Building a large-scale chronometer

Discussion in 'Clock Construction' started by akaash, Jul 23, 2014.

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

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Hello all,

    I am building a mechanical device that functions like a stopwatch for a school project.

    I have several months to do this, and (please correct me if I'm wrong) I believe the device that does this in a watch is called a chronometer.

    My idea is to build a very barebones chronometer on a large scale because it will be stationary on a table; it doesn't have to fit on a wrist.
    The plan as of now is to hook up a big hairspring to a balance wheel and a swiss-lever escapement. I'll have a gear train connected that can do all the indicating.

    Precision must be down to tenth of a second and I won't be timing anything longer than a half-hour.

    Is this doable to any degree of accuracy?

    Thanks!!
     
  2. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    In today's world, it would be simpler to use a computer
    for this operation. There were a number of large dial stop watch
    like clocks made for schools that were electric.
    The need to be 1/10 second requires a small balance wheel with
    a beat of a maximum of 1/10 second. This is a small balance wheel, not
    a large one.
    It is doable but not a practical project. If I were going to do this and
    it needed a mechanical dial, I'd start with a electric synchronous motor
    rather than a balance wheel. These motors can have RPM's in the range
    of 3600 rpm and can have clutch/brake arrangements to give you the
    accuracy your looking for.
    To make a large dial with that small a balance wheel and have it start
    and stop, without stalling the movement would be a significant project.
    A computer display would be more practical.
    Tinker Dwight
     
  3. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Thanks for your input.

    I would absolutely LOVE to use a computer, but I can't use anything electronic.

    As in, no electronic components period. Not even wires and a battery that I solder up myself.

    That's why it must be mechanical.
     
  4. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Also, the dial doesn't necessarily need to be too large.

    Everything will be mounted on bearings as well.
     
  5. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    Building anything that can start and stop in 1/10 second is tough.
    Most hand stop watches are only good to about 1/4 second at best.
    Mounting on bearings, isn't the issue, inertia is.
    Tinker Dwight
     
  6. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Would a stronger energy supply help in that regard?
     
  7. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    No, it only gets worse. Lighter, smaller and finer parts would be the way to go.
    Tinker Dwight
     
  8. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Thanks for your help.

    How do you recommend I begin building the escapement?
     
  9. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    I don't think it is a project for anyone but an experienced watch maker.
    It is way beyond what is practical for a good clock maker.
    If I were to attempt this, I have ideas that might be used, using the
    concepts of what would be called a PLL ( phased locked loop ) in
    electronics but done in mechanics.
    It is something that can be done but way more complicated than
    I'd ever try to describe in a email.
    From the questions you've asked, it is definitely a project beyond
    what you could achieve in 2 months.
    Tinker Dwight
     
  10. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Not even something like this?

    [video=youtube;9-ISAFyFMag]https://www.youtube.com/watch?v=9-ISAFyFMag[/video]
     
  11. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    Cute clock. It could do 1 second with that but 1/10 second
    is a factor of 100 finer because of inertia.
    Tinker Dwight
     
  12. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    So should I just abandon hope with this idea then?

    If so, do you have any other ideas as to what I should build to perform a stopwatch-like task?
     
  13. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    Nothing in the 1/10 second per hour that it practical
    with a 3D printer.
    I'm not sure why you want a mechanical clock for the purpose?
    It is such a trivial uP project.
    Tinker Dwight
     
  14. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Electronic and chemical components are not allowed in the device. Why they aren't allowed, I have no idea, but they just aren't.
     
  15. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    Ah! It is an assignment. You could make a clock
    the would escape at 1/10 second with a 3D printer
    but it would not be accurate for more than a minute or
    two.
    The key word is smaller!
    Tinker Dwight
     
  16. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Yes, it is an assignment :D

    I will keep 'small' in mind when building, thanks.

    Would accuracy be improved at all if I used a stiff metal hairspring in conjunction with the other 3D-printed parts?
     
  17. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    Yes, that would be part of it. I suspect the plastic spring has noticeable hysteresis.
    The balance wheel is an oscillator. Accuracy is all about improving the 'Q'.
    Things like friction, are important to control 'Q'. Making the balance wheel have a higher Q
    is the target for accuracy.
    There are a lot of tradeoffs. You have to make it small to make it faster. Making it faster
    increase the effects of friction.
    You might think ball bearing on the balance wheel would reduce
    but it doesn't compared to a pin point pivot.
    You need it faster to have a beat of 1/10 second.
    Tinker Dwight
     
  18. John MacArthur

    John MacArthur Registered User
    NAWCC Member

    Feb 13, 2007
    347
    29
    28
  19. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    An interesting project but the Tourbillon was
    to solve a problem of a clock having slightly different
    rates depending on the angle of the clock caused by
    the hair spring pulling or pushing on the pivots.
    The OP will have a fixed clock and will gain little
    by the complication of a Tourbillon escapement.
    Still, it is fantastic the things people are creating with
    their 3D printers. I don't know how much longer I
    can resist the temptation myself.
    Tinker Dwight
     
  20. John MacArthur

    John MacArthur Registered User
    NAWCC Member

    Feb 13, 2007
    347
    29
    28
    Yes, I didn't think that it fit his parameters, but it shows what can be done with 3d printing in horology. Me, I'm sticking to brass and steel, being kind of stuck in the past.

    Johnny
     
  21. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    http://www.nicholasmanousos.com/ is incredible!

    I am not required to use a 3D printer, though, I just thought it would be useful in the process.

    Do you guys think if, when I printed the gears, using a 25% or even lighter infill would help the inertia problem with any significance?
    To clarify, infill is the % of the printed object's volume that is solid plastic - 100% is a solid plastic piece, 5% is a plastic shell with wide honeycomb structures inside that provide support but let most of the object be air.

    Also, thanks for all the help.
     
  22. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    I think small is more important. The inertia
    of a rotating item is the square of the diameter, of the mass.
    If you have to start and stop it, small is the most
    important. Small size and small weight.
    Tinker Dwight
     
  23. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Right. I suppose a combination of shrinking everything and using a low infill will have the best chance of working.
     
  24. Tinker Dwight

    Tinker Dwight Registered User

    Oct 11, 2010
    13,666
    69
    0
    Calif. USA
    The energy contained of a moving object is MV^2. Your
    stuck with moving 10 times faster to get 1/10 seconds.
    Your only variable is M.
    Tinker Dwight
     
  25. tok-tokkie

    tok-tokkie Registered User

    Nov 25, 2010
    317
    2
    18
    Cape Town, South Africa
    Country Flag:
    This has been an interesting and informative discussion. It occurred to me that an alternative strategy that will help with the inertia problem would be to let the balance wheel run continuously while engaging & disengaging the hands to record the time. It will require a ratchet wheel and click which needs to be very fine to be able to record 1/10 sec . It will have to have as little inertia as possible so as not to slow (let alone stop) the balance wheel. Actually increasing the inertia upstream of the ratchet & click will be an advantage but the flywheel must be after the escape so it is running constantly at uniform speed.
     
  26. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Tok-tokkie,

    Thanks for the input.

    However, I am not following your train of thought. I have no idea what a ratchet wheel is. I also don't know of flywheels in clocks - maybe the mainspring?

    Sorry if this stuff is obvious. Novice here.

    Thanks
     
  27. tok-tokkie

    tok-tokkie Registered User

    Nov 25, 2010
    317
    2
    18
    Cape Town, South Africa
    Country Flag:
    I am surprised I have not been shot down in flames for that post. Mechanical cloaks are all (almost all?) intermittent motion devices - they start when the escape frees the mechanism but stop as soon as the escape locks the mechanism.

    Firstly to clarify. You refer to a chronometer. A chronometer is a precision watch or clock. http://en.wikipedia.org/wiki/Chronometer_watch. What you have been asked to design & make is a stop watch - a watch for recording the elapsed time between two events; like the start and end of a running race.

    My mistake was to assume that the clock mechanism was running constantly at a fixed speed. My suggestion was that when you start the stop watch the hands get coupled to the clock mechanism & they start moving. When you press the button again the hands are disconnected from the mechanism and thus show the elapsed time. Your problem is you are required to measure the elapsed time with great precision = 0.1 seconds.

    In a conventional mechanical stop watch the watch is started and stopped when the button is pressed. To resolve down to 0.1 second then the balance wheel must be oscillating at least as fast as 0.1 seconds per beat. To make a balance wheel oscillate that fast requires that it be small.

    My idea was to use a larger balance wheel and let it run constantly and simply connect and disconnect the timing hands from the running mechanism so they recorded the elapsed time. What I overlooked is the mechanism is not running constantly so after you connect the hands to the clockwork they will not move until the clockwork mechanism moves (when the escape activates it). You will not get any better resolution of the time than the time intervals of the balance wheel.

    The thing about increasing the inertia was intended that there be plenty of energy stored in the system so that when the timing hands were connected the suddenly applied load of the hands would not slow the running clock. My mistake since the clock is not running constantly. The clock is an intermittent motion device. That idea is only applicable if the clock was running constantly at a fixed speed.
     
  28. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    Now I get it.

    How could it run constantly though? Wouldn't free unraveling of the mainspring be erratic and uncontrollable?
     
  29. akaash

    akaash Registered User

    Jul 15, 2014
    14
    0
    0
    And instead of chronometer, I should have written chronograph.

    From wikipedia, "A chronograph is a specific type of watch that is used as a stopwatch".
     

Share This Page