Announcement

Collapse
No announcement yet.

Impressive $20 brake rotor for tramming the mill

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

  • #46
    Originally posted by wierdscience
    Basically it consisted of a sqaure block fixed to the bottom of the B-port mills belt case under the motor.Bellow that were two indicators fixed to the ram set at 90* to each other.
    Very clever -- that would be an easy afternoon project with some ground flat stock, although I'm having a hard time picturing where to bolt it (I'm at work right now).
    "Twenty years from now you will be more disappointed by the things that you didn't do than by the ones you did."

    Comment


    • #47
      Originally posted by lazlo
      Very clever -- that would be an easy afternoon project with some ground flat stock, although I'm having a hard time picturing where to bolt it (I'm at work right now).
      looking at the patent and the stuff on the net its got something to do with the eyebolt. $350 gets both dial gauges according to the link.

      http://www.nolansupply.com/bysubcate...lse&specs=True

      http://www.freepatentsonline.com/5908994.html

      Comment


      • #48
        delete this post?

        Comment


        • #49
          Cool -- thanks Derek.

          Looks pretty simple:

          "Twenty years from now you will be more disappointed by the things that you didn't do than by the ones you did."

          Comment


          • #50
            Forgive my lack of anatomical detail of BP but are those parts of the motor housing suitable for measuring 0.0005 deflections? btw this gives about 1 part in 30,000 approximately 1' of arc. But even home brew with ebay bits and no royalty to the patent holder its more than £30. or $60
            Last edited by derekm; 08-18-2008, 03:04 PM.

            Comment


            • #51
              Tramming and tilt

              That's all true Derek.

              But people should be aware that it is possible - even when the mill seems correctly trammed.

              Correcting it may be as simple adjusting the knee gibs. It can vary with the tightening of the knee "Z" clamps. With either of these cases it can be the difference between tramming with the knee clamped and then having the clamps released to "edge-find" and then leaving them. If the mill is in good condition it will or may be insignificant.

              Having a large weight such as a large rotary table left on the end of a mill table while tramming and then either removing it and/or moving the rotab under the spindle may cause it in the "X" plane.

              It is relatively easy (less difficult??) to adjust (for) with a mill/drill as the vertical dove-tail can be adjusted (for "X" and "Y") relative to the table and carriage.

              An instance (very extreme case - for example).

              If the vertical dove-tail (column) is OK in "Y" buy tilted 1 degree "left" (anti-clock-wise) in "X" it can be seemingly correctly adjusted for tram by applying 1 degree "right" (clock-wise) at the left-right tilt at the milling head. The tram will seem perfect - which it is.

              If the milling head is raised the axis of the spindle, which may be visualised with a laser "spot" aligned to and inserted in the quill chuck/collet, will move left on the table as the head is raised up the dove-tail. The converse also applies.

              Consequently a hole, located under the quill axis in the lower position will be incorrect in a higher/lower position of the head as will be shown by the laser pointer moving left-right on the table as the head is raised/lowered.

              This lateral movement will be equal to the sine of the tilt angle x the head movement up/down the tilted column.

              In the case of 1 degree it is 0.0175" inches left/right for each 1" the head is raised/lowered.

              In the case of 0.1 degree it is 0.0017" per inch.

              I chose 0.1 degree as that is about the calibrated scale on a vernier or digital protractor.

              A good "Class/type 2" (shop use) machinist square will pick up 0.002" per 6" (0.0003" per inch).

              The same case can be made for an error on the "Y" plane.

              The solution is to have the dovetail (or round in the case of many mill-drills) vertical column travel at 90 degrees to the table in both "X" and "Y" and for the head to be correctly trammed as the head spindle axis will then be parallel to the dove-tail or round column (as the case may be) axis. In this state the "laser pointer" will not (or should not) "move" on the table.

              The verticality of the mill-drill dove-tail slide or the round column with regard to the mill table is adjusted by packing/shimming the column base where it is bolted (4) to the mill base on which the "Y" slide/way is machined.

              In the case of a mill-drill where the column can be tilted easily (as on many small mini/micro mills) the left/right ("X") adjustment is quite easy. The forward/back ("Y") adjustment will/may be by shimming.

              This is done much easier by using either a good test dial indicator (TDI) and either or both a:

              "master square":
              http://i200.photobucket.com/albums/a.../Squaring1.jpg

              http://i200.photobucket.com/albums/a.../Squaring3.jpg

              http://i200.photobucket.com/albums/a.../Squaring4.jpg

              or a machine "frame" level (square):
              http://i200.photobucket.com/albums/a...st_Square1.jpg

              http://i200.photobucket.com/albums/a...st_Square2.jpg

              Comment


              • #52
                Good Info!

                oldtiffie 08-18-2008 03:44 AM

                Blending

                I've heard it all with regard to "tramming" the vertical milling spindle to the milling table.

                So let's take that as a "given" irrespective of what method/s you use or don't use.

                When was the last time you checked your mill table for parallel to your "X" and "Y" ways? Or that your "X" and "Y" ways are accurately at 90 degrees to each other. Or that your table top is accurately at 90 degrees to your "knee" vertical ways (dove-tail or square/flat?).

                If your knee has "sagged" say 0.10 degree (6 arc minutes) and you tram your spindle to your table it will be fine and you will get a nice even over-lapped "circle/arc" pattern on your job.

                (arctan 0.1 arc minute = 0.0017 = 0.0017 per inch = 0.0140" per 8").

                So if you trammed your spindle over 8" and then moved the table 8" in the "Y" direction you would still be trammed correctly. But two cuts 6" apart using "X" using a fly-cutter set to 8" would have a significant "step" of (0.0017 x 6 = 0.0140") - or there-abouts - instead of the smooth over-lapping "blended" cut you may have expected (or been hoping for??).

                Ever wondered about that "unexplained" "saw-tooth" effect you have been getting between parallel cuts (usually in the "X" direction) when your "tram" has been "perfect"/"dead nuts"/"spot-on"?

                I only used the "0.1 arc degree" as an example but it is quite possible that similar smaller errors exist that no amount of "tramming" will detect.

                I will/may address vices later.


                oldtiffie 08-18-2008 03:33 PM

                More tramming.

                Now excuse me while I butt in and get back to the "right tram but tilted table" bit.

                I will discuss the "tilting" later.

                Back to the scenario where (let's say I - me) have my mill spindle trammed to my table and that my table is tilted forward.

                The previously-mentioned "saw-tooth" effect between adjacent nicely over-lapping circular patterns is not the only "hidden" problem.

                If the knee is raised to, say, locate an edge or edges or the centre of a hole (in "X" and "Y"), and if the knee is dropped/lowered, the axis of the - lets say hole, the principal is the same for anything - the coincident centres will remain parallel but will diverge to the extent that the quill spindle axis and the axis of the hole are no longer coincident. This will increase as the distance between the initial location height of the knee and the machining positions widen. The converse similarly applies.

                In this case there will or may well be an unexpected and unexplained alignment error such that an otherwise apparently correctly aligned/located detail/position may not be accurately repeated with the lowering/raising of the "knee". This may well result in a hole drilled or bored by using the quill down-feed, or the knee up-feed not having the same relationship to the referenced edges (used by the edge-finder etc.) as previously aligned or as expected.

                I have only use a tilt/"sag" in the "knee" in the "Y" plane. The same can occur if the mill saddle or table tilts/"sags" in the "X" plane. There may well be both "X" and "Y" components in which case they may well compose a vector that is between but neither wholly within either or both the "X" nor "Y" planes.

                A similar effect and situation arises with an improperly aligned and set up column and/or spindle on a vertical mill-drill.

                All of this matter is directly applicable to situations that may arise in what may for all intents and purposes, a correctly "trammed" milling machine.

                And yes, a brake disc rotor or similar is quite in order to be used - or not used (your choice) in any of these posts that I make

                I will discuss these topics further later

                oldtiffie 08-18-2008 10:54 PM

                Tramming and tilt

                That's all true Derek.

                But people should be aware that it is possible - even when the mill seems correctly trammed.

                Correcting it may be as simple adjusting the knee gibs. It can vary with the tightening of the knee "Z" clamps. With either of these cases it can be the difference between tramming with the knee clamped and then having the clamps released to "edge-find" and then leaving them. If the mill is in good condition it will or may be insignificant.

                Having a large weight such as a large rotary table left on the end of a mill table while tramming and then either removing it and/or moving the rotab under the spindle may cause it in the "X" plane.

                It is relatively easy (less difficult??) to adjust (for) with a mill/drill as the vertical dove-tail can be adjusted (for "X" and "Y") relative to the table and carriage.

                An instance (very extreme case - for example).

                If the vertical dove-tail (column) is OK in "Y" buy tilted 1 degree "left" (anti-clock-wise) in "X" it can be seemingly correctly adjusted for tram by applying 1 degree "right" (clock-wise) at the left-right tilt at the milling head. The tram will seem perfect - which it is.

                If the milling head is raised the axis of the spindle, which may be visualised with a laser "spot" aligned to and inserted in the quill chuck/collet, will move left on the table as the head is raised up the dove-tail. The converse also applies.

                Consequently a hole, located under the quill axis in the lower position will be incorrect in a higher/lower position of the head as will be shown by the laser pointer moving left-right on the table as the head is raised/lowered.

                This lateral movement will be equal to the sine of the tilt angle x the head movement up/down the tilted column.

                In the case of 1 degree it is 0.0175" inches left/right for each 1" the head is raised/lowered.

                In the case of 0.1 degree it is 0.0017" per inch.

                I chose 0.1 degree as that is about the calibrated scale on a vernier or digital protractor.

                A good "Class/type 2" (shop use) machinist square will pick up 0.002" per 6" (0.0003" per inch).

                The same case can be made for an error on the "Y" plane.

                The solution is to have the dovetail (or round in the case of many mill-drills) vertical column travel at 90 degrees to the table in both "X" and "Y" and for the head to be correctly trammed as the head spindle axis will then be parallel to the dove-tail or round column (as the case may be) axis. In this state the "laser pointer" will not (or should not) "move" on the table.

                The verticality of the mill-drill dove-tail slide or the round column with regard to the mill table is adjusted by packing/shimming the column base where it is bolted (4) to the mill base on which the "Y" slide/way is machined.

                In the case of a mill-drill where the column can be tilted easily (as on many small mini/micro mills) the left/right ("X") adjustment is quite easy. The forward/back ("Y") adjustment will/may be by shimming.

                This is done much easier by using either a good test dial indicator (TDI) and either or both a:

                "master square":
                http://i200.photobucket.com/albums/a.../Squaring1.jpg

                http://i200.photobucket.com/albums/a.../Squaring3.jpg

                http://i200.photobucket.com/albums/a.../Squaring4.jpg

                Thanks for going to the trouble Mick. Especially since I am reassemling my old mill and will have a lot of checking to do.


                Rgds
                Michael

                Australia

                Comment


                • #53
                  Originally posted by lazlo
                  Cool -- thanks Derek.

                  Looks pretty simple:

                  That's the one size fits all version.One I am thinking of is more compact with shorter overhangs on the dial mounts.It also doesn't pick up on the motor or belt housing,but instead a 1" sqaure block under the belt housing near the motor shaft center.

                  I lost the pic I had of it on another computer.
                  I just need one more tool,just one!

                  Comment


                  • #54
                    More alignment stuff

                    Originally posted by miker
                    ................................................

                    Thanks for going to the trouble Mick. Especially since I am reassemling my old mill and will have a lot of checking to do.

                    Rgds
                    Thanks Michael - if it helps someone I am glad to have helped.

                    There will be more yet - a day or so.

                    As I recall, you have horizontal mill.

                    You might like to take a look here - the rest of the website is well worth a look too - its a good start:
                    http://shopswarf.orconhosting.net.nz/hormill.html

                    http://shopswarf.orconhosting.net.nz/sindex.html

                    Comment


                    • #55
                      Originally posted by wierdscience
                      Cummalitve error is but a small concern,the mill going out of tram while in use eclipses it.Bridgeports and their related kin are as rigid as wet noodles.It is very much possible for a two flute 3/4" endmill to chatter a BP head out of tram.

                      Putting a cheater pipe on the wrench when tightening the tie bolts won't solve that either.


                      Rather than having to re-tram everytime I use a two flute endmill OR going from my RT table to my vise OR using my flycutter, or having to use the knee in a high position VS a low position OR using the table X axis off to one side more than the other OR using the Y axis in the more for or aft position OR basically all of which I just covered and more which equates to just about every milling operation I make (whew)
                      I just pay attention to the parts im building, I look at the tool patterns and also stop short of my final destination for measuring

                      Comment


                      • #56
                        Originally posted by Fasttrack
                        I know what AK means, but at school I tram the head to whatever I'm working on every time I use it. I am very suspicious of those machines, since they are often crashed or abused before I get there. There is never any guarentee that the vice is square/parellel to table travel or that the head is trammed to the vice (or table or index or whatever happens to be neccessary)

                        FT, I would be doing exactly the same as ALL of your equipment your using is in a constant state of "flux" --- although this has nothing to do with a one man shop and someone knowing their equip. its still important that you bought up another "variable" as with your situation it would be take all the precautions and still keep your fingers crossed.

                        Comment


                        • #57
                          What tiffie has explained is a simple concept that I have noticed is usually ignored in discussion about this sort of thing. It's the concept of dependent adjustments. If something is out of adjustment such as the alignment of the head to the table on a mill then is does no good to check the very last adjustments first in a long string of dependent adjustments.

                          One must go to the start of the dependency chain and start checking there. Then you work your way along ensuring that each step is correct before doing anything to the next dependent adjustment. This was the source of no end of problems when I worked for Xerox. We had many machines that had dependency chains 10 to 15 steps long for some assemblies and each step was checked by micrometer, gauge blocks or other standards before the next. Some service reps would just go in and try to correct a problem by adjusting the end point first and this would simply cause further problems.

                          Adjusting mill head tram is one of these types of adjustments. Even though on many machines there are no explicit adjustments for the alignment of table to column or similar it should still be checked.

                          It's important to use a datum that can be relied on for all of the dependent adjustments in a chain. It's no good to use the last item adjusted as the reference for the next. If you do that then tolerances will stack and you will have no true idea where anything is.

                          I spent a lot of time thinking about this problem when I designed my mill. After over two decades of dealing with these kinds of adjustments I had a very good idea of what does and doesn't work. The number one most important item is that adjustments need to be easy to make. Second is that once made some provision must exist to securely lock the adjustment. Another is that interdependency should be avoided if at all possible so that you don't have to go back and forth to iterate toward a proper setting. Each adjustment in the chain should only depend on the ones before it and not be affected by the ones after it.

                          As an example, on my mill I have round rod ways in pairs to support both the table and the head. Tramming consists of starting at the bottom and working your way up. The X axis ways are the bottom so they are adjusted first. Parallelism is not adjustable as it is set by the original machining process and cannot vary. The roll, twist and slope is adjustable by a pair of height adjusters at each end of the ways.
                          The datum for all adjustments is the bed of the machine which was hand filed and then scraped to better than .001" across the entire surface.



                          The Y ways are not adjustable in relation to the X ways as that would create an interdependent adjustment for the table position relative to the mill bed. Instead, what is called a hardpoint is created. The bearings blocks that carry the Y carriage were milled as a single unit while clamped to a way shaft so the height is the same for all 4 bearings blocks.

                          The next adjustment is the top and bottom of the Z ways. They use opposing setscrews, a pair on the front and back at bottom and the same at top of the ways. The plates that carry the way shafts are line bored together so that parallelism is assured and fixed. Planarity, tilt and twist of the ways is adjustable and the set screws are locked against each other to fix the adjustment. The plates are secured further by clamping bolts. It's important that clamping alone not be depended on for an adjustment to remain fixed.



                          Last is the spindle head which can be adjusted in forward/aft tilt and in roll around the Y axis. The set screws oppose the clamping bolts and prevnt relative motion when clamped. This is the final adjustment in the sequence when tramming this machine.



                          edited for typos
                          Last edited by Evan; 08-20-2008, 07:45 AM.
                          Free software for calculating bolt circles and similar: Click Here

                          Comment


                          • #58
                            Nicely done, Evan. My mini-mill is this: http://grizzly.com/products/Mini-Milling-Machine/G8689 and as simple as it is it has a large number of alignment points at both ends of the column. It's such a kludge I've not yet completed a full alignment. In part because I keep adding new things to it - most recently the x-y table. It had been mounted on the Grizzly lathe as a 3-in-1. The whole damn rig is one big bungie.

                            Comment


                            • #59
                              Thanks.

                              Nicely said and done Evan - top job!!

                              I had your machine in mind too dp (Dennis).

                              I have more to go on this topic and would appreciate all the constructive and informative comment I can get.

                              I have more use for that disk rotor as well!!!

                              More later.

                              Comment


                              • #60
                                More "not OT" "tramming" etc.

                                Originally posted by dp
                                Nicely done, Evan. My mini-mill is this: http://grizzly.com/products/Mini-Milling-Machine/G8689 and as simple as it is it has a large number of alignment points at both ends of the column. It's such a kludge I've not yet completed a full alignment. In part because I keep adding new things to it - most recently the x-y table. It had been mounted on the Grizzly lathe as a 3-in-1. The whole damn rig is one big bungie.
                                Thanks Dennis.

                                I will deal with your mill first.

                                The basic and most fundamental requirements on which so much else depends, are, but not necessarily limited to:
                                - way/guides: "flat" and true with "X" and "Y" co-planar and "Z" mutually normal to "X" and "Y" and with "X" and "Y" mutually normal to each other;
                                - the quill axis being parallel to the "Z" plane;
                                - the quill and spindle axis being coincident; and
                                - the lead-screws and nuts being consistently accurate.

                                Realistically, the apron would have to be removed to check that "X" and "Y" are parallel. We can assume that they are for this purpose.

                                Checking that "X" and "Y" are normal to each other can be done with a good Test Dial Indicator (TDI). A TDI calibrated to 0.001" or 0.0005" will do as you can detect and interpolate to 0.0001". A 0.0001" TDI would be handy - of course. You will need a good angle plate with the two abutting outer faces of the "L" being flat and accurately square to each other, and with one end at least flat and square to each of the outer legs of the "L" .

                                A "master square" or a "frame square" might/will be better - but the angle plate will do.

                                First of all, clean all guides/slides/ways and lead-screws and nuts, oil them (way oil) run them to their limits and adjust all gibs.

                                Put the angle plate down on the mill table with the selected "good and square/d" end down, clamp and set it so that one of the outer faces of the 90 degree "legs" is checked (with a DTI fixed to the non-moving part of the mill, as rigidly as possible) to be "zero-ed in" to the "X" slide/way - just as you would if setting the "fixed" jaw of a vice.

                                Now put the DTI on the other "leg" of the angle plate, lock/clamp "X", traverse "Y" and read off any deviation of/in the DTI. Any variation will be the "out of square" component between "X" and "Y". This may have to be "lived with" if there is any as rectification as it may well be "quite a job".

                                Next, put the DTI on the milling head (which is fitted to the "Z" slide/way) and adjust the DTI onto a vertical face on the angle plate.

                                Let's start with the "left/right" motion of the column ("Z").

                                Run the milling head up and down "Z" and record any variation of the DTI. This variation can be adjusted out by changing the "tilt" (left/right) of the column.

                                Repeat for the "front/back" (aka "nodding") motion by changing the DTI to the other face of angle plate (ie the face facing the "Z" slide/column). Adjust by shimming the "tilt" (forward-back/"nodding") at the base of the column.

                                Re-check that there is no DTI variation on the front and side of the "Z" column. If all is OK, the motion up/down on the column is mutually square to the mill table which runs at 90 degrees on the "X" and "Y" slides.

                                Next, all that remains is to check the axis of the quill and spindle is likewise mutually normal to the mill table (ie "trammed" - which will be (re)checked later).

                                With the "Z" slide locked/clamped, place a DTI in the quill chuck or collet. Run the quill up and down through its limit, with the DTI on each vertical face of the angle plate in turn and check for any deviation. If there is any deviation for "tilt" (ie left/right) adjust the quill tilt on the milling head. If there is any font/back ("nodding") deviation adjust with shimming between the rotating and fixed faces of the milling head tilt faces.

                                Re-check, if all is OK - good, if not, repeat and re-adjust as required.

                                Theoretically, but not necessarily so in practice, the "tram" of the quill spindle should be correct and require no further adjustment - but as we are more concerned with the practicalities than the theory, we will check and if necessary, (re)adjust the "tram".

                                Place the OP's very accurate disk brake rotor (or your preferred method of tramming) on the mill table and check the "tram". If all is OK, good, if not it will have to be compensated for (adjusted) at either the base of the "Z" column or at the "tilting" faces of the milling head.

                                If it were me, I would use a very good angle-plate and a "master square" - or just two master squares - and follow up with a final check with the framing square.

                                I have deliberately picked on Dennis's small "tilting column" mill as it embodied everything that I needed to address.

                                All of this applies equally to adjusting any mill-drill - including round, square and dove-tailed varieties as they are much easier than a "turret and knee" mill (read: BP and its clones/"knock-offs") and horizontal mills (both with a "plain" and "universal" table).

                                I will address them as well as milling machine vices in due course - later.

                                This may seem complicated and tedious - but it isn't. It becomes "second nature" after a couple of "goes".

                                You will do well to go back and re-read Evan's advice about the theory and application of phased adjustment as it is a very good article indeed.

                                Please note that I did both address tramming and used the OP's disk-brake rotor, so I was not OT after all!!

                                Comment

                                Working...
                                X