On 10/28/2014 9:53 PM, Paul wrote:
rickman wrote:
On 10/28/2014 6:14 PM, Ralph Mowery wrote:
"rickman" wrote in message
...
I have a project in mind that would need a very good antenna in the
frequency range of 60 kHz. Originally I looked at loop antennas and
liked
the idea of a large shielded loop made of coax tuned with a
capacitor. My
goal is to get as large a signal as possible from the antenna and
matching
circuit to allow the use of a receiver with very low sensitivity...
in fact
an all digital receiver.
I spent some time simulating antennas in spice and was able to get a
bit
of a feel for the circuit, but I'm not convinced it would work the
way I
want. Just before I set the project aside I was told I needed to model
the radiation resistance. That has the potential of wrecking the Q
of the
circuit. I am counting on the high Q to boost the output voltage.
If the
radiation resistance is at all appreciable I would lose the high Q and
need to start over.
I don't think I would try and reinvent that type of antenna. There are
several designs on the web that use a loop about 3 feet in diameter and
several turns of wire inside the shield. In most cases a low noise
preamp
is needed, but that shold be simpleand inexpensive to build.
Go to this page and go toward the bottom for some loop antenna ideas.
http://www.w4dex.com/lf.htm
I have known Dexter for around 40 years.
I am not sure what you mean by "reinvent" that type of antenna. Every
antenna can be optimized for a given design. My requirements are very
unique. I need as much voltage from the antenna as possible. My
receiver input impedance can be very high (~1 Mohm) which is very
different from a typical receiver.
I have already gone down the road of looking extensively at loop
antenna designs. I have not found a significant difference other than
the ease of construction. That is one reason why I chose to use coax
rather than wire within a shield like pipe or a bicycle rim (as I
found in one project).
My current design is 100 feet (the 50 feet I said originally was due
to my poor recollection) wound on a 2 foot diameter spoke arrangement
of wood which turned out pretty well for a first pass. I have yet to
characterize the antenna which may be the easier path than trying to
construct a good model from theory and the known details.
Several people have suggested that a preamp will be required. That
may be possible. But this is not an analog receiver and don't need a
lot of SNR for it to work. The time code signal is modulated at 1 bps
using both phase and amplitude modulation and pulse width bit
encoding. I will need a resolution of no worse than 100 milliseconds
to decode the bits. So I figure a bandwidth of 10 Hz should be plenty
enough. This means I can vastly over sample the signal and get lots
of gain digitally.
So the tricky part is to overcome the poor analog characteristics of
the differential digital input. I only need it to turn the input
signal into a one or a zero, but it needs to be sensitive to a very
small signal. With the various imperfections of input offset,
hysteresis, etc., I will be lucky if it works with very low voltage
signals at all. I could rig up a test circuit and see just what
signal levels are needed.
The other part is that the purpose of this design is to receive the
signal digitally on as low a power level as possible. The entire
power budget is a couple hundred microwatts. I have yet to find an
amplifier that will fit this power budget. Oddly enough some folks in
s.e.d told me that transistors don't work well with low bias currents,
but that may only apply to bipolar amps. They make time code receiver
chips to do this on a few hundred microwatts and have an internal
amplifier. So obviously it can be done. I just can't find a low
enough power opamp for a 60 kHz signal.
Also this a learning exercise for me. So reinventing something would
be ideal!
For commercial designs, I keep seeing references to a
ferrite core with a winding on it, as an antenna.
Yes, a ferrite antenna is commonly used because of it's small size. But
when I crunched the numbers a larger loop produces a larger output
voltage than did the small loop of a ferrite antenna. The ferrite only
increases the output by the relative permeability, a constant of the
ferrite material that is relatively small compared to the gain of a
larger loop which goes by the the area of the loop proportional to the
square of the radius/circumference or for a constant length of wire is
inversely proportional to the number of turns. In other words you can
do more by making your loop larger than you can by using a ferrite
core... assuming you are not restricted to your loop size.
The length of the antenna wire is important because it determines much
of your losses and so the Q. The Q of the antenna is the ratio of the
total loss resistance to the inductive reactance. Since the Q depends
on the inductance things get complex.
L ∝ N^2 * A where N is the number of turns and A is the loop area
The output voltage of the tuned antenna circuit is the product of the
effective height, Q and field strength or
V ∝ he * Q
since the field strength is constant.
Effective height is the number of turns times the area divided by the
wavelength.
he ∝ N * A
since the wavelength is constant. This gives
V ∝ N^3 * A^2 / Rloss
Looses are from wire resistance with skin effect and radiation
resistance. Assuming Rloss is mostly from the resistance of the wire
with skin effect which will be related to the wire length we can hold
that constant and look at V as a function of the tradeoff between A and N.
N ∝ 1/r and A ∝ r^2. So replacing both N and A we have
V ∝ (1/r)^3 * r^4 or r, so a larger radius gives the strongest signal
everything else being equal. While the permittivity may affect the
signal from the antenna, the typical ferrite antenna is many small if
not tiny loops while fewer, larger loops without a ferrite should give a
stronger signal.
I think this is the first time I have done this all as one line of
thought, so I may have made a mistake somewhere. But I'm pretty sure
the result is correct. It may be mitigated by the small gauge of the
wire normally used for ferrite coils allowing more turns to be used.
But again, that same wire can be used with a larger loop size even if it
does lower the Q.
More interesting is the impact of wire diameter on the whole thing. The
RG-6 wire I chose is about optimal regarding the conductor diameter with
the skin affect making anything larger not of much value. Of course the
fact that it is coax makes it a lot larger when using lots of turns.
This page has a very good drawing of the circuit showing all the
elements about a quarter of the way down the page.
http://sidstation.loudet.org/antenna-theory-en.xhtml
The article here, describes two kinds of receivers. One
is sensitive to AC pickup, so would only be a candidate
in special physical circumstances. The other uses the
high impedance input.
http://home.pon.net/785/equipment/build_your_own.htm
It suggests to me at least, you want plenty of gain
on the input stage, plus enough filtering to reject
louder noise sources. Your digital processing section
can provide the selectivity. But if spurious out of
band signals saturate your gain stage, you might not
get the desired result.
It would all depend on the tradeoffs you want to make.
You'll always require a gain stage.
I'm not sure what you mean by AC pickup, I guess you mean stray power
line signal? The E field receiver is pretty much what I don't want.
The antenna picks up very little signal because of the small physical
size while being very large. The E field is allegedly the source of a
lot of near field interference from appliances. The (again alleged)
advantage of the magnetic antenna is that the shield blocks the E field
and reduces many interference sources. I say alleged because I have not
seen much verifiable info on this and at least one source I found (and
have since lost) disputed the claim of reduced interference by the shield.
The only thing I found of value from this link was the emphasis on low
pass filters, which in my case will be band pass filters, first in the
antenna itself and then in the receiver.
Perhaps the antenna of your choice (not your final design)
and a spectrum analyser that works in that range of
frequencies, you can do a survey to see what is possible.
What noise sources are immediately evident, and so on.
No big antenna here. The antenna is one of these.
http://www.maplin.co.uk/p/ferrite-rod-aerial-lb12n
http://www.burningimage.net/clock/20...0khz-receiver/
I think by "sensitive" what they meant was "it picked
up the signal I wanted". The circuit diagram would
have been labeled "insensitive" if no signal was
found. Or if it didn't oscillate at 60KHz on its
own (like a couple amplifiers to drive speakers
have done here) :-) I think some audio circuit
I built, checking with a scope later on, indicated
a nice fat signal at 500KHz. Great.
Perhaps using your big loop of wire, you get to
remove one of the op-amps.
*******
The circuit above uses TL-081, with gain bandwidth product
of 3MHz. So I guess that's why there is still a bit of gain
at 72KHz.
In school, were were shown an example of a filter that
used only resistors. An example is seen on Fig 2.27(c)
on PDF page 70. The neat thing about this topology, is it
was working at 50KHz on a pair of $0.25 opamps. It uses the pole
of the output stage of the opamp, as a filter element. We
had some afternoon lab to do, with this circuit as part
of the work.
http://www.springer.com/cda/content/...022-p174507347
9780817683573-c1.pdf 3,791,230 bytes
The book table of contents is here. It's by Mohan, P.V.A.
With ISBN 978-0-8176-8357-3. I was hoping the topology
had a name, but I don't see one.
http://www.springer.com/cda/content/...069-p174507347
So the circuit could be in range of some opamps. And then
you might not need a huge antenna.
Thanks for your suggestions. My purpose in building this is not to
receive the WWVB signal. If it were I would just buy one of the small
kits that do it with two chips and a ferrite antenna. My purpose is to
receive the WWVB signal with a digital receiver that is close to the
power consumption of the analog receiver.
--
Rick