On Tuesday, 18 November 2014 23:23:50 UTC, Terry Casey wrote:
In a recent thread, component tolerances were discussed and there was a brief
reference to E6 and E12 series.

The E series (not E numbers) have little to do with tolernaces the E series is teh number of differnt values per decade.


The concept of acceptable tolerances seems to be alien to many people,
especially those who dabble with electronics for the first time.

We have a lab about that.


The availability of accurate digital meters compounds these problems. It is
quite common to see posts in some forums on the lines of "I replaced R72
(100k) because it was reading high at 106.8k".

What's worse is when student comes to you asking for a 733.48K ;-)


Of course, if the resistor had a tolerance of ±10%,

I've found it difficult top buy such a resistor with such a low tolernace
most are 5% and 2% are quite cheap when brought in bulk 100+



In the days when everybody used analogue meters they would probably have
noted that the pointer indicated 100k, near enough, and moved on ...

Some would have, but in those days temperature drift could also a problem.


Then there are these mysterious E numbers ...

In the early days of electronics - or should I say the wireless? -
manufacturing tolerances were so high that a simple 1,2,5,10... sequence was
about the best that was reasonably possible. In fact, although improvements
in resistor technology moved on quite rapidly, there is still a lot of
vintage equipment about with capacitors that follow the 1,2,5 sequence* ....

When it became viable to consistently produce resistors with a ±20%
tolerance, a logarithmic or exponential series of values appeared. This was
the E6 series, with values of 10, 15, 22, 33, 47 & 68 ohms and multiples of
10 thereafter.

In time, as tolerances improved still further, the E12 series (±10%) and the
E24 series (±5%) appeared.

I was asked to buy a set of E192 a cople of months ago.

I do have a 50R 0.01% that cost 15 quid !
and I've brought some 97.6K at 1% for a lab.

Anybody who thinks that the ranges of resistor values follow some weird
random sequence might like to look at this drawing I produced which shows how
neatly the values in the various ranges neatly dovetail together:

https://dl.dropboxusercontent.com/u/24301298/Res_Tolerance.png

* Some people seem to have great difficulty in grasping the concept that the
0.2µF and 0.5µF capacitors that they wish to replace can no longer be found
and that all they are are offered are 0.22µF and 0.47µF components ....


True another problem I'm finding is that when google tells a studetn they need a 0.01uF they tell me I haven't any in the lab, but there's a draw full of 10 nF caps.