Announcement

Collapse
No announcement yet.

Evaluating a surface plate

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Evaluating a surface plate

    Hi Folks,
    I picked this used surface plate with stand out of a pile at a surplus store, I believe it has been sitting outside for 3 or 4 years at least. All of the labels and nameplate all wore off due to weather so I dont know what grade it is. It had a stack of calibration marks on it (saved by the tape over them) that date back as far as 2000 but it said nothing about the spec either, just the date, size of plate (24"x18") and the checkers signature.
    I gave it a wipe down with varsol but what it really needs is a good scrub with a plastic brush and soapy water to get off all the outdoor grunge (and a rust stain) from sitting outside in Hamilton for so long, at least thats what I figure is the thing to do?
    Its got some pock marks and dings too.
    How do I evaluate this thing properly?
    I suppose I would need another precision surface and some blue to evaluate it? What about a piece of glass?
    As you can see the stand it came with had it supported with a few rough cut boards, lol How should I support it? I have done some searching around here and see its recommended to support it in 3 spots, would hockey pucks suffice for this or what?
    I only paid 40 bucks for it and dont expect to be doing any orbiter layout work on it so I am hoping it will be sufficient for my needs, which will be rebuilding my lathe and hobby type layout work...
    Don't mind my lousy photog skills...



    Any advice would be appreciated!
    Cheers,
    Jon

  • #2
    What will be your intended use for it? If it's to be used in an ordinary open shop for general layout and inspection, it's probably acceptable as is. The dents and dings don't matter if items slide smoothly over its surface.

    If you intend any serious metrology (0.0001" or 3 micron resolution) I strongly suggest you get the plate inspected and lapped to the grade best suited for your needs. A 18 x 24 palte may not be worth lapping given that import plates are so cheap. Wait. Ontario? You Canucks got industrial tool importers don't you? I'm in Wash State USA and I bought three plates the same size as yours for a recent class and they cost @ $55 each or $134 each delivered.

    Does you lady bake bread? If so, don't let her know that granite surface plates make the world's finest pastry and candy surfaces
    Last edited by Forrest Addy; 03-01-2013, 10:49 AM.

    Comment


    • #3
      hockey pucks would be fine.

      If you really want to know, you need some fancy equipment (auto coliminator or interferometer) that only the keenest enthusiasts are likely to have. Or you send it it for calibration and pay several times the cost of the plate.

      next best is checking it with something you know or believe to be flat; another plate, camel back etc. Glass won't tell you anything, you don't know how flat it is and its (somewhat) flexible

      What I think most do is play the odds and take our chances.

      Lots of reason to believe it came from industry and from an era before cheap imports, odds are it was a quality plate

      Odds also are that if it was heavy abused that would show up in visible dings, chips etc.

      last part of the mix is what do you want it for; those are darn good odds if its for general layout work. There is one home shop activity where you do want a plate to be very accurate and that's scraping. If that's the intent its somewhat a bigger gamble that if its for layout work.

      Over time i've traded up, but was able to good things with even the China plates I started with. you're probably ok, but if you really want to know, calibration is the answer
      in Toronto Ontario - where are you?

      Comment


      • #4
        Thanks guy's!
        The surface feels and looks very smooth with the odd pock and ding here and there.
        Its mostly going to be used for layout work and indicating odd things like the heads off the old Ruston Hornsby diesels I plan on rebuilding. However, the main reason for getting it was because I intend to use it for scraping when I rebuild my lathe, its a cheap King Canada chinese lathe and I am just a backyard hack so I am hoping it will be close enough for that? I guess its pretty hard for you to tell from over there though eh? lol
        A camel back is flat? Must be an inside joke...
        Cheers,
        Jon

        Comment


        • #5
          Originally posted by Jon Heron View Post
          A camel back is flat? Must be an inside joke...
          Cheers,
          Jon
          Nope, its a straight edge. Try here: http://1.imimg.com/data/4/K/MY-64727...68_250x250.jpg

          Comment


          • #6
            I see! That thing looks pretty spendy...
            Thanks doc!
            Cheers,
            Jon

            Comment


            • #7
              http://m.youtube.com/watch?v=NWObTpn6dTk
              Precision takes time.

              Comment


              • #8
                If the surface is at all reflective, a quick qualitative (not quantitative) test that I often use is to view a reflection of a straight object (a 4' or 8' fluorescent tube is nice) at some distance (across the room) at a very shallow angle (less than one degree). If there are even small irregularities in the surface, then the image will appear curved instead of straight. This method does not give you any direct measurements of the size of the irregularities, but it does tell you if they are there or not. These observations should be made from several different sides/angles to give a good idea of the total surface.

                I was making a steel drilling and tapping guide block earlier today and used this method to roughly evaluate the quality of the surfaces after filing them parallel and "lapping" them with sandpaper. I know that they have slightly rounded edges but the central portion is very flat. And I also know that they quite symmetric.

                If the surface finish is not highly reflective, a very light coat of oil will temporarily improve this and allow the method to function. It is necessary to keep this oil coat very light to avoid viewing a false surface from a thick coat. The oil should only partially fill in the valleys between the peaks of the surface.

                My apologies to the scraping crowd. This method is NOT intended to replace more quantitative methods that are necessary for producing high precision surfaces. It is only a quick check that will spot problems without putting any numeric value to them.
                Paul A.
                SE Texas

                And if you look REAL close at an analog signal,
                You will find that it has discrete steps.

                Comment


                • #9
                  Well, you want to do scraping......

                  OK, FINE......So DO SOME SCRAPING.........

                  possibility 1:

                  Get two items of cast iron of some length which have one flat and not too wide surface of as close to 24" as possible, 18" is OK. Older cast iron levels are a good item for this, but you probably need to skim the surface on a mill to cut the work involved. If they have minor damage not including at least one surface, they are fine.

                  Lay out a location on the plate that crosses the middle, and crosses any other area that you can identify as "suspect" by evidences of possible wear, lots of dings, etc.

                  SCrape each of the two items 'to" the granite flat at the marked location. Do NOT compare them...... get each to a good spotting on the plate. You do not need 34 points..... you DO need a good even coverage of spots showing that you "have a surface".

                  Now clean one, blue the other, and spot 'em with one end of the pair for-sure 'together". Be careful not to rock them, you ONLY want one end to be for-sure together. to transfer, move one a SHORT distance, like only 5mm or so, back and forth.

                  The granite flat may have a worn area, in which case you will observe a twice-as-bad "belly" in the pair when mated up, so that one end is together, and the other has a gap.

                  It is less likely, but the granite could have a "belly", in which case the two surface will not spot in the middle.

                  As a tie-breaker, check both with the "spin test" using a different place on the plate, preferably near an edge.
                  .
                  .
                  .

                  Possibility #2:

                  Get ONE piece of CI of the size mentioned.... scrape it to a swath across the middle as before noted...... move it to along an edge, and spot it. the print will "probably" tell you what's up.

                  Again, do a spin test as a tiebreaker.


                  In both cases, be very careful about heat from hands..... with a lightweight item like a CI level, hand heat will make it squirm like a fresh-caught worm.



                  If your plate passes this test without looking silly, or failing the spin tests dramatically, it is good for scraping to a fairly decent confidence level. You won't have numbers, but you WILL have indicative results.
                  CNC machines only go through the motions

                  Comment


                  • #10
                    If there are any inaccuracies in it you will not have any measuring devices or uses of it that will show it. Assume it perfect, which it is not, of course. Dings, etc. excepted. Should you think you have found lack of flatness, suspect your instruments or technique. I spent a great many years in a reference calibration lab where we inspected these things and they do have minor errors but it takes lab equipment to see it.
                    Don Young

                    Comment


                    • #11
                      AS he intends it for scraping, he will be a bit more interested in its surface than if he is "just a hobby user" (whatever that is).

                      While I actually agree that a flat not showing evidence of abuse or wear is likely OK, there is no reason for him to guess when it is relatively easy to get a good "reading" of its suitability by the actual process of scraping parts to it.

                      That has a couple very large benefits:

                      1)
                      It gives an actual answer, instead of a guess. Any opinion any of us can offer as to what the surface is like is just so much BS..... it's us guessing, and should be ignored. The OP can TEST the surface and needs no fancy equipment to do it.

                      2)
                      It directly answers the question of whether the thing is suitable for scraping....... if scraping cannot turn up a problem, then you KNOW it is fine..... Any errors are not large enough to adversely affect scraping and spotting.

                      But if parts scraped to it are SHOWN to be not flat by the spotting process I gave above, then you KNOW it isn't good enough. You can SEE that you can't trust it as a flatness reference.

                      No guesses, no "opinions from afar"...... the OP can directly determine everything he needs to know, AND in the process hone his scraping skills preparatory to attacking his projects.

                      I think it makes sense.... apparently that is not a universal opinion............
                      CNC machines only go through the motions

                      Comment


                      • #12
                        Well, so far, my vision is at least correctable to 20/20 so it works for me. Yes, you do need to see the reflected image.

                        The human eye is a remarkable thing. Coupled with the brain, it is capable of correcting for many distortions in the simple lens system that it employs to put an image on a roughly spherical surface. We do see straight things as straight. And if a relatively simple glass or plastic lens is introduced before the eye's own lens (eye glasses, prescription or otherwise) the brain quickly adjusts the perception of straightness to account for any new distortion introduced.

                        I find it fairly easy to see the deviation from a straight image the even a 0.001" turned down edge will produce. With a bit of effort, smaller deviations can also be seen. You just have to know what to look for.

                        I am not familiar with all forms of vision defects so I can not speak for everybody.

                        Paul A.

                        Originally posted by oldtiffie
                        Paul that may work if your vision is perfect and if you wear spectacles/glasses that neither they nor your eyes introduce any distortion.
                        Paul A.
                        SE Texas

                        And if you look REAL close at an analog signal,
                        You will find that it has discrete steps.

                        Comment


                        • #13
                          Thanks for all the good ideas folks!
                          I gave it a proper scrubbing with a plastic scrub brush, hot water and plenty of my home made glycerine soap, what a difference!
                          Its as smooth as a baby's but! There is a spot on the one side about 4" in where some pig with hands was beating on it and left a few chips, its only a small area though and shouldn't cause me much of a problem...
                          Are the sides of the plate precision ground also?
                          Once I get a scraper and start practising I should be able to get a feel for if this thing is consistent throughout or not...
                          Currently I have an aerosol can of blue and it is not so easy to control, is there a better way? Buy a can of it and a paint brush maybe? Or just spray it on and use a brush... What do the pro's do?
                          Cheers,
                          Jon

                          Comment


                          • #14
                            Originally posted by Jon Heron View Post
                            Are the sides of the plate precision ground also?
                            no, precisions costs big bucks so its only present on the side the purchaser bargained for, the top


                            Currently I have an aerosol can of blue and it is not so easy to control, is there a better way?
                            what you have is (likely) layout fluid/spray. This is not what you want...you want prussian blue, think oil paint that doesn't dry out. kbc has it if you can find somewhere with reasonable prices
                            in Toronto Ontario - where are you?

                            Comment


                            • #15
                              Thanks!
                              So would I want the dykem in the squeeze tube, or the stuff in the bottle with the applicator brush?
                              From this page: http://kbctools.com/usa/Navigation/N...m?PDFPage=0377
                              Any advice where to get a scraper locally too?
                              Cheers,
                              Jon

                              Comment

                              Working...
                              X