In all kinds of engineering, we build on abstractions in a kind of inverted pyramid. Lots of people can, for example, design a system using ready-made building blocks on printed circuit boards. Fewer people can do the same design using ICs. Fewer still can design with components. But who designs the components? Even fewer people. Then there are the people designing the constituent elements of those components. [Learnelectronics] wanted to break one of those abstraction layers so he shows how to make your own wire-wound resistors.
Wire-wound resistors are often used when you need resistance with a higher power dissipation than a common film or composition resistor. Using nichrome wire makes this more practical since a meter of it has nearly 20 ohms of resistance. A regular wire has much less resistance. The video shows a drill winding a coil of wire neatly, but this also highlights one of the problems with wire wound resistors.
Actually, the winding introduces two problems: First, the coils could touch each other which would change the resistance. We would have been tempted to wind with a definite spacing on a form and then epoxy it to the form to give it mechanical stability. He builds his resistor into a plastic tube and does mention he’d pot it if he were really going to use it, but that would still make it hard to get the resistance exact.
The other big problem with a resistor like this is that it not only looks like a coil but it acts like one, too. In some applications that doesn’t matter. But in many, you don’t want the giant voltage spikes that comes with a quick change in current flow. We’ve seen this cause mysterious power transistor failures when a wire wound resistor served as an emitter resistor. In addition to inductance, the adjacent windings give the resistor a good bit of capacitance, as well.
Does that mean you can’t use wire wound resistors? No, you just have to get the right kind. Turns out you need noninductive wire wound resistors that use something called Ayrton-Perry windings. The idea is simple. Suppose you need one meter of wire to get to 20 ohms, but you want to minimize the unwanted inductance and capacitance.
You double the amount of wire and make a 40-ohm resistor with good gap between the coils. Then you essentially wind another 40-ohm resistor in the opposite direction inside that gap (or, sometimes, on top after insulating the first winding). With the same number of turns, the magnetic fields cancel out, meaning there isn’t much inductance. And the adjacent turns of wire have about the same voltage on them, so that minimizes capacitance.
Of course, now you have two 40 ohm resistors, but when you put them in parallel you get the 20 ohms you desire. You also used four times the amount of wire, but there’s no free lunch.
You might think wire wound resistors are old fashioned, but they just went high tech. If you’ve ever used a metal film resistor, these are very similar, but instead of wire, they use a very thin film of nichrome that can be as small as 50 nanometers thick. Usually, they don’t wind these, though. They deposit a uniform film and then use some process — these days, usually a laser — to cut a spiral or other pattern into the film, effectively making it a coil. Yes, they do exhibit some inductance as well, depending on the material and the cut.
We’ve talked about making your own capacitors and even your own inductors. We hope when we post about projects where people bought ready-made components, we don’t get a rash of “not a hack” comments. After all, we stand on the shoulders of giants.
No comments:
Post a Comment