I'm fortunate to have a Hewlett Packard 312B Selective Voltmeter. You
can see it on the far right in the first photo on this page
http://qrp.kearman.com/html/homebrewing.html The 312B is a radio
receiver tuning from 100 kHz to 18 MHz. Instead of a speaker it has a
direct-reading signal level meter, and attenuators. It's easy to compare
signal levels from broadcast stations or other signal sources.

One thing I've learned from the 312B is that signal levels from
broadcast stations, even local ones, can vary quite a bit. CMBQ-530 in
Havana puts in a fairly constant 40 uV off my large loop during the day,
but that's an averages. Minute-to-minute the level can change by more
than 25%. Even the level of local pest WSTU-1450, a scant 2 miles (3 km)
away varies depending on soil moisture. I wanted to be able to compare
antennas over the long term, so I needed a way to ensure the signal
levels seen by them was more repeatable. Enter the Faraday box.

I happened to have a rack cabinet, steel, with aluminum front panel. I
replaced the back panel with aluminum window screen, so I could see
inside. Placing a radio on a couple of plastic containers taped to the
bottom of the box, I can repeatably position the radios. I use a short
probe antenna connected to a BNC connector on the original front panel,
connected to my signal generator. The box pretty effectively wipes out
the local pest, and everything else. An hour spent taking the radios (G8
and PL-380) out, putting them back in, retuning and resetting the signal
generator (using the 312B to confirm its output level), convinced me I
had a repeatable experimental setup.

The box is large enough to accommodate 7.5-inch ferrite-rod antennas. I
had two rods on hand, and did a few runs. My results were similar to Roy
Dyball's. A 330-340 uH inductance gives best results across the 530-1710
kHz band. Larger inductances improve performance below 530 kHz, but
performance at the high end of the MW band is degraded.

Another interesting observation pertains to positioning the coil on the
ferrite. As long as the inductance is in the range of 250-340 uH, it
doesn't make a lot of difference where the coil is located. You have to
tune the radio off frequency and back on to get it to retune the loop
after repositioning the coil. Other experimenters have found advantages
in centering the coil, and I believe the pattern will be more balanced
if the coil is centered. If you're coupling the large ferrite into yet
another antenna, that's less critical. For stand-alone use, a skewed
pattern could work to your advantage, or not, depending on the relative
bearings of desired and undesired stations.

I've also been looking at coil Q, using methods published by Wes
Hayward, W7ZOI. Ben Tongue did some research into using ferrite-rod
inductors in crystal sets, and found some advantage in using a spacer
between the winding and the ferrite. Using sheet plastic of unknown
composition to space the windings off the core about 0.063-inch (1.6 mm)
I can measure higher Q, but the effect on on-center signal strength is
below my measurement capability. There is, however, an advantage to
increasing Q, as it helps attenuate off-frequency signals. This can be
proven by theory, and I have independently confirmed it with several
regenerative receivers. A regenerative receiver uses active circuitry to
increase Q. My MW regen can achieve circuit Qs high enough to clip the
sidebands off an AM BC station. This simple receiver (3 field-effect
transistors and an audio amplifier) easily receives TAs. At its 1-kHz
bandwidth setting the PL-380 has a slight selectivity advantage, but the
difference is not as great as you might think. If you'd like to discuss
this, please contact me off-list, at jkearman _at_ att _dot_ net

The SiLabs data sheet mentions using an impedance transformer to couple
an air loop to the chip. I tried this with a 4-turn loop about 2 feet on
a side, and it does work. I used a 4:1 turns ratio instead of the 5:1
mentioned in the data sheet. Yesterday I was browsing the auction site
and noticed some untuned loop antennas about the same size as the Terk.
They're intended for use with Bose and other consumer AM-FM tuners. They
don't have tuning controls. I suspect this is the type of antenna SiLabs
refers to in the data sheet. I'm going to get one and see. The advantage
of this type of air loop is, the radio tunes it for you. I have some
data on connecting such an antenna so the internal antenna is
disconnected when it's plugged in. Contact me off-list for info.

The only disadvantage of this system comes if you want to locate the
antenna at a long distance (more than 5 feet/1.5 m) from the radio. The
cable connecting the antenna to the radio is part of the antenna tuned
circuit (the chip provides half of a resonant circuit in the form of a
voltage-variable capacitor or "varactor;" the other half of the tuned
circuit is the antenna itself). Coaxial cable has a capacitance of about
30 pF per foot (1 pF/cm); if the cable is too long the chip may not be
able to tune the antenna to the top of the MW band, the same effect
obtained when the antenna has too much inductance. A fixed series
capacitor will compensate for reasonable lengths, but if you plan to put
the antenna in the attic or back yard, this method probably isn't for
you. The advantage of having the radio peak the antenna tuning
automatically every time you change frequency is hard to beat, though.
Others have had to use spotting receivers to tune their loops, and then
swap in the ULR. I've always believed in letting the machines do the
work!

My breadbox isn't large enough to accommodate a 10-12 inch (250-300 mm)
loop and radio without potentially detuning it, and the prospect of
lining a closet with aluminum screen is not appealing! (I really did
consider it, though.) Fortunately, I can use CMBQ and WSTU to help test
the larger loops, with a little extra effort. Using a separate antenna
connected to the HP 312B I'll have reference signal levels. If one of
them is 2 dB stronger one day than the day before I can factor that
difference into the radio's displayed signal strengths.

I'd be happy to discuss my experiments and experimental system with
anyone, off-list. As my results vary from what others have obtained,
there has been some contentious debate, which can best be avoided by
keeping it off-line! I did want to alert anyone interested to what I'm
doing.

73,

Jim, KR1S
http://qrp.kearman.com/