In addition to what others have said, the most field you can
generate with the ferrite rod antenna will occur when it is
almost reaching saturation, and that takes a lot of ampere
turns. You can deliver more ampere turns to the rod than
your transmitter output can deliver if you resonate the coil
with a capacitor. That way, you have the current bouncing
back and forth through the capacitor added to the current
from the amplifier. If the coil-capacitor Q is, say, 100,
there will be 100 times more current through the coil than
the transmitter is delivering. This will probably take a
coil with a considerable mass of copper in it.

John, that is what I have seen! I resonated the antenna coil and driven it
with it's resonance frequency. Seems that the achievable distance was a
little more than the circuit without resonating capacitor.

You say, that driving the ferrite rod into saturation will force it to leave
more power into air? Why?

- Henry

Henry Kiefer wrote:

Hello all -

I built a simple ferrite antenna communication system. Unfortunately it
won't work if I set the sender more distanced than about a meter. That is
even true with different transmitter configurations.

Here the details:
Transmitter:
ferrite antenna: diameter 8mm , 50mm long
frequency is 77.5KHz, digital modulation is AM 25%
bit-rate is 1 bit/sec (0 is 100ms carrier 25%, 1 is 200ms carrier 25%)
insulated copper wire coil 10 turns

The transmitter is self-constructed and delivers a very good signal.

Receiver:
same antenna copied, but a built-in resonating capacitor.
ready-to-use WWVB 77.5KHz receiver. Demodulated signal goes to scope.

The transmission works over about one meter without any shortage.


Now the problem is that I can change the transmitter parameters but I
cannot reach a substancial greater distance. I changed:
- the coil wound times
- output current to the antenna (measured across a series resistor)
- added an antenna current sensor coil to sense the antenna current and to
see if the ferrite antenna saturizes (NO! Very clean sinusoid)


Googling around to find theoretical aspects of ferrite antenne got no good
results. I spent several hours and read all I can read.


Have someone suggestions to try or good links to read? Especially for:
- when a ferrite or iron powder rod/bar goes in saturation?
- optimal rod dimensions
- optimal coil design (I suggest single layer, resonating with good Q
capacitor, about 3 to 10 turns)
- LNA design for such a low frequency?
- antenna field theory in near-field.

If you need further details please ask.

Thanks in advance.

Regards -
Henry

Efficient antennas at that frequency are effectively very long
bits of wire. The ferrite rod is small compared to the wavelength
and very inefficient at generating a far field.

This is the antenna of the DCF77 transmitter (same frequency):
http://de.wikipedia.org/wiki/Bildcf77.jpg

Kind regards,

Iwo

Have someone suggestions to try or good links to read? Especially for:
- when a ferrite or iron powder rod/bar goes in saturation?
- optimal rod dimensions
- optimal coil design (I suggest single layer, resonating with good Q
capacitor, about 3 to 10 turns)

So there is a resonant circuit at the transmitter and not just a coil?


I tested it as resonating circuit using the original time-code receiver
antenna AND a second time without the capacitor.
Maybe I got a little more power in the air with the resonating circuit, but
it was not very distingiuable.

With such low number of turns (and hence low inductance), the
capacitor would have to be huge to resonate it at 77.5 kHz. Where do
you get high Q capacitors with such capacitances ?

I don't know the exact manufacturer of the time-code receiver ferrite
antenna but I comparable model reads:
L=900uH
bandwidth=700Hz
n=94
see original data
http://www.hkw-elektronik.de/pdfengl...00-77,5-DE.pdf
It is not the same antenna but very similar.

The original foil-capacitor is 682 labeled. I don't measured it but I think
it should be 6800pF reading.

For my second experiment I used no capacitor and turns=10.

If I would find a PSPICE model for an ferrite antenna ...


The resonant circuit impedance levels are quite low in this
configuration (small L/large C), how do you effectively couple power
from the transmitter to this low impedance level at the resonant
circuit ?

Hm. I thought he just trying different turns value to achieve this. The
coil is the impedance transformer for the ferrite rod (=antenna). I'm wrong
here?


The skin depth at this frequency is about 0.25 mm, so any wire thicker
than 0.5 mm will not utilise the full copper wire, so some kind of
Litz wire with separately insulated strands could be used to keep the
coil resistance low.

The original coil is thinner than 0.3mm. If I compare it to my 0.3mm wire
maybe it is 0.18mm. The second experiment with the 10 turns coil is 0.3mm
enamelled copper wire.
I will give Litz wire a try if the system as such works...


The inductance of some ferrites varies if there is some DC field
present. This inductance change could detune the resonant circuit and
drop the radiated power. Are you sure that the transmitter coil is not
carrying any DC components or some even harmonic distortion, which
would cause an unbalanced magnetic field in the ferrite rod ?

Good question. I series blocked DC with a WIMA MKS4 1.0uF 100VDC
high-quality capacitor. As measured the "big" capacitor is outside the
bandwidth of the antenna.

I don't think there is any DC component left. And yes, there is no magnet on
my desk laying around :-)
Is there any internal rectifiation phanomen in the ferrite possible?


- LNA design for such a low frequency?

The band noise is the dominant (compared to "white" amplifier) noise
when listening to the band with your transmitter switched off, the
receiver noise performance should be adequate.
How much band noise should I expect?

- Henry

I can't really help you with ferrite antennas for transmitting, but can
tell you that if you google around for "lowfer" and the Longwave Club
of America http://www.lwca.org/ you will find a lot about antenna
designs that are suitable for this band. They will also might have
recommendations for frequencies of operation that are legal for
transmission in your home country (I don't even know what that is!)

Thanks. I will look there.

I'm a little surprised that your achieved range was so small from a
ferrite rod antenna, actually. Did you really tune both antennas, in
place and in circuit, for resonance? The resonance is so so super
narrow that strays between design and circuit make a big difference. I
mean, CRT screens with flybacks, and faulty flourescent lamp ballasts,
and incadescent dimmers radiate all sorts of crap around the LF
spectrum for blocks, and they aren't even trying to be intentional
transmitters! And don't get me started about induction heaters and
welding machines, those can be heard across several states!

Maybe the time-code receiving IC is a bad design. I don't know. It's
operating current is 500uA only. That is very small. It can receive the
time-code over 2000km with such an antenna with an transmitter EIRP of 30KW.
The receiving antenna is 700Hz bandwidth. I don't think this is super
narrow. Even if we look at the time-code receiver quartz filter with a
bandwidth of about 10Hz I can met it with my stable wave generator. It is a
PLL-design with a clock quartz. Should be typical 10ppm. I don't have a very
good frequency meter to verify it.

In my second transmitter experiment I used a not-resonated driver design. So
there are no problems with detuning the transmit antenna expected. It is
just driven by the 77.5KHz power signal.

CRT screen is off if I experiment. Otherwise I seen a very big CRT signal at
the receiver...

If the two ferrite rods will detune because of the close proximity I cannot
control it. I don't think so.

If you can hear induction heaters or something this is surely with a very
big antenna and a resonable good receiver design.

- Henry

Henry Kiefer wrote:
and that takes a lot of ampere
turns. You can deliver more ampere turns to the rod than
your transmitter output can deliver if you resonate the coil
with a capacitor. That way, you have the current bouncing
back and forth through the capacitor added to the current
from the amplifier. If the coil-capacitor Q is, say, 100,
there will be 100 times more current through the coil than
the transmitter is delivering. This will probably take a
coil with a considerable mass of copper in it.

John, that is what I have seen! I resonated the antenna coil and driven it
with it's resonance frequency. Seems that the achievable distance was a
little more than the circuit without resonating capacitor.

You say, that driving the ferrite rod into saturation will force it to leave
more power into air? Why?

You misunderstood what I said. It was, " the most field you can
generate with the ferrite rod antenna will occur when it is
almost reaching saturation,"

If you saturate the rod, the field you generate will have lotsof 3rd
harmonic components in it, but little more of the fundamental. I was
trying to emphasize that you will need as strong a magnitic field as
possible aat the transmitting antenna, and just below saturation is
that limit, when a ferrite core is involved.

If the rod has a large lenght to diameter ratio (say , above 10) then I
think the uptimum coil arrangement on the rod also doffers considerably
for the transmitting and receiving cases, since the receiving case does
not deal with saturation.

In the receiving case, the end sections of the rod act as flux
collectors, and only the middle thirs or so has almost all the
collected flux passing through it, so this third is the optimum place
for the coil. /in the transmitting case, the rod has a tendency to
saturate at the center, first, with this arrangement, and you want
essentially the whole rod to approach satuation at the same ampere
turns. This will produce a field that acts as if it has been produced
by the full length of the rod. You can achieve something close ot this
by spreading the turns out, all over the rod, with an extra
concentration (a second or third layer layer, perhaps) at the ends.
Something like this (shown in cross section. View with fixed width
font i.e. Courier, so charcters are on grid pattern):

* = wire in cross section
# = rod

*** ***
****** ******
************************
##########################
************************
****** ******
*** ***

Efficient antennas at that frequency are effectively very long
bits of wire. The ferrite rod is small compared to the wavelength
and very inefficient at generating a far field.

This is the antenna of the DCF77 transmitter (same frequency):
http://de.wikipedia.org/wiki/Bildcf77.jpg

OK Iwo. But why a small receiving ferrite antenna works here? A
non-saturated (=linear, and that means the superposition theorem works)
antenna system is reciprocal as antenna theory predicts. So you should
explain where the difference is!

- Henry

If you saturate the rod, the field you generate will have lotsof 3rd
harmonic components in it, but little more of the fundamental. I was
trying to emphasize that you will need as strong a magnitic field as
possible aat the transmitting antenna, and just below saturation is
that limit, when a ferrite core is involved.

I understand that. I added a second coil on the ferrite rod to measure the
antenna current and set it just below the point where I saw harmonics (or
say non-sinusial) waveform on the scope.


If the rod has a large lenght to diameter ratio (say , above 10) then I
think the uptimum coil arrangement on the rod also doffers considerably
for the transmitting and receiving cases, since the receiving case does
not deal with saturation.

In the receiving case, the end sections of the rod act as flux
collectors, and only the middle thirs or so has almost all the
collected flux passing through it, so this third is the optimum place
for the coil. /in the transmitting case, the rod has a tendency to
saturate at the center, first, with this arrangement, and you want
essentially the whole rod to approach satuation at the same ampere
turns. This will produce a field that acts as if it has been produced
by the full length of the rod. You can achieve something close ot this
by spreading the turns out, all over the rod, with an extra
concentration (a second or third layer layer, perhaps) at the ends.
Something like this (shown in cross section. View with fixed width
font i.e. Courier, so charcters are on grid pattern):

That is a very interesting configuration. Never seen such a design. I read
about a old-fashion remote controller system having a ferrite antenna
transmitter. There someone wrote, the transmitter antenna was a mignon
battery-shaped ferrite rod. e.g. much shorter but wider than mine. So an
optimum ferrite transmitter antenna is maybe more like a fat battery shaped.

- Henry

On Wed, 25 Oct 2006 17:57:21 +0200, "Henry Kiefer"
wrote:


Yes, a ferrite stick antenna works quite well for receivers, but not for
transmitters. Try winding a few dozen turns around the whole room - i.e.,
up the wall, across the ceiling, down the other wall, across the floor,
and so on. Or, you could wrap a piece of 50-conductor ribbon cable, and
make loops by soldering the ends together offset by 1. ;-)

You made my day
BTW: Your idea with the ribbon cable gives you a very easy made
transformator if using clamping connectors. This works very good. I
practiced it 10 years ago.

The problem with loop antennas made of ribbon cable (or other
multiconductor cable connected this way) is the stray capacitance
between turns. The self resonance frequency (without external
capacitor) may be below the band of interest, so you can not resonate
such antenna with an external capacitor.

Paul OH3LWR

The problem with loop antennas made of ribbon cable (or other
multiconductor cable connected this way) is the stray capacitance
between turns. The self resonance frequency (without external
capacitor) may be below the band of interest, so you can not resonate
such antenna with an external capacitor.

I found it very useful for design of hard-driven Power MOSFET driver
transformator till 400KHz without problems.

- Henry

Henry Kiefer wrote:
Efficient antennas at that frequency are effectively very long
bits of wire. The ferrite rod is small compared to the wavelength
and very inefficient at generating a far field.

This is the antenna of the DCF77 transmitter (same frequency):
http://de.wikipedia.org/wiki/Bildcf77.jpg

OK Iwo. But why a small receiving ferrite antenna works here? A
non-saturated (=linear, and that means the superposition theorem works)
antenna system is reciprocal as antenna theory predicts. So you should
explain where the difference is!

At LF, rarely is the problem "not enough amplitude of received signal",
so past a certain point there isn't much need to make the receiver
antenna more efficient. The problem is always "too much noise!". So
antenna designs are usually built around nulling out local noise, and
loop antennas will get rid of the mostly electric-field local noise.
And they have some directionality (notably sharp nulls) which can help
get rid of specific further-away noise.

Tim.