PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2020-07-04, 22:14:28
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: 1 2 3 4 5 [6] 7
Author Topic: Magnetic Delay Transformer  (Read 6100 times)

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Quote
Thanks again Itsu.  The dip in input power clearly shows up at 240KHz, but no OU there.  IMO it is not necessary to use a CSR for the output power, just use rms voltage squared divided by the load resistor value.  A quick check taking the Pk to Pk voltage from the screen shots doesn't improve things.  I still can't explain why the circuit losses (input power minus output power) reduce at that resonance, whereas in my mind they should increase.  Have to think more on that.
Smudge


Smudge,

i was wondering too if i could use the output rms voltage squared divided by the load resistor value in this case, so i can.

Using that methode on the 10K load (actually measured 9880 Ohm) i get:

Freq. Vrms  Pout (mW)
200   3.14   0.99
210   3.24   1.06
220   3.33   1.12
230   3.40   1.17
240   3.45   1.19
250   3.49   1.24
260   3.51   1.24
270   3.52   1.25
280   3.52   1.25
290   3.49   1.24
300   3.46   1.21

Similar rms voltages are measured using the 100K (104.7K) equating to 100uW output range figures

 
let me know if i can do some more testing.

Regards Itsu
« Last Edit: 2019-11-28, 23:53:33 by Itsu »
   
Group: Tech Wizard
Hero Member
*****

Posts: 596
Hi Itsu,

Would like to ask what kind of resistors do you use for the 10 k and 100 k Loads?
I mean that their values may be different from their expected values in the 200 to 500 kHz frequency range (even if they are carbon resistors).
Even though your circuit output is surely reactive, the parasitic components of those resistors may be embedded in the circuit in an unknown way.
Somehow the resistor values should be checked at those frequencies separately from the circuit to learn about their real values.

Gyula
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

Hi Gyula,

these 10K and 100K resistors are indeed carbon ones as i do not have any high resistance inductionfree precision resistors.

So you might have a point as their parasitic components (inductive reactance) might influence the resistance in this frequency range.

Not sure however how to measure these resistors at that frequency range, but i will look for a solution.

Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Using a known 270pF SMD capacitor parallel to the 100K (104.7K) resistor and using my picopulser to pulse
this LC (adding 8pF for my probe) i measure a damped resonance at 167MHz.

Using this resonance calculator ( http://www.1728.org/resfreq.htm  ) i calculate the inductance to
be 0.00326uH (3.26nH).  Input 167Mhz and 278pF.

This inductive reactance calculator (http://www.66pacific.com/calculators/inductive-reactance-calculator.aspx)
calculates the reactance of this 3.26nH @ 200KHz to be 0.0041 Ohms.

Guess its of no influence on the resistance.

Itsu 
   
Group: Tech Wizard
Hero Member
*****

Posts: 596
Okay Itsu, it sounds good enough.  I will ponder on some other simple means to check this if possible or still needed.   

The main reason I mentioned this as possibly problematic is that I find little difference in the swept voltage amplitudes between the two resistor loads in the scope shots you included in your Reply #123 (previous page). 
This small difference may come from a certain low output impedance of your circuit, very likely much lower than 10 kOhm.  So it is very likely that the 50 Ohm function generator at the circuit input transforms to the circuit output because of the 1:1 input - output turns ratio between the coupling coils and the delay circuit would not influence this significantly.  So now I think this explains why the very small amplitude difference for the two highly differing load resistance values.   

Thanks for doing these tests.

Gyula
   
Newbie
*

Posts: 7
Hi Itsu

Considering Gyula comments, perhaps the experiment should be repeated with a 50 Ohm output resistor.
Regards
Cortazar.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

Hi Cortazar,

I did use a 100 Ohm resistor earlier, see post #121.

But using a 50 Ohm 1% induction free resistor as load parallel to the secondary LC (C again 270pF), i get
the following input / output relation (output calculated from rms voltage across the 50 Ohm (P=U²/R)):

Frequency (KHz)    input (mW)    output (mW)

    200                  56.34           46.20
    210                  56.88           46.82
    220                  57.33           47.43
    230                  57.72           48.05
    240                  58.11           48.67
    250                  58.45           49.30
    260                  58.76           49.93
    270                  59.30             "                         
    280                  59.54           50.56
    290                  59.75             "
    300                  59.95             "
    310                  60.11             "
    320                  60.28             "
    330                  60.41             "
    340                  60.53             "
    350                  60.64             "
    360                  60.67             "
    370                  60.72             "
    380                  60.75           49.93
    390                  60.74              "
    400                  60.69           49.30


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378
Itsu,

Have you still got your delay line transformer handy?  After finding the PM motor idea shown below on my computer, it struck me that if your primary and secondary coils were connected in series opposing this could show up some interesting features.  Could you set this up as shown in the modified image of yours and measure input voltage and current at different frequencies?  If there is any sign of the phase going above plus or minus 90 degrees then that could be significant.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

Hi Smudge,   yes i still got my delay line transformer.

Let me set up the prim. and sec. in series opposing tonight and do the measurements.............


Itsu     
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

I swepped from 1KHz to 20Mhz (20Vpp) sine wave, but never saw the phase between I and V be more then 90°, see video:  https://youtu.be/G_sN7h_HRJs


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378
Hi Itsu,
Thanks for doing that work.  Unfortunately I think you misunderstood the test.  You have a 100 ohm "load" resistor there and that will dominate the phase between input current and voltage.  The series opposing connected coils will have a peculiar input inductance/resistance characteristic, so I am interested in the phase between input voltage and input current of that strange inductor.  But you are not measuring that, you are measuring that input characteristic in series with the 100 ohms.   Can you please repeat with only the CSR in series with the FG input.  The CSR will affect things so you really need to measure the voltage only across the coils, and not across the series CSR plus coils.  Maybe use a current probe, or take a differential voltage measurement to eliminate the voltage across the CSR.
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Sorry about that Smudge.

I redid the test, now with my battery operated FG and a 10 Ohm csr in the return line.

FG return line (floating) on left side csr, probe grounds on right side csr, so measuring
on yellow probe the voltage across the series opposite coils, and on blue the voltage
across the csr (inverted), see diagram.

I also used the current probe (green), but as expected in the higher frequency range
it starts to be very optimistic.

The phase between voltage (yellow) and current (blue) is for a big frequency part at about
90°, but from 1.5MHz and up it starts to grow higher topping at about 101° on 2.6Mhz.

But this is closely before a resonance point, so current drops and the blue signal gets small
and unstable (the current probe (green) shows abnormal phase).

Hope thats what you wanted to see.


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378
Thanks again Itsu, that looks very promising.  I think there is a slight improvement that can eliminate the floating FG, see modified image below.  The FG will have a significant self capacitance to ground that will shunt the CSR, and that may affect the data.  The differential method for obtaining the input voltage could be better.  In any case the phase exceeding 90 degrees does indicate a negative input resistance, which is what we are looking for.  Please carry on with this work.  If you can give me some voltage, current and phase measurements over the higher frequency range I'll do some data analysis.  You mention the blue signal becoming unstable, and instability is itself an indication of something good.
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189


Smudge,

with "the blue signal becoming unstable" i mean that the signal gets weak (low amplitude) and
thus noisy so the scope has problems to accuratly calculate data like amplitude, frequency and phase.

Setting up the measurement as proposed by you made the "exceeding 90°" dissappear  :(
Below 1st screenshot shows the same 2.6MHz frequency as yesterday, but now only having 86° phase shift.

Yellow is the FG input voltage (for reference).
Red is the math: yellow minus blue for input voltage.
blue is voltage across csr thus current (10 Ohm)


There is a slight(er) above 90° range, but more down in frequency, see screenhot 2

Here we are at 70Khz but the blue (and yellow) signal is low in amplitude (still 20Vpp
input from the FG, but the impedance seems low).

This low in amplitude again has an effect on the math calculations as the sigs become noisy
and thus inaccurate, so the shown 96° phase shift here is doubtfull.

Is this shown data OK (V rms, frequency and phase shift) or do you need different data and/or more range?

Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378

Is this shown data OK (V rms, frequency and phase shift) or do you need different data and/or more range?

As I was hoping for the effect to show up at the frequency range where your delay line transformer was showing a big shift in V-I phase, and this new experiment didn't live up to that expectation, I do not see any further work along those lines as being of use.  However if you are interested in doing some more work I would like you to look into continuing work that was being done by Chava and then got cut short.  This showed an anomaly at about 14MHz which is way above the frequency for that 3F4 ferrite,  but it did show a negative input resistance.  It used the toroidal core that you have with primary and secondary on opposite sides.  Can you do measurements around that frequency?

Thanks again for trying.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

Smudge,

no problem in doing some further tests, but i don't quite understand which toroidal core you are referring to now for this 14Mhz test.

Is it the 3F4 core with the delay line setup i was using or the big red T520-2 toroid i did use earlier which is without any delay line?

The 3F4 delay line toroid has a -40° phase shift as it has a resonance point (so 0° phase) at 14.2Mhz.
Anything higher or lower reverts back to about 88° phase shift.

The T520-2 toroid without any delay line has a 80° phase shift around 14Mhz and does not change much when going higher or lower.

Using the same setup as proposed by you, using a 10 Ohm csr.

Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378
Hi Itsu,

The previous work was done on a 3F4 ferrite as shown in the image below, hence no delay line wound onto the core.  This core is 107mm OD, is your 3F4 core that size?  There are two phase shifts involved.  One is the phase between input voltage and input current and this core exhibited an anomaly near 14MHz.  The other is the phase delay across the transformer from primary voltage to secondary voltage.  At that frequency the phase delay was close to 180 degrees.  If you look at figure 6 in the progress report you will see that measured phase delay for a 1K load shunted by various capacitor values, the one of interest being the 20pF scope probe result.  This almost straight line of phase v frequency indicates a fixed time delay across the transformer, and it would be interesting to have this measurement confirmed.  Note that at 14MHz the core is very inefficient, so the secondary voltage is small.  Maybe the transmission from pri to sec was by a lossy surface wave, who knows?  But somehow that transmission created the anomaly where the input resistance went negative, and it would be good to track that down.   Graham's set up was quite large, he automated the system so as to automatically switch in different load resistors and capacitors and to take measurements at different frequency steps, thus providing a huge data set that I had to analyze.  It may be that the anomaly was an artifact associated with that set up, in which case you will not be able to replicate it.  But on the off chance that it was real, I think it is worth exploring further.  So the first step is to measure the phase delay from primary to secondary using say a 1K load and just the scope probe as capacitance, and repeat that curve in figure 6.
Smudge
   
Group: Moderator
Hero Member
*****

Posts: 1378
For anyone interested here is another form of transformer that uses electric fields as the coupling mechanism.  Basically it is a ring "core" made of high K dielectric material.  The primary is a small magnetic ring core acting as a single magnetic turn, and so is the secondary.  Coils wound onto those magnetic rings become the primary and secondary connections to our external electric circuit.  We could use multiple magnetic ring cores to increase the number of magnetic turns.  The possibility of this being a delay transformer interests me as the TEM transmission wave along the dielectric "core", like its magnetic equivalent, has an imaginary characteristic impedance (E and H in phase quadrature).  And transmission line theory tells us that lines with that impedance can exhibit negative input resistance, hence can be a source of energy.  Turning that "core" into a series of capacitors does not alter that imaginary impedance.  Is anyone interested in making one of these?

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Hi Itsu,

The previous work was done on a 3F4 ferrite as shown in the image below, hence no delay line wound onto the core.  This core is 107mm OD, is your 3F4 core that size?  There are two phase shifts involved.  One is the phase between input voltage and input current and this core exhibited an anomaly near 14MHz.  The other is the phase delay across the transformer from primary voltage to secondary voltage.  At that frequency the phase delay was close to 180 degrees.  If you look at figure 6 in the progress report you will see that measured phase delay for a 1K load shunted by various capacitor values, the one of interest being the 20pF scope probe result.  This almost straight line of phase v frequency indicates a fixed time delay across the transformer, and it would be interesting to have this measurement confirmed.  Note that at 14MHz the core is very inefficient, so the secondary voltage is small.  Maybe the transmission from pri to sec was by a lossy surface wave, who knows?  But somehow that transmission created the anomaly where the input resistance went negative, and it would be good to track that down.   Graham's set up was quite large, he automated the system so as to automatically switch in different load resistors and capacitors and to take measurements at different frequency steps, thus providing a huge data set that I had to analyze.  It may be that the anomaly was an artifact associated with that set up, in which case you will not be able to replicate it.  But on the off chance that it was real, I think it is worth exploring further.  So the first step is to measure the phase delay from primary to secondary using say a 1K load and just the scope probe as capacitance, and repeat that curve in figure 6.
Smudge

Smudge,

I have the 107mm OD 3F4 core, but it has the delay line around it which toke some work to install and needs to be removed then.

I did use this core earlier on some tests in this thread without the delay line, but on a lower frequency range.

I have received my newer MDO3000 series scope which i need to get acquainted to, so i will need some time here.

Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1378
Itsu,
My brain is suffering from old age.  I had completely forgotten about the work done by Peter on my "A closer look at a simulated negative resistance coil" thread
https://www.overunityresearch.com/index.php?topic=2773.0
which was devoted to the bucking coils with time delay idea.  This was 5 years ago.  For some reason the work ceased.  But it did throw up some links (supplied by Orthofield) to other related work.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Hi Itsu,

The previous work was done on a 3F4 ferrite as shown in the image below, hence no delay line wound onto the core.  This core is 107mm OD, is your 3F4 core that size?  There are two phase shifts involved.  One is the phase between input voltage and input current and this core exhibited an anomaly near 14MHz.  The other is the phase delay across the transformer from primary voltage to secondary voltage.  At that frequency the phase delay was close to 180 degrees.  If you look at figure 6 in the progress report you will see that measured phase delay for a 1K load shunted by various capacitor values, the one of interest being the 20pF scope probe result.  This almost straight line of phase v frequency indicates a fixed time delay across the transformer, and it would be interesting to have this measurement confirmed.  Note that at 14MHz the core is very inefficient, so the secondary voltage is small.  Maybe the transmission from pri to sec was by a lossy surface wave, who knows?  But somehow that transmission created the anomaly where the input resistance went negative, and it would be good to track that down.   Graham's set up was quite large, he automated the system so as to automatically switch in different load resistors and capacitors and to take measurements at different frequency steps, thus providing a huge data set that I had to analyze.  It may be that the anomaly was an artifact associated with that set up, in which case you will not be able to replicate it.  But on the off chance that it was real, I think it is worth exploring further.  So the first step is to measure the phase delay from primary to secondary using say a 1K load and just the scope probe as capacitance, and repeat that curve in figure 6.
Smudge

Hi Smudge,

i made some measurements on my 107mm OD 3F4 core (2x 6 turns) using my voltage probes (3.9pF @ 10MOhm)
with a 1K resistor on the secondary.

I measured:

Frequency
input voltage rms
input current rms
phase input voltage / output voltage

is this what you had in mind?
Do you need more frequency steps and/or other measurements?

be aware, i switch at 1Mhz from 100Khz steps to 500Khz steps

The picture goes to 18Mhz,  the xls file till 20Mhz.
be aware,  my decimal point is a comma.

I added a graph frequency / phase:

Regards Itsu
« Last Edit: 2020-05-05, 09:29:32 by Itsu »
   
Group: Moderator
Hero Member
*****

Posts: 1378
Thanks Itsu for doing that.  I have charted your results showing the phase as a negative one (i.e. a delay) for comparison with Graham's earlier measurements.  I am not clever enough to merge the two results onto a single graph.  In Graham's case his phase delay reached 180 degrees at about 14MHz and that was the area where his input power went negative.  In your case this occurs at about 17MHz.  His scope probe was 20pF whereas yours was much smaller.  His output power was tiny at his anomalous frequency, but of course his negative input power gave a negative COP result (energy fed back to source).  It would be interesting to know the phase between your input current and voltage to see whether you also get a negative input power that occurs when that phase exceeds 90 degrees.  We never discovered whether his anomalous result was an artifact associated with his complicated set up (like come external RF coupling between output and input) and I was hoping that your work would clarify this.
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
OK Smudge,

There is a slight phase dip around 13.5Mhz, i will zoom in there to see how deep it is.


Further i will add some 16pF to the probes to get around 20pF and redo the measurements using 500KHz steps including the input voltage / current phase.


I added Graham his data (from the above graph, so very little data points) to my delay data graph.


Itsu
« Last Edit: 2020-05-05, 19:59:44 by Itsu »
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189
Smudge,

here the second batch of data:

secondary was loaded with 1K Ohm and 20pF (16pF cap + 3.9pF probe)

Input frequency range from 100KHz to 20MHz in 500Khz steps

frequency
Input voltage rms
input current rms
input voltage / current phase in °
input voltage / output voltage phase in °

Again, decimal point is a comma.

Blad 2  (Page 2)


EDIT, i added one more colum being the negative delay phase and created a graph, see below.

Regards Itsu
 
« Last Edit: 2020-05-06, 10:09:15 by Itsu »
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 2189

I did some work using he input voltage / current and their phase relationship to calculate the input
impedance using this (AC part) website:  https://www.rapidtables.com/calc/electric/ohms-law-calculator.html


Results shown in the below graph:

Itsu
   
Pages: 1 2 3 4 5 [6] 7
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2020-07-04, 22:14:28