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

Re: Another obnoxious lumped theory supporting post ;-))

Hi All,

	As Richie and Antonio pointed out, there must have been something wrong
with my experiment, and there was...

First, my ailing signal generator made it difficult to find the exact
resonant point.  Replacing it made the tuning much more accurate.  

Second, the flying wire leads I used on the scope probes have terrible
phase accuracy.  The information they gave was pure garbage.  You simply
can't get phase information from flying wires hooked to scope probes!!  I
remember someone writing a paper on this once like that one at:

http://users.better-dot-org/tfritz/site/papers/planant/waveant3.html

I should have read that guy's paper before I started this. ;-)))

Third, given all this erroneous data, I was able to quickly and logically
justify it to myself.  I really WAS convinced it was true...  So after 2+
years at this, not only can "I" still be tricked by secondary behavior, I
can even trick myself! ;-)))

One thing I did do right was have other people check my results.  Richie
certainly suspected a problem and Antonio was convinced I had done
something wrong.  I knew right away that it was time to go rechecking
things!!  I am glad I didn't go on for years without realizing the problems!

So the secondary experiment I did yesterday was "all messed up".  I think
Richie may have been able to repeat the results if he had a sine generator
but he probably would have figured the problems out quickly.  The coax
experiment seemed to work but that may have been sheer luck or the phase
distortion between my two probes was matched better in that case.

So, another secondary theory experiment bits the dust...  However, it was
hardly a failure.  Not only did it still manage to support lumped secondary
theory, it also demonstrated a number of mistakes and sources of error in
such work.  

I guess if this was easy, everyone would do it ;-))  I hope Richie joins in
this game too!  I can assure him, that it is pretty challenging!!

Cheers,

	Terry



>Hi All,
>
>	Inspired by Ken and Robert I did an interesting experiment.  I measured the 
>top and bottom phase shift of a coax cable at 1/4 wave open load resonance 
>and a secondary coil top and bottom at open load resonance.
>
>The coax is just 100 feet of RG-8 TV cable driven from a sine generator.  I 
>grounded the negative terminal of the generator (It can normally float) and 
>ran the output through a 100 ohm resistor to the coil.  The generator has a 
>one ohm output but the load at resonance draws too much current from it, so 
>I need a resistor.  I then placed a scope probe next to the wire from the 
>generator feeding the coax to capacitively couple the voltage (it has a 
>short wire on it).  This picks up the input voltage.  I put a short wire on 
>the other end of the coax next to the other scope probe to pick up the 
>output voltage.  I had the sensitivity on the scope turned way up and I was 
>able to get a good signal off both probes.
>
>A picture of the cable and probes is here:
>http://users.better-dot-org/tfritz/site/misc/quarter/Image2.jpg
>
>I then adjusted the generator to get resonance.  It was a little tricky 
>since the cable seems to like harmonics and the input signal level was not 
>great but I think it was right.  
>
>The scope picture is here:
>http://users.better-dot-org/tfritz/site/misc/quarter/Image1.jpg
>
>Notice how the phase at the open end (top) is 90 degrees later than the 
>phase at the generator end (bottom).  See, that transmission line stuff 
>really does work for transmission lines! ;-))
>
>I then disconnected the coax and hooked the generator to the base of a 
>secondary (1180 turns #24 3.25 Dia. 26.125 long).  The same scope probe was 
>laid next the input wire just like before with the coax as shown:
>
>http://users.better-dot-org/tfritz/site/misc/quarter/Image5.jpg
>
>The other probe was laid on the top of the coil as shown:
>
>http://users.better-dot-org/tfritz/site/misc/quarter/Image4.jpg
>
>I adjusted the generator to get resonance (first harmonic) as usual and 
>snapped the scope picture:
>
>http://users.better-dot-org/tfritz/site/misc/quarter/Image3.jpg
>
>Geee.  Is not zero.  Maybe 5 degrees off.  I'm still not getting a warm 
>fuzzy about all this transmission line stuff! ;-))  I moved the probes 
>around and all that looking for the nice 90 degree phase shift to coax had 
>but I just can't find it.
>
>The 3F0 harmonic was there in all it's glory at -180 degrees and the 5Fo was 
>back at zero degrees again.  Another thing to try is to put one probe 
>monitoring the top of the coil and take the other probe and move it along 
>the coil (you will have to retune).  That 90 degrees phase shift was not 
>there.  The coax worked just as it should.
>
>So electricity really does have to run the length of the 100 feet of coax 
>and the velocity factor stuff about coax cable is true.  The wave really 
>does take 1/4wave to get to the other end and the phase shift is 90 degrees 
>just like it should be.  But in a repeating the test as closely as possible 
>with a secondary coil, the 90 degree phase shift was not there.
>
>So give it a try and tell me what's wrong.  I think if I did all that RGLC 
>transmission line stuff I could explain the 5 degree phase shift (as I have 
>before) but there really isn't much that can go wrong with this test and I 
>invite anyone to give it a try...
>
>I point to Malcolm's example of taking a steel ruler and clamping one end in 
>a vice and thumping the top end to get it oscillating.  Hardly any phase 
>shift in that situation and I think he is right in that all the mechanisms 
>there are analogous to or Tesla coils.
>
>In a transmission line, you have inductance and capacitance per length that 
>slows the wave propagation some.  But you really need pure wire length to do 
>the majority of the wave delay and phase shifting.  You can take a very 
>short line and come up with super high values of inductance and capacitance 
>per length to get that slow wave (the key behind delay lines) but those 
>values are simply not real life.  Once the wire length aspect of Tesla coil 
>theory fell apart, the 1/4 wave transmission line theories were doomed too.  
>You have lots if inductance for your RLGC values but there just isn't enough 
>capacitance to slow the darn wave down to get a 90 degree shift (5 degrees, 
>yes).  You can make up an imaginary number to get there but unless you pack 
>barium titanate around your coil, that is not realistic.
>
>No doubt that the secondary self capacitance is "distributed".  However, the 
>caps in my MMC are rather "distributed" too and they don't cause phase 
>shift. They are far closer to each other than the coil's wavelength.  by the 
>same argument, I would submit that the top and bottom of a secondary coil 
>are also far shorted than the wavelength despite the distributed
capacitance...
>
>Whew!  Terry's giving the 1/4 wave guys a rough time!  :-))
>
>Ok, their turn...
>
>Cheers,
>
>	Terry
>
>
>