gauge blocks , micrometer screw … translation/ translating the microscope accurately that critical dimension can be found. [Other than taking a test cut as some have wisely said.].

    I know that's very vague but pretty sure it can be done... shop microscope that can mount to machine or go in spindle (like a Bridgeport) that some have used in the past might get it within a tenth/ two tenths .

  7. #26
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    7,601
    Post Thanks / Like
    Likes (Given)
    378
    Likes (Received)
    6350

    Default

    I wonder what you could resolve to on the latest phone camera given this field of view?
    Telecentric optics would be nice but you can partially correct in the soft side for distortions.
    At one micron and white light there are other problems but how about a few microns.
    A add on lens?
    When I started in "camera measuring" we had 256x256 sensors... yea a whole 64k.
    Bob

  8. #27
    Join Date
    Feb 2013
    Country
    UNITED STATES
    State/Province
    Idaho
    Posts
    654
    Post Thanks / Like
    Likes (Given)
    132
    Likes (Received)
    711

    Default

    Quote Originally Posted by litlerob1 View Post
    He was joking dude.

    I figured someone would bite on that one

    In all actuality, I kinda wonder what someone could do with a real vision system mounted in a machine. In theory, you could spin a boring boring bar in the spindle and have good enough resolution to pick up the tangency point on opposing shadows. Of course this completely ignores tool and part deflection.

  9. #28
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,719
    Post Thanks / Like
    Likes (Given)
    3231
    Likes (Received)
    717

    Default

    Quote Originally Posted by CarbideBob View Post
    I wonder what you could resolve to on the latest phone camera given this field of view?
    Telecentric optics would be nice but you can partially correct in the soft side for distortions.
    At one micron and white light there are other problems but how about a few microns.
    A add on lens?
    When I started in "camera measuring" we had 256x256 senors... yea a whole 64k.
    Bob
    That's a good thought / thoughts.

    Telecentric lenses are really expensive.

    It is possible to calibrate the snot out of more regular lenses for radial and tangential distortion as well as magnification and principal distance (effective focal distance) + principal point (intersection of optical axis) with sensor as well as determining the rear nodal point of an imaging (perspective center) or point of projection.

    That's kinda my day job / part of the systems I have to roll out.

    There is an neat old school photographic trick is that you can reverse the lens (there's special mounts for that) … So you can image near microscopic objects, (different from a lens extension for macro stuff).

    If you are not doing photogrammatery then regular images suffer from what's called relief displacement... Objects at different heights and distance radially displace away from the perspective center of an image. Think of aerial photography of sky scrapers in New York (camera pointing down) the tops of the sky scarpers radially displace away from the bottoms of the shy scrapers relative to the principal "center" point on the photograph, that's WHY photographs typically are not "Planimetrically" correct.

    There's a lot you have to do to make imagery "Metrology" grade (that's one of the things I do / engineer).

    HENCE your suggestion / general use of Telecentic lenses (it's sort of an industrial hack that works for some situations especially if everything is on one plane.

    TRUE TRUE about diffraction effects of light around edges (like a tool point or sharp edge) [Well done / thumbs up (without being patronizing.). ].

    Phone cameras are weird beasties and there's a LOT of digital processing also that kinda makes some metrology a bit invalid.

    Some of the stuff we do we can fit accurately to an 1/8 th of a pixel / "Photo element".

    On some Medium format sensors that are roughly 10K by 6K "Pixels" … 60 Million + pixels on a bayer sensor.

    Each pixel / photo site is approx. 5 microns each. With the right optics and monochromatic light you can certainly image down to one micron.

    But overall the cheaper easier approach is translate (mechanically) $20 webcam hooked to a microscope. The effective long narrow angle view is repeatable when translated.

  10. #29
    Join Date
    May 2011
    Country
    UNITED STATES
    State/Province
    New Jersey
    Posts
    1,923
    Post Thanks / Like
    Likes (Given)
    173
    Likes (Received)
    1043

    Default

    Dial it in and send it!

  11. Likes cameraman liked this post
  12. #30
    Join Date
    Nov 2014
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    2,719
    Post Thanks / Like
    Likes (Given)
    3231
    Likes (Received)
    717

    Default

    Quote Originally Posted by G00 Proto View Post
    I figured someone would bite on that one

    In all actuality, I kinda wonder what someone could do with a real vision system mounted in a machine. In theory, you could spin a boring boring bar in the spindle and have good enough resolution to pick up the tangency point on opposing shadows. Of course this completely ignores tool and part deflection.
    I think three dimensionally there's a lot that can be done more generally especially for 5 axis snafuu prevention. With cameras placed variously in the machine capturing three dimensional information real time.



    ^^^ Check this out , kinda relevant to Op's question , (the music is beyond stupid and distracting worth turning off lol ).

    But I LIKE the 5 axis calibration routine and corrections thereof through test cuts verified in process to converge on the most probable fit for that machine (i.e. better kinematics + process)…

    Need to watch this again,

    COOL with what I would call a dynamic fixture that the spindle could be used as a ROBOT with an Allen head / socket head and make computed adjustments on the specialized fixture plate (+ rotary table moving to new positions for adjustment / access to mechanic adjustments made by the spindle with hex socket tool) .

  13. #31
    Join Date
    Apr 2013
    Country
    UNITED STATES
    State/Province
    West Virginia
    Posts
    59
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    31

    Default

    Quote Originally Posted by G00 Proto View Post
    I figured someone would bite on that one

    In all actuality, I kinda wonder what someone could do with a real vision system mounted in a machine. In theory, you could spin a boring boring bar in the spindle and have good enough resolution to pick up the tangency point on opposing shadows. Of course this completely ignores tool and part deflection.
    It was fairly well written and I really thought it was someone grossly overestimating the power of their phone. The ISO compliance should have tipped me off I suppose. Alas, I was taken in.

  14. Likes G00 Proto, litlerob1 liked this post
  15. #32
    Join Date
    Feb 2013
    Country
    UNITED STATES
    State/Province
    Idaho
    Posts
    654
    Post Thanks / Like
    Likes (Given)
    132
    Likes (Received)
    711

    Default

    Quote Originally Posted by IrritableBadger View Post
    It was fairly well written and I really thought it was someone grossly overestimating the power of their phone. The ISO compliance should have tipped me off I suppose. Alas, I was taken in.
    All in good fun, thanks for being a good sport.

    It is astonishing how often I take a picture of something with my cellphone and import it into CAD. Then I diligently draw lines, splines and arcs around the often times hand carved part to make a solid model. So much so that I've gotten scientific and use a tripod for perpendicularity and use a 6" scale to get the aspect ratio correct... I figure I can hold plus or minus .030" but I might be optimistic

  16. #33
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    4,408
    Post Thanks / Like
    Likes (Given)
    805
    Likes (Received)
    2374

    Default

    Quote Originally Posted by G00 Proto View Post
    .....I take a picture of something with my cellphone and import it into CAD. Then I diligently draw lines, splines and arcs ....
    The modern take ^ on how to convert old PCM drawings to CNC programs. Sure beats my old method of laying the PCM over graph paper and plotting points, estimating radii, and arc centers from that.

  17. #34
    Join Date
    Feb 2013
    Country
    UNITED STATES
    State/Province
    Idaho
    Posts
    654
    Post Thanks / Like
    Likes (Given)
    132
    Likes (Received)
    711

    Default

    Quote Originally Posted by Vancbiker View Post
    The modern take ^ on how to convert old PCM drawings to CNC programs. Sure beats my old method of laying the PCM over graph paper and plotting points, estimating radii, and arc centers from that.
    You're bringing back nightmares of using Mylars… now that was fun.

  18. #35
    Join Date
    Feb 2014
    Location
    Sunny South West Florida, USA
    Posts
    2,699
    Post Thanks / Like
    Likes (Given)
    10320
    Likes (Received)
    3118

    Default

    Quote Originally Posted by G00 Proto View Post
    I figured someone would bite on that one

    In all actuality, I kinda wonder what someone could do with a real vision system mounted in a machine. In theory, you could spin a boring boring bar in the spindle and have good enough resolution to pick up the tangency point on opposing shadows. Of course this completely ignores tool and part deflection.

    So they actually make this vision system for centering tools in a Swiss lathe, which isn't identical, but shows what you can do with a good vision system in the machine. I had a demo in my shop. Neat stuff. Not worth it to ME, but not worthless, either.

    Perfect Zero Visual Alignment System - GenSwiss

  19. #36
    Join Date
    Apr 2013
    Country
    UNITED STATES
    State/Province
    West Virginia
    Posts
    59
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    31

    Default

    Quote Originally Posted by G00 Proto View Post
    All in good fun, thanks for being a good sport.

    It is astonishing how often I take a picture of something with my cellphone and import it into CAD. Then I diligently draw lines, splines and arcs around the often times hand carved part to make a solid model. So much so that I've gotten scientific and use a tripod for perpendicularity and use a 6" scale to get the aspect ratio correct... I figure I can hold plus or minus .030" but I might be optimistic
    I’d like to see a tilt/shift lens for phones, like they use for fine art and architectural photography.

  20. #37
    Join Date
    Nov 2013
    Country
    UNITED STATES
    State/Province
    Utah
    Posts
    4,374
    Post Thanks / Like
    Likes (Given)
    1170
    Likes (Received)
    2399

    Default

    Quote Originally Posted by IrritableBadger View Post
    I’d like to see a tilt/shift lens for phones, like they use for fine art and architectural photography.
    There are phone apps, to "tilt/shift" images. But I think an actual lens might be asking a little much of my phone. I hate all the cell phone BS though, so the resentment might be bleeding through.

    R

  21. #38
    Join Date
    Sep 2015
    Country
    CANADA
    State/Province
    Ontario
    Posts
    92
    Post Thanks / Like
    Likes (Given)
    43
    Likes (Received)
    35

    Default

    Quote Originally Posted by G00 Proto View Post
    All in good fun, thanks for being a good sport.

    It is astonishing how often I take a picture of something with my cellphone and import it into CAD. Then I diligently draw lines, splines and arcs around the often times hand carved part to make a solid model. So much so that I've gotten scientific and use a tripod for perpendicularity and use a 6" scale to get the aspect ratio correct... I figure I can hold plus or minus .030" but I might be optimistic
    Quote Originally Posted by Vancbiker View Post
    The modern take ^ on how to convert old PCM drawings to CNC programs. Sure beats my old method of laying the PCM over graph paper and plotting points, estimating radii, and arc centers from that.
    FWIW I used to run a FLOW waterjet at an old shop and the software it came with had a function called "rasterizing" where it could draw lines and arcs over the gradients in a JPEG photo.

    It wasn't perfect but it worked pretty damn good and if you can find something similar it can save a lot of time when you need to do this kinda thing

  22. #39
    Join Date
    Feb 2013
    Country
    UNITED STATES
    State/Province
    Idaho
    Posts
    654
    Post Thanks / Like
    Likes (Given)
    132
    Likes (Received)
    711

    Default

    Quote Originally Posted by 007Rob View Post
    FWIW I used to run a FLOW waterjet at an old shop and the software it came with had a function called "rasterizing" where it could draw lines and arcs over the gradients in a JPEG photo.

    It wasn't perfect but it worked pretty damn good and if you can find something similar it can save a lot of time when you need to do this kinda thing
    Unfortunately, rasterizing either in cad cam or in Photoshop yields unsatisfactory results for what I’m doing (same with digitizing). I am generally transitioning a prototype concept into something that can be machined in high volume. By the time I’m done with the cad model it needs to be fully defined and resolved as a solid model that can be opened with no errors by any cad system in the world. It sounds a lot more fun than it is

  23. #40
    Join Date
    Nov 2013
    Country
    UNITED STATES
    State/Province
    Utah
    Posts
    4,374
    Post Thanks / Like
    Likes (Given)
    1170
    Likes (Received)
    2399

    Default

    If I remember right Mastersuk V9 still had "rasterizing". When they switched to X in like "98? they got rid of it.

    R


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
  •