Nice project. People interested in learning more about the diode
compensation circuit can find additional information in John
Grebenkemper's article "The Tandem Match -- An Accurate Directional
Wattmeter", QST, Jan, 1987, and a good deal more in my article "A Simple
and Accurate QRP Directional Wattmeter" in February 1990 QST.. It turns
out that you can reduce the diode drop to an arbitrarily low value by
terminating it with an arbitrarily high resistance. But that's not the
problem in detecting small signals. The problem is that the diode
reverse leakage current becomes a bigger and bigger fraction of the
forward current as the current gets smaller. That is, the diode becomes
more and more like a resistor as the signal gets smaller and smaller.
The problem, then, isn't a lack of forward conduction, but that the
charge conducted into the load during the positive half cycle is sucked
back out during the negative half cycle. The compensation circuit used
in these two articles and in Jason's circuit only approximately match
the RF conductivity characteristics of the diode, and the operating
points of the diodes have to be carefully chosen to insure tracking with
temperature changes as well as signal level.

Roy Lewallen, W7EL

Jason Hsu wrote:
Last semester, I designed and built the QROP Meter for my
graduate school independent study project. Thanks to those of you who
helped me with certain thorny issues, such as transformers and DC
amplification. Go to my web page at http://www.jasonhsu.com/ee.html
to learn how to build the instrument.

What makes my SWR/wattmeter different from the numerous versions
available from MFJ and the numerous homebrew designs out there?

1. Works from 200mW to 100W! You get the best of both worlds! MFJ
SWR/wattmeters don't work at QRP, and QRP SWR/wattmeters can't handle
100W. Tuning up at 1W instead of 100W reduces the QRM you cause by
20dB, or over 3 S units! You can also safely change the tuner's
inductance taps WHILE transmitting at 1W.
2. GOOD resolution at SWR values above 3 and even above 5! This can
be very handy at 160m and 80/75m.
3. 10-LED display for the SWR meter and another 10-LED display for
the wattmeter: No more squinting at cross-needle meters!
4. More accuracy at low power levels: Most SWR meters understate SWR
at low power levels due to the infamous diode drop loss. My design
compensates for this and allows accurate SWR measurements at QRP power
levels.

Let me know what you think, and feel free to contact me if you have
any questions or suggestions. If you decide to build the QROP Meter,
let me know how it goes.

Jason Hsu, AG4DG