Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] CMB and TV



On 03/28/2012 09:40 AM, Anthony Lapinski wrote:
I'm teaching about cosmology soon in my astronomy class. I know that the
CMB (cosmic microwave background) is about 2.7 K. Using Wien's Law, this
radiation is most intense around 1 mm. This corresponds to an EM frequency
of about 300 GHz. I once heard that some of the snow/static in between TV
channels is made up of the CMB. How much? Does anyone know what percent
this would be? I figure UHF/VHF channels are around 300 MHz, so maybe
divide to get 0.001 = 0.1%. Does this make sense?

Well, suppose you were named Penzias or Wilson and you wanted
to build something that was capable of observing the background
radiation. Then:
a) You need an antenna sufficiently directional that you can aim
it at the cosmos, not at the earth. If there is even a tiny side
lobe aimed at the earth, the noise power you get the side lobe
would overwhelm the cosmic background radiation. The earth is
100 times hotter, and at frequencies well below the peak of the
black-body spectrum, the power per unit bandwidth goes like T.

b) You need a low-noise amplifier. This is similar to item (a),
in the sense that the amplifier needs to "see" the signal
coming in from the antenna and not "see" the Johnson noise in
any resistors internal to the amplifier.

c) You need a way to /measure/ the noise level, sensitivity, and
bandwidth of the receiver.

Let's apply these ideas to your TV receiver, in reverse order.

Starting with item (c), suppose you take a piece of coax cable and
terminate it in a 50 ohm resistor. You use that as the input to your
TV. You observe nothing but snow on all channels. So far so good.
Now stick the resistor into a bucket of liquid nitrogen. This
reduces the noise from the resistor by a factor of 4. Do you
the snow pattern to change by a factor of 4? I don't expect it
to change much at all. That's because the noise temperature of
the TV receiver (item b) is way too high. The noise coming in
via the input cable adds in quadrature to the endogenous noise of
the receiver.

Penzias and Wilson used exactly this technique -- called a "cold
load" -- to check the noise temperature of their electronics ...
except that they soaked the resistor in liquid helium instead of
nitrogen, and they went to a lot of trouble to build a maser
amplifier that was so good that its noise did /not/ dominate
the noise coming from the cold load.

There would be no reason to waste money lowering the noise
temperature of the TV receiver, because the signal it is
looking for is already dominated by terrestrial noise
sources at or above 300 K.

Now we come to item (a). I'm pretty sure the rabbit ears on
your TV are not aimed at the cosmos, with no side-lobes that
see thermal radiation coming from the earth. If it were aimed
at the cosmos, it wouldn't work as a TV receiver. Similarly
I doubt that the unused channels in the cable-TV system are
aimed at the cosmos.

Does anyone know what percent
this would be?

Much less than 1% for rabbit ears, and 0% for cable TV.

Neither the antenna nor the electronics is anywhere near
suitable for observing the background radiation.