> how does thickness affect ribbon ratio's/frequency/power?
Build two ribbons. One is twice the thickness of the other.
The thicker ribbon has twice(*) the mass, so it will move half as fast, make half the output voltage.
However it also has half the resistance, so we can use a 1.414 times higher step-up transformer.
If you change both the thickness and the step-up, double thickness means 3dB less output.
Therefore thinner ribbons are better.
(*) Ah, but there is a layer of air that clings to the ribbon. And it has mass that won't change with ribbon thickness. An infinitely thin (but air-tight) massless ribbon would still have mass when used in air (using mikes in a vacuum is generally not done, though it can help quantify the air-effects by measuring the difference).
So the total mass is air-mass plus ribbon-mass. If the ribbon mass is higher than air-mass, motion is low, and output falls faster than the reduced resistance helps. But if the ribbon mass is lower than the air-mass, motion and voltage is limited by air-mass, and resistance rises which reduces output power.
So as a first-crack, try to split the difference. Consider setting ribbon mass equal to air-mass.
As a VERY rough approximation: the air-mass on a circular diaphragm is about like a hemisphere of air (on each side) the same diameter as the diaphragm. For a long rectangle like a ribbon, I suspect it is roughly a semi-cylinder (both sides) with diameter about like the narrow side of the rectangle.
Assume the ribbon is infinitely long, so we can treat it as an extended cross-section and just figure the 2-dimensional problem. If the ribbon width is 1 unit, the area of a circle the same diameter is 3.14*(0.5^2) which is 0.78. So we can approximate the air-mass as a rectangle of air 1 unit wide and 0.78 unit deep (0.39 on each side, but they have to move together).
Aluminum is 165 pounds per cubic foot. Air is 0.08 pounds per cubic foot. The density of Al is 2,000 times greater than air.
Therefore to mass-match, the ribbon thickness/width ratio should be 2,000/0.78= 2,600. For a 0.25" ribbon, 0.000,1" foil.
This is very slender stuff, hard to work with, easy to blow-out. However I believe this is the general range of the old 1/4" ribbon mikes; or rather, they picked 1/4" because they could get 0.1 or 0.2 mil foil.
But that is hardly the end of your worries. Take a sheet of roofing-tin or hobby aluminum, 24" wide and 0.01" thick. Shake it. Sounds like thunder. Scale it down to 0.25", it will sound like small high-pitch thunder. Corrugated roofing-tin thunders less, though it has higher mass/area than flat tin, and still "tinks" on the flats between bends. The floor-pan of your car is also roughly 24" by 0.01" tin, but bumped-up in both directions, and thunders even less. (Gripe: Honda knows this, but made my Accord's exhaust-shield out of near-flat tin, and it clangs on bumps. I assume I have to pay more for the Accura to get a well-dented non-clang heat-shield...) The "tink" in a floor-pan can be killed with added soft mass: not a big deal in a car, but deadens a ribbon which has to move to earn its cost.