Grasshopper skeleton Clock

[PavelK]

Hi,

This is my first post on this forum, although I have been reading the forum for several years. My dream was to build a skeleton clock from scratch and this forum convinced me to give it a try. In particular the Pin wheel skeleton clock construction thread by Allan Wolff has encouraged me.

I would like to share the clock I made. It is a grasshopper escapement clock based on plans by W. R. Smith with some modifications. The maintaining power is taken from plans by John Wilding. In contrast to the original plan, which uses a single ball bearing, and contrary to clockmaking convention, I have fitted the clock with ball bearings at all points where the parts make a rotary or rocking motion. I used a total of 29 ball bearings (plus 2 in the case hinges). Some more statistics: the clock (including the case) consists of 320 parts, it can be disassembled into 233 parts (each ball bearing is counted as one part).

The clock is far from perfect, I see the polishing and mainly the timekeeping as the biggest problems; my tries to regulate it are hopeless. But in any case, I'm glad it's finished and functional. The construction took me about 3 years (with some breaks) and several more years of preparation.



More photos of the finished clock can be found here:

Grashopper Skeleton Clock

by pkopriva | | 17 photos

eu.zonerama.com

Here are some photos from the construction:

Grashopper Skeleton Clock Construction

by pkopriva | | 145 photos

eu.zonerama.com

And some videos of the clock in motion:



Thank you all for the inspiration.

 

[John MacArthur]

VERY nice Pavel. I'm going to look through your pictures of construction. I wish you had posted them in progress - a "build log", but I'm sure I'll enjoy them as I look through.
Johnny

 

[PavelK]

Thank you. Unfortunately, I was not disciplined enough to systematically photograph every step of the construction. The photos are more or less random, or moments that I found interesting at the time.

 

[bpd]

That clock is beautiful and fun to watch it work.

 

[tok-tokkie]

Excellent. Congratulations.
The clock face - were the numerals etc etched?
Grasshopper escape is quite something to start off with.

 

[PavelK]

Thank you very much. I started with the grasshopper escapement because I wanted something interesting and there was quite a high probability that my first clock will be also my last one.

The clock face is etched, then filled with wax, silvered and lacquered. I had trouble lacquering it because the lacquer immediately dissolved the wax. I repeated the lacquering several times, being worried about the silvering, and the result is not quite what I was hoping for.

 

[Allan Wolff]

Pavel, Your clock is beautiful! Excellent craftsmanship. I really like the dial and hands. I bet it took a long time to file and polish the hands to that intricate shape; and great job on the blueing!
Some of your tooling and work holding is very ingenious! The filing attachment on your lathe is very clever! Is that your design?

Don't worry too much about the timekeeping of your clock. Even W.R. Smith said it is more of a conversation piece than a regulator. Everyone I have talked to about this clock say that it does not keep time well. Mine typically runs within 5 minutes a day, meaning it may gain 5 minutes one day and lose 5 minutes the next. I did read something on an old Yahoo Group about a guy who attached a regular pendulum to this grasshopper escapement and it was fairly accurate. The compound pendulum seems to be the problem, but it is certainly fun to watch!
Great job!
Allan

 

[PavelK]

Thank you for your kind words, Allan. Your construction of the clock you have in your avatar was also a thread I watched very closely and drew inspiration from it.

The filing attachment is not my design, it comes from Harold Hall. I followed his plans, I only left out the belt drive.

As to the timekeeping, I wasn't expecting great results, but it's still worse than I expected. First I tried to modify the fusee shape to get the clock to run evenly over the week (the running time per winding). In this I was not very successful but it is not so important. The more severe problem is, as I found out later, that the total variation during a week can differ greatly week to week (up to 50 minutes). The results are not repeatable. Perhaps it is caused by the compound pendulum, another reason that came to my mind is insufficiently lubricated mainspring (sometimes I can hear the spring unbinding) or the pendulum sometimes hitting the mechanical stops (my theory is that the ball bearings lowered the friction so that the mainspring is now too strong). But never mind, like you said, the clock is mainly a conversation piece and a moving decoration. And really fun to watch.

What surprised me was that the cover acts as a resonator and when I put it on, the ticking was so loud that it was distracting after a while. It got better when I put soft pads under the cover.

 

[Hessel Oosten]

Pavel,

It's superb !!!
Thanks for sharing.

Allen already mentioned before the beautiful hands.

If I'm right (looking at the photo with the wooden trunk behind them) you did file them from the beginning with a flat surface till they are right?

//

Once I did see (in my clockmakers club) another method, also clever one i.m.o.

The maker soft-soldered 2 (!) raw parts with their flat backsides on each other.
With some extending ends this (for now soldered, temporary part) could be easily put into a vice/rotating device etc.
Especially making the bulging is made comfortable in this way.

After completing the bulged shapes, he "desoldered/heated" the 2 parts and could give the extra hand of the two, to another builder.

Hessel

p.s. I'm sure you did see Tommy Jobson's maintenance of "this" clock? Here

 

[PavelK]

You are right, Hessel, I glued a paper template to a steel gauge plate (1 millimeter thick) and filed to the marked edge, leaving the handles flat. Then I removed the paper, glued the hands to a wooden block a started rounding the edges.

The method you described certainly sounds interesting. I'm only not sure how would hold the hands. Holding them just on the end doesn't seem like enough support to me.

I didn't know the Tommy Jobson's video, thanks for sharing.

 

[Hessel Oosten]

I'm only not sure how would hold the hands.

Pavel,
Will ask him when I see him, and report back.
Hessel

 

[GregS]

Very nice work Pavel. Thank you for sharing!
Greg

 

[Hessel Oosten]

Pavel,

I asked the "hands-builder (mentioned above) about his method for clamping.
He clamped increasing parts of the hands in a four-jaw chuck between two small steel plates.
When reaching half of the (double) hand, he turned it to file the other side.

Hessel

 

[gmorse]

Hi Pavel,

Holding the hands securely to shape them without damage becomes more of a problem the smaller they are. For watch hands I usually leave them attached to the stock for as long as possible to provide a 'handle'.

For hands that are intended to be significantly rounded, one way to approach it is to turn the hand in the lathe to produce as much of the required profile as possible, and then make the back flat by milling or filing. This would also work with two separate blanks soldered back to back.

A versatile vice like an instrument type such as an Eclipse 180 is a great help, (if you can find one!).

Regards,

Graham

 

[timryan.trr]

Hi Pavel,
I have been away from the site for quite a while, so I was happy to see that you, like me, have completed this project.

I have seen some of the same characteristics as you:
- This clock is loud. The pallets swing in and hit the escape wheels with some significant force.
- The mainspring is very prone to sticking regardless of the lubricant used. One clue may be that the original design apparently had a larger spring barrel with a longer spring that would allow the clock to operate more in the middle of the spring's range. I suspect Mr. Smith reduced the spring barrel to accommodate more commercially available mainsprings.
- One observation I have made is that if the pendulum is allowed to impact the mechanical stops, the rate will dramatically increase, so I have made sure that this doesn't happen by slightly modifying the stops.

My clock runs very well but has driven me quite insane trying to get it to keep good time. I suspect that the problem is inherent with the compound pendulum and extreme recoil escapement. It seems that even small variations in the driving power of the spring result in large changes in the rate. I don't understand why this should be so. If anyone has an ideas, please share.

Regards,
Tim Ryan

 

[PavelK]

Hi Tim,

Thank you for your insights. I'm glad to know that I'm not the only one having similar problems with this clock, and that it's probably more a design issue than a build fault. I think, at least in my case, a much weaker spring would have sufficed. However, I doubt that it would improve the running of the clock.

Regards
Pavel