Maintaining Power Spring

I am still working on my Congreve. Having disassembled the fusee for cleaning, I discovered the maintaining power spring, has broken. It's a very strong spring!

My chances of sourcing a replacement are probably vanishingly small and I don't think any repair I could manage would be sufficiently strong. It appears to be steel (its magnetic), 1.8mm thick. The spring section is 2.75mm wide and the recess in the great wheel is 48mm in diameter and 2.5mm deep.

The spring does not appear to be hardened - I can mark it with a file. Is it likely to be mild steel, or more likely to be some sort of spring steel? How can I tell?

Does anyone have any suggestions as to how to make it, and suggestions as to what from. One idea being to cut it out of a steel disk on the mill. Will mild steel be a problem? The spring only gets flexed once a week when the clock is wound.

Would it be possible to make one from a suitable size spring ring like used to retain bearings in housings. They are not hardened but can be filed?

Thanks Clive, I had not considered a circlip as the basis. A 45mm circlip has an external (unsprung) diameter of 48.5mm which is close. I will go and see if the hole alignment can be made to work.

I am guessing that it's high carbon steel, if so, a small section of gauge plate could be rolled up and tempered? Have a tour of suppliers catalogues, model engineering and clock material suppliers, the latter being a best bet. Alternatively, go back through old volumes of model engineers and see what was recommended by the writers of clock making articles.

Good luck.

John

Edited By DMB on 16/06/2022 15:26:28

I have made quite a few of these. I start with gauge plate of suitable thickness, cut a rough disc on the bandsaw, machine the inside and outside diameter on the lathe then finish on a rotary table on the milling machine followed by hand tools. Then harden, and temper to blue. Here is one I have just finished for a bracket clock project.

I am guessing your spring was not tempered to the correct temperature?

Just to report back, and in particular to thank Clive for the idea.

44mm nominal circlip arrived and has been fitted with the pins to make a replacement spring.

Seems fine, fits the inside of the wheel perfectly. Some quick math using an online beam calculator suggests it is not far off the original in stiffness. It will be tested when I get the clock back together.

Great repair. Would never have thought of using a circlip. That spring though, it looks disproportionately stiff. Much bigger than a typical maintining power spring. I guess the clock needs a lot of power to keep the tilting plate moving.

I know next to nothing about clocks but was interested to see from Britten's Handbook that the device was invented by John Harrison.

The Handbook points out that "it is important to see that on the one hand the [maintaining power spring] is not stronger than the mainspring, yet on the other hand it is sufficiently strong to drive the watch".

I hope the circlip does the trick!

I didn't think of it, to give credit it was Clive Hartland's suggestion (post 2).

The replacement spring is definitely "stiff", however I have used an online beam calculator as a check. Treating the circlip and the original as simple cantilever beams where the length is the arc length of the thin parts in both cases, and using the true cross sections, it transpires that the circlip ( thicker but longer) has about 15% more deflection than the original for the same force.

That said, in answer to ega's comment about the relative strengths, the mainspring is big. The barrel is 42mm wide and 62mm in diameter. I haven't had it out (yet!), but it's probably .45mm thick and about 3 metres long! It scares the living daylights out of me to think about the forces involved when fully wound.

Glad to see my idea was useful Peter, hope it all works well when set up.