EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: george graves on April 08, 2014, 11:29:40 am
-
So switches usually have three specs, current and voltage, and resistance. In this case I'm not too concerned with voltage(that must be a voltage creep thing? FOr safety?)
But I assume resistance contact rating, and current rating interplay a bit.
So, I've been working with some unknown rotatory switches. And after 200 mA, they seem to become possessed. Getting worse and worse. Below that, they are "perfect" conductors - just like a wire.
So I know a bigger switch will have a larger contact area....but what is the electrical model of a small switch?
And for that matter, a small conductor/trace vs a large conductor/trace. A inch long piece of 30 gauge of wire will have a low ohm rating. But a 12 gauge wire will ohm out the (about)same for a small current(such as a DMM supplies) when you measure it.
But.....Unless you are passing amps through it, then they have a very different characteristics...no?
I assume that conductors don't act as perfect resistors? Duh! And resistors don't act as perfect resistors outside their spec'ed range.
So I guess I'm asking, how are low rated current devices(a switch, a PCb trace, or a conductor, or a resistor outside of it's specs) modeled over an amperage range? I know there is a temp increase, but how does resistance change?
-
Conventional metallic conductors are ohmic to the tune of, geez, I don't know, but it must be more than nine orders of magnitude. Caveats: assuming the temperature remains constant, you're measuring at DC, the surfaces are clean, the metal is pure (and not one of those weird alloys or compounds that doesn't behave itself) and so on.
DMMs that don't read microohms obviously cannot be expected to measure microohms, and certainly not with a two wire test. A regular handheld DMM usually measures a low resistance as somewhere between 0.0 and 0.4 ohms, the same as the probes shorted together. Such measurements are simply erroneous.
Surface oxides (especially of copper, iron, nickel, and the sulfides of silver and copper) can form semiconductor junctions, causing low level errors (distortion in audio circuits, harmonics and IMD in RF connections), while bulk effects like thermal heating or (in semiconductors) velocity saturation and breakdown, and surrounding effects like breakdown of the insulator*, contribute various effects at extremely high levels.
*Especially at high frequency, and especially at low pressure. If you go high enough in frequency and voltage, eventually you can ignite plasma right out of thin air, and pump energy into it. This is used in chemistry for some of the most sensitive analytical tests, such as ICP-MS (Inductively Coupled Plasma Mass Spectrometry -- a sample is fed into the plasma, which breaks it down atomically, then the MS measures the atomic weights of the leftovers; sensitivity is in the parts-per-trillion!).
Resistors as such do exhibit a voltage coefficient outside of their ratings. Metal film types are usually quite stable (imagine that! :) ), whereas carbon composition are the worst, reading, high I think, above 300V or so (depending on size).
Tim
-
In this case I'm not too concerned with voltage(that must be a voltage creep thing? FOr safety?)
The max voltage rating is mainly because of contact arcing, I think. Shouldn't be an issue if you're testing the switch at low voltage.
So, I've been working with some unknown rotatory switches. And after 200 mA, they seem to become possessed. Getting worse and worse.
What are the symptoms of this demonic possession? Unreliable switching? High resistance when closed? Switch turning on & off when your back is turned? :scared:
-
how are low rated current devices(a switch, a PCb trace, or a conductor, or a resistor outside of it's specs) modeled over an amperage range?
Depending on what effect you are trying to model, and what amperage you are talking about.