[Elecraft] K2 Resistor Tolerance

Sat May 17 14:44:00 2003

The myth persists...

That the dB was scaled to match minimal human perceptions. The dB is a
simple child of a base ten logarithm. Picked for mathematical ease of
use, nothing more.

Watershed work on this done by Bell Laboratories before WWII gave a
minimal figure of roughly 3 dB for human perception. Tone at a certain
level. End tone. Tone at another level. End tone. Question: second
tone same, louder or less? A large variation in the threshold.

In fact, we know now in certain comparison situations, e.g., tone
embedded in white noise, with the tone near the noise level, some
people can discern less than a dB, others not even the three. And even
in this kind of comparison, the KIND of noise changes the results
violently.

The ability of an electronic detection to be affected by changes in
level is an exercise in physics, math, and algorithm programmatic
sensitivity, no relation to 1 dB whatsoever.

There is NO easy, simple five word concept to summarize the usefulness
of another dB. It just depends on what you are doing. It's as complex
as it gets.

73, Guy.

----- Original Message ----- 
From: "Ron D'Eau Claire" <[email protected]>
To: "'Bill Henderson'" <[email protected]>;
<[email protected]>
Sent: Saturday, May 17, 2003 12:31 PM
Subject: RE: [Elecraft] K2 Resistor Tolerance

Good questions, Bill...

The reason for suggesting that you check the resistors with a DMM is
to
ensure you aren't mis-reading the colors. It's easy to do in a lot of
cases.
The idea is to avoid your putting a 100k ohm resistor in where a 10k
ohm
resistor belongs.

The accuracy of the K2 power readout at 5 watts is 10% in any case,
which is
typical of this type of power meter. That's because 10% is still very
accurate in terms of the needs of a communications system. A change of
10%
is less than 0.5 dB. When you consider that, under ideal laboratory
conditions, 1 dB is the smallest change in a sound level that a Human
is
capable of detecting by ear, and that in most radio communications it
is
considered that the smallest change a Human can detect by ear because
of
QRN, QSB, etc., is 3 dB, measuring within 0.5 dB is very accurate
indeed.

The potential error in the K2 power measuring system is more dependent
upon
the characteristics of the diode used than anything else. The
recovered d-c
the diode provides does vary with frequency slightly, even across the
HF
range. That particular diode type was chosen over for its fairly flat
response, but it's not THAT flat, nor are the diodes all exactly the
same.

Even the popular OHR wattmeter (http://www.ohr.com/wattmeter.htm)
isn't any
better than that. It's spec'd at 5% of full scale, but "full scale" is
10
watts. So the accuracy at 5 watts is also 10%.

IF you really want a more accurate power measurement system, you'll
probably
need to go to a much more expensive system such as the Bird
instruments. But
they will cost you a large fraction of what you paid for the K2 just
to
measure the power output a little bit more accurately and allow you to
make
adjustment not detectable in normal operation.

Ron AC7AC

-----Original Message-----
From: [email protected]
[mailto:[email protected]]
On Behalf Of Bill Henderson
Hi

Thanks to all who took the trouble to answer..

I am still stuck though. Why would the manual tell you to check the 1%
resistors with an ohmeter if a resitance outwith the tolerance wasn't
mportant.

Bob AG5Q says that it effects the accuracy of the power readout. Since
I
will be using the K2 for QRP this is important to me.

I'll stick the resistor on the board but will probaly change it later.

Thanks for your help.

_______________________________________________
Elecraft mailing list: [email protected]
You must be a list member to post to the list.
Postings must be plain text (no HTML or attachments).
See: http://mailman.qth.net/mailman/listinfo/elecraft
Elecraft Web Page: http://www.elecraft.com