using 5% resistors vs using 1% resistors

[mattron]

is it better to just always use 1% resistors? or are there any reasons to not use them in any circumstances?  like let's say i want to build an overdrive pedal and use all 1% , would the tone not be as good? or less warm?

[R.G.]

Tone is not the issue. The internet has produced its own myths, one of those being "precision resistors" ( or film, or non-carbon, or non-carbon-comp) sound harsh or sterile.

Tolerance (1%, 5%, etc.) says nothing about the underlying resistor's composition and manufacture, except accidentally. At one time, the cheapest way to make usable resistors was carbon composition. Today it's carbon film. And at one time metal film resistors and the sorting/adjustment time made 1% resistors expensive. No longer. The gap between 1% metal film and 5% carbon film resistors is really quite small. Today it's carbon comp that fetches higher prices, largely because of the internet myth that they produce magical musical tone.

Metal film resistors, as a property of their composition of a metal film, do have near-zero excess noise. Metal film and wirewound resistors are to be preferred where low noise is an issue. It is a quirk of history and manufacturing costs that "metal film" gets the knee-jerk response "1%!".

Carbon film resistors shed much of the excess noise of carbon comp resistors. They are quieter than carbon comp, a little bit noisier than metal film, and cheaper than either metal film or carbon comp.

Carbon comp resistors *do* have a tone advantage in very, very tightly limited circumstances; these circumstances exist in only about two places even in tube amps. They do not exist to a noticeable degree in low voltage pedals.

So
1. The tolerance (1%, 5%, etc.) does not matter to tone if the actual resistor size is proper for the circuit.
2. The material may matter to circuit operation and noise, and ought to be considered.
3. Using all 1% metal film for an overdrive pedal is a good idea for lowest noise, as overdrives tend to be noisy. But it's the metal film that matters, not the 1%.
4. The tone of a resistor type is effectively not an issue in low voltage pedals, so no, it will not be more or less warm or good, even if you could define what "warm" or "good" meant in this context.

[Seven64]

so is it basically personal preference to use metal film or carbon film in madbeans dirtbag deluxe?  i was planning on using carbon film to give it more of an "analog sound" because i have read your posts before about them producing slightly more noise. 

i just don't want to regret building one with 'noisier' resistors.  on the other hand, i could order another board and use metal film and see which one i like better.

[wavley]

There isn't really anybody that can explain it better than R.G. so I wouldn't even bother trying.

So I'm going to give you my opinion.  In the quantities I buy things for home use the extra cost of 1% resistors is negligible, so it's what I buy (for the most part) that way when something actually needs to be matched I don't have to worry and if I'm troubleshooting I don't have to worry if tolerances are working against me (of course 5% isn't really an issue here either, but some old stuff is 10 and 20% and that's how I got in the habit).  It's mostly just my preference to overbuild things and the fact that metal film is quieter.  I stay away from carbon comp for the most part, except for my tube amps where if it's original and isn't a source of noise I tend to leave it alone, but generally I don't replace them with carbon comp if I change them for any reason except the places RG mentions in his carbon comp article (even then I'm pretty sure the sound difference doesn't make or break my amp, but I do try to eek out every bit of good tone I can get)


edit: cleaned up something that didn't make sense

[therecordingart]

Great topic and responses. Thanks to all!

[mattron]

thanks for the response, i actually have a hole back of metal film resistors that i bought on ebay a few years ago, havent made a pedal in a while, but i guess ill just use those.  i also built a Ross compressor a while ago with 5% carbon resistors and its a bit noisy, i think ill swap them out for 1% metal film.

[R.G.]

i also built a Ross compressor a while ago with 5% carbon resistors and its a bit noisy, i think ill swap them out for 1% metal film.

That's a reasonable concern, but you should remember that compressors are by their nature noisy, so don't be disappointed if the improvement is minor. The high gain of the compressor circuit/IC itself at low signals may overwhelm the noise of resistors.

[CodeMonk]

Tone is not the issue. The internet has produced its own myths, one of those being "precision resistors" ( or film, or non-carbon, or non-carbon-comp) sound harsh or sterile.

Tolerance (1%, 5%, etc.) says nothing about the underlying resistor's composition and manufacture, except accidentally. At one time, the cheapest way to make usable resistors was carbon composition. Today it's carbon film. And at one time metal film resistors and the sorting/adjustment time made 1% resistors expensive. No longer. The gap between 1% metal film and 5% carbon film resistors is really quite small. Today it's carbon comp that fetches higher prices, largely because of the internet myth that they produce magical musical tone.

Metal film resistors, as a property of their composition of a metal film, do have near-zero excess noise. Metal film and wirewound resistors are to be preferred where low noise is an issue. It is a quirk of history and manufacturing costs that "metal film" gets the knee-jerk response "1%!".

Carbon film resistors shed much of the excess noise of carbon comp resistors. They are quieter than carbon comp, a little bit noisier than metal film, and cheaper than either metal film or carbon comp.

Carbon comp resistors *do* have a tone advantage in very, very tightly limited circumstances; these circumstances exist in only about two places even in tube amps. They do not exist to a noticeable degree in low voltage pedals.

So
1. The tolerance (1%, 5%, etc.) does not matter to tone if the actual resistor size is proper for the circuit.
2. The material may matter to circuit operation and noise, and ought to be considered.
3. Using all 1% metal film for an overdrive pedal is a good idea for lowest noise, as overdrives tend to be noisy. But it's the metal film that matters, not the 1%.
4. The tone of a resistor type is effectively not an issue in low voltage pedals, so no, it will not be more or less warm or good, even if you could define what "warm" or "good" meant in this context.

Oh yes.
Very much agree RG.

I do the majority of my large parts purchases at a local electronics parts surplus store ( Awesome place if you live in or near the San Fernando Valley, go there : http://www.apexelectronic.com/index.htm THEY DO HAVE AN EBAY STORE), which consists mainly of NOS stuff.
Its not the kind of place you go to buy 2 or 3 pieces, you get 2 or 3 HUNDRED.

They want 15 cents each for Allen-Bradley carbon comps. But I can generally get a handful (about 100) for about $10.
Metal films, I bought 500 for that same $10.
But damn, 15 cents for one resistor? Screw that.

Odd stuff like this:







I don't buy these parts cause I think they will sound better, but because they look cooler.
And I usually get them for a whole lot less than parts of the same value online.

[wavley]

I'm guilty of using/buying parts because they look cool too!  Sometimes it makes something cool, sometimes not.  For example, I just took a Kustom 4x10 combo I got in exchange for teaching the guy at the pawn shop some basic electronics and made it quiet (if you've ever played an old tuck and roll then you know the blast of white noise they have)  I changed the carbon comp mixer resistors to metal film and lowered the values and swapped out all the transistors in the signal path with lower noise stuff with lower gains in some places and now I have a pretty much studio quiet amp, though I'm not quite done yet.

The point of that story is, I know danged well that those transistors are noisy as heck, but I saved them to make a fuzz pedal anyway because they look cool and I kinda want to make a fairly low gain fuzz pedal that kinda mimics the character of the amp.  Will it be useable without a wash of noise?  Probably not.  Will it be fun? Most likely.

[R.G.]

Excellent point! It's a question of objectives. If what you want is for it to look cool, that's a perfectly valid objective. So is seeing if you can make a collection of parts do something. So it designing for lowest THD, or whatever else.

Where you get into trouble is finding out after you build it that you had unstated objectives.

[CodeMonk]

I've been pretty lucky with these parts.
Most of the resistors are 1% or less.
Some 1% caps as well.
And so far, all that I have tested or used did the job to my satisfaction.
The transistors have been a mixed bag, but overall a pretty low rejection rate.

Can't always find exactly what I go looking for either but usually find something just as good or better.
Example...went looking for some 10k resistors. Found none. But found plenty that were 9.31K 1%

I don't really go looking for the cool stuff, but If I see 2 bunches of caps that are both the value I am looking for and tolerance is comparable, I usually get whatever one looks the oddest. But if one is 10% tolerance and the other is 1%, I go with the 1%.
As long as its not massive in size.

[Paul Marossy]

Tone is not the issue. The internet has produced its own myths, one of those being "precision resistors" ( or film, or non-carbon, or non-carbon-comp) sound harsh or sterile.

Would this be called "resistor sag"? (or lack thereof)

[stezza]

Carbon comp resistors *do* have a tone advantage in very, very tightly limited circumstances; these circumstances exist in only about two places even in tube amps.

Could you please elaborate? For example, what resistors in the 5F1 circuit?

[petemoore]

  Use em if ya got 'em...
   Otherwise knowing when and where they make a difference is something we'll know more about afterwords...don't expect them change much of anything in say a dirtbox.
   These questions are relatable to:
  How precision does a shoe string need to be ? Will it influence traction ?

[R.G.]

Could you please elaborate? For example, what resistors in the 5F1 circuit?

Specifically, you can only get noticeable carbon comp distortion where the signal swing is over about 50V, and it gets more noticeable the bigger the signal swing. There's an article on this at geofex. What is happening is that the resistance of the resistor actually changes with the voltage across it. It's a tiny effect, miniscule fractions of a % per volt, so you need 100V or more to get it up to the few percent.

Where it really shows up is on a resistor with, say, 100V to 300V DC across it with no signal, and a signal component that swings that DC close to 0 and much higher than the DC condition. A really good one would be a 150V DC with +/-100V from that DC condition. This happens around the phase inverter, where you get 70-100V signals on a 300V B+.