Electronically Impulsing a pendulum for a clock

[markelectro]

bobledoux

Seems to be a neat little project,just the thing that might start me with the understanding of where to go.

Cheers!!!

 

[markelectro]

Thanks for this
all sounds very interesting and could be what I'm after But!! sory there is a lot I don't understand in the explination.If you have the time and could expalin in laymans terms I would be very pleased in hearing from you.

No-one seems to have mentioned that the natural frequency of the pendulum (ie. it's frequency when it is swinging without the electrical assistance) needs to be as close as possible to the frequency that it is to run at when electrically driven.

Otherwise, it will be impossible to make it sync with the electrical signal from the crystal oscillator (divided down of course).

Instead of a 4017, you could use a presettable counter such as the 74HC161.

This has a synchronous preset. Configure it to preset to 10 (ten) when it reaches 15 (ie. when Tc (pin 15) goes high. It will then be preset to 10 at the next clock pulse. Thus it is a modulo 6 counter.

The advantage of this is that you can use the Tc output (gated with say 0.2 sec) to drive the pendulum pulse.

If you wanted say a modulo 7 counter, then preset to 9, etc.

 

[markelectro]

Ray
I can sympathise with the fact that Children and Families do take up lots of our time o finding time to work on project can be a little hard.The real vain is the dreaded work,it seems that through your work you have the opportunity to work with Electronics unfortunately I don't.

Again this explination does seem to be closer tho what I'm looking for like (ljcox')explination but could I ask you to explain in laymans terms the parts you dercribe and what they do and are capable of?

Many thanks Mark.
Ps.don't forget about the swap (uk-New zealand) I will throw in a gpo.36

Also I'm working at the moment on a external very accurate timebase upon a 4.19 MHz crystal, in Crystal oven.

( Pendulum clocks with invar rods are very accurate, but occasionally may drift a couple of seconds a week. )

4.194.304 MHz divide by 22 gives one Herz.

For a 3/4 seconds pendulum i need 4 complete, or 8 half swings for one impulse per 6 seconds. ( I choose this value because i may use it for 2/3 seconds clocks as well )
I have not got a 1 second pendulum clock at the moment because of the hughe shipping costs to New Zealand $1000 NZ or 700$ US .

I'm thinking along the line of using a CMOS 4017 which counts the 1 seconds impulses from the 4017. decode the 6 seconds, and provide a synchronising impulse to the pendulum.
The 1 second impulse is about 0.2 seconds long.
By AND gating the 0.2 and 6 seconds impulse together to a driver transisitor (2N2222) a impulse can be put into the coil to drive the pendulum.

I haven't got this far yet because of 2 young children at home and long working days, and too much time on the forums i admit, but I will get onto it now sooner because of your interest in this great topic too.

This impulse can't be to strong, because the pendulum is already driven through the graham escapement from the electrically wind clockwork.

What country are you from BTW ?

Regards, Raymond

 

[bobledoux]

I got interested in master/slave clocks when I started looking at the British Synchronome, far away and expensive.

Recently, I purchased an American Standard Electric Company slave clock and built a very simple PIC controller to impulse it once a minute. Over the last couple of months it has been accurate to about a quarter second per month.

The impulse unit operates on a standard 4 mhz crystal at room temperature. I have corrected variations between calculated and actual times by adding and removing clock cycles from the timing loops.

The Microchip simulator has allowed me to count the number of loops required to complete a minute. The current count is 60,000,789 instruction cycles per minute; the PIC uses 4 clock cycles per instruction cycle.

I monitor the clock status every morning by comparing the clock advance pulse to the WWV minute tone. The clock is so accurate that I'd need to compare it to WWV with an oscilloscope to improve it. I'm waiting to see how summer temperatures affect time keeping when the indoor temps rise by about 5 degrees F.

It seems redundant to use both a pendulum and electronic time keeper to constantly pulse a pendulum which is a good time keeper in its own right. I'd prefer to use Shortt's method of setting the pendulum to run a few seconds slow per day and then use the electronic standard to provide the extra push to keep the times in sync.

See her for more info:

http://www.geocities.com/old_electric_clocks/page02.htm

 

[ljcox]

Thanks for this
all sounds very interesting and could be what I'm after But!! sory there is a lot I don't understand in the explination.If you have the time and could expalin in laymans terms I would be very pleased in hearing from you.

Do you have the data sheet of the 74HC161?

You need this in order to understand what the IC does and my post.

So I suggest you download one and re-read my post.

Then ask any questions that you can't resolve.

Of course, you can do it all in a PIC if you have PIC knowledge and a PIC programmer.

The Modulo of a numbering system is the number of digits it uses.

eg. decimal is modulo ten since it has 12 digits 0, 1 ~ 9.

Binary is Modulo 2 since it has 2 digits, 0 & 1.

Modulo 6 has 6 digits, 0, 1, ~ 5.

The IC counters count in binary but you can make them count in any "Modulo". The digits are coded in binary but can be displayed as numeric digits using a 7 segment display.

For example, a digital clock is working in binary, but it displays the time in decimal. So to make a digital clock you need 2 Modulo 60 counters (one for the seconds and one for the minutes) and either a Modulo 12 or 24 counter for the hours.

There have been plenty of threads on counters in the forum. So do a search for "counter".

 

[RODALCO]

Hi, interesting conversation going on re this very interesting topic.

A pendulum masterclock is referred to, in general in the amount of beats per second.

Say for a one second master, one swing is one second, a complete forward and backwards swing takes two seconds.

A 3/4 seconds master takes 1.5 seconds for a complete swing.
To get that to a round seconds value i have to impulse that every 6 or 12 seconds to keep the crystal oscillator in time with the natural swing of the pendulum.

The Hipp toggle has an off set butterfly switch, which operates on the left hand side swing. The contacts close when the arc of the pendulum decreases to the correct amount to close the Hipp Toggle switch via a notch in an attachement to the pendulum rod.
The coil underneath the pendulum attrackts a soft iron bar underneath the pendulum rod, and attracts it at the correct time, increases the arc so the pendulum in motion is clear to swing freely for the next 20 or 25 swings.

An other variety is described below.
The Electrique Brillié master clock impulses every second into the 1/2 second beating pendulum by means of a hairspring contact. Details may be on viredaz electric clocks website Google.

I take a detailed photo off the Hipp toggle on my Favag masterclock and will post it to make it clearer to other interested members.

I work in the Power industry as registered electrician, technician and deal with cable faults, substations, high 11, 22. 33. 66 and 110 kV and low voltage 110, 230, 400 Volts switch gear, control, scada, metering and instrumentation.
Electronics is a hobby of mine and read Silicon chip, Elektor magazines and build things with trial and error sometimes.
I don't know much about PIC's, and like to build projects with discrete components and TTL and CMOS IC's which are cheap and readily available.

Kind regards, Raymond

 

[RODALCO]

Link to great electric clock web site

www.mypage.bluewin.ch/electric-clocks/


Hope this link works

Regards

Raymond

 

[agent420]

Another clock fanatic besides myself... I'm not sure if I should be amused or concerned LOL.

I have been thinking of a similar project, using the natural pendulum cycle to drive a nixie display...

Anyaway, I am wondering if one of those inexpensive 'perpetual motion' kinetic art toys (**broken link removed**) could be hacked for the drive assembly... It seems to me they operate on the same principle, and using a correct length pendulum would result in the desired period. Myself, I'm hoping to use seconds escapement.

I'm not concerned with absolute accuracy, so long as it is usable.

edit -

If anyone involved with pendulum projects could explain to me the purpose of the pendulum spring, I'd be grateful... Is it simply to avoid the friction associated with a bearing?

 

[RODALCO]

Another clock fanatic besides myself... I'm not sure if I should be amused or concerned LOL.

I have been thinking of a similar project, using the natural pendulum cycle to drive a nixie display...


If anyone involved with pendulum projects could explain to me the purpose of the pendulum spring, I'd be grateful... Is it simply to avoid the friction associated with a bearing?

The pendulum spring has indeed to with avoiding the friction.
A normal bearing or pivot would wear out to fast, or the holes become oval and friction will increase.
The spring steel is well within it's range of flex with the 2 or 3° of the pendulum and not prone to wear.
Considering that one year has 31,536,000 seconds for a one second pendulum, a clock with a 1/2 second pendulum will double that to 63 million.

The pendulum has to be hung with caution and avoid twisting the spring which may break or shear it off.

Under normal working conditions these suspension springs have an almost unlimited lifespan.

 

[RODALCO]

Nixie

Another clock fanatic besides myself... I'm not sure if I should be amused or concerned LOL.

I have been thinking of a similar project, using the natural pendulum cycle to drive a nixie display...

To drive a nixie display would be a great idea !
a mechanical drive to a classic Nixie tube system is certainly a different approach for a clock.

Have a correct length pendulum, 1, 3/4, 2/3, 1/2 seconds beat.

Drive the pendulum via an electronic Hipp Toggle or Fedchenko system and put output pulses into your 7413 schmitt trigger and counting decades with 7400, 7490 and 74141 IC's or eq.

I'm also still looking for a good and reliable pendulum trigger circuit so any members with knowledge, please post us a link.
The links on Google give you more a completed product to buy instead of any usefull schematics.

Thanks, Raymond

 

[Cabwood]

Instead of forcing the pendulum to synchronise with an electronic timebase by accelerating and decelerating using electromagnets, how about this idea: Have the bob shift up or down along a worm screw driven by a small motor. If the pendulum oscillates to fast, lower the weight, if too slow then you raise it.

 

[RODALCO]

Cabwood

Thanks for your idea, very good thought but not really practical.
A pendulum clock is in general very accurate, if a minor adjustment is needed the course adjusting screw may have to be turned 1/16 of a revolution, or the fine adjustment screw 1/32th.

On heavy bobs a small weight ( 1/2 gramme ) may be added for a period of time or removed.
We talk bobs between 1 and 10 kg.

A small motor with it's associated wiring would cause bigger errors in pendulum accuracy because of microns of endplay in its armature shaft between different activations.
Also the controlwiring will fail at some stage due to repetitive flexing.

Thanks, Raymond

 

[agent420]

I'm also still looking for a good and reliable pendulum trigger circuit so any members with knowledge, please post us a link.

Have you seen this one? It seems to infer the kinetic art toys I referred to earlier use the same idea... I may pick one up at the local dollar store just to open it up...

I'm thinking of incorporating a decent sized relay (open frame for good aesthetic measure) that will be synced with the impulse circuit (or other pendulum counting sensor) to get the audible 'tick-tock'.

I also woke up this morning with another clock face idea based on the old color-changing **broken link removed** from the 70's... Honestly, I should probably see a therapist

Thanks for the info on the suspension spring.

 

[bobledoux]

The Solarbotics.com "SunSwinger Pendulum Kit" adds a transistor and uses a single coil for sensing and pulsing. You can download the manual which has a schematic diagram of the circuit.

 

[RODALCO]

Thanks agent 420 and bobledoux for your links.
They could be useable, especially when a external crystal oscillator is used.

Here are a couple of pics taken from the Hipp Toggle mechanism in my Favag 2/3sec masterclock.

https://farm1.static.flickr.com/242/457283586_75faebeb65.jpg Hipp toggle coil

https://farm1.static.flickr.com/237/457283584_5206c6a76b.jpg Hipp toggle coil

https://farm1.static.flickr.com/242/457283582_6ea363324f.jpg Hipp toggle switch

https://farm1.static.flickr.com/175/457283576_9a5e3c900d.jpg Hipp toggle switch

Regards, Raymond

 

[sam2]

Have you seen this one? It seems to infer the kinetic art toys I referred to earlier use the same idea... I may pick one up at the local dollar store just to open it up...

I'm thinking of incorporating a decent sized relay (open frame for good aesthetic measure) that will be synced with the impulse circuit (or other pendulum counting sensor) to get the audible 'tick-tock'.

I also woke up this morning with another clock face idea based on the old color-changing **broken link removed** from the 70's... Honestly, I should probably see a therapist

Thanks for the info on the suspension spring.

I have a room full of cuckoo clocks, I had to sound proof the room so it didn't drive my wife crazy. 125 clocks all going off at once is a great sound to me.
sam

 

[agent420]

They could be useable, especially when a external crystal oscillator is used.

I'd like to avoid any electronic timebase in my project. While I'm sure that such a circuit would no doubt improve the accuracy of the clock, there is just something about a 'natural' timebase that appeals to me. Occasional time (re)setting is just another reason to go play with it LOL. I will however try to make pendulum assembly fairly accurate, probably using invar. I'd love a merc compensated pendulum - how many old thermostats do I need?

I am quite jealous of your Favag

I have a room full of cuckoo clocks, I had to sound proof the room so it didn't drive my wife crazy. 125 clocks all going off at once is a great sound to me.

Music to my ears as well. I sometimes play the intro to Pink Floyd's Time just for that effect

Although it started as a side project, I am thinking of incorporating my solenoid driven xylophone (similar to the **broken link removed**) into the clock to serve as a chiming mechanism.

edit -

I note the polarity of the impulse in the referenced schematics that repel the pendulum is opposite in comparison to the older mechanical designs that attract the pendulum. It seems to me that the repelling method may introduce error because the lines of force are probably not always the same, as well as they won't always be tangent to the pendulum arc. I would think the attraction method is better in that respect...

 

[markelectro]

Ray

Nice neat set up.Probably close to where I want to be with the electromagnets below the pendulum.The hip and toggle set up must have to be at critical points in relation to the pendulum??? I suppose basically this works when the pendulum tiers enough for the little toggle to engage with the V slot thus causing a circuit .So no electronics required to impulse the magnets!!!

I think I would like to take this further and use combined electronics and Mechanics.The Mechanics being the pendulum and the electronics being what keeps the pendulum swinging.I think I would like to go down the route of placing a Permanent Magnet below the pendulum and use a sensor and a PIC. do do the maths.

Anyone with any ideas of sensing and creating a circuit usng PIC's to see where the pendulum is at and what it requires to keep it running I would like to hear your comments.Please try and explain in Layman's terms just until I learn the Electro Jargen!!

By using PIC'S to control the pendulum and because the pic should always know where the pendulum is and at what rate it should be swinging at I.E 1 beat per second We should be able to produce a slave contact from the circuitry.


Any help will be very appreciated.


Regards Mark

Thanks agent 420 and bobledoux for your links.
They could be useable, especially when a external crystal oscillator is used.

Here are a couple of pics taken from the Hipp Toggle mechanism in my Favag 2/3sec masterclock.

https://farm1.static.flickr.com/242/457283586_75faebeb65.jpg Hipp toggle coil

https://farm1.static.flickr.com/237/457283584_5206c6a76b.jpg Hipp toggle coil

https://farm1.static.flickr.com/242/457283582_6ea363324f.jpg Hipp toggle switch

https://farm1.static.flickr.com/175/457283576_9a5e3c900d.jpg Hipp toggle switch

Regards, Raymond

 

[agent420]

My physics is a bit rusty, but I'm wondering how a microcontroller or other 'smart' impulse method affects the pendulum accuracy... If I understand correctly, a pendulum has a natural oscillation, and it's period is not necessarily related to the length of swing in the arc, but only the length of the pendulum itself.... Meanining that as the pendulum 'slows down' or swings less side to side, it still takes the same amount of time to complete a cycle.

For a one second pendulum, that equates to approximately 39" (**broken link removed**)

 

[markelectro]

Agent420

I see what you are saying about the pendulum and you are correct with the length of the pendulum and the arc being fundamental to accuracy but what I think I'm trying to say is-

If we are using a 39" 1 sec beat pendulum we still need to give an impulse to keep the motion going while this is a simple operation as in hip/toggle movements where the pendulum receives an impulse via contacts as the length of the swing decreases and this is as you rightly said not an arc issue but a continuous motion issue but still the impulse is at a given rate.

I think when using Electronic impulsing Mathematics have to take over.We can use the fact we require an impulse no matter what and we can use this fact to impulse at set points. I think these points have to have a source from which information is gathered and this would be the pendulum.While we cannot use the electronics to control the pendulums rate (this is best left to the rating nut) we can determine when impulse is required.

PIC'S and circuitry could take over what the contact set did for hipp/toggle clocks but we keep the Electromagnets. The PIC may as well be used to it's potential being used to calculate the pendulums position we can then use these calculations to run off slave indicators.

Basically the Pic would be the clocks motion work as in std Mechanical clocks and motion work only works with the correct gearing on the wheels together with pendulum length and constant motion and would replace contact set as in Hipp/Toggle clocks.

Having said all this this is only theory, practicalities may be another issue. I am not an expert in electronics and my explanation is only vision. I would very much welcome suggestions other than Pic's to do the job.

Regards Mark

My physics is a bit rusty, but I'm wondering how a microcontroller or other 'smart' impulse method affects the pendulum accuracy... If I understand correctly, a pendulum has a natural oscillation, and it's period is not necessarily related to the length of swing in the arc, but only the length of the pendulum itself.... Meanining that as the pendulum 'slows down' or swings less side to side, it still takes the same amount of time to complete a cycle.

For a one second pendulum, that equates to approximately 39" (**broken link removed**)