diy surface plate lapping - Page 2
Close
Login to Your Account
Page 2 of 6 FirstFirst 1234 ... LastLast
Results 21 to 40 of 102
  1. #21
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by Pete F View Post
    You scraped your straight edge from this plate right? I'd suggest the issue is the plate is not being properly supported
    Yes, i scraped the SE along the plates length. Im not so sure that the mounting is the problem though. Given that the mounts (according to specs) are placed too close to the edge of the table, id expect any sag to occur across the tables length, rather than as the level readings (and prints) suggest across its width, but do i know. Its a shame theres no mount id marks on the bottom of the plate.

    Anyways, ive mapped the plate as best i can on its current supports. Tomorrow ill rig up some timber with mounts placed 20% in from the edge of plate, then ill take another set of readings. Be interesting to compare.

  2. #22
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    I took some measurements today in a british flag type style as suggested, the only thing i changed was adding an extra row of measurements across the plates width (3 rows along length, 4 across the width). This was so i could try and base the level off the same points on both axis.

    imag1711.jpg imag1712.jpg imag1713.jpg

    I split each graduation of the level into 5ths (it felt comfortable), each graduation over 230mm = 11.5 microns (0.00045") so a 5th comes out at 2.3 microns (0.0001)

    The largest top to bottom measurement i got was across a diagonal,. This came out at 9 microns (0.0004) ish. Got to take all of this with a jumbo pinch of salt given the context. Also, for the size of the plate, its a pretty low resolution reading due to the length of the level.

    Just to be clear, im most interested in the level results, the prints are to back up what the level tells me.

  3. #23
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2056
    Likes (Received)
    1283

    Default

    Quote Originally Posted by Demon73 View Post
    id expect any sag to occur across the tables length
    Correct. You then scraped your straight edge to that bowed surface. I'd suggest your straight edge is not straight.

    I was just going off your initial posting and the bluings you showed, where I believed the surface plate was supported at its ends and you had created an inaccurate straight edge from it. That was indeed the case, and what was needed was to support it correctly and then scrape your straight edge to the plate then. I expect you would have found it to be relatively flat. However since you've now got stuck in to it with abrasive sorry I can't offer any more clues as anything is now possible.

    Pete

  4. #24
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2056
    Likes (Received)
    1283

    Default

    Double post. Sorry

  5. #25
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Well if the table had sagged across the width (turned concave), id expect any SE scraped along the plates length to be convex. The print across the plate (the first one in post 3) suggested that the SE was concave, which it isnt, according to testing it with a level and checking (just for the hell of it) across the width of the plate.

    The initial readings i took using the level showed a clear sag (some 0.0005") running through the centre of the plate along its length, not across its length.

    I 100% agree with you tho that the SE isnt straight! Will be the first thing i sort when my plates straight

    Cheers for the input

  6. #26
    Join Date
    Jul 2007
    Country
    UNITED KINGDOM
    Posts
    4,331
    Post Thanks / Like
    Likes (Given)
    2045
    Likes (Received)
    1847

    Default

    The stand is a three point support with two additional outriggers that are adjusted for minimal clearance and minimal load. This stops the plate tipping over when a large load is placed on a corner at the single-support end. All the decent UK stands I've seen (mostly Crown Windley) are like this, including my 2x3' and my 3x4' ones. For a a 2x3' 4" surface plate this is more than adequate and the spacing will be pretty close to the ideal minimum distortion spacing (Which is a compromise on a rectangular plate).

    Silicon carbide will cut granite quite slowly, which is good, it'll take you longer to ruin the surface plate. Diamond powder on a cast iron lapping plate really cuts quite fast.

    You're right, a half thou/10" level is nominally ten arc second resolution. If you make a sled for it with three feet spaced about 2.5" apart and 3/8" diameter, you'll get 4 or more times the resolution. Bear in mind that it'll fall over if you look at it crossways! Get some 800-1200 grit paper. Lay it on the surface table, abrasive up and lap the feet nominally flat against it if they don't come out perfect from the construction effort.

    Bear in mind that the difference between good cover and partial cover with blue on a surface can be less than a tenth of a thou and you can get a lot more movement than that with a bit of sunshine on top of the plate. The spec that Bebop mentions, calls for a temperature gradient of less than 0.5°C from top to bottom of a 4" thick plate and gives .35 thou per °C distortion on a plate of your size. If you Google for that standard, there are a number of grainy pdfs about.



    I've currently got a 0.2 second Hilger&Watts autocollimator. I haven't constructed any mirrors yet, but I've got cast-iron stock and HSS bar to make the sleds out of and know which Edmund's optics mirrors I need to save up for. I've got 50 carats each of 5-7micron and 1-2 micron diamond powder; an originally grade A 48" granite straight edge and I just managed to get a Mahr Extramess 2001 10 micro-inch digital+semi-analogue gauge. I feel that I'll be ready to start working on my surface tables in about 18 months time.

    I'm aiming at the higher end of what's possible for an amateur (got a surface grinder to refurbish that needs better standards than the lathe and mill did), but I think you're already on the edge of what's possible with the kit you've got and aiming far higher than most are capable of.

    Bloody good luck to you, but don't be surprised if it all goes to hell in a handbasket!


  7. #27
    Join Date
    Jul 2008
    Location
    Sydney, Australia
    Posts
    3,753
    Post Thanks / Like
    Likes (Given)
    2056
    Likes (Received)
    1283

    Default

    The length spacing if that stand is nowhere near the airey point lengths, assuming thats where it was supported when made. It is being supported virtually at its edges. If the outrigger points are incorrectly adjusted and instead support the plate, it will dish across its width.

    With a single level, how did you confirm that the readings you were getting were from the plate itself and not movement of the stand as the weight of the level was being moved around it thus tipping the whole plate as a plane?

  8. #28
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by Mark Rand View Post

    Bloody good luck to you, but don't be surprised if it all goes to hell in a handbasket!

    Thanks for encouragement (and good tips!) Mark. Especially the bit about silicon carbide cutting slow! The diamonds wouldnt have shown any mercy to my heat banana'erd SE blunder!

    PS: Youve just made me realise ive no goal atm. Im gona say something better than B grade, hopefully near A grade. Goto when to stop eh!
    Last edited by Demon73; 05-31-2013 at 05:44 AM.

  9. #29
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by Pete F View Post
    The length spacing if that stand is nowhere near the airey point lengths, assuming thats where it was supported when made. It is being supported virtually at its edges. If the outrigger points are incorrectly adjusted and instead support the plate, it will dish across its width.
    Yeah its a shame theres no id marks on the plate. Might try and contact the name on the plate (if there still in business). The outrigger are wound down and are well clear of the plate.

    Quote Originally Posted by Pete F View Post
    With a single level, how did you confirm that the readings you were getting were from the plate itself and not movement of the stand as the weight of the level was being moved around it thus tipping the whole plate as a plane?
    Ahh good points! Maybe thats why Forest suggested using two levels. Im pretty sure the setup is rigid enough, it took my bodyweight in an unsupported corner to move the bubble 1 graduation. Could possibly do a paper test under the outriggers. Set it up so the paper just snags between the outrigger and the plate. Would at least be some indication for the top of the setup.

  10. #30
    Join Date
    Feb 2005
    Country
    UNITED STATES
    State/Province
    Utah
    Posts
    327
    Post Thanks / Like
    Likes (Given)
    371
    Likes (Received)
    118

    Default

    To answer your question ... yes I have done a bit of plate lapping in years past ... i talked to a buddy yesterday that still laps plates he now uses diamon powder 300-400 mesh.
    Sprinkel diamon powder lightly on the cast iron plate, roll in with a hardened roller (an old bearing with a handel will work) dust off the excess then move plate randomly over the high spots.
    areas that are cutting will show as a mate surface finish, low spots will stay shinny. it will be helpful in your endevor if when you "map" your plate you draw it on a paper, list the readings you've gotten, mark the high spots and "connect them like you were drawing a topographic map. this will help you to visulize the highs and lows so you dont create more work for yourself
    :-} As Mark Rand notted you do need to have the plate suported corectly before you begin.
    Cheers Don

  11. #31
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    16,911
    Post Thanks / Like
    Likes (Given)
    1444
    Likes (Received)
    2607

    Default

    With the various people guessing problems, such as distorted plate, etc, it may be time to step back a bit.

    What would have been a good idea to begin with, would have been to scrape another surface to the plate exactly as the original SE was scraped. The second surface would need to be stable, but by no means does it need to be a useful SE as far as size and weight (wouldn't hurt, though).

    With those two, then, the SE and the "temporary surface", scraped to the same granite "flat", you can compare them by bluing one to the other. If they blue up, there isn't likely to be much wrong with the 'flat", as they would both have the same error, and comparing them then doubles the net error. That is the same idea used when making three identical flats or SEs.

    That would be a good "tiebreaker" as far as identifying the existence and type of faults in the plate.

    At this point, you probably are not able to do that without a lot of extra trouble. I don't know if the "flat" is worth it, depends on what you paid. It isn't really practical to scrape two things to the flat every time you want to check a new round of lapping.

    One thing you can do is to find another smaller flat that can be borrowed, and used to check the bigger one. If there really are problems with the larger flat, the two will not print together all over the surface. I say smaller, because it helps a lot if the "checking" flat can be actually lifted and maneuvered, but ideally it would be the same size, and known good.

    My understanding of the support problem is that with most "flats", a small error in position of the three supports makes relatively little difference in flatness, errors smaller than you are able to quantify with the 0.0005" level, although they may be "visible" as slight changes in position between marks.

    with a 3' x 2' flat, reasonably correct position vs a very wrong position (like edge support) should likely show up, but even there I don't know what a position error of a hundred mm or so would turn out as far as errors. After all, the flat has to be stiff enough to support items which would reasonably be put on it to check flatness, and it isn't always possible to put them over support points with perfect balance.

  12. #32
    Join Date
    Dec 2000
    Location
    Bremerton WA USA
    Posts
    10,644
    Post Thanks / Like
    Likes (Given)
    40
    Likes (Received)
    4070

    Default

    Be careful of the straight edge Vs flat cross-check. You can get both to agree perfectly but you will have proved only a spherical shape having a large but unknown radius that may be either convex or concave..

    You need either a separate means of checking via a level (ensures the plane's radius is equal to the Earths) or optics (which ensures a flat surface provided the intersection of any folded optical path is precisely at right angles to the surface being checked.)

    A third option is a third reference surface which is employed in "averaging of errors" technique.

    BTW: flat is a surface can be said to have a theoretically infinite radius.

    I know we're splittinghairs here but flatness of surface plates require split hair accuracy.

  13. #33
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by bebop View Post
    To answer your question ... yes I have done a bit of plate lapping in years past ... i talked to a buddy yesterday that still laps plates he now uses diamon powder 300-400 mesh.
    Sprinkel diamon powder lightly on the cast iron plate, roll in with a hardened roller (an old bearing with a handel will work) dust off the excess then move plate randomly over the high spots.
    areas that are cutting will show as a mate surface finish, low spots will stay shinny. it will be helpful in your endevor if when you "map" your plate you draw it on a paper, list the readings you've gotten, mark the high spots and "connect them like you were drawing a topographic map. this will help you to visulize the highs and lows so you dont create more work for yourself
    :-} As Mark Rand notted you do need to have the plate suported corectly before you begin.
    Cheers Don
    Got you. I did read some stuff about using diamonds rolled into a plate. As Mark says, probably for the best im using paper as its a lot slower. if id gone wrong as i did with diamonds in that SE id expect very rounded edges by now! . Ive got 8x10 surface plate rigged and a 4x5 angle plate set up with a wood grips.
    After getting all about airy points and such i decided to try and measure any movement. I reinstalled the plate onto its original supports after a nonsense mess around with timber (dont ask!) 3 points to the floor and 3 points to the plate. With the level placed along the length of the plate I placed 40kg in the centre over an 8x4 area. The bubble stayed still! ie no detectable bend in the plate. Im calling that good.
    Then i mocked up a dummy level and moved it around the plate as you would use 2 levels (properly, sorry Forest, i understand now). i managed to get a very slight, almost unmeasurable movement in the level as i moved the dummy it across the single support to the other side. Im calling that good.
    The general plan i have is to work from a known flat area (a plane that is flat according to the level) and work from there.
    Cheers again
    Last edited by Demon73; 06-01-2013 at 05:04 AM.

  14. #34
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by Forrest Addy View Post
    You need either a separate means of checking via a level (ensures the plane's radius is equal to the Earths) or optics (which ensures a flat surface provided the intersection of any folded optical path is precisely at right angles to the surface being checked.)
    Ah please, ive just got my head around the twin level thing!

  15. #35
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by JST View Post
    What would have been a good idea to begin with, would have been to scrape another surface to the plate exactly as the original SE was scraped. The second surface would need to be stable, but by no means does it need to be a useful SE as far as size and weight (wouldn't hurt, though).
    .
    I do actually have a 5ft SE scraped across the diagonal of this plate, its only roughed but it did hit everywhere. To be honest im loathed to use any SE to qualify things except as a final check. That might change though

  16. #36
    Join Date
    Oct 2003
    Location
    East Peoria, IL, USA
    Posts
    5,270
    Post Thanks / Like
    Likes (Given)
    302
    Likes (Received)
    708

    Default

    today is the last day of encos free shipping for this month and at the same time 12x18 surface plate is on sale for $46 with free shipping you cant go wrong.

  17. #37
    Join Date
    Jul 2007
    Country
    UNITED KINGDOM
    Posts
    4,331
    Post Thanks / Like
    Likes (Given)
    2045
    Likes (Received)
    1847

    Default

    What's the code for free shipping across the Atlantic?

  18. Likes bebop, baldwin, Demon73 liked this post
  19. #38
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Quote Originally Posted by Demon73 View Post
    Well if the table had sagged across the length (turned concave), id expect any SE scraped along the plates length to be convex. The print across the plates width (the first one in post 3) suggested that the SE was concave, which it isnt, according to testing it with a level and checking (just for the hell of it) across the width of the plate.

    The initial readings i took using the level showed a clear sag (some 0.0005") running through the centre of the plate along its length, not across its length.

    I 100% agree with you tho that the SE isnt straight! Will be the first thing i sort when my plates straight

    Cheers for the input
    Just noticed i muddled length / width, the corrections are in bold.
    (beyond edit time)

  20. #39
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    A quick update of yesterdays efforts.

    The first thing i had to do (especially as i only have one level) was make sure that the plate and and stand were solid. Using things that were a bit heavier than the level (the dummy), i left the level in place and shifted the dummy around all ways. The bubble stayed still, so i called it good.
    imag1717.jpg

    Next job was to get the long sides parallel to each other. The maps id taken suggested that the rear right corner (as viewed from the front) was the lowest area, so the rear left edge of the plate received the most attention. Both edges needed slight straightening.
    imag1722.jpg

    When i was satisfied with the long edges i turned my attention to the plates width. I started a rulers width in from the edge (as id done with the long edges but forgot to mention) and worked my way along the plate in 5" intervals, in grid like fashion. When i was happy with one i moved to next and so on.
    imag1724.jpg
    The width was very slightly convex (thats a first!) on the left side, was to be expected given the material id removed setting the long edges straight. Things got straighter as i got to the right edge. Heres a walk through the first effort after id worked it (this is what i aimed for in all of the measurements).
    imag1725.jpg imag1726.jpg (thats strange, should be two more pics but they wont upload! they looked the same anyway lol))

    Next i carried out the same procedure along the plates length. It needed no work that i could see, as expected if id done my work across the width correctly i guess. Finally i checked the diagonals, which measured up good too.

    Thats all i can do for now i think, will have to cobble up something to check the repeat side of things. Forrests surface gage sounds like an option but i need to get some kit (which im hoping to find at a friend whos having a clear out, tuesday with any luck).
    Cheers all
    D
    Last edited by Demon73; 06-01-2013 at 10:54 AM.

  21. #40
    Join Date
    Mar 2013
    Location
    Surbiton, surrey, UK
    Posts
    1,236
    Post Thanks / Like
    Likes (Given)
    1364
    Likes (Received)
    731

    Default

    Interesting. These guys use a guide coat to show the progress. Reminds me of my car bodywork days!
    calibration and resurfacing service of granite surface plates - Cleveland Ohio, Chicago Illinois Optiaml Calibration

    Edit:
    Actually. Taking another look at the pics. It looks as though they rub the plates dry, without water. The light parts are the places that are high!
    areas that are cutting will show as a mate surface finish, low spots will stay shinny.


    I wonder what caused that ring in the corner?

    dry-rub-1.jpg dry-rub-2.jpg dry-rub-3.jpg
    Last edited by Demon73; 06-02-2013 at 04:27 AM.


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
  •