Would it be possible to construct a compensating pendulum thus: With metals having different coefficients of expansion, construct the "arm" of the pendulum with, let's say steel. At the base of the arm have some levers attacched. Also have some brass rods going down the arm which are attached to the ends of the levers. At the other ends of the levers you would have the weight attached so that the levers could raise or lower the weight. As the temperature increased, the length of the steel would increase, but the lenth of the brass would increase more which would activate the levers wfich would push the weight upwards, thus nullifying the lenth increase of the arm. I imagine this has been thought of before, it took me about 30 seconds to come up with the idea. Has it been tried and if so, what are the flaws to the idea?
I don't know about any clock pendulums that work like that, but in watchmaking you never know. I think it is a viable option, but maybe tuning it would be tricky, like the mercury compensated pendulum. And I think the biggest reason why we never many iterations on the compensating pendulums is because Invar steel made it obsolete apart from it in the form of a gimmick on modern clocks. Just to give you an idea, a 20m rod of Invar extends about 0,5 mm if you increase the temperature by 20°C.
Both George Graham and John Ellicott made compensating pendulums exactly like this. Graham thought it wasn't satisfactory so he used mercury. Ellicott's pendulum was published in 1752 by the Royal Society, it's on the net.
Would it be possible to construct a compensating pendulum thus:
With metals having different coefficients of expansion, construct the "arm" of the pendulum with, let's say steel. At the base of the arm have some levers attacched.
Also have some brass rods going down the arm which are attached to the ends of the levers.
At the other ends of the levers you would have the weight attached so that the levers could raise or lower the weight.
As the temperature increased, the length of the steel would increase, but the lenth of the brass would increase more which would activate the levers wfich would push the weight upwards, thus nullifying the lenth increase of the arm.
I imagine this has been thought of before, it took me about 30 seconds to come up with the idea. Has it been tried and if so, what are the flaws to the idea?
I don't know about any clock pendulums that work like that, but in watchmaking you never know. I think it is a viable option, but maybe tuning it would be tricky, like the mercury compensated pendulum. And I think the biggest reason why we never many iterations on the compensating pendulums is because Invar steel made it obsolete apart from it in the form of a gimmick on modern clocks. Just to give you an idea, a 20m rod of Invar extends about 0,5 mm if you increase the temperature by 20°C.
Both George Graham and John Ellicott made compensating pendulums exactly like this. Graham thought it wasn't satisfactory so he used mercury. Ellicott's pendulum was published in 1752 by the Royal Society, it's on the net.