Resistor noise? 1/4w Metal Film VS. >1/4w Carbon Film???

[km-r]

Ive read somewhere about noise in these resistors. Metal film < carbon film noise blah blah...

Anyway, since resistor noise is inversely proportional to size, what power rating of carbon film can equal the noise profile of metal films?

Hope someone can shed a light on this issue... hehehe

Thanks guys,

[John Lyons]

Depends on the circuit and gain as well as resistor value.
Unless you have huge resistor values in series with the signal
I'd doubt you will notice any difference.

[rogeryu_ph]

Hi Km-r, long time not been here and long time no see.
Were both pilipino guys, and he's my best DIYer friend here in the Philippines
Anyway, In my experience and only opinion, if you're making a modulation pedals like those delay and reverb etc better use a metal film BUT in dirt pedals nevermind using metal film coz your going just to make supposedly dirt sound and it would help sometimes   
Noise is inherent in dirt pedals obviously when into mid to high gain level BUT for a simple overdrive pedal, I would suggest use a metal film.
This is my own just experienced.
Roger

[Ben N]

A nice discussion of resistor noise in amps: http://www.aikenamps.com/ResistorNoise.htm

[Ronsonic]

For my pedal building I'm pretty much all metal film. For the cost why bother. Compared to the other costs the difference in price is negligible.

[km-r]

HEY MR ROGER! i had a lengthy disappearing act here! its for a tube amp actually, and well, i get lots of noise, the unhealthy ones...
if only theres an "equation" to equalize thermal noises of these two resistor types.

@lyons, that is also a topic i would like to stress out... like between opamps, rule of thumb says that NE5532 is quieter than the TL072. but craig anderton [IIRC] studied that the NE5532 is only quieter than the TL072 with input impedances of about 20k and below. so, question is how huge is huge?

@ben, thats my refence but they dont provide a side-by-side comparison between resistor types.

@ron, its too bad that metal films have to come in expensive packs here and i recently ordered a lot but only those >50k ohms

[askwho69]

Nice thread my pre!

need to watch this thread very educational "pedal making"

[km-r]

yes dude, i want to shed light on this subject. if only theres an instrument that can give empirical data on white/pink noise content...

[Zapp Brannigan]

In low voltage circuits it does not matter whether you use CF or MF resistors, even if you build hi-gain overdrive, BSIAB for example. Noise won't be noticeable. In hi-gain tube amp/preamp CF resistors are not preferable as plate resistors, where they work under high voltage, 'cause you'll get hiss-like noise.

[askwho69]

no youre wrong i tried metalfilm in my dirt boxes and it has big different ... metalfilm flow like beer and carbon flow like rhum

[Ben N]

In hi-gain tube amp/preamp CF resistors are not preferable as plate resistors, where they work under high voltage, 'cause you'll get hiss-like noise.

Sorry about the OT, but my experience was different. I replaced the plates in my Deluxe Reverb with CF (because I couldn't find any MF 100k/1-2W), and experienced an immediate and noticeable drop in hiss from the original CCs. Not gone altogether, but a lot better.

[Zapp Brannigan]

In hi-gain tube amp/preamp CF resistors are not preferable as plate resistors, where they work under high voltage, 'cause you'll get hiss-like noise.

Sorry about the OT, but my experience was different. I replaced the plates in my Deluxe Reverb with CF (because I couldn't find any MF 100k/1-2W), and experienced an immediate and noticeable drop in hiss from the original CCs. Not gone altogether, but a lot better.

Try MF, they are less noisier than CF, which are less noisier than CC (according to your experience) And try to build something like Engl e530, SLO100 etc and compare different types of resistive material. And don't forget about power dissipation value - the bigger the better (I mean the less noise you get).

askwho69, well I never tried beer or rhum in my sompboxes, but never noticed any difference between my stuff where I used different types of resistors. Or maybe beer and rhum were inside of me?  Really - no difference on 9 volts.

[askwho69]

Lol its up to the builder to choose

[ocg]

i built two tube screamers, one with carbon film and the other one with metal film.... the differences are shocking, the metal film screamer have a very pleasant smooth feeling.

i also noticed that my MXR distortion III have metal film resistors...and yep it sounds smooth too.

[rogeryu_ph]

HEY MR ROGER! i had a lengthy disappearing act here! its for a tube amp actually, and well, i get lots of noise, the unhealthy ones...
if only theres an "equation" to equalize thermal noises of these two resistor types.

When you said noise is it Hum or Hiss? there's difference between them, first is the spelling  second is the sound    .  I knew you built Marshall 18watts can you share us the schematic and what grounding scheme you did coz I read some where RG has his own grounding Star suggestion against vox pacemaker Its in his GEOFEX site, maybe its not the resistor. First those vintage classic marshall doesn't use metal film yet and in 9v pedal why some of those modern smd resistor use in pedal seems not metal or carbon but sound the same with discrete solid state resistor 

[merlinb]

Anyway, since resistor noise is inversely proportional to size, what power rating of carbon film can equal the noise profile of metal films?
Hope someone can shed a light on this issue... hehehe

If there is no DC voltage across the resistor, then both carbon and metal film produce essentially the same amount of noise, regardless of power rating.

However, if there is a DC voltage across the resistor then there will be extra "excess noise", and any advantage you get by reducing the temperature of the carbon film will probably still be offset by the excess noise, which is a lot worse with carbon film resistors. Metal film resistors produce perhaps 0.01uV/V excess noise, whereas carbon film produce something like 0.2uV/V.

Now, suppose you have 5V across the resistor; if it's metal film that makes 0.05uV excess noise. Across a carbon film resistor that makes 1uV excess noise, so you need to make the carbon film sufficiently cooler that the reduced Johnson noise at least cancels the extra excess noise.

Now let's say the metal film resistor is 100k ohms at 27 celsius with a 20kHz bandwidth; using the standard formula
http://www.sengpielaudio.com/calculator-noise.htm  it produces 5.8uV Johnson noise, plus 0.05uV excess noise making 5.85uV noise. To match the metal film's performance we need the carbon film resistor to produce no more than 5.85-1 = 4.85uV Johnson noise, so what must its temperature be? Rearranging the formula I get 213 kelvin, or minus 60 celsius!

In other words, you're never gonna make a carbon film less noisy than a metal film. Unless you pack it with dry ice...

(OK OK, excess noise decreases with frequency, unlike white noise. Sheesh. Make it more complicated why don't you!)

[culturejam]

Thanks for that great explanation, Merlin.