Welcome to Issue #8 of the Radio-Sky Journal
Copyright 2003 by Radio-Sky Publishing
All rights reserved.
Wading through Integration
Amateur Tip #10
Wading through Integration
I try to make it to the beach at least once per week. One of my favorite
beaches requires about a twenty minute walk along the coastline to get there.
At one point along the trail, you need to wade through water when tide is high.
This little wading area is in a protected inlet. A natural circle of rock surrounds
this wide body of water, broken only by a narrow channel on the side farthest
from the shore. You can look out and see the waves crashing into this channel
but at the wading spot there is very little wave to contend with. Still, at the
wading area the effect of the tide is very evident; low tide and you stay dry,
high tide and you get wet all the way up to level which the moon dictates.
After arriving at the beach you note that situation is much different. There is
no protective wall of rock to limit the movement of the waves from open ocean
to the beach. The waves come crashing onto the shore unobstructed. As a
result, it is very difficult to determine from watching the interaction of the
ocean with the beach whether or not it is high tide or low. The waves mask
the average level of the ocean. When I start my hike home, I never know
whether or not I will have to again wade until I reach the inlet.
In case you wondering if you had signed up for the wrong newsletter,
of course, this all applies directly to radio astronomy. All of the radio signals
we can expect to receive from the cosmos fluctuate. Like the ocean, radio
waves are a complex mixture of little waves riding on big waves. Suppose
you want to detect the relatively faint signals coming from a celestial object.
The radio waves from the object can be much smaller than the other waves
from the region of space to which your antenna is pointed. In addition to
this unwanted noise (these waves) from space is the noise from the electrons
bouncing around in the electronics of your receiver. These local wave
contributions to your total received signal are further modulated by the
effects of temperature changes. It might all seem like a hopeless cause to
try and figure out what part of this roaring ocean (total signal) is the tide
The simplest detector is a diode. By only allowing electrons to pass in
one direction, the diode can produce an output voltage that corresponds
to the AC signal level on its input. We won't get into the specifics of this
correspondence this time, but generally, if the radio frequency AC signal level
is stronger on input of this detector, the voltage seen at the output is higher,
as you would expect. However, if you looked at the output of the detector
with a fast oscilloscope you would see a field of spikey waves that is
just as complex as the as the RF signal at the input. This corresponds in my
analogy to the waves breaking into the narrow channel of the inlet.
The body of the inlet is a pool of water surrounded by rocky walls. Under
ordinary circumstances this pool of water cannot be filled instantly by the
ocean. It takes time for the waves to push through the narrow inlet. These
waves are meeting with "resistance" at the opening to the inlet. By the same
token, when the ocean outside of the inlet begins to lower as the tide ebbs,
the pool cannot instantly lower to the new average level of the ocean.
In our radiotelescope, we could really use a device to simulate this behavior.
Two simple components, a resistor and a capacitor can do this for us quite
nicely. Together they form an "integrator". The resistor corresponds to the
narrow channel at the mouth of the inlet, preventing instant equalization of
the voltage (water level) inside the capacitor with the voltage we are feeding
it. The capacitor, corresponds to the pool that is the body of the inlet.
Electrons can build up within the capacitor the way water builds in pool. By
looking at the total charge that builds within the capacitor, we can see the
average level of the raging sea of waves passing through our detector.
We can take this just a little further. Remember I said that at the wading spot
"there is very little wave to contend with"? By narrowing the mouth of the inlet
or by increasing the size of the inlet pool we could theoretically reduce these
tiny little waves to almost zero. If, however, we make the pool too large and/or
the channel too narrow, we would find that our tide measurement becomes
too smeared out to be useful. The pool level cannot keep up with the change
of the tides.
The same is true of our radiotelescope integrator. We ideally want to set it's
level of integration so that it can keep up with the expected change we hope
to see from the celestial source passing through our antenna beam. Luckily,
this is not hard to calculate. We simply multiply the value (in Ohms) of our
resistor times the capacitance of our capacitor (in Farads) to achieve a value
called the time constant. Suppose we want a time constant (T) of 5 seconds
and that we have a nice stable capacitor of 8 microfarads that we want to use
to achieve this. What value of resistor would we use in our integrator?
T(time constant) = R(ohms) * C(Farads)
5 = R * 8*10^ -6 [ a microfarad is a millionth of a Farad ]
5 = R * .000008
R = 5/.000008
R = 625000 ohms
If we use an integrator with this time constant we can expect that fluctuations
in our radiotelescope output that occur faster than 5 seconds will be suppressed.
Slower fluctuations will not be suppressed significantly. There is a danger that
if we use too high a time constant we can see a delayed response to the move-
ment of the object through the beam. Precise measurement of a transit is not
usually a task for amateur radio telescopes, however.
Note that integration does something more than suppress unwanted noise in
your observation, relatedly, it increases the sensitivity of the observation. In
fact, it increases the sensitivity proportionally to the square root of the time
constant , SQR(T). Said a slightly different way, the random fluctuations seen
at the output if the radiotelescope will be proportional to 1/(SQR(T).
Another way to increase sensitivity is to average multiple observations. This is
helpful because as I expressed above, there is a practical limit to how large
we can make our time constant for a given type of observation. To do this we
take values measured at corresponding times during the multiple observations
and add them together then divide them by the number of observations. If you
do this remember to re-align the observations in sidereal time when it is
appropriate, that is, when the observed object is very distant and transits
according to a sidereal clock.
Next Issue: A shark just bit my preamp.... (Just Kidding)
For those of you who have had problems building or getting your MAX186
eight channel ADC to work, I have added a page to the Radio-SkyPipe help
pages . See:
One thing you will see in there is a reference to what kind of signal you
are feeding to the ADC. I want to reiterate that you cannot feed an audio
signal directly to the ADC from a radio and get a meaningful reading. ( At
least not using Radio-SkyPipe software. ) The audio signal must first be
detected and passed through an integrator to produce readable results. There
are simple suggestions for how to do this.
For people who do not want to build their own detector/integrator,
RF Associates has a new product that you can get through Radio-Sky
that solves this problem . The RF2040 is the standard MAX186 ADC with two
channels dedicated to audio detectors. These are sensitive independent op
amp based detectors with selectable time constants and gains. They present
high impedance inputs to your radio audio line and the added gain means that
you can make full use of the range of the ADC. The remaining 6 ADC input
channels can be used for standard DC measurements. Beginning with
SkyPipe 1.2.10 you can remap the MAX186 ADC channels to any of the chart
channels. Thus, you could use one detected sound channel and two DC
inputs or any other combination you like.
Radio Astronomy Projects 2nd Edition
I am pleased to announce that the revised and expanded 2nd Edition of
William Lonc's popular book is available on our website. This book takes
a non-mathematical approach that is suitable for the beginner of
experienced amateur radio astronomy. Most of the projects are in the
VHF through microwave realm and use either off-the-shelf or surplus
equipment. Experiments accompany the projects, imparting a solid
understanding of the basic principles of radio astronomy. This edition has
almost 50 additional pages and a helpful index. The price is $24.00
Radio-SkyPipe Version 1.2.12 Available
A completely new way of mapping data channels is introduced in this
version of Radio-SkyPipe. Pro version users will find an awesome new
feature at their disposal, user defined equations can be applied to
live data or data files. You can even create channels that show the
results of equations applied to other channels. For example, you can
multiply the values of channel one and channel two and display the
results in channel three. Equations are not restricted to simple arithmetic.
You can use log and complex trig functions. To download go to:
HTML Radio-Sky Journal?
I am considering issuing this newsletter in HTML format. As it seems I am
one of the few people left on Earth with a dial up connection, the reason for
a text only newsletter seems to be diminishing. I have concerns that an
HTML newsletter might be filtered out by spam filters. I would really appreciate
your feed back on this subject.
Amateur Tip #10
Sealing your outside connectors against the weather is good fundamental
practice. There are a number of ways to do it. A product call Coax-Seal has
been available for many years. Unfortunately, the versions of Coax-Seal
that I have worked with form a sticky mess making it unsuitable for temporary
connections. Hardware stores are now carrying products with names like
"Liquid Electrical Tape" or "Liquid Rubber" which can be painted on to provide
good protection. 3M makes a Insulating Spray Sealer recommended for
If you are like me, and are very often changing what is on the end of the coax
(new antennas, preamps, etc.), you may find these products a bit too permanent.
You can keep temporary connections dry for short periods by using a
small housing. For semi-permanent installations I build a open bottomed box
for the preamp and its connections from aluminum flashing. This is a very easy
material to work with using a couple of pieces of wood as a bending jig. Pop
rivets secure the box. Of course, you wouldn't use a metal housing if
it needed to be placed in an area that would affect the antenna radiation
pattern. For that I have used margarine tubs and even 2 liter plastic soda
bottles. If the test is going to last more than a couple of weeks
you should also wrap the connections tightly with electrical tape.
I want to re-iterate that electrical tape is not a good solution as a sealer
for a permanent installation. for a couple of months of an experimental
setup it seems to work fine.
Look what Jim Hardy has done with some creative plumbing...