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

Bang the rocks together harder lumpophiles

Hi to  all the slow wave helical resonator (SWHR) believers and bang the
rocks harder to every one else.

-----Original Message-----
Subject: Another obnoxious lumped theory supporting post ;-))

>Original Poster: Terry Fritz <twftesla@uswest.net>
>

No Terry this is not a rough time its  basic circuit theory  but it is time
consuming. I will give it one more posting on the assumption that your
having genuine difficulty understanding the concepts involved and you
need help with the experiments.

I assume your coax experiment is an attempt to disprove my assertion that
there is zero phase in a standing wave by constructing 1/4 resonator with
the coax and performing a test. OK . You dont have the correct set up. You
have
the driven end terminated in 100ohm. Thats more like an open circuit than a
short. Connect say a 4.7ohm or 1ohm resistor across the input and make sure
YOU HAVE A 1/4 WAVELENGTH.  Cut the cable in half and reconnect it so you
have access to the central conductor. Repeat the experiment and check the
phase at the junction. It will be inphase with the end just a lower
amplitude.  Half way down the cable the forward wave will lead the end
by 45deg and the reflected signal will lag by 45deg.the sum is zero.  In
fact the phase will be zero
where ever you cut the coax except at the driven end.  Just like your ruler
or a Tesla coil.  .

Lets focus our attention on the main object of our interest the secondary
coil. You suggest the analogy of a vibrating clamped steel ruler. I agree
but this is a almost perfect example of a 1/4 wave resonator. Its shorted at
one end by the clamp.  The other end is free to move that's an open circuit.
It also has a finite propagation time for transfers waves . The vibration
you see is a standing wave resulting from the superposition of the forward
and backward travelling waves propagating along ruler.  As you correctly
pointed out no apparent phase shift only a quarter of a sine wave changing
in amplitude.

Hopefully you cut your coax so you know you cant to measure the propagation
delay from the phase shift. measurements in your coil when its resonating.
So you must remove the reflection  or use a low frequency square
wave as I previously suggested .  I now think (a minor conversion for me)
that the coil is significantly dispersive (different frequencies travel at
different velocities)  so this could complicated the interpretation of an
experiment using square waves.  So  stick to sine waves and  you want
have to worry so much  about terminating the generator and other things and
hopefully the result may be more convincing.

So what you must do is terminating the coil in its characteristic impedance
at both ends.
I have tried to think off a quick  and foolproof (well you screwed up the
coax test) way of you doing this but have not come up with one.

The best I can do is as follows.  Use the isolator described below with say
470ohm in place of the 4.7ohm. Starting at 50 ohms and increasing in say 30%
steps until you get close (or use a 1k pot), connect various resistors
across the output of the coil to ground until the output voltage is equal to
the input voltage. At that point I think you have no reflections from at
least one of the ends.and hopefully its the terminated end. Replace the
divider at the input with a divider that has an output impedance equal to
the
value you found above. check the input and output voltages are the same if
not you may have to adjust the divider again.  You only need to be within
say 10% if you only want to see the delay and not make accurate
measurements.

Now your ready if you are using the resonant frequency of the coil you will
be able to observer the 90deg phase shift.  QED and conversion. (assuming
you don't invent a spurious reasons to account for this and the coax)

You can now perform various experiments that you must have missed at collage
like open circuit resonance, short
circuit resonance, effects of C or  L and  determine the dispersion i.e.
velocity against frequency. You can also switch back to a square wave, drive
the coil from a high impedance source, say ten time the characteristic
impedance, and leave the other end open circuit you will be able to observe
at the driven end the progressive step up of the square wave edges  to the
source voltage as each reflection returns from the open circuit.  Similarly
if the end is shorted you will be able to observe the decaying transients
in voltage this time caused by reflections that are inverted at the short
circuit. Just like the primary transients when the gap closes.
You could also determine the characteristic impedance at different
frequencies.

Note that the square wave must be at least ten time the resonant frequency
or there will not be time to observe the effects before the next edge
occurs. Note also that the edge may be smeared out because of the dispersion
i.e. the harmonics will travel at different velocities.

Some potentially important points:

1) You don't appear to have a ground plain this could produce all sorts of
spurious responses that could be very confusing. Ideally you should do the
experiment over a large metal plate or aluminium foil.  Preferably although
not essential with the coil laying parallel to it. You could use the open
circuit primary. The generator ground, scope probe grounds  and any
terminating resistors or short circuits should be connected to it preferably
at a common point.

2) You should have an isolator on the output of the generator at the point
it is connected to the coil. Particularly if your using coax to connect to
the generator and the generator has a 5 ohm output.  You should also
terminate at the generator if its output impedance does not match the cable.
Again use a 50ohm termination assuming 50ohm coax.  This may be essential if
you doing square wave experiments as the reflections in the cable will
degrade the edges of the signal and can upset the generator. Its interesting
to speculate which  point  in your set up is equivalent to the point in the
coax experiment where you measured the 90deg phase shift.

3) Use the scope probes on X10 and a direct connection you only need 0.1
volt in or less depending on the ambient electrical noise.

The isolator can be made in one of two ways depending on how you want to
drive the coil from a high impedance source or low.  In the low impedance
case which you need to determine Tesla resonance (the coil is supposed to be
shorted at one end) and assuming 50ohm coax, use a pie net work consisting
of a 50ohm(or 47) resistor across the generator and a 470ohm and 4.7 ohm in
series across that. (for the coax experiment you my need to use 1ohm to get
better results but it will make tuning sharp).The coil is connected to  the
junction of the 470 and 4.7 resistor.  For a  high impedance drive use a
50ohm (or 47) across the generator and a 10kohm resistor from it to the
coil.

Note that for 50 ohm coax  4.7ohms is  only 10% of the cable impedance so
it does not represent a good short.  A better value would be 1ohm with a
series 100ohm.  This should give you a good standing wave but the tuning
will be more critical

I did say it was tricky but it should teach you a lot. Suddenly all those
spurious effects can be easily be explained.  The funny ringing on the spark
is produced by waves bouncing round the primary.  A new layer of complexity
and wonder will be open to you.  Then all you will have to do is convince
the other non believers.  Well I have done my bit.
.
Regards Bob