My mechanical clocks

[Davidz90]

Another attempt at calendar that doesn't need huge wheel with all 365 days.

The biggest problem to solve is to build a 12-sided cam that encodes the length of the months (or more specifically, how many days to skip at the end of 32-day dial; 1 for january, 4 for february etc.). Also, the design would be so much simpler if 32 tooth gear existed...

Edited December 28, 2022 by Davidz90

[aeh5040]

2 hours ago, Davidz90 said:

Another attempt at calendar that doesn't need huge wheel with all 365 days.

The biggest problem to solve is to build a 12-sided cam that encodes the length of the months (or more specifically, how many days to skip at the end of 32-day dial; 1 for january, 4 for february etc.). Also, the design would be so much simpler if 32 tooth gear existed...

That's amazing! Looking forward to further embellishments of this...

[Davidz90]

36 minutes ago, aeh5040 said:

That's amazing! Looking forward to further embellishments of this...

Thanks! I hope this design won't be a dead end.

[aeh5040]

For the 12-sided cam, one starting point might be to use these at 4 different orientations.

Edited December 28, 2022 by aeh5040

[Davidz90]

That's great idea! Thanks!

[Davidz90]

Still working on a proper Lego cam, in the meantime I tested the rest of the mechanism with a cardboard cutout.

 

[Davidz90]

I think I got a workable design for 12-sided cam:

20230101_230920 by David_Z1, on Flickr

it is based on the small ring gear. Radius can be easily controlled by adding plates. The ring will double as attachment point for the dial with month names.

The ring has 60 teeth, on the back side I have a 20T gear and knob wheel on a common axle. Advancing 20T wheel by quarter of a turn moves the ring by 1/12.

 

[Davidz90]

The ring gear idea worked. Now the calendar is finished. I believe it's world's first mechanical Lego calendar.

 

[lcvisser]

Does it do leap days? 

Very impressive, I need a few closer looks on a proper screen to figure out how it actually works.

[Davidz90]

2 hours ago, lcvisser said:

Does it do leap days?

Currently not, but that would be pretty straightforward to do here. February is represented by a gap on the ring, allowing the skipping arm to fall as far as it can go. A small cam attached to an axle rotating once per 4 years would stop it 1 day earlier on leap years.

[2GodBDGlory]

Nice work! I like seeing original ideas that take some mechanical cleverness to make work!

[Davidz90]

6 hours ago, 2GodBDGlory said:

Nice work! I like seeing original ideas that take some mechanical cleverness to make work!

Thank you! That's why I love mechanical clocks - they allow me to build all sorts of interesting contraptions unlike anything else. 

[Davidz90]

I have improved my calendar mechanism. First, I hevised a way to make a compact, 12-sided cam to encode lengths of months (by adding plates)

cam_design_2 by David_Z1, on Flickr

Then, I built the rest of the mechanism to match. Now it is way more compact.

 

[aeh5040]

Brilliant!

 

[Sentinel]

I’ve been working on building a clock on and off for  the past couple months, I have finally got the really simple and smooth, but now I need a way to power it. I’ve tried weights unsuccessfully, and I think I have an idea for how to power it indefinitely, but it’s going to require some custom parts

[Davidz90]

12 hours ago, aeh5040 said:

Brilliant!

Thanks!

9 hours ago, Sentinel said:

I’ve tried weights unsuccessfully

Hmmm.. what was the problem? Not enough power? The power consumption of various types of escapement mechanisms can vary greatly, from tens of miliwatts to tens of microwatts.

Anyway, indefinite power sounds fascinating.

[Sentinel]

13 hours ago, Davidz90 said:

Thanks!

Hmmm.. what was the problem?

Maybe I’m just too ambitious, I had trouble integrating the weights into my already finished mechanism, and I had a lot o trouble trying to figure out the escapement.

and I want it to run as long as possible without rewinding

This is my third and best design for a simple mechanism. Time can be set without winding. Hands are arranged bottom to top: seconds, minutes, hours the two 20t 1/2 thickness gears are there because i needed a 1-1 ratio at exactly that distance, and the full thickness gears are ever so slightly too largethere is a 5-1 gearing on the back as well

[Davidz90]

Nice! Quite a compact way to implement 3 hands. I suppose that the part with 3-sided star connector on the second photo is the escapement?

You might want to try this one:

or this one:

or this one:

All of them are super efficient (necessary power source is on the order of one kilogram falling 10 cm per hour), reliable and need relatively little fine tuning to make them work right.

[Sentinel]

3 hours ago, Davidz90 said:

 I suppose that the part with 3-sided star connector on the second photo is the escapement?

You might want to try this one:

I knew you would see that.

yes that was a prototype that did not work, I needs some tweaking but I didn’t put a lot of work into it.

I did manage to make one that worked using a 60t turntable (thinking I could get one oscillation per second on the pendulum), but I was misunderstanding how a pendulum actually worked on a clock and made it so the pendulum swinging stopped the rotation instead of being pushed by the rotation. This put pressure on the pendulum and consequently it only oscillated for a couple seconds

Clockmaking takes time

btw thanks for the escapement designs

Edited June 18, 2023 by Sentinel

[aeh5040]

For the weights, I think the best arrangement is a continuous loop of chain, with a self-winding mechanism in the weight: the weight consists of a battery box and motor and switch, which crawls up the chain to the top when it hits the ground...

Like in this one: https://ideas.lego.com/projects/5f13f0c2-284c-4cb2-b649-f554f29e1514

https://flic.kr/p/H9Wvnk

Edited June 18, 2023 by aeh5040

[Sentinel]

found this the other day

[Davidz90]

On 6/18/2023 at 9:16 PM, Sentinel said:

found this the other day

Yeah, seen this one as well. Astonishing.

 

Update to the calendar mechanism, now it has its own power source to avoid straining the clock mechanism.

 

[Davidz90]

A new large project. Clock powered by a synchronous motor:

A 230V, 50Hz motor in the base provides very steady 5 rpm output. That is used to lift two levers. The falling levers power the clock. The system is self-regulating: if the pendulum is lagging, levers are lifted higher, producing more torque and speeding up the pendulum. As a result, the accuracy is astonishing; I couldn't measure any error after 5 days of running (I don't have seconds hand so it is possible that there was some error of under 30 seconds).

Edited July 12, 2023 by Davidz90

[Davidz90]

I'm proud to present a strong contender to the title of the most accurate Lego clock in the world. It all started from a stable base:

Lego tower sandwiched between wooden boards and the base contains ~15 kg granite slab. This sort of extreme build was needed to keep structure vibrations in check with ~1 kg swinging pendulum. The final (for now...) clock is here:

Just now, I finished a 23 hour measurement of clock accuracy. The pendulum is intended to have 2 seconds period. Here's how it actually is:

period_24h by David_Z1, on Flickr

basically 2 seconds +-300 microseconds. However, what we care about is the total error of the clock (how much it is early/late), which is a sum of the errors of all periods: 

error_24h by David_Z1, on Flickr

The clock started on time, at ~2 hours it was 0.6 seconds early, near 9 hours it was 0.8 seconds late. Less than 1 second error in 24 hours is a Rolex-level accuracy (in fact, a little better than any mechanical watch).

And now some technical details:

The key component of the clock is grasshopper escapement (invented by John Harrison, it is one of the most accurate clock mechanisms). It's characteristic feature is that within some limits, clock speed is independent of amplitude. Due to the technical limitations of Lego (too much friction), I couldn't get that - the clock speed depended on amplitude/driving force, which is always a little variable due to friction. In order to combat this, I devised a magnetic compensation system - two magnets on the sides of the pendulum, pulling it away from center. As the amplitude increases, the distance between magnets at the pendulum at full swing decreases. By pulling the pendulum away from center, the magnets fight the gravity, and thus slow the pendulum down. The slowdown depends on magnet distance, so it is a function of pendulum amplitude. This amplitude-dependent slowdown counters the amplitude-dependent speedup of the mechanism. With this system in place, the clock speed looks like this:

mag_s4 by David_Z1, on Flickr

Clock rate is the standard way of measuring speed - rate of 1 seconds/day means that after 1 day of working, clock will be 1 second early or late. You can see that near 4 degree amplitude, rate is not changing much at all. This is astonishing stability - 1 second/day means 1/(24*3600) = 1/86400 - about 10 parts per million! 

Second key component is the compensation of thermal expansion - as it gets hotter, pendulum expands and clock slows down. This is especially bad with ABS plastic. I fixed this by hanging the pendulum weight on steel wires, and then using the expansion of bricks to compensate the smaller expansion of wires. The system is described in the video and currently is over 90% efficient (the dependence on temperature is decreased 11 times).

 

 

[Davidz90]

Final design of my high accuracy clock. In short: less than 1 second of error in a day, less than 3 seconds in a week. Beats high-end mechanical wristwatches and medium quality grandfather clocks, and approaches the accuracy of low-end quartz mechanisms.