PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2019-10-20, 23:11:50
News: Registration with the OUR forum is now by invitation only.

Pages: 1 [2] 3 4
Author Topic: Magnetic Delay Transformer  (Read 1694 times)

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Hi Smudge,

Quote
Are you measuring the output from the CMC using two probes and taking the difference?

No i did not as can be seen in the pictures.
But doing so does not change much.
In the last setup (50 Ohm termination of the CMC), the differential output still shows about 160mVpp.

Going back to the complete setup with the test toroid attached it also makes no difference, output on
the CMC 100mVpp and no output on the test toroid load (10Mohm / 8pF).

Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Got my T107/65/18-3F4 Ferrite Toroid today.

Will start with the setup as described by Smudge in the below PDF..............


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1258
Itsu,
Looking back at an earlier post I see you said
Quote
I was playing with the twisted magnet wire on a toroid as iso xformer as suggested by Smudge and that seems to work out ok.

The twisted wire thingy is for the CMC which I regard as a balun, not the iso transformer.  I should have picked this up from your image which clearly shows the twisted wire toroid as the iso transformer.  That is not what I meant for you to do.  May I suggest you use the twisted wire idea for the CMC instead of classical coils.   I think you will then solve the problem as the signal now travels along the twisted wire transmission line.  At least you now have a good iso tranformer which was your earlier problem. 
Smudge

Edit.  I might add that the signal flows down the twisted pair in the same manner that broadband internet signals travel down the twisted pair telephone cables.  So there should be very little loss in the CMC.
« Last Edit: 2019-10-07, 10:48:38 by Smudge »
   
Group: Moderator
Hero Member
*****

Posts: 1258
Just to clarify here are images of a wide bandwidth balun, unbalanced to balanced transformer or CMC (with thanks to Itsu for use of his image).
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834
Thanks for the clarification Smudge, sorry for the confusion.

So i keep the ISO Xformer (galvanic isolation) and build a new balun/choke (T106-2, twisted wire,
straight through), see picture below.


I followed the mentioned PDF and have my T107 toroid with 1 turns / 10 turns, a 10 Ohm series resistor
and a variable air capacitor (45 - 345pF).

Please check this is what you meant to do.

FG set to 1MHz sine wave 1Vpp
Cap tuned to resonance (157pF)
Input voltage (red) done via differential (2 probes) measurement show still some loss/loading (806Vpp).  Input measured at 1 turn prim. of the T107.
Output voltage (purple) at resonance 9.22Vpp at almost no phase shift (1 - 2 °)   Voltage measured across variable cap C1, see diagram below.
 
Inductance measured:
Primary 1 turn 1.93uH @ 100KHz
Secondary 10 turn 139.2uH @ 100KHz

Regards Itsu
« Last Edit: 2019-10-08, 15:33:10 by Itsu »
   
Group: Professor
Hero Member
*****

Posts: 2590
  Thanks for the updates!   O0
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Here the diagram from my last setup above with value's:


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1258
Hi Itsu,

I would refer you to my "MDT paper" in post #9 of this thread where the story is somewhat clearer.  You will see that the capacitor and load are not in series, they are in parallel as this is looking for an effect where the secondary is loaded by a capacitor.  And you will see that the effect occurs above resonance.  We are looking for the real component of the input impedance going negative, i.e. a negative resistance.  If that occurs then the system feeds energy back to the source and the COP is infinite.  It is predicted by transmission line theory.  Below is figure 4 from that paper where the zero crossing is circled in red.  Graham's measurements did not obtain that zero crossing, but interestingly the ratio of Vout/Vin which you would expect to peak at resonance did not do so, the peak occurred at just that point where theory predicted the zero crossing, so the magnetic delay plus capacitive loading did produce an effect.  This tells me it may be possible to succeed perhaps if the magnetic delay could somehow be artificially increased.  That I think is the exercise.  So if I were doing the work I would measure the ratio of Vout/Vin to see if it peaks above the resonant frequency.  Try using the highest frequency you can where the time delay has greatest effect.  You might also measure the input resistance to see whether this gets to its lowest value at that point where Vout/Vin is maximum.  This all points towards the system nearly getting to a zero crossing.   If you have some success there I would then be inclined to try to increase the magnetic delay, but more on that later.

Not sure where the one turn primary came from, Graham used I think six or ten turns for both.  The load resistance across the capacitor is somewhat arbitrary for now, I would be inclined to not have one and just use the capacitor.

Smudge   
   
Group: Moderator
Hero Member
*****

Posts: 1258
Here is an old paper where I show an experiment I did using a small toroidal core with input and output coils diametrically opposite.   Putting a series of additional coils each shunted by a capacitor produces a measurable delay as shown.  This could be something worth trying.
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Smudge,  you see, the info you put into this thread is enormous, so i find it hard to get a clear
understanding on what the correct schematic to follow is.

I understood from your post #13:
https://www.overunityresearch.com/index.php?topic=3847.msg78101#msg78101

Quote
Here is one of my earlier papers before the MDT program got started.  I modeled a  1 turn primary and a 10 turn secondary.  That could be the basis for a simple experiment.
Smudge.
------------------------
* Magnetic Delay Effects.pdf (28.61 kB - downloaded 17 times.)

you pointed there to an easy start, but i now see its just a "modeling", not a real circuit.


Looking at your post #9 i see the schematic, but no real data on used components.

So for starters, putting 1 + 1 together, i need to change to 2 x 6 turns on the T107, get my load resistor
parallel to the secondary LC of the T107.


What about the value of this parallel resistor, does it matter or is the 10 Ohm OK?

 
Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1258

Quote
What about the value of this parallel resistor, does it matter or is the 10 Ohm OK?
I think I would use a high value like 1M Ohm.  Not bother with output power which would be small, but look for the input characteristics I mentioned.  That does mean taking measurements over a frequency range.  I might mention that Graham built a rig that did things automatically so he could just leave it running and gather vast amounts of data, that I then analyzed.  I don't expect you to do that  :).
Smudge

   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834
I changed my setup to have 2x 6 turns on the T107 and to have a 1MOhm resistor parallel to the secondary
LC of the T107, see diagram / picture


Went up to 2.5Mhz to have my 45-345pF variable cap somewhere midway (~260pF).

I have set the FG to have 1Vpp to the primary of the T107.

Resistor measures 1.020MOhm
Both T107 coils measure 49.2uH @ 100Khz.

Please check the diagram for errors/misunderstandings.


Using the FG in sweep mode, i sweep the setup between 1KHz and 5Mhz, see screenshot 1.
Purple is the signal across the 1MOhm resistor when in resonance (2.5Mhz)
Red is the differential signal across the T107 primary (set at 1Vpp).

We see that the resonance output voltage peak does not coincides with the input dip.

Screenshot 2 is from the signals non sweeped at 2.5Mhz.


Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834
Below screenshot is an input power sweep (white) compared to an output power sweep (red).

I used a 1 Ohm 1% inductionfree CSR for these power (current) measurements.

Again the sweep is from 1KHz to 5Mhz.

I first did the input measurement / calculations in White (across the T107 primary) then the output measurement / calculation (across T107 secondary)
as my scope only can do 1 math function at a time.

Not sure how to interpret these power sweeps, but i hope to be able to pinpoint an area of interest.

(using max. voltage (20Vpp) from FG)
 
Itsu
« Last Edit: 2019-10-09, 15:19:36 by Itsu »
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Hmmm,   when i swap the FG leads (red / black), then the input power measurement/calc (white) at the T107 primary gets more symmetrical; guess thats more according to how it should be.


Itsu

   
Group: Moderator
Hero Member
*****

Posts: 1258
Hi Itsu,

The Area of Interest (AOI) is just above resonance where the input power dips.  I think you will find that the COP there is a maximum, and will be greater than the COP at resonance.  In fact just doing a quick look on that last screen shot, assuming both scope traces are at the same settings and measuring the powers with a ruler I get the COP (output power/input power) to be about 0.92 at resonance and about 4.3 at the AOI.  With a COP of 4.3 that is certainly an area of interest  O0.

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

I made an update yesterday evening, but somehow it did not went through.


My problem is that the measurements are not consistent.
Even at resonance i get sometimes a COP = 2 then lateron 0.5 or so.

So i still think the problem is in the scope leads ground probes, especially
when measuring the input at the primary of the T107.
Then i have to "ground" the primary of the T107 bringing it to the same potential as the FG black lead.
so effectively shorting out the Iso / balun.

I can try to use a battery operated (simple) FG and skip the ISO / balun combo and/or try to use my
current probe for current readings and differential probing for voltage, but with my scope it will
take some manipulations.


Itsu
   
Newbie
*

Posts: 7
    itsu:
    Can your scope read the voltages, if the scope's ground lead is left disconnected?
    Have you found a frequency where the input power drops?
    That is what I've had to do on some of these tests on my device, not to use the negative scope probe.
    In any case, consistency of the reading needs to be found.
    Hoping that your tests on this device will show some gains.
       NickZ
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Hi Nick,

my scope can read voltages with its probes ground leads disconnected, but it would fluctuate and i
would not trust these readings much.

As you can see in the picture below from Smudge, the "Area Of Interest" line would be the frequency
around 2.8MHz (while keeping the circuit tuned to 2.5MHz resonance) where the input takes a dive while
the output still remains significant and thus hopefully above COP=1.

Checking around 2.8Mhz up till now shows fluctuating values mostly on the input (phase difference
involved) and thus no consistent readings.

Regards itsu
   
Group: Moderator
Hero Member
*****

Posts: 1258
Hi Itsu,

I am not sure how important the ISO transformer or the balum/choke are.  If you get consistent readings without using them I would concentrate on that.  You use the math channel to calculate power, what formula do you use?

Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834

Smudge,

i will try tonight to use the battery operated FG directly to drive the T107 primary and see what readings i get.

The TDS3054B scope uses its internal math function to calculate the power (mean) from the instantaneous
values given by the 2 probes used, in this case 1 measuring voltage and an other to measure current.

As these 2 probes used need a ground reference there lies the problem i think.


Itsu
   
Group: Moderator
Hero Member
*****

Posts: 1258
If you have your CSR in the grounded end of the winding giving voltage V1  and the other end V2 then use math V1 x (V2-V1) does that solve the problem?
Smudge
   
Group: Moderator
Hero Member
*****

Posts: 1258
Also, how do you ensure that the math channel averages over whole cycles?  Do you set cursers to do this?
Smudge
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834
If you have your CSR in the grounded end of the winding giving voltage V1  and the other end V2 then use math V1 x (V2-V1) does that solve the problem?
Smudge

The problem is in the ground lead connections of the probes.
See the diagram below, no matter where i measure, i need to put a groundlead at any of those red circled points and thus grounding there where no grounding was/is.

I can see on the scope that there is an influence.


Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1834
Also, how do you ensure that the math channel averages over whole cycles?  Do you set cursers to do this?
Smudge

My scope has 3 options,  across the full record, screen size or between 2 vertical cursors, see screenshot.
Normally i use the "full record" and put up many cycles to get a good average.

Itsu

 
   
Group: Moderator
Hero Member
*****

Posts: 1258
The problem is in the ground lead connections of the probes.
See the diagram below, no matter where i measure, i need to put a groundlead at any of those red circled points and thus grounding there where no grounding was/is.
I can see on the scope that there is an influence.
Itsu
OK I see your problem.  Graham had the benefit of an expensive multi channel scope so he had all scope probes permanently connected, he didn't have to swap ground connections about.  Also he could use differential measurements using two probes so that ground loops were not a problem.

In your case in order to keep things the same for both input and output power measurements the ISO and balun/choke are redundant and you may as well just connect the FG directly to the device.  Also I can't see how you can use two ground points at opposite ends of the CSR, surely those two scope ground connections are shorting out the CSR.  You need a different set up with just one ground point.  For the output power you don't need a CSR.  My two images below show a permanent ground connection between primary and secondary, with the scope probes grounded there.  For the input power you need the math shown on the image there, CH2-CH1 of course gives you the voltage across the primary to be multiplied by the current.  For the output power the math channel uses the load resistance.  (If you are worried about the inductance of the load resistance then use the CSR and connect in the same manner as for the input power).  I show dummy probes which are simply a resistor and capacitor in parallel (e.g. 10M and 20pF or whatever your probes are) to simulate the presence of the absent probes.  All this ensures that to the best of your ability the system is identical for both measurements.

I see your power measurements are noisy.  Maybe this is because you are using a 1M load which will not consume much power.  I think that 1M is OK when you are exploring input impedance, but maybe a lower value would be better for power measurements.

Hope this helps.

Smudge
   
Pages: 1 [2] 3 4
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2019-10-20, 23:11:50