An esoteric noise question

[scott wurcer]

In any electro-mechanical system losses are usually modeled as resistance. My thinking is that if the model is correct there should be a noise associated with it. The damping material on the cantilever, at the microscopic level, generates heat. So thermodynamically I think it is sound to assume there is a noise associated with the damping action. I have never seen any mention of this, not that I have looked a lot. What got me thinking of this is that I usually use a cartridge with the stylus removed (if possible) as a source impedance when testing preamps. This would be one noise source not measurable, in fact I don’t know how I would measure it. Possibly it is so below the noise of the coil resistance that it does not matter. In small microphones these losses can matter a lot and can contribute considerable noise.

[sreten]

You are correct to use a cartridge body to measure noise.
A 1kR will give entirely different results.

The inductance of a MM effectively means that the major noise
source at high frequencies is not the cartridge R but the input
R of the amplifier.

Given vinyls surface noise and resistances causing noise I
don't think any other noise sources need to be considered.

sreten.

[scott wurcer]

Measuring the unmeasurable is one of my work/hobbys. It probably doesn't matter in this case but I did find a reference that shows it matters in gravity wave detectors.

[SY]

Well, I would suppose that it would depend on the nature of the damping material. If you were using a polymer run above its glass transition (normal for elastomers), it would seem appropriate to model the T dependence using something like a Rouse model. You end up with relaxation times proportional to 1/kT, which might not be so different than the free-gas-with-collisions model that gives us the Johnson noise equation.