In article , Ian Iveson
wrote:
Trevor Wilson said


MM meters are generally not linear, if I remember, because the magnetic
field is not constant over the travel of the coil. With a moving coil,
you can have a magnet with a hole in it, or two magnets, or a ring
magnet with a gap, with a much more constant field in the space
between, in which a small coil can move. It would be impractical to
have such a magnet assembly as the moving part.

TBH neither MM nor MC carts are, electrically, inherently linear. Indeed, I
think that most 'MM' carts are actually induced field 'variable reluctance'
sensors which are nominally unstable without a mechanical restoring force.

However the 'geometric' and mechanical causes of nonlinearity tend to mean
that the above doesn't matter much as you'd end up with distortion
regardless of the magnetic design.

A problem for MC dynamos is the need for commutator or slip-rings. A MC
cartridge with two coils needs at least three connections brought to a
stationary point, with attendant compromises in suspension.

I assume MM cartridges were initially high output, simple, cheap, not
very linear, and had a high mechanical impedance. MC were the opposite
in all respects. Advances in magnetic materials and their fabrication,
and smaller-scale boutique production, will have allowed carts to use
more complex and lighter magnets, and hence MM with more linearity, and
MC with larger coils perhaps? Or how else is linearity and high output
achieved simultaneously?

The 'linearity' of MM and MC carts largely depends on the displacements
being 'small', and on "hiding un-noticed behind the mechanical and dynamic
nonlinearities"... :-)

Does one tend to have more crosstalk than the other?

BTW, has anyone mentioned the problem of noise in the early stages of
amplification from a low output source? Largely solved now, perhaps, but
it was a real problem during most of the development of cartridges.

This problem had two causes.

1) The engineers involved were used to designing low-noise amps with a
relatively high input impedance, so had becomed used to designs suitable
for that.

2) The devices widely available at the time tended to give optimum noise
performance with source impedance well above those of the MCs.

Again, though, I found that a common-base BP design worked just fine, even
20+ years ago.

The interesting thing was that some power transistors delivered very low
noise in such circumstances due to having multiple parallel emitter areas.
They also sometimes worked better if used 'the wrong way around' - i.e.
using the collector as the input and the emitter as the output! :-) This
was a trick I'd discovered some years before when making low noise amps for
low impedance sources in instrumentation. Careful selection could give
lower noise than using any of the devices sold as 'low noise' signal
transistors at the time. Not much use for production, though, because of
the need for selection.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html