Announcement

Collapse
No announcement yet.

Building the Trent Pinion Mill

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

  • #31
    good progress! As one who has done just wee bit of scraping, you are going to have a horrible time of it if you don't get some relief in the corners. Here's a shot of cutting some in a maximat cross slide, and the result. the scraper needs this clearance to get to the bottom of the surface. you can use any cutter (the one shown is for fluting taps so leaves a nice round profile) or even a thick power hacksaw blade....but there has to be clearance



    in Toronto Ontario - where are you?

    Comment


    • #32
      Originally posted by mars-red View Post
      II'd welcome any recommendations in that regard.
      Here's what I use very small bits of carbide that let you get into the the relieved corners. clamped to an adjustable angle plate with a movable light source is almost a requirement - makes it so much easier. Then just keep at the tedium until you get good bearing. (photos from the very small dovetail on a mini mill - note the relief slot in the corner)





      in Toronto Ontario - where are you?

      Comment


      • #33
        Dovetail scraper. The end has been beveled down until only one edge remains, so it will fit down in the dovetail.



        As for getting anal about stroke length, etc, it's not really just a matter of "academic correctness" or "getting anal", but is actually how you get a good result. The stroke length and the resulting surface are directly related. For a way surface, you want about 15 points of contact per square inch, and about 40-50% contact area, for good performance.

        It is very hard to get that without short strokes, and without doing "intentional scraping", i.e. seeing the print, and deciding that certain higher spots need to be scraped, while other spots are lower, and do not. If you simply go on the principle of "scrape off all blue", you stall-out at an "unfinished" surface, and will not improve anything beyond it, you will only move the spots around.

        Not what you wanted to hear, but it's the plain truth.

        Scraping takes a good deal of explaining.... bit I have places to go today.
        CNC machines only go through the motions

        Comment


        • #34
          One thing I have always found helpfull whe scaping dovetails or smaller surfaces is to use a scraper with a fairly narrow tip. On the order of 1/2 or 12mm. The grind on the working edge is is as per normal. When I was doing scraping at work we did almost everything with Biax's. Power scrapers when set to low speed and short strokes are capable of some very delicate work. The problem with grinding one of these blades( we used the Sandvik ones almost exclusively) is while you can rough them out on a green wheel you need a good diamond set-up to finish. But then if you oing scraping thats a given anyways. I've been thinking that maybe one of those small recip saws that are called "one handed" might be a nice starting point or a small power scraper. They even come in cordless models.
          Forty plus years and I still have ten toes, ten fingers and both eyes. I must be doing something right.

          Comment


          • #35
            Originally posted by Mcgyver View Post
            good progress! As one who has done just wee bit of scraping, you are going to have a horrible time of it if you don't get some relief in the corners. Here's a shot of cutting some in a maximat cross slide, and the result. the scraper needs this clearance to get to the bottom of the surface. you can use any cutter (the one shown is for fluting taps so leaves a nice round profile) or even a thick power hacksaw blade....but there has to be clearance



            Hm, my plan was to truncate the corners of the male dovetails instead, shouldn't that work?

            Sent from my BLN-L24 using Tapatalk
            Max
            http://joyofprecision.com/

            Comment


            • #36
              Kinda, but it leaves a wear ridge in the corner. Later that has to be removed in re-scraping.

              Better to have the ridge left on the more accessible surface, which will be the top of the male dovetail.

              It depends what are the surface in contact. In the picture you "quoted", the top of the male dovetail is not even involved.

              For the vertical column ways on the Trent, the flat of the male dovetail IS a baring surface. So if you let IT overlap the inside of the knee say surface, any ridge will be out where you can deal with it.

              But, yes, if you stop the male doovetail short, that works.
              CNC machines only go through the motions

              Comment


              • #37
                Originally posted by J Tiers View Post
                Dovetail scraper. The end has been beveled down until only one edge remains, so it will fit down in the dovetail.



                As for getting anal about stroke length, etc, it's not really just a matter of "academic correctness" or "getting anal", but is actually how you get a good result. The stroke length and the resulting surface are directly related. For a way surface, you want about 15 points of contact per square inch, and about 40-50% contact area, for good performance.

                It is very hard to get that without short strokes, and without doing "intentional scraping", i.e. seeing the print, and deciding that certain higher spots need to be scraped, while other spots are lower, and do not. If you simply go on the principle of "scrape off all blue", you stall-out at an "unfinished" surface, and will not improve anything beyond it, you will only move the spots around.

                Not what you wanted to hear, but it's the plain truth.

                Scraping takes a good deal of explaining.... bit I have places to go today.
                For roughing down past the machining marks, it really doesn't matter to me what the stroke length is, I'm just quickly getting the surface down. I may not have formal scraping training like a lot of guys on here but I've done enough of it to know how to get a good bearing surface without endlessly chasing blue dots. The new territory here for me is with scraping in the fit of the dovetails. What scraper to use, and methods for checking parallelism of the male column dovetails are what I've been mulling around in my head lately.

                Seems most everyone who has chimed in so far uses carbide, unfortunately I don't have a way to grind carbide. I'll continue to weigh my options.

                Sent from my BLN-L24 using Tapatalk
                Max
                http://joyofprecision.com/

                Comment


                • #38
                  Carbide can be ground using $6 diamond coated disks from Kent Blades on Amazon. I have them and they work just fine.

                  Three corner bearing scrapers are not used for flat surfaces, rather for scraping the ID's of tapered plain bearings which would be found on old surface grinder spindles for instance.

                  HSS works fine for scraping, but you do have to touch it up every pass. I'll get pictures later tonight or tomorrow, I'm away from the shop currently.

                  Sent from my XT1053 using Tapatalk

                  Comment


                  • #39
                    Originally posted by Andre3127 View Post
                    Carbide can be ground using $6 diamond coated disks from Kent Blades on Amazon. I have them and they work just fine.

                    Three corner bearing scrapers are not used for flat surfaces, rather for scraping the ID's of tapered plain bearings which would be found on old surface grinder spindles for instance.

                    HSS works fine for scraping, but you do have to touch it up every pass. I'll get pictures later tonight or tomorrow, I'm away from the shop currently.

                    Sent from my XT1053 using Tapatalk
                    I'll have to look up those disks, that sounds promising. What are they meant to be mounted on? I've always been wary of inexpensive diamond tooling so it's great to have a specific recommendation from someone who has used the product.

                    I know the traditional use of three cornered scrapers, but my case here is a bit unusual in that all of the internal dovetails are only a few inches long. It seems to me like being able to come into them at a sideways angle might be helpful, but then again I've never tried it. I've always been the sort of person who has a hard time believing something doesn't work or shouldn't be done just on someone else's say-so. I tend to learn lessons through my own trial and error, at least then I only have myself to blame, lol.

                    I hope my replies don't come off as too ungrateful, it's just my nature that I need to try things to really understand why they might not work or might not be ideal.

                    Sent from my BLN-L24 using Tapatalk
                    Max
                    http://joyofprecision.com/

                    Comment


                    • #40
                      Roughing you just need to avoid making big hollows or leaving big lumps, and to leave a surface that is reasonably flat to begin with, so it is easy to finish. And to have the various directions and angles close.

                      Originally posted by mars-red View Post
                      .... The new territory here for me is with scraping in the fit of the dovetails. What scraper to use, and methods for checking parallelism of the male column dovetails are what I've been mulling around in my head lately.
                      ....
                      A thin scraper, that will reach to the back of the dovetail, or one ground as the one I showed.

                      Another issue with NOT relieving the corner, is that you cannot scrape TO the corner. That generally results in the corner having more-or-less of a rise on the surface as it reaches the corner. It's only natural as you cannot quite scrape up to the end, and that point just naturally moves slowly back out of the corner as you scrape.

                      As for the fit, you have three issues. Flatness, angle, and direction. Without being too obvious here, the surface must be flat, which your straightedge can handle. It must match the angle of the mating dovetail, which can be assured by scraping the mate first, and then using that as the reference for flatness AND angle. The angle is not too critical, so long as the two match.

                      The hard one is "direction", by which I mean getting the two sides to be parallel as to direction, and having that direction be such as to have the OTHER surfaces in the correct relation.

                      With the FEMALE dovetail, and an "American" gib, pushed in by screws (not a tapered gib), the ONLY dovetail that is important is the one NOT having the gib. The gib will supply the other surface, and conform to the angle. Therefore the only issue is to get the angle and flatness by comparing to the male part, and to check the relation to other alignments so as to have them come out right.

                      But that transfers the problem to the MALE dovetail, which actually is more difficult to get right in terms of measuring.

                      The easiest way to do male dovetails is with a "Kingway" device. I made mine, they were invented by Richard King's father. Here I am measuring a male dovetail for parallelism, same type dovetail as your column. The indicator is riding on the dovetail angle, the cylinder is riding on the opposite angle and front, and the ball is riding on the nearside front. That gives a stable base on two surfaces, so you can measure the last one. (both front parts must be in one plane, so I count them as the same surface)



                      Another view in use. The cylinder has a slot cut in it, so that it fits over dovetails and V-ways. As a cylinder, it also fits IN dovetails and reverse V-ways, and there are mounting holes at various angles for those.




                      A view of the ball and the slider to use with it (bridges scrape marks)



                      People have made special purpose devices to measure dovetails like your column, but the Kingway is a device that does nearly every measurement you could want as far as the relation between surfaces is concerned.

                      For INSIDE (female) dovetails that mate with this type of male dovetail, the technique is to measure between identical pieces of dowel pin or reference cylinders, if you have them. The pins go inside the dovetail, and contact the two sliding surfaces.

                      You need the Kingway for the other type of female dovetail, but for those you can use pins on the MALE part.
                      Last edited by J Tiers; 04-16-2017, 05:32 PM. Reason: correct tool name to "Kingway"
                      CNC machines only go through the motions

                      Comment


                      • #41
                        Originally posted by mars-red View Post
                        Hm, my plan was to truncate the corners of the male dovetails instead, shouldn't that work?
                        No is the short answer, or at least I wouldn't want to.... and putting relief is SOP for scraping. Every time you go into the corner you'll leave a stop that the next iteration will cause the scraper to stop short of on the next pass. Far, far easier to have it go off the edge into the relief.

                        Try it without.....or trust me
                        in Toronto Ontario - where are you?

                        Comment


                        • #42
                          Originally posted by J Tiers View Post
                          Roughing you just need to avoid making big hollows or leaving big lumps, and to leave a surface that is reasonably flat to begin with, so it is easy to finish. And to have the various directions and angles close.



                          A thin scraper, that will reach to the back of the dovetail, or one ground as the one I showed.

                          Another issue with NOT relieving the corner, is that you cannot scrape TO the corner. That generally results in the corner having more-or-less of a rise on the surface as it reaches the corner. It's only natural as you cannot quite scrape up to the end, and that point just naturally moves slowly back out of the corner as you scrape.

                          As for the fit, you have three issues. Flatness, angle, and direction. Without being too obvious here, the surface must be flat, which your straightedge can handle. It must match the angle of the mating dovetail, which can be assured by scraping the mate first, and then using that as the reference for flatness AND angle. The angle is not too critical, so long as the two match.

                          The hard one is "direction", by which I mean getting the two sides to be parallel as to direction, and having that direction be such as to have the OTHER surfaces in the correct relation.

                          With the FEMALE dovetail, and an "American" gib, pushed in by screws (not a tapered gib), the ONLY dovetail that is important is the one NOT having the gib. The gib will supply the other surface, and conform to the angle. Therefore the only issue is to get the angle and flatness by comparing to the male part, and to check the relation to other alignments so as to have them come out right.

                          But that transfers the problem to the MALE dovetail, which actually is more difficult to get right in terms of measuring.

                          The easiest way to do male dovetails is with a "Kinway" device. I made mine, they were invented by Richard King's father. Here I am measuring a male dovetail for parallelism, same type dovetail as your column. The indicator is riding on the dovetail angle, the cylinder is riding on the opposite angle and front, and the ball is riding on the nearside front. That gives a stable base on two surfaces, so you can measure the last one. (both front parts must be in one plane, so I count them as the same surface)



                          Another view in use. The cylinder has a slot cut in it, so that it fits over dovetails and V-ways. As a cylinder, it also fits IN dovetails and reverse V-ways, and there are mounting holes at various angles for those.




                          A view of the ball and the slider to use with it (bridges scrape marks)



                          People have made special purpose devices to measure dovetails like your column, but the Kingway is a device that does nearly every measurement you could want as far as the relation between surfaces is concerned.

                          For INSIDE (female) dovetails that mate with this type of male dovetail, the technique is to measure between identical pieces of dowel pin or reference cylinders, if you have them. The pins go inside the dovetail, and contact the two sliding surfaces.

                          You need the Kingway for the other type of female dovetail, but for those you can use pins on the MALE part.
                          That device looks great! I don't think it would be much more difficult to construct than any of the purpose-specific contrivances I could imagine, so seems like a no-brainer that's what I should do.

                          Sent from my BLN-L24 using Tapatalk
                          Max
                          http://joyofprecision.com/

                          Comment


                          • #43
                            I think scraping the dovetails on this might be too tricky to worry about, because they're so small. If you get a good finish off the mill I think some careful lapping with a fine lapping compound would work plenty well. You won't need much either, just a few strokes to smooth things out. Then you can add some flaking to the unexposed way surfaces to hold oil.

                            Sent from my XT1053 using Tapatalk

                            Comment


                            • #44
                              Originally posted by Andre3127 View Post
                              I think scraping the dovetails on this might be too tricky to worry about, because they're so small. If you get a good finish off the mill I think some careful lapping with a fine lapping compound would work plenty well. You won't need much either, just a few strokes to smooth things out. Then you can add some flaking to the unexposed way surfaces to hold oil.

                              Sent from my XT1053 using Tapatalk
                              That's actually what the build instructions suggest for an optional step. Other than just wanting to get some practice scraping dovetails, I was also not confident what abrasives could be used that would not charge into the iron. I'm going to carry on with the scraping, but for the sake of anyone else who might come across this, it would be good to have some suggestions for abrasives that will not embed into iron. I would like to also mention that in one of the upcoming videos in the series too. It's definitely the path of least effort if the right abrasive is used.

                              Sent from my BLN-L24 using Tapatalk
                              Max
                              http://joyofprecision.com/

                              Comment


                              • #45
                                That is a concern with cast iron, so you will need a non-embedding abrasive. I believe garnet is one type that will not stick into the surface, timesaver is a name brand that many people like.

                                Clover and silicon carbide compounds will embed though.

                                Sent from my XT1053 using Tapatalk

                                Comment

                                Working...
                                X