About the Moon
  • Hello,
    I'd really like that my Sextus shows the moon phase too. I studied the mathematical problem, from 2 turns a day of the hour hand to 1 turn in about 2 month (using a classic semi-circle dial), and I solved with four arbors and this gear: 8, 40-10, 36-25, 32-8, 41. If I'm right, this set of cogs should realize a really good simulation of the Moon phases (synodic month=29.53059 vs. 29.52). I quickly checked the obstructions of this gear on the clock, I can manage it, (ehmm... not still sure), but for now the most important question is: in your opinion, the added friction of 8 cogs, will be too much in order to be offsetted by added weight?
    Do you know other set of cogs to realize the moon phase?
    Dave, are you interested to the problem? ;-) (help!! need experience...)
    Thank you
  • I must admit I have no experience of adding a moon phase dial to a clock, so anything I say should be taken with a pinch of salt.
    I can understand you wanting to add the gear train to the hour hand as it is the slowest turning in the clock, but it is conceived as a low load gear train, and I am not at all certain it would be able to turn a further 4 arbor gear train. However as the gear train ends in just a dial it really is just a case of increased friction, as opposed to a gear train that powers a pendulum.
    One method that occurs to mind is to introduce a separate drive weight somewhere in your gear train, this would relieve the strain on the hour hand and would enable you to keep the clock drive weight as is. I would think it would only have to be a small weight and with such a slow turning gear train would not need frequent winding.
    And lastly, as the Sextus is not designed to incorporate a moon dial I can only think the conversion would involve quite a few compromises as far as construction and design are concerned, but having said that I am very interested to see where you take this.
    Have you got any drawings we could take a look at?

    Dave
  • Not yet, the good idea of a separate weight make me think another bit. Roughtly my original idea was to remake a bigger dial, supporting the arbors and the gear train. The first cog would be a copy of the hours cog, in order to go righward and not have inteference with the drive arbor, then the gear train downward and finally the moon dial-cog, in the bottom of the frame. I don't really know if this asymmetrical design will like me... the right pendulum can swing freely, but it is hided for a half of the period. Maybe "the armony" is lost at the view.
    Another possibility: taking the movement from the drive cog, but it need another arbor at least. I'd like it because the gear train would be "inside" the frame, rear the pendulum.
    With this solution I have to constrict the dimension of the teeth as much as possible, not plywood probably.
  • Hi, here is a 3D sketch.
    I imaged it as an adjunct connected to the drive cog of Sextus, with similar frame design (but silver fiber veneered...)
    http://postimg.org/image/fk53e1j6n/
    http://postimg.org/image/8i75rufkv/
    Very happy for finding a new gear train, an arbor less then what I thought before.
    I'm going to use pins because I have a lathe but not a CNC milling machine, so I can do the same with half external work and cost.
    But not now, I'm afraid I'll dare this job next winter, with the snow on the streets :-)
    In the meantime, if you have other idea and opinions...
  • Hi
    Being a long time astronomical gearing aficionado my gut reaction here would be to keep it simple and stick with a 29.5 day lunar cycle and have it easy to adjust. Why? 1) I agree with Dave's warnings 2) this will be a wooden clock, subject to slight fluctuations, and will not be a high precision temperature & humidity compensated regulator and 2) even more important from what you are saying this will be the only astronomical gearing and so adjusting it will be very easy if you allow it to be. You are only talking about a 43 min 11.5 s correction over 29.5 days. You likely won't notice the difference in phase over a year (your dial is only an approximation to what you see anyway). With 29.5 days the gearing is easy. Use 1 tooth on a 12 hour rotation to a 59 tooth, or, better yet a gathering pallet run off the 12 hour pulling a tooth of a 59 tooth ratchet. With the latter arrangement adjustment is a breeze - every 2 years roughly you bump the 59 a tooth. If you forget to wind the clock you lift the pallet, spin the 59 tooth to where it should be and set the pallet back.
    High precision gear ratios are typically used when there is much to be displayed and it's integrated in with much other astronomical gearing leaving only the input (where everything is adjusted to compensate for the timing input error), or, dissassembly as the only means of adjustment.
    Lastly, There are still minor variations due to the moon's orbital inclination to earth (which also rotates ~ 18 years) and due to its elliptical orbit around earth (near/far) which also rotates with time. Unless you are factoring it *all* in I would keep it simple & reliable. Doing things like transposing ecliptic velocity (and other factors) to equatorial velocity takes a fairly complex set of gearing.
    Daryl
  • Hi Daryl,
    thank you for the answer, you're right, I don't need hight precision, expecially because in a year I suppose I'll have to adjust it many times, when i'll not be at home and so...
    Well, 1-59 gear ratio runs at double speed on a classic lunar dial, because the last cog must rotate 1 time in about 2 months... but this way seems very simpler anyway ;-)