Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Percent% Errror Quoted for Electrical Components. What is it?

  1. Feb 9, 2009 #1
    Errror Quoted for Electrical Components. What is it?

    (Where do I post electrical engineering homework questions? I'm home. I'm working. OK, maybe I'm not really working.)

    Manufacturers quote a percent variation from nominal value for their parts. What does this percent error even mean??

    For a 5% error, does this mean that each and every part is within 5% of the quoted value. Is the distributions flat, or is it Gaussian in shape?

    If the manufacturer also sells a 2% version, does this mean that the 5% parts will have the middle 2% missing as these parts are selected-out?
    Last edited: Feb 9, 2009
  2. jcsd
  3. Feb 9, 2009 #2


    User Avatar
    Science Advisor

    Re: Errror Quoted for Electrical Components. What is it?

    I believe that if you're quoted 5%, the manufacturer guarantees that the resistors are within 5% of the nominal value (at some given temperature)--otherwise, they'd go into the next resistor size! I seem to always get better though.

    The reason you don't see the 2% chopped out of the 5% (and the 1% chopped out of the 2%, and so forth) is that resistors are sold in various series. There are more 2% nominal resistor values available (E48) as opposed to 5% (E24) and 10% (E12):

    With better manufacturing tolerances, you can always sell higher-tolerance resistors as lower ones.
  4. Feb 9, 2009 #3
    Re: Errror Quoted for Electrical Components. What is it?

    That makes some sense. It may depend on the manufacturing process; upon whether a part is tested or not. Without testing, periodic process checks would be preformed to see that process parameters are within limits, such that tolerance limits are met in the final product. In this case you'd expect a small fraction of items to leak outside the tolerence limits. So the curve might be more or less a normal distribution. With 100% testing, those that exceed spec would be kicked out. In this case the distribution would be normal, with the tails of the curve truncated.

    It also depends on the quality of the people. I've worked for some crap companies, whos specs were bogus.

    But this is a lot of semi-educated guess-work on my part. Wish I knew better.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook