[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Plane wave antenna making



Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

Hi Charlie,

There are certainly problems with extending the chart "too" far.  As the 
distance is increased, the effects of the floor, ceiling, and walls begin 
to distort the fields:

http://hot-streamer-dot-com/andrewb/models/models.htm

I have a bit of an advantage over what Niels Jonassen talks of in that the 
source voltage target is relatively low impedance and it is AC.  The 
antenna sitting say 15 feet away is not going to "load" the coil (voltage 
source).  And unlike a DC charge, The AC voltage is measured with an 
effective capacitive divider.  Once calibrated in a given position, the 
field profiles do not really matter.  Streamers may certainly dynamically 
distort the field so it is best to fire them off the far side away from the 
antenna.  I extract real AC current off the system so there are no 
"delicate charges" to worry about as in the case of measuring static DC 
charges.  The antenna's back plane not only fixes ground for the probes 
impedance, but it also tends to keep the voltage profile around the probe 
constant reducing the effects of local E-field probe distortion (basically, 
keeps it a constant).  I really don't care how things get distorted as long 
as they are distorted the same way during calibration.

My antenna's real 'claim to fame' is that it can pick up the fields over a 
wide frequency band (but not DC) and deliver them to the scope with flat 
frequency and phase response.  Thus, a signal's low and high frequency 
components are recorded equally and without distortion.

The problem with say a wire antenna is that it has a variable gain 
depending on the frequency.  That makes voltage measurements hard since 
Tesla coils may have a pretty messy spectrum of frequencies in the 
signal.   If the probe has all kinds of resonances and a messy frequency 
response (and a messy phase response), the signals it gives are no good for 
voltage measurements.  You can determine quenching and other rough things, 
but not any accurate voltages.

E-Tesla6 can plot the field contours and give a pretty close estimation of 
the field attenuation at a given distance.  But in a really messy Tesla 
coiling lab (mine ;-)) all the junk in the room violates E-Tesla's 
assumption of a perfect world in which the coil operates.  So, I find it a 
bit better to calibrate the antenna with real voltages in place.

I guess DC Cox has a 270kV power supply (everyone needs one of those ;-)) 
so he could calibrate the antenna very accurately in it's actual position 
far off in the distance.  No need to get close to measure a "small" high 
voltage signal when you can make "big" high voltage signals that the 
antenna can pickup well at a far distance.  The probe does go down to 60Hz, 
so even an NST could calibrate it.  Really, the probe is more of a 
capacitive divider than an "antenna".  Thus, it can dive into very low 
frequencies were a real antenna would fail.  But few real antennas are 
meant for broad band use...

An "official" E-field antenna may do the same thing too.  But I "think" 
they are calibrated at a series of specific frequencies where they are 
"very" accurate for regulatory standards testing.  In our case, super 
accuracy is not as important and good wideband response.  I do know my 
antenna is a lot cheaper that its TUV certified cousins ;-))  Come to think 
of it, for as much money as they put into it, those TUV setups should be 
able to do the same thing...  I walk by it all the time, should stop and 
ask ;-))  I bet they can't go down to low frequencies though...

I should say that the plane antenna's function and theory is pretty 
trivial.  It is just a capacitive divider with a large (in this case 10nF 
is large) base capacitance that is impedance matched to the cable and 
scope.  Physically, it is laid out well to work with "our stuff".  But 
there is no highbrow science involved.

I should say that I actually rarely use the antenna for absolute voltage 
measurements.  Usually I am just trying to "see" the signals without 
concern for the absolute voltage.  I don't care "what" the absolute voltage 
is, I only care how to make it "more" ;-))

Cheers,

         Terry




At 07:16 PM 6/2/2003 -0400, you wrote:
>Hi Terry,
>
>I don't think you can assume that the voltage falls off linearly with
>distance.  Take a look at this paper:
>www.ce-mag-dot-com/archive/01/11/mrstatic.html
>
>Charlie
>
>Tesla list wrote:
> >
> > Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>
> >
> > Hi,
> >
> > There are a lot of ways.  I usually just charge the coil to a known high
> > voltage and short it to give a known voltage step.  The voltage is very
> > proportional to distance.
> >
> > Relative distance   Relative voltage
> > 1                       1
> > 2                       1/2
> > 4                       1/4
> > 8                       1/8
> > 16                      1/16
> >
> > So you can use a smaller voltage close and then back off a known distance
> > for a known attenuation.  A lot depends on how accurate one cares to
> > be.  At a far calibration distance (~5 feet) and a high calibration voltage
> > (50.0kV) it can be extremely accurate.  I prefer using a high voltage
> > supply and such as opposed to "figuring it out" on paper (computer) since I
> > have the stuff.  It is easier to calibrate it with known high voltages and
> > I don't have to worry with ceiling, floor, and other room effects messing
> > with the fields.  E-Tesla-6 is pretty good at doing that, but the room has
> > to be pretty big without much nearby stuff.
> >
> > In many cases, I don't care what the voltage is, I just need to know what
> > the waveform looks like so I don't worry about calibration.
> >
> > Cheers,
> >
> >          Terry
> >
> > At 10:05 AM 6/2/2003 -0400, you wrote:
> > >Terry,
> > >
> > >How do you calibrate your voltage measurements?
> > >
> > >Charlie
> > >
> > >Tesla list wrote:
> > > >
> > > > Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>
> > > >
> > > > Hi All,
> > > >
> > > > I am going to loan my plane antenna to DC for awhile:
> > > >
> > > > http://hot-streamer-dot-com/TeslaCoils/MyPapers/planant/waveant3.html
> > > >
> > > > The most obnoxious part about making these is the etching of the grid
> > > pattern:
> > > >
> > > > http://hot-streamer-dot-com/TeslaCoils/MyPapers/planant/Image133.jpg
> > > >
> > > > http://hot-streamer-dot-com/TeslaCoils/MyPapers/planant/Image136.jpg
> > > >
> > > > So I thought this would be a good time to try it with a solid plane
> > > > instead.  So I whipped one up:
> > > >
> > > > http://hot-streamer-dot-com/temp/030601-01.jpg
> > > >
> > > > it seems to work fine:
> > > >
> > > > http://hot-streamer-dot-com/temp/030601-02.gif
> > > >
> > > > The wave form is still very true without any problems.  I was 
> originally
> > > > worried about eddy currents and such floating around in a solid 
> plane, but
> > > > it does not seem to mater.  One "could" put a few simple slits in 
> it too I
> > > > guess...
> > > >
> > > > In this case, I am just driving a toroid with a signal generator 
> and the
> > > > attenuation is 5000:1 at 5 inches.  Of course, for a real operating 
> coil
> > > > that distance might be 15 to 20 feet so the attenuation might be 
> more like
> > > > 100000:1.
> > > >
> > > > So if one wants to make one of these and you can't etch the nice 
> pattern
> > > > easily, just leave it solid.
> > > >
> > > > I also fine I hardly ever use the current sensing side for 
> anything.  Just
> > > > making the simple voltage side probably covers 99% of the uses.  It 
> is also
> > > > best to use very small leaded resistors and capacitors for the
> > > > antenna.  Trying to solder those delicate surface mount components 
> to try
> > > > and squeeze out better performance is a real pain and probably not 
> needed.
> > > >
> > > > Cheers,
> > > >
> > > >         Terry