Dividing and gear cutting equipment for clockmaking

Discussion in 'Horological Tools' started by NigelW, Jan 5, 2019.

  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  1. NigelW

    NigelW Registered User

    Jan 2, 2015
    315
    23
    18
    London
    Country Flag:
    I repair clocks as a hobby and have now started to build clocks from scratch. In my tiny workshop in London I have two lathes: a Cowells and a Myford ML10, but no drill stand, milling machine or means of dividing. I am thinking of investing in some additional tooling, especially for gear and pinion cutting.

    As as a member of a horology club I have access to an old wheel cutting engine where I have cut missing wheels for clocks under repair and the initial set of wheels for the going train of my current clock. The club also has a drill stand and a milling machine which I can use to cut my pinions.

    My latest clockmaking ideas however involve cutting some quite high prime number wheels not available on the club's wheel cutting engine. My ambition is to make an equation of time differential, which requires a year wheel (365 = 73 x 15 or 366 = 61 x 6), and possibly a clock with both a sidereal and mean time train which would also need quite a few exotic primes. I am thinking that acquiring a smallish but high quality milling machine (such as a Cowells) together with a dividing head coupled to an electronically driven servo motor could be the answer rather than, say, a dividing head to be fitted to either of the lathes.

    Any thoughts or suggestions would be appreciated.
     
  2. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    Hello Nigel!

    While I’m not in your exact situation, I’ve been pondering the exact same thing for quite some time now. What milling machine to get and how to divide for wheel and pinion cutting.

    At long last (I think it’s been five years now where I’ve put off getting the gear due to not knowing what to get, financial situation from being a student and space issues) I’ve finally settled on a Sherline 5410 milling machine and their stepper motor equipped rotary table. They sell it paired with a controller but I’m going to put off getting that as the table can still be used manually and I think building a controller could be a fun project.

    CNC dividing could be viewed as being no sport, but looking back on horological history, accurate diving has been in constant evolution. I think the precision and ease with which you can use modern stepper motors would be perfect for making one off wheels with a high number of teeth such as you describe.

    I’m in a situation where the machines I’m choosing between are far between, geographically and I’ve not been able to try them out. However, out of the few I’ve seriously considered the Cowells is the smallest. If you intend to work on larger clocks maybe a larger machine would suit you better?

    Best of luck!
    Karl
     
    NigelW likes this.
  3. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    10,241
    1,031
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi Nigel,

    I think a Sherline mill plus their 8700 CNC rotary table should fit your requirements very well and come out about £700 cheaper than a Cowells mill by itself if you're buying new.

    Regards,

    Graham
     
  4. Harry Hopkins

    Harry Hopkins Registered User
    NAWCC Member

    Nov 16, 2011
    543
    37
    28
    Retired Instrument Technician
    Mason, IL
    Country Flag:
    Region Flag:
    I highly recommend the Sherline CNC controlled rotary table with controller for clock wheel cutting but when buying beware that if you decide to purchase the Manual Rotary Table with thoughts of upgrading to the CNC controlled model at some later date you cannot upgrade the manual rotary table to CNC. At the minimum, if you wish to upgrade at some point, I believe you need to buy the CNC Rotary Table with Stepper Motor initially and add the controller at a later date. I have not seen on Sherline's website where the controller can be bought separately so I would confirm all of this with Sherline before commiting.
     
  5. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    Sherline does offer the controller separately. And you are right when it comes to the rotary table with stepper motor, the worm wheel and shaft are different. Getting one that is either CNC ready or with motor mounted is a good idea I think. The perk with the one with motor mounted is that they supply it with a hand wheel so that it can be used manually as well.
     
  6. wefalck

    wefalck Registered User

    Mar 29, 2011
    453
    18
    18
    Male
    Paris
    Country Flag:
    I like Karl's comment of "CNC dividing could be viewed as being no sport" ;)) Yes, within the physical limitations of stepper resolution and backlash, one can now cut virtually any number of teeth error-free without having to resort to gear-trains and dividing plates. A whole set of books on how to calculate the necessary set-up of a dviding head has become obsolete with this.
     
    NigelW likes this.
  7. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Nigel
    Been down this road many times.

    One of the things that you quickly learn when cutting teeth, is that the cutter depth of cut is a visual thing to form a tooth under optics in almost all cases, as opposed to a calculated depth.
    As such, if the machining process is not clearly visible to optics when setting the depth/form of the first tooth or so, the process can quickly become trial and error.

    For work pieces within its envelope including wheel/pinion, I favor the Sherline 5400 in that it is easily configured for very clear visibility of the tooth cutting process per the first couple of photos,

    By adding the Sherline CNC rotary table, testing and adjusting tooth forms is a quick and easy procedure. Once a full rack of teeth have been cut, the rotary table can be set into continuous rotation and tested by holding the pinion against the wheel per third photo. If any resistance or sticky function is detected at full depth, the tooth can then be adjusted as required. When starting the cutting process you start at zero setting on the R/T. At this point, you return the controller to zero and turn it off. You then ever so slightly rotate the rotary table manually and shave the tooth to allow ever so slightly more clearance. You can now turn the controller on and program for the number of teeth adjusting each tooth identical to the first. Repeat if required. Forth photo shows cutting a watch pinion in the same manner.
    While not practical or possible in every circumstance, Having the options in most work, sure beats none at all with most other systems.

    In addition, this arrangement also allows one to machine a wheel crossing that is faster and far more accurate than by hand. The Mill headstock is returned to the vertical position and straight or radius cuts can be taken by manual or in part by CNC using a small center cutting endmill allowing entry to the cavities. Setup is per the fifth photo and performed again before the wheel is removed from the setup.

    Jerry Kieffer

    DSCN7634.JPG fullsizeoutput_95.jpeg fullsizeoutput_96.jpeg fullsizeoutput_8a.jpeg fullsizeoutput_97.jpeg
     
  8. NigelW

    NigelW Registered User

    Jan 2, 2015
    315
    23
    18
    London
    Country Flag:
    #8 NigelW, Jan 6, 2019
    Last edited: Jan 6, 2019
    Thank you very much for your posts and recommendations. I can relate to Jerry's points about visibility and the need to open up the spaces between teeth. My last major project was to reconstruct the motion work of a 17th C lantern clock. In the end I had to resort to quite a bit of hand filing to get the wheels to mesh properly. This was probably acceptable in a clock of this kind which would originally have made with the most basic of tools but is not really good enough for more modern pieces.

    I also rather incline to the view that making things using computer controlled machinery is not really cricket, so having the option to use dividing plates or a stepping motor is an attractive one.
     
  9. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    I don't want to burst your bubble but the only way to control the stepper motors is with some sort of computer.
     
  10. wefalck

    wefalck Registered User

    Mar 29, 2011
    453
    18
    18
    Male
    Paris
    Country Flag:
    Can't you joggle forward a stepper motor with electric pulses from a switch ? I am not really an electronics person, so I am not sure. However, this would mean that you have to count your pulses, which is prone to errors. One could probably add an electric counter, which is a sort of primitive computer ... where is the boundary to a real one ?
     
  11. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    You would wear through switches and your patience rather quickly I think... The number of steps on the motor combined with the reduction in the worm gear would mean you would have to do an awful lot of flipping that switch. Kind of defeats the purpose of the stepper motor too, if you count yourself. Then it’s probably just better to count graduations on the handwheel as you turn it.

    But yes, I suppose you could provide the motor with single pulses more or less manually.
     
  12. Harry Hopkins

    Harry Hopkins Registered User
    NAWCC Member

    Nov 16, 2011
    543
    37
    28
    Retired Instrument Technician
    Mason, IL
    Country Flag:
    Region Flag:
    There are 28,800 steps per revolution of the CNC rotary table which equals .006 degrees per step so counting would be beyond tedious.
     
  13. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Nigel
    Keep in mind that the most successful craftsman of the distant past, used the highest quality and most efficient tools of their day. If they were standing in front of you today, I doubt very much they would encourage any type of disadvantage to be "cricket" as you mention.

    When repairing or reproducing parts of hundreds of years old, I personally use modern equipment to very quickly remove about 90% of the metal to be removed. This allows me to utilize almost all of my practical time allotment to the project to duplicate fit, function and appearance with hand tools.

    Wefaick

    CNC controls vary by brand, but the Sherline has a forward and reverse jog button that can be pushed at anytime. This is designed for machining and positioning purposes and is not calibrated to be exact at any given time or movement.

    For indexing, the table can be programmed so that it will very precisely divide by divisions or degrees. For each programed division or degrees you push a forward or reverse button once and each movement is counted out on the display. For forward movement, the controller simply moves the table forward. However, if the reverse button is pushed, the controller moves the table back past its set point and then back forward to remove any backlash in the system.

    Factory specified accuracy of the system is 1/2 step in the 29000 or so steps in one rotation of the table.

    Jerry Kieffer
     
  14. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    I’ve always wondered about that with backlash compensation for CNC. When doing it manually you have a visual reference on the hand wheel. But with a stepper motor you only have steps. So the system would only “see” the movement it in itself provides. Unless the backlash is known to a very high degree (and with a worn lead screw it could vary along its length) is there a way to compensate for it?
     
  15. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Karl
    On the Sherline system, (Others can be checked) backlash compensation can be set at whatever you desire, typically you want far greater than will ever be realized.

    Jerry Kieffer
     
  16. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    Sorry, I didn’t quite get that. I was asking on a general level about backlash in CNC systems. I realize that most dedicated CNC machines have ball screws and virtually no backlash. But do I understand you correctly that the kind of backlash compensation that you can do by hand and eye can’t be accomplished by stepper motors?
     
  17. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Karl
    Sorry about the confusion. My explanation only pertains to rotary tables with a worm and worm gear that will require a ever so slight backlash to operate smoothly. It is compensated for in the method mentioned.

    Backlash, if any with ball screws, is determined by the ball screw manufacturers method of dealing with it.

    Jerry Kieffer
     
  18. wefalck

    wefalck Registered User

    Mar 29, 2011
    453
    18
    18
    Male
    Paris
    Country Flag:
    There is mechanical and operational backlash compensation, both have been implicitly mentioned already.

    In mechanical backlash compensation a spring, for instance, ensures that always the same two reference surfaces are in contact with each other, e.g. the same sides of a spindle and the corresponding spindle nut.

    Operationally, you do the same thing by first moving the slide etc. beyond the desired point and then back, i.e. you approach a desired point always with the same surfaces of spindles/nut in contact. This is good practice in manual operation of a lathe or mill, and can be programmed into CNC programs.

    Jerry, I was not really suggesting that one should use the manual joggle for dividing, I was merely responding to the question, whether computer control is absolutely necessary.
     
  19. Jim DuBois

    Jim DuBois Registered User
    NAWCC Member Sponsor

    Jun 14, 2008
    2,461
    374
    83
    Male
    Magnolia, TX
    Country Flag:
    Region Flag:
    Karl, most people who get serious about CNC work will very highly recommend going with a zero backlash approach. Generally the easiest way to get to zero backlash is to use ball screws on all axis, save the rotary axis. On the rotary axis it is often possible to keep the axis rotating one way and with the gear train optimised backlash can be pretty well reduced to near zero, if not eliminated. There are ways to make zero backlash nuts for conventional threaded screws, but in most cases they are subject to quick wear and require much more torque to move an axis than do ballscrews.

    I have only had 1 machine that used conventional lead screws with CNC, and I used that in a non critical fashion where a bit of extra motion was not an issue. I still don't recommend it however. But, if you are interested in smooth finishes, easy to move axis/stages, repetitive accuracy of parts from the machine, and less hassle in your machine work I would recommend spending more money and either buying or constructing a machine with a zero backlash solution. With careful shopping a machine can be found or made that does not break the bank. Oh, and in the world of CNC, ridigity of the machine and mass of the machine also come into play. All depends on what you want to do with it.

    I have 2 CNC mills, one Chinese, the other American, and they both will deliver .0005" part after part. I still produce too much scrap but that is because of my limitations, not those of the machines...
     
  20. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Wefalck
    I didn't`t think so , but missed your question anyway. Sorry About that, but I will blame it on the dog scratching on my leg to go outside.

    Jerry Kieffer
     
  21. NigelW

    NigelW Registered User

    Jan 2, 2015
    315
    23
    18
    London
    Country Flag:
    Thanks for this. I had thought about this aspect but am still a little hazy about how it will all fit together. My understanding is that you can either link the servo motor to a computer or buy some kind of stand alone electronic controller. I guess one should be able to connect to a CAD programme. I don't currently use one (when I studied engineering in the 1970s such things did not exist and we drew everything by hand) but I have started trying to use Freecad, which I find a bit flaky. The differential I have in mind for driving a pair of hands showing solar time on the same dial as mean time could benefit from being drawn out properly before I plunge in I feel!
     
  22. wefalck

    wefalck Registered User

    Mar 29, 2011
    453
    18
    18
    Male
    Paris
    Country Flag:
    In a nutshell you would need to transpose your CAD-based drawing into a CNC-code, which calculates tool-paths and work-movements from the CAD vector graphics. I don't have CNC-experience (only CAD), but understand that the CNC-programs would do most of that work for you, provided that your CAD can export the drawing in a format the CNC-program can understand. There are specific formats for this, such as HGL. So it is important to pick the right CAD-program.

    I suppose in the early days of CNC on would have programmed individual tool movements sequentially, which is probably what you are referrring to as 'electronic controller'. I don't think you would find many machines like this anymore. The computer allows you to run the whole sequence of machining steps in one go. It also possible (and needed) to simulate the whole process without actually cutting any metal.
     
  23. NigelW

    NigelW Registered User

    Jan 2, 2015
    315
    23
    18
    London
    Country Flag:
    I am sure you are right about this. My understand also is that many craftsmen of the past also bought in part finished components.

    My own feeling is that the use of modern technology is fine, up to a point. Since mechanical clocks and watches are now technologically obsolete one has to ask the question why we are either keeping old ones going or making new ones. The answer seems to me to be largely to do with aesthetics and historical curiosity than functionality. Restoring a very old clock which was rather crudely made by today's standards is probably best done by using a similarly "crude" approach, otherwise the new parts will look out of place . On the other hand going too far in that direction could verge on fakery. If a clock is to be converted back, say, from a later anchor escapement to a verge along the lines of what it may have had originally, should it be done so you wouldn't know, or in such a way that is sympathetic but clearly visible to an expert? I think the latter is preferable.

    As for building a new mechanical clock, there would be no fun in the whole process if it was laser printed and assembled by robot. I am still a beginner, but the kind of clocks I would like to build are not those which look like they had been made by Breguet or George Daniels (whose skills I could never aspire to in any case) but ones which have more of the touch of the human hand about them. For similar reasons I like the idea of non-linear, solar time which takes us back from the bureaucratically convenient mean time to time based on the rising and setting of the sun as used by our ancestors in pre-industrial times.
     
    Duncanbootmaker likes this.
  24. wefalck

    wefalck Registered User

    Mar 29, 2011
    453
    18
    18
    Male
    Paris
    Country Flag:
    This develops into an interesting discussion about the aesthetics of the products and the tools we use - and why we do what we do ...

    You are right, mechanical clockworks are utterly obsolete from a practical and precision point of view. From an utilitarian point of view a cheap digital watch/clock does the much better job with less resources required to run and maintain it.

    I gather what drives people here is the desire to explore what is mechanically possible. In this way, we impose ourselves a boundary, namely the 'mechanical' one - as we could easily go beyond these boundaries by going electronic.

    Following the same line of thought one can also impose on oneself another boundary, for instance, to explore what is mechanically possible without CAD and CNC. One could go even further and not use any machine tools, as was the situation of early clockmakers.

    So it all boils down to what kind of boundaries and constraints we impose on ourselves for aesthetic or other reasons. In all these cases we sacrifice precision with respect to the underlying objective, which is measuring time as accurately as we can.

    Personally, I am imposing myself the aesthetic and haptic boundary of working with antique machinery as part of a 'holistic' past-time experience. I am very well aware that with other tools and strategies a much higher quality of the product may be achieved, but I consciously forgo this for the pleasure of the challenge of using the old, mechanical tools.

    It is all about understanding what objectives we have. Jerry, for instance, has the objective of making or repairing mechanical clockworks with the highest precision possible and in the most time- and cost-efficient manner. This is very different from making a tool for measuring time most precisely - if that was the objective, one would have to dump mechanical watch- and clockmaking altogher ...
     
    Duncanbootmaker likes this.
  25. Jim DuBois

    Jim DuBois Registered User
    NAWCC Member Sponsor

    Jun 14, 2008
    2,461
    374
    83
    Male
    Magnolia, TX
    Country Flag:
    Region Flag:
    You might like to check out the work being done by "clickspring" on youtube. He is currently making a reproduction of the Antikythera device. He uses many approaches that may have been used in the devices original construction, circa 30 BC. He shows how to make files to do the work.....he shows how to lay out gears using something besides indexing devices we normally might use, they had not been invented yet...extremely well done series, informative to say the very least.
     
    Duncanbootmaker and karlmansson like this.
  26. John MacArthur

    John MacArthur Registered User
    NAWCC Member

    Feb 13, 2007
    302
    24
    18
    Wefalk - This is what CAM (computer-aided manufacturing) is for; it takes the output from the CAD program, typically in .dxf (digital exchange file) format, and converts it into .nc (numerical control) format, by doing the toolpath and work-movement calculations that you mention above. This is what the machine controller then takes as instructions as to what moves to make. Thus, you have three distinct programs and filetypes. This is not exactly trivial, but is incredibly powerful when you get used to it.

    For the moment I'm staying clear of the "to CNC, or not to CNC..." discussion. It's really just a better hacksaw.....

    Johnny
     
  27. Jim DuBois

    Jim DuBois Registered User
    NAWCC Member Sponsor

    Jun 14, 2008
    2,461
    374
    83
    Male
    Magnolia, TX
    Country Flag:
    Region Flag:
    I have been using CAD CAM with CNC machines since 1992. I have no desire to go back to manual controlled machines for much work. On the other hand my most used mill is one of those "horrible" Chinese made mills. (Egads, even worse, it was from Habor Freight too) It gets used every day for the mundane tasks of roughing out this or that, or sawing or slotting, or squaring up a cut, or drilling this or that, and yes, I do use it for bushing work too. It is at least as accurate as the machinist connected to it. It is not a great mill, or even one I would rate as good overall.

    I replaced a much higher quality mill (Prazi mill) with this one because I wanted something that would use R-8 collets. A number 2 morse taper is no substitute for tool holding in a milling head in my opinion. But, to each his own, and smaller Sherlines use that taper with appropriate adapters to good ends....I just don't like adapters added to adapters, accuracy will suffer in one way or another in most cases. The tooling supplied with the Prazi extended the spindle near 2" which certainly did not aid in rigidity, run out/accuracy, or size of the machining envelope.

    But, CNC does the heavy lifiting of gear cutting, engraving, crossing out, machining movement plates, on and on.....

    20140901_153040.jpg 20140902_152126.jpg 20140906_121324.jpg 20140910_080813_Android.jpg
     
  28. karlmansson

    karlmansson Registered User

    Apr 20, 2013
    2,486
    83
    48
    Doctor
    Linköping, Sweden
    Country Flag:
    When I wrote "computer" I meant any software equipped microcontroller or computer system. Sherlines controller is a stand alone system with a controller and stepper motor driver board in it.
    You need a program that converts your model into G-code. I recommend watching the CNC basics video by a youtuber called This Old Tony if you are interested in learning more. He keeps it simple and fun. Still pretty in depth though.
    I agree with you in part. You should strive to achieve the same level and type of finish. But you still want the mechanism to function. For something like dividing, you really need to be accurate. If it comes down to me getting a dedicated wheel cutting engine where I need to shell out a substantial amount of money, and do a lot of tool hunting to get the proper dividing discs for it, I'd prefer to do what the watchmaker of that time TRIED to do and get a tool that divide very accurately at any tooth count and at a fraction of the cost for gearing up like in the 1800s. That's my take on it. Still, when it comes to finish I agree with you completely.

    Bergeon still makes and sells a wheel cutting engine if you are interested. It's several times the price of the Sherline mill complete with CNC rotary table + controller.

    Best regards
    Karl
     
  29. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Wafalck
    You are correct in regard to many of my projects, But in Horology, my Goal is to return a movement back to its original state as close to possible in regard to function and appearance. Developed mechanical efficiency is utilized to leave time to address appearance so that both can be done in a timely fashion.

    Unfortunately, I have not been able to master methods of avoiding the Mother-in-law

    Jerry Kieffer
     
  30. DeweyC

    DeweyC Registered User
    NAWCC Member

    Feb 5, 2007
    1,384
    351
    83
    Watchmaker
    Baltimore
    Country Flag:
    Totally. Bryan Mumford, the person who designed that table, also made the MicroSet timer. Very creative guy.
    The most important tool for this work is a book: Malcomlm Wild's book on wheel and pinion cutting. Covers it all including at least 3 methods to center the cutter!

    I use the Sherline and my Habegger interchangeably. I keep the attachment (minus dividing plates) mounted on my Habegger which results in low setup time. There are differences in being able to see depending on what you are doing. The Sherline has the advantage of being able to be moved all around; the 102 has the advantae of space to get your head and microscope in.

    The biggest problem with a lathe milling attachment is that on most setups the milling attachment is in the way. I had picked up a Schaublin double slide rest whch allows me to mount the milling attahcment behind the lathe and facing me.

    Sometime ago I sold my Derby Micromill for 3 times what a new Sherline cost. Banked the remainder. Then I sold my Levin dividing plates. There are many less expensive solutions than those most often sought by amateurs, Read the Tubal Cain (Argus Press) books. Modelmakers have been doing this for decades with no sweat.
     
  31. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    10,241
    1,031
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi Dewey,

    That essential part I already have, the rest I hope will come later this year.

    Regards,

    Graham
     
    DeweyC likes this.
  32. AJSBSA

    AJSBSA Registered User

    Nov 24, 2009
    541
    19
    18
    Male
    Clockmaker
    Cheshire, UK
    Country Flag:
    Region Flag:
    This setup has been working well for me for years I can see all round it and I am not winding a heavy spindle and motor up and down with and against gravity


    WheelCutt_00011.JPG

    DSC00475.JPG
     
  33. John MacArthur

    John MacArthur Registered User
    NAWCC Member

    Feb 13, 2007
    302
    24
    18
    This was my first lathe and gear cutting apparatus, which I made up in the early '80's. It is an 1890's era Sebastian lathe, with my watchmaker's lathe mounted on a slide attached to an angle block. The indexer I found in a junk heap behind a machine shop, and was able to get it freed up without too much trouble. The whole thing is pretty cobbled together, but shows what can be made from largely junk. The whole thing cost me less than $150. The clock I made with it is running quite well in my hall, after having ticked over a billion beats, and keeps time within a couple of seconds a month.

    I'm sure darn glad of having more and newer equipment - I doubt I would go through the agony and handwork of that first one again.

    Old lathe 1.jpg Old lathe 2.jpg
    Johnny
     
  34. P.Hageman

    P.Hageman Registered User

    Jul 20, 2014
    733
    38
    28
    Forensic science
    The Netherlands
    Country Flag:
    I used an arduino micro computer with a steppermotor attached to my manual dividing plate. Quite easy to make with the parts bought from china. It works perfectly, now I never have to be afraid of losing count with the manual dividing plate. Just look on youtube, there are some very good explanations.
     
  35. AJSBSA

    AJSBSA Registered User

    Nov 24, 2009
    541
    19
    18
    Male
    Clockmaker
    Cheshire, UK
    Country Flag:
    Region Flag:
    On a normal dividing head there are two arms on the dividing plate which you move to the stop pin after each cut so it is very difficult to lose count in fact there is no counting at all
    WheelCutt_100008.JPG
     
  36. P.Hageman

    P.Hageman Registered User

    Jul 20, 2014
    733
    38
    28
    Forensic science
    The Netherlands
    Country Flag:
    Except if you have to turn the crank a few times round for the next hole. I managed to get it wrong once in a while!
     
  37. Jim DuBois

    Jim DuBois Registered User
    NAWCC Member Sponsor

    Jun 14, 2008
    2,461
    374
    83
    Male
    Magnolia, TX
    Country Flag:
    Region Flag:
    I too have made more than a few mistakes using manual indexing, no matter how careful one works, phone calls, and other disruptions, as well as late night work all seem to generate scrap. I built my first CNC indexer in 1992. Since then I think I have manually indexed two or three wheels. That is it. But I have cut many hundreds using full CNC control.

    And back to manual indexing no matter how many plates I had for direct indexing and using both 40:1 and 120:1 indexing heads too it always seems like the next job was one that didn't have a proper plate to complete the work. Wheels with 365 or 366 teeth come to mind, or the stray that called for 257 teeth, or for that matter 57 or 59 teeth, or 163 teeth. I have cut wheels for all those sorts of wierd requirements. Difficult to do with manual indexing. Not impossible, just hard to do. CNC, not so much.
     
  38. AJSBSA

    AJSBSA Registered User

    Nov 24, 2009
    541
    19
    18
    Male
    Clockmaker
    Cheshire, UK
    Country Flag:
    Region Flag:
    If you are in cutting wheels every other day sure buy or make a cnc rotary table, but if you making wheels for clock repair a 60:1 dividing head with standard plates should cover everything with no counting once you set the arms certainly that has been the case for me and I have cut a full set of wheels and pinions for several clocks as well as normal clock repair. I downloaded a Excel spreadsheet added my plates and it does all the hard work for me
    How ever I take on board that CNC is getting less expensive and normal precision tooling is getting more expensive not sure this is a good thing though
     
  39. Jim DuBois

    Jim DuBois Registered User
    NAWCC Member Sponsor

    Jun 14, 2008
    2,461
    374
    83
    Male
    Magnolia, TX
    Country Flag:
    Region Flag:
    My first CNC indexer cost me right at $16. I already had the index head, I located a stepper motor at the salvage yard that cost me $.28 (why do I remember that when I forget why I went to the shop frequently?) I spent the $16 at Radio Shack on the very few electronic parts I needed. I had an old PC doing nothing already, and I had a 12 VDC power supply and a parallel port cord to connect the indexer to the PC. The SW package a friend wrote in very short order in exchange for a couple of old clock movements. I already had a milling machine, so it all worked out quite nicely. If I were to do it on the cheap today I would most likely use a Rasberry PI or arduino micro computer as suggested above. And as you can see neatness did not count in my wheel and pinion cutting work. When cutting several a week and using coolant and lubrication it gets messy in a hurry....


    Today, I use more modern and commercial solutions but neatness has only improved a bit....
    100-0004_IMG_2.JPG 100-0007_IMG_2.JPG 106-0632_IMG.JPG
     
  40. AJSBSA

    AJSBSA Registered User

    Nov 24, 2009
    541
    19
    18
    Male
    Clockmaker
    Cheshire, UK
    Country Flag:
    Region Flag:
    In my past life as an Engineer dragging very skilled draftsmen from their drawing boards onto CAD and as a Software Engineer writing cad/cam software for others to use I find using tools in the more traditional manner just as I did when I was an apprentice refreshing but can understand the attraction of CNC I really can. Learning CNC because you want to is one thing but for the jobbing repair person I am not convinced it is cost effective.
     
  41. ChrisCam

    ChrisCam Registered User
    NAWCC Member

    Dec 9, 2017
    1,126
    19
    38
    Male
    England
    Country Flag:
    Region Flag:
    Hi Jerry,
    I am interested in learning how to cut wheels and have a Sherline 5400 mill and a 4500 lathe both with imperial calibration. Could you give me a list of components as far as you can that will be needed. I take it the 4inch rotary table is a must and looking at the pictures the mill is altered so exactly what is required for this. Obviously cutters will vary but please give details of type from which one is selected.
    Kind Regards
    Chriscam
     
  42. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Chris
    Assuming you are referring to the photos in post #7, you do not need much other than the rotary table. While the manual table will work, you would prefer the CNC version if used side by side.

    Items needed as follows

    (1) The Mill and Rotary Table.

    (2) The spindle rotates and locks in place at 90 degrees per Sherlines instruction guide. (360 degrees if you wish)

    (3) An arbor to hold your cutter can be purchased through Sherline or machined yourself.

    (4) I machine a Wheel blank arbor and the blank diameter on the Lathe and then transfer the Chuck/blank to the rotary table on the Mill.

    Thats It.

    Cutters are a little more complicated. Each manufacturer has their own opinion on what a tooth form should be and will provide calculations on creating matching wheels and pinions.

    Duplicating existing tooth forms is best done by comparison of a cutter to an existing tooth form such as the instructions provided with the example set shown in the attached photos.

    Personally, I prefer to duplicate existing tooth forms exactly for many reasons. For this reason, I machine cutters to that end as covered in the NAWCC workshop WS-119 that will not help you based in the UK. As such, I would suggest attending some of the larger Model engineering shows in the UK and solicit gear cutting info/demo`s from those that display and venders.

    Good Luck
    Jerry Kieffer

    fullsizeoutput_288.jpeg fullsizeoutput_287.jpeg
     
  43. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    10,241
    1,031
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi Chris,

    The book mentioned by DeweyC in post #30 is a good investment; 'Wheel and Pinion Cutting in Horology' by J Malcolm Wild FBHI, (ISBN 1 86126 245 0).

    Regards,

    Graham
     
  44. ChrisCam

    ChrisCam Registered User
    NAWCC Member

    Dec 9, 2017
    1,126
    19
    38
    Male
    England
    Country Flag:
    Region Flag:
    Jerry, thank you for the generosity of your reply. I am able to take on board most of your post apart from ' while the manual table will work, you would prefer the CNC version if used side by side.,. My problem is surely a CNC rotary table needs a program to run and I can see from the photographs no way to set up a program, I guess I am asking what does the CNC rotary table do that the manual table doesn't, it must come down to either electrics or electronics (albeit software)?

    Kind Regards
    Chris
     
  45. ChrisCam

    ChrisCam Registered User
    NAWCC Member

    Dec 9, 2017
    1,126
    19
    38
    Male
    England
    Country Flag:
    Region Flag:
    Graham, I shall invest, thank you.
    Chris
     
  46. Jerry Kieffer

    Jerry Kieffer Registered User
    NAWCC Member

    May 31, 2005
    2,425
    232
    63
    Male
    wisconsin
    Country Flag:
    Chris
    While I initially used the manual table it requires math calculations and attention to hand wheel settings. This and mistakes are avoided not to mention saved time with the CNC version.
    The CNC version is called 8700 and requires the included control box as shown on the Sherline site. The basic control is as easy to use as a calculator and requires no programing skill.
    Other uses for the CNC version are outlined in post #7 that includes testing and tooth profile adjusting etc.

    Good Luck
    Jerry Kieffer
     
  47. ChrisCam

    ChrisCam Registered User
    NAWCC Member

    Dec 9, 2017
    1,126
    19
    38
    Male
    England
    Country Flag:
    Region Flag:
    Thanks Jerry, with your information and my research I have enough to go on. As always pros and cons. Pro once learned certainly easier with the CNC. Con if as expected I could cut only rarely can I justify the expense? Also you started with the manual rotary table which must give you a good understanding. Lastly CNC relies on electonics which means after a few years reliability issues and possible replacement. So the deciding factor for me is one final question (hopefully) how well did you manage cutting wheels using the manual table in terms of time it typically took relative to the CNC table?
    Kind Regards
    Chris
     
  48. ChrisCam

    ChrisCam Registered User
    NAWCC Member

    Dec 9, 2017
    1,126
    19
    38
    Male
    England
    Country Flag:
    Region Flag:
    Well I suppose my last question was silly as there cannot be a typical time saved as very hard to work out. Have decided to go the CNC route and will leave using it for many months down the road.
    Regards
    chris
     
  49. Technomaniac

    Technomaniac New Member

    Dec 16, 2009
    4
    0
    1
    Hi. I'm using a very old Victoria milling machine , a home-made dividing attachment which bolts to the bed, and a fly cutter. The hardest part of the job is manually grinding the quarter inch square toolbits to the correct profile. If I have an existing wheel its easy enough to grind the tip close enough, but working from scratch is a bit difficult and takes a few hours. I don't have a proper dividing head and do everything by direct division. But I made up another gadget I use on a drill press, from a 50:1 worm and wheel gearbox, so I can make up division plates with odd numbers of holes. For Pinions I usually use a Chronos Pinion Mill, but I see that cutters are now per cutter, $90 odd to buy and I don't have a full set. With the fly cutter you can't use thin blanks unless you have a thick piece of waste behind, or the job bends over. But I get a real polished finish with the fly cutter, even with all of the teeth on the rotary cutters, the speed is a bit slow, and the finish not so smooth. Can anyone recommend a good book of tables which gives the best tooth profiles for different sizes of wheels and numbers of teeth ? I've been thinking for a while regarding how to accurately grind a fly cutter bit so that both sides match each other, and haven't come up with a sensible method which allows for grinding wheel wear. etc.
    Another difficulty I have is with my means of holding a blank on the milling machine attachment, and getting it absolutely centred relative to the rotation of the division plate spindle.I use just a 6mm high tensile bolt, but I find I have to centre the blank with a dial gauge at the start of the job. I need to use some method with conical washers or something so that step is eliminated. Haven't decided on anything yet in that connection. I suppose I could fit an independant 4 jaw chuck to my spindle. Haven't bought any CNC gear to date, don't think I want a computer in MY machine shop unless I can make room for a dustproof cupboard. There's a pub over the road that runs rodeos twice a week (of all things) and stirs up lots of dust.....have to clean my solar panels at least once a week (couldn't have them on the roof, they have to be easily reachable! ) Brass is extremely expensive over here, when you get up in diameter, but it gives the best result, that's for sure !
     
  50. motormaker

    motormaker Registered User

    Apr 5, 2010
    91
    6
    8
    Tucson, AZ
    Technomaniac,
    Here are the best sources that I have in my library:

    Wheel Cutting Practical Notes for Clockmakers from Chronos Limited 1979
    Clock Wheel and Pinion Cutting by J. Malcolm Wild 1983 Argus Books Ltd.
    The Clock & Watch Makers Guide to Gear Making..."and other machines" by Robert D. Porter 2006
    Wheel and Pinion Cutting in Horology by J. Malcolm Wild 2001 The Crowood Press
     

Share This Page