Table of Contents
Advertisement
db
10
.
P db
(
)
10
.
=
P 21.74
(
)
6.7 10
P r
1
P f
VSWR
P r
1
P f
0
20
.
=
P
1 10
2
P 0 ( )
P 20
(
)
=
2
simply summing all of the dBm readings and taking the average; you cannot
average power by averaging dBm! For example let's say we have two read-
ings of 0dBm and –20dBm. The average dBm would be –10dBm, which is 0.1
miliwatts but the average watts is actually 0.5 miliwatts. Quite a difference,
and this is with only two samples. Imagine how far off we would be after 200
all averaged together. So to take average power, we need to convert the dBm
from the samples to watts each time, sum the watts, then take the average
watts, not dBm. This slows things down a bit, but at least the accuracy is
there. Luckily the processor is fast enough to average 100 samples before up-
dating the display, giving you a pretty good idea of average power.
Another useful thing the PM10DC does is show percent of AM. Since we
can take more than 200 samples between screen updates, we can take all of
the samples and look for the maximum level and the minimum level. Then
when we calculate the difference between the two, we can derive the percent
of AM modulation! Rather than do some really fancy math to find AM, we
cheat and use a lookup table. It is much faster and it can compensate for
slight error in the AD8307s not responding fast enough to high-frequency AM.
We wind up with an accuracy of +/-1% of AM, which is not bad at all! Since we
only look at the difference in level from peak to valley of the reading, the AM
modulation readings are good from 100kHz up to 1GHz, even though the ac-
curacy of the power meter falls off after 450MHz.
Well, if your eyes aren't glazed over I hope you have learned something
from this; the math is a real challenge! Imagine doing all of this math on a little
calculator with just Add, Subtract, Divide, and Multiply. Now you begin to un-
derstand, grasshopper, why you are getting a LOT for your money!
see what we get:
0.001
As you can see the result is in scientific, but if we look
close, we see the result is 6.7 micro watts! A pretty small
6
value to be sure.
Now if we need to display VSWR, we need to use the follow-
ing formula. This will give use the ratio of reflected power to
absolute power. The problem here is we need to convert
both forward and reverse powers to watts, divide the reverse
power by the forward power, take the square root of the re-
sult, find 1—answer and 1 + answer and divide those also. It
is a bit of a daunting task, and square roots are a bit tricky in
a small micro-controller when you have to write
your own math routines. Imagine doing this by hand
4
2 times a second!
Since this is an advanced meter, we not only
.
4
5.05 10
wanted to display peak power, but also average
power. Average power cannot be calculated by
PM10DC 7
Advertisement
Table of Contents
