Apparently fluorescent tubes continuously emit a lot of other microwave signals.  I once built a  homodyne doppler "speed" radar kit (used a coffee can for the antenna).  The way you calibrated it was to point it at a florescent tube and and adjust the reading to a specific value.

On 4/11/17 11:09 AM, Mark Sims wrote:

That's not because the tube is emitting..  It's a target reflector
turning on and off at twice line frequency.
In most homodyne radars, you filter out the DC (the reflections from
stuff that's not moving), so anything that pulses on and off creates
nice output.

At 10.525 GHz, the Doppler is about 70 Hz/ (m/sec), 31 Hz/(mi/hr)

at 24GHz, 160 and 71 Hz, respectively

Most simple speed guns just have an audio frequency counter on the
output of the mixer diode(s).  Older units use a Gunn oscillator, newer
ones use a DRO.  Some have a pair of detectors so you can distinguish
motion towards and away.

The old "calibrate with a tuning fork" for police radar wasn't
calibrating the RF frequency (a 1000 ppm change of the gunn oscillator
isn't a big deal.. this is a "3%" kind of measurement) - it was
calibrating the audio frequency counter, which, in very early units,
used an RC timebase. (or an actual analog meter reading) I cannot
imagine a crystal oscillator bad enough that a tuning fork  would be
better.  - if XO based speed guns were checked with a tuning fork it's
for one of two reasons:

1. the purchasing requirement said "A tuning fork for calibration shall
   be provided" (based on an older design)
2. it provides a "functional test" and you don't really care what the
   frequency is, as long as it lights up anything reasonable

Homodyne/Doppler radars are fun
(http://home.earthlink.net/~w6rmk/radar10g.htm) and can form the basis
of a business that saves lives
(https://www.nasa.gov/jpl/finder-search-and-rescue-technology-helped-save-lives-in-nepal)

I investigated police radar stuff a long time ago, and for a while had
an old X band unit shaped just like a searchlight, with analog meter.
What I learned then was that even on the newer units, the tuning fork
was specified to provide an independent means to verify the accuracy of
the unit in the field.

I also remember hearing that the radar sales reps had a special gift
they would hand out to friends: a tuning fork marked "60 MPH" but that
rang at a lower frequency, so you could use it to "calibrate" the unit
to read higher than actual.  I heard that it was popular among the
speed-trap crowd.

On 04/12/2017 09:25 AM, jimlux wrote:

Some incandescent lamps can emit RF.

http://www.rexophone.com/?p=1081
http://www.radiomuseum.org/forum/rustika_lightbulb_fm_measurements.html

On Tue, 11 Apr 2017 18:09:52 +0000, you wrote:

Hi

All incandescent lamps emit RF ….. They are a resistive device that is heated to well above
room temperature. People do use them in simple noise figure meters. The inductance of the
filament in a typical bulb restricts the bandwidth a bit. The are designs from at least the 1960’s
running around. I suspect the approach is much older and that’s just when I was introduced to
the technique (= built one).

Bob

Many decades ago, QST had an article about the
"Monode" RF noise generator.  No it wasn't an
April Fools joke; the Monode is simply a light
bulb.  You can probably download the article
from the arrl.org web site.

HP used to sell a fluorescent tube embedded in
a waveguide as a noise source.

These days, noise comm has a much easier
solution off the shelf.

Rick N6RK

On 4/12/2017 1:23 PM, Bob kb8tq wrote:

On 4/12/17 1:23 PM, Bob kb8tq wrote:

The advantage of the diode noise source (a typical one is 7000K, a bit
higher than a 3200K incandescent bulb) is that it's more stable - the
temperature of a lightbulb filament depends on a lot of things - and
that it can be "hotter"

A  15 dB ENR noise source is something like 9500K  - that's a lot hotter
than something incandescent, and well above the 3700K melting point of
tungsten

Hi

Back in the day (1960’s) noise diodes were quite a bit more expensive than they are today. Even
today, a miniature lightbulb is quite a bit less expensive than a noise diode. You don’t go that way
because it’s better. You go that way because it’s cheap ….

Bob

On 4/12/17 6:33 PM, Bob kb8tq wrote:

yeah, but I'd think something like a NE2 might be better, but maybe the
noise properties of a glow discharge aren't as good.

Hi

You likely would get more noise density out of the NE-2 than the incandescent bulb. The
advantage of the incandescent is that you can estimate the temperature fairly accurately
and thus come up with the noise by calculation. We did the build as part of a radio club
meeting back in the 60’s. I was still in high school and will not in any way claim I followed
the zigs and zags of the physics between the various candidate sources as that was all presented.
My only focus was on “what goes where to make it work” :)

Bob