I suppose I ought to watch the whole video, but gave up when he said B equals H. They are not the same thing. The unit H is the magnetic field strength in amps per metre, the unit B is the magnetic flux density in teslas. They are proportional in a non-magnetic medium like air. But as soon as a ferromagnetic material is introduced the relationship becomes non-linear and complex, in a mathematical sense, ie, complex numbers.

He is right that pure iron has a higher permeability than carbon steel. If I recall my lectures the problem with carbon steel is not the carbon per se, although it is diamagnetic, but when the carbon precipitates out in the form of carbides.

As hinted at by MG there are specific steels intended for magnetic applications, known as electrical steels. In simple terms the carbon is replaced by a few percent of silicon. Permeability is almost as good as pure iron, BH characteristics can be tailored and it has a much higher electrical resistance than iron. This reduces losses due to eddy currents.

Andrew

I think the motivation for this work was the poor performance of a Eureka clock reproduction using mild steel cores.  Having looked at the work and the Eureka design I actually think the problem is poor magnetic design of the Eureka which makes it susceptible to lower permeability.

https://groups.io/g/Horological-Science-Newsletter/message/548 – Brian Mumford suggested something similar.

Andrew’s point on air gaps is well made, there are several joints in the magnetic circuit which could add to the reluctance.

Just a thought; why concentrate on iron?  Ferrite materials are available in both soft and hard grades, the soft grades having low remanance. Worth trying the ferrite rods used as antennas in old radios?

Russell

The Eureka magnet which forms the “spokes” of the balance wheel is actually not just a plain rod core with a coil, it’s an E-shape with a bar each side, magnetically joined at one end, and the air gap that interacts with the stator at the other.  It seems to be possible to buy soft iron rod but flat strip as would be needed for the “arms” doesn’t seem to be around.

Try this for flat iron

dave8

https://www.aliexpress.com/item/1005006016395605.html?src=google&aff_fcid=5e76adcaa2234d5396979c5de72cc201-1708547603501-04540-UneMJZVf&aff_fsk=UneMJZVf&aff_platform=aaf&sk=UneMJZVf&aff_trace_key=5e76adcaa2234d5396979c5de72cc201-1708547603501-04540-UneMJZVf&terminal_id=14e706d9ab3b4488852737047f36d9ce&afSmartRedirect=y

Rather than the design shown by JH, could the side pieces be omitted and a second fixed pole diametrically opposite the existing one be provided. These 2 poles should be magnetically connected. There would then be much less sideload on the balance wheel.

The modern way to do it would be to have the coil on the stationary bit, save having flex wires to the balance wheel.

In the first few minutes of the AHS video the presenter refers to ‘copies of the Eureka clock’, which makes JH’s supposition entirely reasonable.

The clockdoc link above has design details from which the weight could be calculated if it’s relevant.   Other docs there may give oil details.