Announcement

Collapse
No announcement yet.

Threading problem, threads to long

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

  • It should have been, but it wasn't. Most likely because of price constraints. As Peter pointed out, the conversion would have been even closer with a 47T spindle gear, they chose to use 48 and fudge it a bit on the chart. Most likely to keep the gear count down, enabling the same set of gears for both systems. Much cheaper that way.

    I was actually looking at this model before I found my current South Bend 9A. I'm glad I waited, because now I don't have to compromise on gear selection, I can get true conversion with 127/100. IMHO a true conversion will change both the lead screws and the gear set.


    Originally posted by Paul Alciatore View Post
    But what I do not understand is, if the lead screw is 10 TPI they did not need a metric to English conversion. So why did they need a "half-assed conversion" in the first place? It should have been straight forward.
    Last edited by nickel-city-fab; 10-02-2021, 11:27 PM.
    25 miles north of Buffalo NY, USA

    Comment


    • And your 9A has a "real" QCGB, not a "partial" one.
      2730

      Keep eye on ball.
      Hashim Khan

      Everything not impossible is compulsory

      Comment


      • Originally posted by Peter. View Post
        The problems is:

        Most of the Chinese lathes of that size have 3mm lead screws. That chart is for a 3mm lead screw, you can tell because the 3mm pitch chart would give a perfect pitch thread on the gearing specified. This lathe has a .100" pitch leadscrew which is 2.54mm pitch.

        3mm is coarser than .100" (or 2.54mm) by a factor of 1.181

        To make the chart 'nearly' work they have increased the 40T gear to 48T - a factor of 1.2

        The diffrence between 1.181 and 1.200 is the error you are seeing (1.6%).

        So what they have done is applied a 'near enough' correction for using a inch leadscrew without realising that the requirement for screwcutting quite critical to get a working thread. Ironically they would have been MUCH closer if they had made the spindle gear 47T instead of 48 then the difference would only be 0.5% instead of 1.6. They also didn't realise that the gear set they supply is actually quite capable of getting close enough to all the pitches for everything but leadscrews and measurement tools if only they made the effort to figure it out (or use a program to do it like I did).

        Anyway, I am glad that you're finally able to make proper threads Johannes. Hopefully the info will also help other people along the way who have bought the same machine.
        Well it will be interesting what they will say and do to fix this problem? I will let you know here what they say . It will also be interesting for DroPros in USA to know, they have shown limited interest in this thought they sell the same machine with 10T lead screw, 1,5 inch Spindle hole witch shut be with 48Tooth the same as I have.This kind of thing will never help sale on any machine.

        Comment



        • Here are some photographs from the project. A friend sent me a new barrel that was chambered for 30 BR cartridges, he only shot 100 shots from it and stopped since the barrel would not shoot. He said that something was wrong with the chamber work, and I could have used it to make tools or whatever since it was useless to him.

          The gunsmith how did the job had no intention to fix this ! Here are the the groups my friend got (2-3 inch groups at 100 meters) Click image for larger version

Name:	Screen Shot 2021-10-06 at 09.38.59.png
Views:	192
Size:	269.9 KB
ID:	1964763Normal groups from 30 BR at 100m is less that 1/4 (,250")

          So after Peter here in this forum found a solution for me to cut threads I could finally start this and cut my first full chamber job in 30 years. I had to cut away the old threads and make a new one. The new threads were good until the last pass, my Borden Rimrock rifle action is glued into the stock and I new that the best way to do the threads was to do it slow and try to to it without having to take it away from the 4 jaw chuck, to avoid having to re index it if the threads would not fit.

          I made a mistake in the last cut, went into to far and got some chatter in the threads. I can not do anything about this. The barrel screwed into the action and tightened up very well.

          So here are some images from this work and the test shoots. It was shot in heavy rain, wind was calm in the beginning 1-2 meter per sec but in the last groups to the left the wind had picked up.

          The Borden action with the new barrel


          Click image for larger version

Name:	Screen Shot 2021-10-05 at 15.13.26.png
Views:	193
Size:	1.96 MB
ID:	1964766
          My home range.
          Click image for larger version

Name:	Screen Shot 2021-10-06 at 10.02.29.png
Views:	195
Size:	3.24 MB
ID:	1964768


          The result from this experiment. 100 meters. We do not place the shot dead center, doing that would ruin the aiming point. This are 25 shoots, 5 in each group The aggregate is 0,265"

          Click image for larger version

Name:	Screen Shot 2021-10-06 at 09.34.15.png
Views:	189
Size:	399.0 KB
ID:	1964767

          So again, thank you all for helping me with this, and specially Peter how found the solution. Now I only need to figure out how to cut the rest of the threads, China is not helping;-)
          ​ ​
          Attached Files

          Comment


          • Super result! Thanks for the follow-up.
            Southwest Utah

            Comment


            • Originally posted by Paul Alciatore View Post
              But what I do not understand is, if the lead screw is 10 TPI they did not need a metric to English conversion. So why did they need a "half-assed conversion" in the first place? It should have been straight forward.
              Excellent question Paul !

              Even number spindle gear 48
              Even number of desired TPI at 18
              Even numbered Lead Screw at 10 TPI
              The 'gear box " is direct (B) or "1/2 revs ( 2 !) or Double revs ( x2 !)
              So "IF" all subsequent gearing is "Even" ( Function of 2 ) then the result is always a "even" output

              Rich

              Green Bay, WI

              Comment


              • I am not so sure about the reason being to keep the gear count down. I have both made and sold the 100//127 conversion gear for my SB-9C. And I have a complete set of manual change gears for that lathe: actually more than the factory set as I have three compound gears and one or two extra single gears. I have calculated all the possible gear combinations that are possible with that gear set when using the metric conversion gears and there are very few standard metric threads that I can not cut. And most metric threads coarser then 0.35mm do not require the use of one of the compound gears. I have cut metric threads and have not needed to purchase or make any additional change gears.

                There may be one or two metric threads that would require an additional change gear, but I have not had to cut them.

                I have not explored the situation with a SB-9A with the quick change gear box so it may be a lot different there. This is because the metric pitches are laid out in a sensible sequence when the thread pitch is measured directly as a linear distance (M1 = a 1mm thread) while English/imperial threads are laid our in a sensible sequence when the pitch is measured by the reciprocal of the linear distance (18 TPI - TPI means Threads / Inches). These two types of mathematical sequences are, by necessity, fundamentally different.

                In spite of those two, fundamentally different types of pitch sequences, standard sets of change gears seem to lend themselves rather well to both. The math involved in converting one lead screw pitch to another (either English/imperial or metric) is heavily dependent on having different PRIME numbers of gear teeth available and standard sets of change gears are designed to have a selection of those prime and non prime numbers. Those sets of change gears are sometimes described as "by fours" or "by fives". This means that a set of change gears will have gears that are different multiples of four or five. An example of "by fours" would be:

                16, 20, 24, 28, 32, 36, 40, 44, 48, 52, 56, 60, 64, etc.

                This sequence comes from multiplying four by 4, 5, 6, 7, 8, etc. That way you get all the prime factors as you come to them in increasing the gear sizes: 20 = 5x4 for the prime number 5. 28 = 7x4 for the prime number 7. Etc. Things work out in the same manner if the set of change gears is "by fives" (15, 20, 25, 30, etc).

                This only leaves problems when a higher prime number is needed or when the standard sequence does not have enough of a given prime in a small enough number of gears that will fit on the banjo arm. A very good example would be a 27 TPI thread. 27 = 3x3x3 so you need three of the prime number 3. Several of the gears in that standard sequence do have 3 as a factor, but none of them have two 3s or a 9. So you would need at least three of those standard sequence gears AND three more for them to mesh with for a total of a minimum of six of those gears on the banjo. It may be possible to find or make an extra long, two arm banjo for this, but the easier solution is to get or make a 27 tooth gear. Another example is the exact metric conversion where a 127 tooth gear is needed. 127 is a prime number and that is why there are no exact conversions that use any smaller numbers. Mathematically, there simply can not be.

                The math for metric threads uses the needed prime numbers on the opposite sides of the compound gear combinations that are dictated by the metric pitch to be cut. My point being, that those prime numbers are still needed. The math is just effectively inverted from the math used for English/imperial threads. And, of course, there will always be that odd example of a metric thread which requires an additional change gear to be added to the standard set, just as my example above for a 27 TPI thread does.

                Of course, all of this ASSUMES that you are using an exact English-metric conversion (100-127) and not one of the approximate ones. With all the approximate conversion gears out there, this means that individual lathes can and WILL vary.



                Originally posted by nickel-city-fab View Post
                It should have been, but it wasn't. Most likely because of price constraints. As Peter pointed out, the conversion would have been even closer with a 47T spindle gear, they chose to use 48 and fudge it a bit on the chart. Most likely to keep the gear count down, enabling the same set of gears for both systems. Much cheaper that way.

                I was actually looking at this model before I found my current South Bend 9A. I'm glad I waited, because now I don't have to compromise on gear selection, I can get true conversion with 127/100. IMHO a true conversion will change both the lead screws and the gear set.

                Last edited by Paul Alciatore; 10-06-2021, 09:27 PM.
                Paul A.
                SE Texas

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

                Comment


                • It seems that the postmortem discussion is assuming that there was some "almost correct" sizing in the leadscrew or change gears. But after the picture of the banjo around post #103 it wass pointed out that he was not using the compound gear properly. It was using only the 70T gear.

                  Then in post #116 Johannes had success when he swapped the final gear and it's spacer so that it used the 70 (driven) and 65 ( driving) to properly complete the drive train.

                  This was not the first time that this problem has been encountered on this forum. Another member had exactly the same situation a year or so ago. Now that he's placing the gears correctly he should be able to cut all the threads without drama.

                  Dan
                  At the end of the project, there is a profound difference between spare parts and left over parts.

                  Location: SF East Bay.

                  Comment


                  • Originally posted by danlb View Post
                    It seems that the postmortem discussion is assuming that there was some "almost correct" sizing in the leadscrew or change gears. But after the picture of the banjo around post #103 it wass pointed out that he was not using the compound gear properly. It was using only the 70T gear.

                    Then in post #116 Johannes had success when he swapped the final gear and it's spacer so that it used the 70 (driven) and 65 ( driving) to properly complete the drive train.

                    This was not the first time that this problem has been encountered on this forum. Another member had exactly the same situation a year or so ago. Now that he's placing the gears correctly he should be able to cut all the threads without drama.

                    Dan
                    Well to be clear danlb the gear I used to finally cut the threads are way from the gear combination given on the threads card on the machine. I now have now way to cut other inch threads correctly other that really on some computer program to calculate a gear combination. If you see some drama in this the I'm sorry.

                    In post "103 Im trying out a combination from Peter and just telling him that the gears he recommend would not fit on the Banjo. When he told me the top gear connecting to the spindle could be any gear it finally worked out.

                    Again, I cannot find the gear combination on the machine to work. If you have a solution for it then please tell me.? The only way to finally do this was because Peter here in this forum found a gear combination that worked. I like to point out the the factory has not find any solution and can not point out to me how to use the cart on the machine to cut threads correctly. Johannes

                    Comment


                    • Wow, nice shooting! I have been wanting a 7x57 for some time now, I hope I can shoot like that. Most of my shooting is shotgun (ducks and geese)
                      25 miles north of Buffalo NY, USA

                      Comment


                      • Originally posted by Johannes Frank View Post

                        Well to be clear danlb the gear I used to finally cut the threads are way from the gear combination given on the threads card on the machine. I now have now way to cut other inch threads correctly other that really on some computer program to calculate a gear combination. If you see some drama in this the I'm sorry.

                        In post "103 Im trying out a combination from Peter and just telling him that the gears he recommend would not fit on the Banjo. When he told me the top gear connecting to the spindle could be any gear it finally worked out.

                        Again, I cannot find the gear combination on the machine to work. If you have a solution for it then please tell me.? The only way to finally do this was because Peter here in this forum found a gear combination that worked. I like to point out the the factory has not find any solution and can not point out to me how to use the cart on the machine to cut threads correctly. Johannes
                        Johannes you won't be able to cut good enough thread pitches from the gear chart because they are plain and simply wrong. The charts I have put up here will work just fine but you might find that the odd one doesn't physically fit. If you can't make that work by changing the idler for a smaller one simply come back and say which ratio you are sturggling with and I will find another combination that will work.

                        All the manufacturer needs to do is either swap the 48 for a 47 then their gear chart errors will become acceptable. You could do the same (make or buy a 47T gear). I haven't spent time working out the actual error but I bet it's small enough not to matter for a fastener.

                        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

                        Comment


                        • Originally posted by Johannes Frank View Post
                          If you see some drama in this the I'm sorry.
                          I used the term "without drama" to mean that you should be able to cut the threads without guesswork or other problems. I'm sorry that I phrased it in a way that it might be considered a criticism. It was not.

                          Dan
                          At the end of the project, there is a profound difference between spare parts and left over parts.

                          Location: SF East Bay.

                          Comment


                          • Originally posted by Peter. View Post

                            Johannes you won't be able to cut good enough thread pitches from the gear chart because they are plain and simply wrong. The charts I have put up here will work just fine but you might find that the odd one doesn't physically fit. If you can't make that work by changing the idler for a smaller one simply come back and say which ratio you are sturggling with and I will find another combination that will work.

                            All the manufacturer needs to do is either swap the 48 for a 47 then their gear chart errors will become acceptable. You could do the same (make or buy a 47T gear). I haven't spent time working out the actual error but I bet it's small enough not to matter for a fastener.
                            Thank you again Peter. Im waiting to see what the factory will do? that will be interesting. I will post it here.

                            Comment


                            • I would not bother with the factory in fact I would be inclined to tell them where they went wrong and leave them to it.

                              There may be a small market in printing and selling alternative gear charts if that sort of thing floats your boat.
                              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

                              Comment


                              • Perhaps you could make a new "gear carrier" in a boomerang shape and in that way you could accomodate big gear combinations more easily.
                                Helder Ferreira
                                Setubal, Portugal

                                Comment

                                Working...
                                X