1
$\begingroup$

In casual conversation, and often in bill of materials, tolerances are specified without the qualifier +/- xx%, often because the specific ± character is hard to obtain (especially in legacy UNIX EDA systems) E.g.

100R 1% 0603 resistor

Is there a convention as to whether this should be interpreted as +/- 1% i.e. resistance value from 99.0-101.0 ohms or as a total tolerance, i.e. resistance value from 99.5-100.5 ohms?

The datasheet should specify this of course, but I was wondering if there is a convention in scientific / engineering speech whether unspecified tolerances are one or two-sided?

Edit: To all: I understand that everyone is trying to be helpful but I not asking for how to solve the problem. My question is literally with regard to whether there are any conventions in interpretation, in electronic engineering, of just "1%" as +/-1% or +/- 0.5%.

Improve this question
$\endgroup$
0
3
$\begingroup$

If it says +/- 1% then it means from -1% to +1%. in everyery standard i have ever seen.

+/- is simply shorthand for writing a symmetric bound one can also make unsymmetric bounds like 0 to 1% or even positive positive bounds like +0.001 to + 0.025. Granted these are rare in electrical components but play a big role in mechanical engineering where its useful to design both shaft and the hole for its fitting as having same nominal value but indicate the fit with tolerance entries. You see a symmetric bound hole with symmetric bound insert would not reliably go together, but would be really tedious and error prone to call a 4mm screw a 3.9 mm screw.

More generally when people talk about percentages they mean within 1% of the target value again it means +/-1%. But one should really try to avoid doing this.

Improve this answer
$\endgroup$
5
  • $\begingroup$ Have seen electrical supply specs of -6% +10% though... still, plus one from me. $\endgroup$ – Solar Mike Nov 25 '18 at 13:31
  • $\begingroup$ Electrolytic capacitors often have unsymmetrical tolerances, sometimes as extreme as -10% +100% on the nominal value. $\endgroup$ – alephzero Nov 25 '18 at 15:02
  • $\begingroup$ Thank you for weighing in, but I think you misread question (or I formulated it poorly). My question was in the case that there is a specification 1% without the +/- qualifier - how should just "1%" be interpreted? $\endgroup$ – Snowas Nov 26 '18 at 14:11
  • $\begingroup$ @Snowas that would be indicates with the +-0.5% class. The problem is that resistor color codes does not know of any other possibilities than ± 1%, ± 2%, ± 0.5%, ± 0.25%, ± 0.1%, ± 0.05% otherwise it can not be color coded. The spec sheet can of course say anything it likes but at the end of the day color codes is what you read on trough hole components sm components have similar standards. So its natural that 1% resitor is supposed to be within ± 1% since thats what color codes do. So if you see anybody talking of 1% then they most likely mean +/-. But this is specifically for resistors $\endgroup$ – joojaa Nov 26 '18 at 16:29
  • $\begingroup$ @Snowas: ...and in the end, if you want to be absolutely 100% sure, e.g. because you're not sure whoever gave you the specs really understood the problem, you can always go with the narrower tolerance. A +0.5/-0.5% resistor does fit in the +1/-1% range after all. It may get slightly more expensive than absolutely necessary but you'll be sure it's within specs. $\endgroup$ – SF. Nov 26 '18 at 17:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.