Output stages; cathode current sense- 1-10 ohm often used in fixed bias. MF types eventually go high as they cannot handle peak transient currents. Use wirewounds for these applications.
richy
I just replaced some 2W carbon composition resistors I'd been using as plate loads on a 6SN7. Even though I'd paralleled a 47k with a 220k to get roughly 38k, and each resistor in each pair was rated for 2W, all had drifted high by different amounts. So much for my carefully matched values! Carbon composition resistors just can't take any heat without drifting way off value. They do seem to sound nice, though.
At this point, I only trust carbon comps as grid stoppers.
At this point, I only trust carbon comps as grid stoppers.
MF resistors are usually made by deposition of a metal on a ceramic cylinder. To get the desired resistance a helix is cut into the metal film by a laser, so its basically a thin strip of metal running around a ceramic rod (which also makes them inductive). Due to the tight interspacing MF resistors have a limited voltage rating. So wherever you have very high voltages or a huge voltage drop you have to be careful with MF resistors.
Apparently carbon film has the worst of two worlds...inductive, high TC and low voltage rating, plus noisy.
Carbon comp virtually have no inductance and a high voltge rating due to the continuous voltage drop over the part.
MF have low noise and very low TC.
(TC is short for temperature coefficient)
I generally only use MF, bulk tantalums, and non-inductive wire wounds in the signal circuitry, and despite the contrary advice have found MF types work just fine as grid stoppers as well.
I use some metal oxide (mo) and carbon films (cf) in power supplies.
I use no carbon composition resistors in audio at all, but do in RF applications (mainly tube tuner repairs and mods)
In addition to Johnson noise which all resistors exhibit as a function of their temperature and value, carbons of any type also exhibit about 20 - 25dB of additional excess noise compared to MF. (Due to current flow in the resistor) The quietest resistors overall for excess noise are wire wounds and bulk tantalums. (IIRC on the tantalum resistors)
Just echoing TheGimp and Rodeodave amongst others who got it right IMHO..
I use some metal oxide (mo) and carbon films (cf) in power supplies.
I use no carbon composition resistors in audio at all, but do in RF applications (mainly tube tuner repairs and mods)
In addition to Johnson noise which all resistors exhibit as a function of their temperature and value, carbons of any type also exhibit about 20 - 25dB of additional excess noise compared to MF. (Due to current flow in the resistor) The quietest resistors overall for excess noise are wire wounds and bulk tantalums. (IIRC on the tantalum resistors)
Just echoing TheGimp and Rodeodave amongst others who got it right IMHO..
Last edited:
Great info here. temp coefficient of carbon comp resistors is low? I didn't know that...
But I have to ask, why do carbon comp resistors seem to drift so much in value? I have carbon film resistors working for years and their value seems to stay put, but I (and others) have found carbon comps drifting quite a lot as the resistors age. Moisture gets in? (Sort of like old paper caps?)
-=|=-
But I have to ask, why do carbon comp resistors seem to drift so much in value? I have carbon film resistors working for years and their value seems to stay put, but I (and others) have found carbon comps drifting quite a lot as the resistors age. Moisture gets in? (Sort of like old paper caps?)
-=|=-
Actually CC resistors have very bad tempcos as is evidenced by significant changes in resistance value due to heating. Measure one just prior to soldering and remeasure immediately after whilst the leads are still hot! Note also that once cool it probably will not return to the same value you previously measured...
WRT to drift, moisture is definitely one mechanism, thermal cycling is another. I bet there are other mechanisms I know nothing about as well.
CCs may be noisier on paper, but in practice I can't hear a big difference to MFs.
One electrical characteristic unique to CCs is the voltage dependant resistance, the resistance varies a bit with voltage applied. This effect isn't much to talk about before hitting >100volts or so, but I've read some designs use that to compensate a tube's non-linearity.
I like CCs b/c they are non-magnetic, non-inductive, and handle surges very well. In some tube guitar amps CCs are essential for the vintage tone.
CFs are just crap in all respects, perhaps less drifty, but have all the negative aspects of a CC and none of the good ones. Some guitar amp builders brag about using CFs to emulate vintage CCs, but they are not the same at all, except for the noise. (Again, the noise is a non-issue for most designs).
I've had WW power resistors blow up, and the little CC I have in parallel kept going.
Unless for guitar, I'll have no probs using CCs where values up to 1kohms are used, and above that I usually go MF. *(The Vishay RN65s are nonmagnetic as well and very nice to my ears)
My fav resistor tweak is a CC in parallel with MF, and power resistors in cathode circuit it's CC in parallel with WW.
*(((my ears may not be the best but I'll kep that a secret
))
One electrical characteristic unique to CCs is the voltage dependant resistance, the resistance varies a bit with voltage applied. This effect isn't much to talk about before hitting >100volts or so, but I've read some designs use that to compensate a tube's non-linearity.
I like CCs b/c they are non-magnetic, non-inductive, and handle surges very well. In some tube guitar amps CCs are essential for the vintage tone.
CFs are just crap in all respects, perhaps less drifty, but have all the negative aspects of a CC and none of the good ones. Some guitar amp builders brag about using CFs to emulate vintage CCs, but they are not the same at all, except for the noise. (Again, the noise is a non-issue for most designs).
I've had WW power resistors blow up, and the little CC I have in parallel kept going.
Unless for guitar, I'll have no probs using CCs where values up to 1kohms are used, and above that I usually go MF. *(The Vishay RN65s are nonmagnetic as well and very nice to my ears)
My fav resistor tweak is a CC in parallel with MF, and power resistors in cathode circuit it's CC in parallel with WW.
*(((my ears may not be the best but I'll kep that a secret
))
*(((my ears may not be the best but I'll kep that a secret
No secrets hidden on this forum......Sounds like' he's been married for too long ............or play louder.
richy
Sure, you probably wouldn't want to use them in critical parts of a phono stage. But for most everything else, their noise is negligible. I mean, if you can't hear any noise at the listening position, what's the problem unless you're just obsessing over numbers?
se
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.