Aged Carbon Film Resistors

bechbox

Member
On the Chase Tone Secret Preamp he says he uses "New production, Accurate Vintage Spec, Aged Carbon Film Resistors".
What does that mean? Is it just a gimmicky way of saying "carbon composition resistor"? Or is there anything else to it?

When looking inside a Secret Preamp he's using a mix of film and carbon comp resistors:

HllApAx.jpg
 
What hokum... Maybe "aged carbon film resistors" means he bought some old stock for cheap. Resistors are resistors, metal film are better quality than carbon - carbon resistors just have higher 1/f noise...
 
"New production, Accurate Vintage Spec, Aged Carbon Film Resistors".
In that picture I don't see ANY carbon  film resistors, only metal film and carbon composition.

But yeah, that's a hilarious description for the resistors. Makes me wonder if he hasn't solved the mystery of time travel in order to age his resistors?
 
The common name is the resistive element in the resistor.

Metal film is just that, a thin film of metal or metal oxide that is sized (cross section & length) to provide the resistance required. The film is applied to a ceramic core spirally and topped with a protective coating. These are the most accurate and quiet resistors commonly available. Typically light blue.

Carbon film are similar to metal film but use a film of carbon. Next in line for accuracy and noise. Typically tan or light brown.

Carbon composition use a mixture of carbon and a binder for the resistive element. Resistance is determined by mixture ratio of carbon and binder. The mixture is encapsulated in an outer case. Lowest tolerance and noisiest. Typically darker brown case that looks like plastic.

The two film type resistors are also produced in a variety of other colors and markings. Some with text rather than color bands.

Tonal characteristics of each type can be a topic of debate.

This is just a quick description, as usual search engines can provide more indepth information.
 
This legit article discusses the voltage non-linearity for carbon resistors is negligible until substantial voltage, and even then I doubt it's audible in consumer electronics (including tube amps).

But there's also a noise aspect, which is much more likely to be audible.

All resistors have an intrinsic thermodynamic noise, called Johnson noise. This noise is dependent on the resistance and the temperature, and is not affected by the resistor's composition. The Johnson noise density is constant versus frequency, and is thus sometimes called "white noise". E.g., a 1MΩ resistor has Johnson noise density of 0.13 microvolts/√Hertz⁠ at room temperature. For a 10kHz bandwidth, the total Johnson noise is thus (0.13µV/√Hz)*(√10,000Hz) = (0.13µV/√Hz)*(100√Hz) = 13µV. ⁠Put some real gain on that and you'll hear it as white noise "hiss". See https://en.m.wikipedia.org/wiki/Johnson–Nyquist_noise

But there's an additional extrinsic noise that is dependent on the resistor's composition, sometimes called excess noise, flicker noise, or "1/f" noise. (This 1/f noise also arises in transistors and opamps, you may have seen a datasheet where the noise rises at low frequencies.). See https://en.m.wikipedia.org/wiki/Flicker_noise although they don't describe it's origin (it can have many). The noise density is higher at low frequencies, so it's most audible there - it's a kind of "pink noise" (if you think of red as lower frequency, then more red added to white gives pink) There is no general equation for this excess noise, as it depends on the detailed construction of the resistor. In a metal which conducts well, this excess noise is typically very low for the audible frequency range. Hence metal film resistors tend to have low excess noise, so their overall noise in the audible range is dominated by intrinsic Johnson noise. But carbon is not a good metal, there's a lot of internal electronic tunneling and hopping going on which generate excess noise. Thus carbon resistors generally have high 1/f noise, which is added on top of the Johnson noise. Add some gain on this, and you'll hear the pink noise "hiss", i.e., it has more low frequency noise than "white" noise.

Bottom line: there just is no mojo for resistors, only excess noise. So go with metal film resistors, because they're quieter.

(Edited a typo)
 
Last edited:
^ Great write-up, I wish I had the patience to do something like this. The contribution of resistor noise is important and often overlooked in audio designs. A seemingly benign, mostly useless "current limiting" 10k resistor in series with a TL072 noninverting input degrades the noise floor by around 3dB, and I see this kind of stuff in a lot of circuits. 3dB extra noise is very audible.
 
Back
Top