resistor percentages? Do they make a big difference?

[Guitartoma]

I am new to the pedal building arena and am loving it.

A question: I've learned that resistors have a tolerance percentage - Should I be trying to get as small a tolerance as possible for building pedals? Does it make a difference?

Thanks in advance!

-Tom

[Processaurus]

5% resistors are fine for pedal building, it is rare that a situation comes up where it is critical to match resistors or that they be highly accurate, as good designers design circuits that will work with wide tolerances on the resistors and that use common values.  But sometimes a circuit will need just the right resitance.  I just go through the 5% resistors with meter in those cases until I find the right one.  Or if I can't find the right one, use a much higher resistor in parallel to reduce the resistance to the right value.

It would be nice to have a collection of 1% metal film resistors though, not because of the tolerance, but they can make some circuits a bit quieter (less thermal noise).  They're just more expensive than the 5% carbon film ones.

[Guitartoma]

That's great information - thanks for sharing!

[cpm]

doesit matters having 950k instead of 1Meg (or 4k47 instead of 4k7)? on most situations it doesn make a difference.
On the other hand, there are applications like n-th order filter where frequencies may shift due to the tolerances, even there, its usually better to care for caps tolerances, since they may be into 10% or 20%

[Mark Hammer]

As already indicated, sometimes they matter, and sometimes they don't.  Some folks find it easier to just keep a decent stock of 1% standard values around so they never have to think about it.  Given that 1% resistors can be had these days for about the price you'd pay for a 5% from Radio Shack, that's not a bad idea.....but it's not critical in more than maybe 5% of the instances where you'll use resistors.  And, as Processaurus notes, if you have enough resistors of a given 5% value, you can often score one of the precise value needed by sorting them with your meter.

[JKowalski]

And, as Processaurus notes, if you have enough resistors of a given 5% value, you can often score one of the precise value needed by sorting them with your meter.

My dad once hypothesized that some factories check resistor values on the assembly line and sort them according to tolerance - say, spot on values go to 1%, okay to 5%, bad to 10%. That would mean that if you got 5% you would never get a 1% or better value. I'm not sure if that's true but it's a plausible theory... All my (new) resistors are 1% metal film so it's hard for me to tell.

[Venusblue]

And, as Processaurus notes, if you have enough resistors of a given 5% value, you can often score one of the precise value needed by sorting them with your meter.

My dad once hypothesized that some factories check resistor values on the assembly line and sort them according to tolerance - say, spot on values go to 1%, okay to 5%, bad to 10%. That would mean that if you got 5% you would never get a 1% or better value. I'm not sure if that's true but it's a plausible theory... All my (new) resistors are 1% metal film so it's hard for me to tell.

I doubt they use that much quality control in their products. One of every batch probably gets tested to make sure the batch was "Good". I don't think most places care enough to check each and everyone one. That would probably be too labor intensive, when they could save all that money and hope it turns out good enough.

[Mark Hammer]

The 5% and 1% uits have a different composition, with the 1% being metal oxide and subject to lower noise.  It's the materials that permit one to aim for, and achieve, different tolerances.

[JKowalski]

And, as Processaurus notes, if you have enough resistors of a given 5% value, you can often score one of the precise value needed by sorting them with your meter.

My dad once hypothesized that some factories check resistor values on the assembly line and sort them according to tolerance - say, spot on values go to 1%, okay to 5%, bad to 10%. That would mean that if you got 5% you would never get a 1% or better value. I'm not sure if that's true but it's a plausible theory... All my (new) resistors are 1% metal film so it's hard for me to tell.

I doubt they use that much quality control in their products. One of every batch probably gets tested to make sure the batch was "Good". I don't think most places care enough to check each and everyone one. That would probably be too labor intensive, when they could save all that money and hope it turns out good enough.

It's not labor intensive at all - at one point in the fully automated assembly line a meter checks the resistance in a millisecond or less and sends it on down the line. No effort at all.

The 5% and 1% uits have a different composition, with the 1% being metal oxide and subject to lower noise.  It's the materials that permit one to aim for, and achieve, different tolerances.

But they do have tolerance sets for the same type resistor, such as 0.1% metal film, 1% metal film. That's what I am talking about, the 5% etc was mainly generalizing. If you look on Mouser or similar you'll find same brand resistors with the same exact specs but different tolerance ratings.

Note that making 1% resistors and picking out all the ones that fall into 0.1% is a lot easier and cheaper then setting up another whole production system to run in parallel for super precision trimming.


Anyways its just a theory. Dunno if it applies to any factory out there.

[Processaurus]

You know, now that you mention it, hunting with a meter for the right resistor, I have seen something like that with batches of 5% resistors from the same purchase, sometimes it seems like they'll be closely grouped on either side of 5%.