William Chops Westfield wrote:

>
>  >Aren't resistors within a particular "batch" expected to come out
> very close to each other these days?  The tolerance value is more to
> handle differences between batches, extremes in temperature/etc, and
> so on.  I'd be very surprised if ALL the resistors on a particular
> reel weren't much closer to one another than they were to the nominal
> resistance.  So if it's a "low" batch, they're probably ALL low.

Yes, I would certainly agree with that.   But when you check a lot of 
resistors, both of different values and of different tolerance ranges 
(some 5% and some 1%), which I did, then they are presumably from 
different batches. 

I raised the issue pretty much tongue-in-cheek, as I think  the OP makes 
clear.  But I was thinking about it some last night, and it really does 
appear that the mean is lower than the nominal value.  From what I 
remember of statistics, a random sample of greater than about 20 should 
result in values from both sides of the mean.  If we assume that the 
mean should be the nominal value (6.8K, 100K, whatever) then values on 
either side of that should appear.  None do.  All of  the values -- 
without exception -- are on the low side.

Therefore, for whatever reason (lower costs to the manufacturer, or 
designed so that as the resistor ages it comes closer to spec, or a 
conspiracy, or whatever)  one has to conclude that  the true mean is not 
the nominal value.  It is something lower than the nominal value.

John


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist