No announcement yet.

Rivett report

  • Filter
  • Time
  • Show
Clear All
new posts

  • #76
    The gear on the Atlas 12 x 24 we are getting working at the museum, covid lockdown permitting, is very similar in design, but made from Zamak, fortunately it still has its teeth.

    That liittle right angle locking feature from post #5 is used on Smart & Brown lathes.
    Last edited by old mart; 02-08-2021, 04:09 PM.


    • #77
      I have noticed that Zamak varies. There are a dozen or so formulations, plus the ones where the foundry foreman got tired of problems and threw in some lead to increase the output of "good" (to complete form) parts.

      Some of the formulations seem to break. I have some zamak gears that teeth broke off of. I have others on which the teeth got "mangled", but did not break. So either formulation, age, or both affect breakage.

      As for Rivett....

      Well, whatever sort of "precision" Rivett accomplished, it certainly was not "German" precision (for lack of a better term).

      That sort would have had all the parts well defined and toleranced, made to spec. And, if fitting (or scraping) were required, it would be on a "selected parts" basis (as ball bearings are made if of higher grades) or at least it would be done to known and defined parts as a part of the defined assembly process.

      Not a "whatever it takes" approach of dealing with the parts however they happen to turn out. I begin to see just why the "numbered parts for assemblies" were used. To me, that is a woodworking approach. "Interchangeable parts" was an old concept by the time of the 608, but does not seem to have arrived at Rivett.

      It had, however, arrived at many of the watch tool companies long before. There are illustrations of that where, for instance, a number of tailstocks could be lined up, and a tailstock ram pushed right through the lot as if they were one piece, the ram being a good snug slip fit in each. See Goodrich's book "The Watchmaker's Lathe" for more.

      I see many parts in the units which are, frankly, rough and ugly. Yes, they are mostly unseen parts, and no, they were not later bodged fixes, as far as I can determine. But "German" style precision would make all the parts to spec no matter where they are in the unit, and then , if needed, adjust some of them in known ways.

      And, yes, "German precision" predates these units, it goes back into the 19th century along with "interchangeable parts".

      I have a considerably different opinion of Rivett than I did a few days ago. Not quite as high as it was. They "ended up with" a good product, but I think they were working against themselves as they did it.
      CNC machines only go through the motions


      • #78

        The Germans were not as "German Precision" as they get credit for. They have made their share of poorly designed and built items over the years. As has everyone else.

        A fun example I enjoy is pointing out that the mighty German manufacturing of tanks in WW2. A Tiger tank could vary as much as 3" or 6" in size from one another. Assembly lines were expected to simply cut off a bit if too big or weld some a bit on if too small. Or weld holes in and redrill to fit. All to get parts to fit a particular machine. Made them extremely difficult to repair.

        So to find your Rivett is somewhat "custom" does not really surprise me at all.
        If you think you understand what is going on, you haven't been paying attention.


        • #79
          Most German products that I have seen are very consistent. They may be poorly made, but they are generally CONSISTENTLY poorly made..... IOW if there is a problem, they all have the same problem in the same place to about the same extent. Leica parts fit any Leica, for instance. What may have happened with "improvised workers" (I do not want to get into that subject here) during the war may be another thing entirely.

          In any case, this is not about Germans, but about Rivett. It is a matter of having a drawing, and making the part to the drawing. I suppose that a drawing might have a tolerance of 0.020", but when another part is expected to fit a space closely, and both are toleranced at +- 0.020", well, you had better look out for squalls, because you have heavy weather headed your way. I do not expect they did that, so I don't know how/why the same parts in a minor variation of the same product would be different enough to not even fit.

          It DOES surprise me that with Rivett there is so little consistency in identical and theoretically interchangeable parts such as the headstock casting and back gear shaft. Oddly, then, Rivett made very accurate lathes, but not "high quality" lathes. Each seems to have been an "artisanal product", having more in common with loaves of bread than with what other lathe makers produced.

          SOME things are consistent.... The index pin from the PB fits the PV headstock perfectly, and will index the PB bull gear, so the hole in the PV headstock is located in the same place as it is in the PB headstock. The PV drive pulley fits the PB bull gear perfectly as to the shaft. The PB bull gear fits the PV spindle perfectly. Other parts fit perfectly.

          It is the glaring mis-fits that stand out as so odd.

          At Southbend, parts was parts, and in general, any part that comes off a 9" will fit any other 9". At least I have dealt with several of them, and never found a part that did not swap without trouble. In certain cases, such as parts that are scraped, there may be a difference, but it will be small.

          With SB, you may not find every tailstock the same height (scraped to fit), but the ram from one T/S will fit any other. Pulleys and gears in the headstock are interchangeable. Ditto with Logan etc. That holds true unless there is a design change, in which case there is a serial # at which the change occurred.

          With the watchmaker's lathes mentioned before, obviously even the tailstocks were highly identical, and the makers were proud of that fact.

          It was otherwise with Rivett, and that really does surprise me, given the way they are presented, and the way they presented themselves.
          CNC machines only go through the motions


          • #80
            Not a bit of shop work done today, aside from some parts soaked in de-rusting solution (I could leave them for a while and come back when I was taking a break). Two clients slammed me with two weeks of work, wanted in one week in each case, of course.

            Aside from a couple breaks taken, worked on them all day.
            CNC machines only go through the motions


            • #81
              Not much action to report...

              I DO have enough parts now to put together the headstock completely, with a good deal of a headstock left over as well. I have a long enough straightedge to handle this.

              And I have (or at least think I have) all the parts needed to build up a complete carriage. There is nothing I know of materials or parts-wise stopping me from finishing off this project. Nothing but ambition to take on the most complicated and largest scraping job I have yet done. One that will probably require some special fixtures to rough in things like the dovetail. And one that may end up with application of Turcite or the like to surfaces of the carriage.

              But I have a new problem..... I have to decide what approach to take. I can either scrape the entire bed, using the bare headstock as a gauge for the top 4 surfaces (out of 10 that must all agree) since obviously it sat on the bed and never moved, so zero wear, OR I can preserve the factory headstock end of the bed, and scrape the rest, which would preserve the original alignment, but leave open the alignment of the scraped part to the original....

              I am leaning toward using the headstock as a gauge, which means not assembling it until the top, at least, of the bed is scraped. That "gravels me" a bit, as I wanted to get another part "done" before attacking the scraping, but I see this as probably a better approach.

              It might delay things even farther, as the top must agree with 6 more surfaces. The carriage touches all 10, so there is a chance that if I "finished" the top 4, some of the others might need a LOT of scraping to bring them in with them. So I kinda need to at least rough-in every one of the 10 in order to know that no more work is needed on the top.

              I see that while I have posted the bed pics before, they are not here. So here is the end of the (rather dusty) bed, from the tailstock end.

              The 10 surfaces are: The top right vertical surface (backside), the RH flat top surface, the RH side of the open V, the left hand surface of the open V, the LH flat top surface, the top left vertical surface, the top portion of the dovetail, the flat vertical surface between the dovetails, the bottom surface of the dovetail, and the bottom vertical flat surface.

              All of them must be "perfectly" in alignment. Any small error of direction might require a fairly large correction to a different surface. So there is a premium on getting it just right, and so it is likely not advisable to do any one part to completion without very careful reference to others. Roughing in everything to where it is known not to need much tweaking is probably what needs to be done. So no easy assembly of the headstock just yet.

              And, there is the question of the leadscrew and feed screw, which go in the semicircular slots on the left. They may not align after scraping, and they are set into the bed, so not much chance to change them. And I have no mill that would take a cut the the entire length of the rather heavy bed. I think I am hand scraping the whole thing.
              Last edited by J Tiers; 02-25-2021, 11:30 PM.
              CNC machines only go through the motions


              • #82
                I would start with some evaluation. Figur out which prts of a flat outside way *never* see any carriage movement and you can be almost certain that there won't be any wear there. Set a straight edge across those points on gauge blocks and measure the gap all along between them with a dial gauge on a sled.

                If you find a way that has little or no wear that's your starting point for measuring. With so many faces in full contact wear should be low all over IMO unless it's been heavily neglected.
                Peter - novice home machinist, modern motorcycle enthusiast.

                Denford Viceroy 280 Synchro (11 x 24)
                Herbert 0V adapted to R8 by 'Sir John'.
                Monarch 10EE 1942


                • #83
                  Looks like a Phd. level scraping project to me.

                  Lot going on there, seems carriage might need gibs on three different surfaces?


                  • #84
                    Evaluation of course. That is not the question. And there is really not much question anyway. It would be nice to be able to use some of the original surfaces, but realistically, there is not much point, as aligning to them will be more trouble than it is worth.

                    Evaluation merely allows one to "find the lathe bed that is hiding inside the lump of metal"...... to discover the minimum amount that can be taken off and still end up with good surfaces. I will probably end up tolerating some scratches or gouges in order to avoid taking off tens of thou of surface (even 0.25mm would be far too much material off any surface, unless the commitment is made to build up the mating parts with turcite, not to mention the hundreds of scraping passes required to remove such an amount)

                    It really is not so much "wear", as it is "damage".

                    Originally posted by Peter. View Post

                    If you find a way that has little or no wear that's your starting point for measuring. With so many faces in full contact wear should be low all over IMO unless it's been heavily neglected.
                    "find a way that has little or no wear".................ROFLMAO

                    The guy I bought it from was using it in this condition, he did not even have a working power switch, he used the plug. At least he oiled the spindle.

                    Yeah, no wear, and no damage......
                    CNC machines only go through the motions


                    • #85
                      If you have the ability to scrape, I would be tempted to re-scrape it all in. And yes, I know that is potentially years worth of work. But I think a Rivett is worth it. On the other hand, there is re-grinding, if you have the means -- it is not cheap. Its sad to see a good machine destroyed by unqualified monkeys -- glad you saved it!
                      25 miles north of Buffalo NY, USA


                      • #86
                        Scraping is a given. Strategy for the bed is not settled.

                        A fair bit has been done, the following and more....

                        Check pic 2 in post 84 above, and then this

                        Can't put metal back where idjits damaged the corners, but the slideways are good, and there is a new screw and some other new (made in the shop) parts.
                        Last edited by J Tiers; 02-26-2021, 03:23 PM.
                        CNC machines only go through the motions


                        • #87
                          Looked over the bed again, and it is more-or-less as I remembered. Pretty decent for an old machine as far as major geometry, but detail problems with scratching, gouging, and corrosion of surfaces that were exposed, plus the not-fully-evaluated wear..

                          I did not have the long straight edge when I originally surveyed it. Now I can blue that up and do some verification. I need to decide what to do with the surface, it won't take blue evenly, so I may need to run over it once just to get it to where it will take and hold blue well enough for an initial spotting. I need to clear the bench for that, however.
                          CNC machines only go through the motions


                          • #88
                            Well, side trip is required...... The bed needs to be up on the bench, but it is essentially a bar of cast iron about 4 x 5 inches cross section, and not quite 40 inches long. It's not completely solid but is nearly so, and likely weighs close to 200 lb even with the portions that are not solid iron.

                            I have picked it up, meaning gotten it off the caster frame it is sitting on. But it was enough of a strain to do that, and I think I would regret picking it up and putting it up on the bench, so I have been looking for ways to lift it, since I have not yet installed the chain hoist. I either need to find time to put up the hoist, cobble together a temporary hoist frame, or find enough cribbing to do it the old fashioned way. The latter is probably what I will do, if I can find all the cribbing that should be here, but may not be.

                            I have been putting off the hoist project, because I'd have to cut the trolley rail, and if I re-arrange the shop area, as I plan to do to fit this lathe in, the cut length I'd need now would be too short for any of the layouts I have figured out.
                            CNC machines only go through the motions


                            • #89
                              FWIW, I got my SB up on the bench by myself, no hoist. It was fully assembled except for the tailstock and motor. Trick was to swing the TS end up onto the bench first, letting the headstock end rest on the floor. Then squat the headstock end up onto a stool. Then swing the headstck end up onto the bench. Didn't feel like doing much after that, so I took a break for a while before I continued setting up.

                              Eventually I want to pour a 2" cement top on the bench for the lathe, so it'll be coming off and going back on again in the future. In case anyone wonders, I'm 54 yrs old and I eat whatever I want. 5ft 11" and 180 lbs.
                              25 miles north of Buffalo NY, USA


                              • #90
                                Yeah, that would work for the SB, the bed is fairly long.. This is not really long enough, and has no legs on the bed. It might work with a chair as the intermediate, a method I have considered, since I can lift an end one-handed to shove cribbing underneath*, so two handed is no issue, as long as the chair will stay upright with the end of the bed at an angle on it. I don't want to find out it will not....... There is not really space to do it as I would like to, and the chairs are relatively light, don't stay put.

                                I need to be careful, though, the rack is on the bottom of the bed, and is fairly fine pitch. Should have bought a M/C lift when I had the opportunity a while back, but I just didn't want another thing that takes floor space.

                                The concrete deal is what I intend to do with the Logan at some point. I'm thinking 2 1/2" or 3" thick, mesh reinforcement, fine gravel or maybe just sand.

                                * I do need to be somewhat reasonable, I'm 68, about your size /weight, and not up to the lift weight I want to get to yet.
                                Last edited by J Tiers; 03-15-2021, 12:57 AM.
                                CNC machines only go through the motions