Metal Film vs. Carbon vs Carbon Composite Resistors

[zacharybroyles]

Is there any real difference?

[petemoore]

  Check the data sheets for R accuracy, 1% Vs 5%, IIRC, does that matter?
  You may notice a difference in creation of thermal noise, then again you may not.
  Which brings the question: Where to pots 'fit' in the thermal noise equation, I suppose thermal drift doesn't matter because they're adjustable anyway?
  Noise matters more near the beginning of the chain or at inputs, if the signal is then amplified. Larger value resistors tend to make more noise [or that's my understanding of it].
  I'm pretty certain most of the time R noise is relatively quite small when compared to say an opamp choice for noise reduction, or guitar cavity shiedling.
  I don't really know how much these things matter, beyond using 5% resistors all the time and experiencing what I consider low noise...if I built everything I have with 1%ers I'd be able to tell more...some people use only 1% resistors .

[R.G.]

Like many questions, the answer is "read the tutorials at GEO - http://www.geofex.com".

There is an article there on the carbon composition resistor mojo.

The short answer for here is:

Metal film: lowest noise of the three, available easiest in 1% tolerance. Only slightly more expensive than carbon film
Carbon film: modest noise, usually undetectable in effects, and cheapest of the three. Only slightly more noise than metal film, and the easiest to find and buy.
Carbon comp: The cheapest kind of resistors in the 1950's and 1960's. Now more expensive because the technically better resistors like carbon film have supplanted them in the market, so the few remaining ones are more expensive. These have the highest cost, the worst tolerances (10% is most common), and the highest noise.

Carbon composition resistors are reputed to have magic mojo tone sound. This is simply not true for effects. There is a slight, possibly detectable difference in sound if you have BIG signals across the resistor - like 75 volts of signal! In that case the voltage coefficient of resistance causes a slight distortion.

[brad]

I think it really depends on what colour you prefer: brown, tan, or blue?