[time-nuts] "Shaking" of magnetic shields in atomic clocks

[Bill Byrom]

Rick, there appears to be interest in "shaking" the magnetic field for both atomic clocks and medical devices which operate at very small magnetic fields (such as magnetoencephalography). Shaking is mentioned here:
https://en.wikipedia.org/wiki/Magnetoencephalography#Active_shielding_system

I believe that the effect of shaking is to change the average portion of the ferromagnetic material B-H curve the weak external field affects. The incremental permeability (change in magnetic flux density B produced by a small change in magnetizing force H) for a ferromagnetic material is often much larger when the H field is not too close to zero. See this curve:
https://upload.wikimedia.org/wikipedia/commons/thumb/3/3a/Permeability_of_ferromagnet_by_Zureks.svg/996px-Permeability_of_ferromagnet_by_Zureks.svg.png 

The incremental permeability (slope of the B-H curve) is highest at a H field value greater than zero. An AC shaking H field component will allow very weak external fields to interact with the high slope portion of the B-H curve. If the shaking field was not present, the external field would interact only with the low permeability area of the B-H curve (near zero). There is best value for the shaking field which corresponds with the peak in the incremental permeability. The shaking frequency should be much greater than the frequencies of interest to reduce interference.

============================
Here are some other references:

Design, construction, and performance of a large-volume magnetic shield
https://www.researchgate.net/publication/3105787_Design_construction_and_performance_of_a_large-volume_magnetic_shield 

> The best way to achieve good shielding at low frequencies is to combine several methods. Ferromagnetic shielding and active compensation are standard methods for very low frequencies. In ferromagnetic shielding, the walls of a shield are made of high permeability material, and active shielding uses a closed loop control system that controls the field within the shield by special coils driven by a magnetic field sensor. The feedback control system counterbalances the disturbing field by generating a field equal in magnitude and opposite in direction to the disturbance. At higher frequencies, it is possible to use eddy current shielding that takes place when the walls of the shield form a closed enclosure of high electrical conductivity. 
> 
> Still, one method, so-called shaking, can be used in connection with ferromagnetic shielding. It effectively increases the permeability of a ferromagnetic material by generating a relatively strong alternating field into it. However, it generates extra interference at the shaking frequency. 

Separable Magnetic Shield with Magnetic Shaking Enhancement
https://core.ac.uk/display/37850118

Effect of magnetic anisotropy on magnetic shaking
https://ui.adsabs.harvard.edu/abs/1999JAP....85.4645P/abstract

Effective shielding for low-level magnetic fields
https://kyushu-u.pure.elsevier.com/en/publications/effective-shielding-for-low-level-magnetic-fields
--
Bill Byrom N5BB


On Wed, Sep 2, 2020, at 6:32 PM, Joseph Gwinn wrote:
> I have not read the paper yet, but this reminds me of the use of AC 
> bias to linearize the magnetic coating of recording tape.  That has the 
> property of causing linear behavior right through zero average field.
> 
> Joe Gwinn
> 

On Wed, Sep 2, 2020, at 4:31 PM, Richard (Rick) Karlquist wrote:
> 
> In the NIST paper available at the URL below:
> 
> http://pdfs.semanticscholar.org/47ac/742de238c0ece5e91ff7d12c515b9173eb60.pdf
> 
> At the beginning of page 2 (4th line) the paper
> states:
> 
> "Note that the shield permeability is a nonlinear function of the 
> magnetization and increases to a maximum value of umax =400,000 at
> higher applied fields. “Shaking” the shields by continuously
> applying an alternating magnetic field is a way to take advantage of umax."
> ... ...
> Rick Karlquist
> N6RK
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>