Ohm's law

We still use the term "Ohm" as the unit of reactance. Maybe we need a different word.

I use to like "mho" for conductance, but those damned ISO people keep naming things for dead European scientists.

The amazing thing about Ohm's law is that this leading-order behaviour works over pretty much the whole range of useful current densities, in the vast majority of conductors.

Cheers

Phil Hobbs

Well, dimensionally X is in ohms, i.e. volts per ampere.

Cheers

Phil Hobbs

(In Gaussian units, B and H are dimensionally the same, though their units have formally different names.)

calculator.html

I still like mhos. But churlys are my favorites, followed by lharrys.

Yes, it's pretty good. Bulk metallic conductors, like wirewound resistors, can have 0.1 PPM/volt sorts of coefficients, which sounds hard to measure to me. Tempcos will dominate that.

But 90 degrees off!

You can do pretty low level measurements if you have enough integration time.

Cheers

Phil Hobbs

You could get 5 identical 1K resistors, and make 4 of them into a 1K config, and compare that to the loner. The resistors in the quad see half the voltage.

Interesting history in Wikipedia:

I remember reading somewhere that before Dr. Ohm came along, people were pretty sure that E = I * R^n, but there was controversy over the value of n, and even whether it was an integer.

Well, even back then somebody ought to have noticed the symmetry properties: if it went as an even power, current would only flow in one direction, so if you reversed the resistor, it would start charging up the Leyden jar spontaneously.

If it went as a fractional power, there's more or less the same problem; in one direction it would look like a resistor, in the other, like an inductor or capacitor.

So you're stuck with odd powers, and it would be odd if the leading-order behaviour weren't linear, because unless its coefficient were exactly zero for all materials, the linear term dominates for small signals.

The thing that justifies calling it a law is that the higher order terms are so small for nearly all conductors. The dominant cubic term is due to heating; if you get rid of the temperature dependence, the cubic term is hard to measure.

Cheers

Phil Hobbs

Resistance is inevitable.

On Sun, 15 Nov 2015 09:34:04 -0500, krw Gave us:

Ohm's Law = Resistance has been quantified and assigned a moniker and valuation method.

Actually, the original "laws" pertaining to resistance (including Ohm's) were about the current that you could drive through a wire as a function of its length and diameter -- Ohm's predecessor (Barlow*) came up with a law that indicated that long-distance telegraphy was unfeasible, and (according to Wikipedia) held beck telegraphy research for decades.

I don't think that the actual notion of the independent lumped component that we call a "resistor" existed until well after Dr. Ohm.

• formatting link
  's_law

On Sun, 15 Nov 2015 15:06:05 -0600, Tim Wescott Gave us:

snip

Black Edison:

OK thanks, I wonder what the physics is behind the voltage coef.? (Does an "electron gas" type model take care of it?)

George H.

Maybe I'm being dense, but if there is a voltage coefficient in resistors (delta R/Volt) then doesn't that imply that the next term is quadratic? (After the very dominate linear term.)

George H.

The history of science is interesting. It's too bad we always focus on the successes and not all the false steps that we took along the way.

George H.

The Michelson?Morley experiment was a huge failure, so it was a huge success. False steps (failures) are sometimes as important as successes. But, you know that.

The nonlinearity is tiny in bulk metals. It's worse in non-homogenous (like thickfilm ceramic) and amorphous or granular materials, so it's a crystal or grain boundary thing. The extreme case is something like a MOV varistor.