Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Lenzless resonant transformer  (Read 185243 times)

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Lenzless resonant transformer
« Reply #165 on: March 07, 2014, 05:24:15 AM »
I think it is important to reiterate one thing about a typical resonance setup, something that is going on in this thread.

Take a look at a secondary that is connected to capacitor as a load.  That can be driven into resonance by a signal applied to the primary.  Naturally you can put a light bulb in series to eyeball the AC current if you want to.

So the secondary circuit looks like a series resonant LC tank with an AC EMF source:   [AC EMF source] -> [Coil + wire resistance] -> [Capacitor]

So when driven at resonance the (Coil + Capacitor) becomes a short circuit and you are left with [AC EMF source] -> [Wire resistance]

That is an AC energy DESTROYER.  At resonance, it will burn off the maximum possible power in the resistance of the winding of the coil.

It seems to me that often people are "chasing after resonance" without realizing that sometimes resonance is a bad thing if you are trying to achieve over unity.

MileHigh

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #166 on: March 07, 2014, 09:37:40 AM »
  To estimate their values I'd need to know if CH3 of the Tek scope can be set to 50Ω impedance.

Programming of the SG and Scope later...

verpies, it does have the posibility to set each channel in 50 Ohm impedance.

Thanks,  Itsu

Jack Noskills

  • Sr. Member
  • ****
  • Posts: 346
Re: Lenzless resonant transformer
« Reply #167 on: March 07, 2014, 09:43:55 AM »

Ok i followed Jack's directions and paralleled the both secondaries with only one (double) cap (1100nF).
Without bulb it resonates at 10.5KHz (signal injected through L3)

After hooking up the bulb in the secondaries LC, its resonance frequency raised to 18KHz.

Then searching the L3 for resonance (injecting a signal through the both paralleled secondaries without cap/bulb) which was 1MHz.
Bringing down this frequency to 18KHz by using 70nF capacitance (where have i seen that value before).

Now both L3 and the paralleled secondaries with cap (1100nF) and bulb resonate at 18KHz.

In this situation,

Input voltage and current (from FG) are in phase
Minimum input current (is this what you mean by blocking the current?)
maximum input voltage
Maximum input power  (144mW) from FG
maximum output power (122mW bulb max. lit   see picture*)

* as i forgot to measure in the video the voltage/current (power) across/through the output bulb, the picture shows the
blue trace the voltage across the bulb and the green trace the current through the bulb with the red math function blue * green = avg power
(times 2 for the currentprobe terminator)
 
Video here:  https://www.youtube.com/watch?v=BE7tAbYRg7w&feature=youtu.be

Regards Itsu


Yes, by blocking I meant minimum input current.


Can you verify that input is not affected if you disconnect the output ?
Now can you test the same setup using the other signal gen that gives 30 volt signal ? The one on the vid used 5 volts, right ? At least this is seen on the scope.


If there would have been bulb in the L3 coil to 'measure' input then that bulb would have been more lit than the output bulb ? If so, then there is still some difference compared to my setup.


From this test I now I remember what happened when I first tested this. I had 1000 nf cap in the output and I was about to quit testing, 'just one more sweep and then I dump these cores in the ocean' and as sweep was about to end at 19 kHz my output bulb started to light up. Not very bright though, so I added second 1000 nF cap and then I got better result as frequency dropped to 11 kHz. I could not add more because then my L3 started to leak as the frequency went down and not very easy to find resonance in L3 using caps for me.


So now I am thinking that one way to increase output power is to add more caps in the output side and at the same time keep L3 in the resonance as you already did. The second way could be to increase input voltage, I am not sure but now you can test this easily with your current 18 kHz setup.


If these two options give positive results then next step is to see how increasing the resonant F by spreading turns, using less turns, using thicker wire etc. affects this. Then bring it down by adding more caps in the output. But lets not think about this at the moment, better to keep variables as few as possible for now.

Jack Noskills

  • Sr. Member
  • ****
  • Posts: 346
Re: Lenzless resonant transformer
« Reply #168 on: March 07, 2014, 09:46:41 AM »
Itsu:

Jack:

I don't know what you mean by "local inductance field."  More importantly, there is essentially no "influence of a wire to a neighboring wire."  That sounds like an old wives' tale.  In the context of winding wires around a toroid to make an inductor, it's the permeability of the core, and the number of turns that really count.  Also, the core can only store so much magnetic energy before it saturates.  There is a culture of experimenters neglecting or being afraid to correct each other on the forums.  That leads to stagnation, people don't learn because nobody corrects them and it becomes a vicious circle.

MileHigh



I need to correct myself here. What I call local inductance field is actually the magnetic loop of a solenoid through air from end of solenoid to beginning, verpies made me realize that. Anyway, end result is the same: not much influence compared to looped core.


There is magnetic field around a wire so it will affect neighboring wire, but effect is barely noticeable below 20 kHz.

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #169 on: March 07, 2014, 09:47:53 AM »
Itsu:

In your latest clip where you have L1 and L2 in parallel and a single capacitor, you still have a very high resonant frequency, implying that L1 and L2 are in a "flux fight" and canceling each other out.  So if you cross one set of wires then L1 and L2 should add together and you should go back to a low resonant frequency.


MileHigh, 

i did cross one set of wires when i started, and found almost no resonance point, untill i cranked up the sensitivity of the scope.
Its like you said, the resonance then is around 175Hz (low again), but very small in amplitude (50mV while inputting 20V pp from the FG).
My thoughts where that they where wrongly connected and now had a "flux fight".

I then changed them to where they are now and found a nice big resonance point around 10.5KHz of about 13V pp which lateron changed to 18KHz when connecting the bulb in series.

So what is the correct connection of the both parelleled coils? Like it is now High Amplitude / high frequency resonance point or low amplitude / low frequency resoance point?


Regards Itsu

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #170 on: March 07, 2014, 10:34:13 AM »

Yes, by blocking I meant minimum input current.


Can you verify that input is not affected if you disconnect the output ?
Now can you test the same setup using the other signal gen that gives 30 volt signal ? The one on the vid used 5 volts, right ? At least this is seen on the scope.


If there would have been bulb in the L3 coil to 'measure' input then that bulb would have been more lit than the output bulb ? If so, then there is still some difference compared to my setup...........................



Jack,

i normally set my FG to the max. amplitude, which is 20V pp for the one i used yesterday.
The other FG with 30V pp i use if for some reason this 20V pp is not enough, which was not the case yesterday.

I can do some further testing with this setup (including crossing the parallel coil wires again) and insert an input bulb in L3 to show the current there etc.

Regards Itsu

verpies

  • Hero Member
  • *****
  • Posts: 3473
Re: Lenzless resonant transformer
« Reply #171 on: March 07, 2014, 06:06:49 PM »
I don't know what you mean by "local inductance field." 
Neither do I  ???

More importantly, there is essentially no "influence of a wire to a neighboring wire."  That sounds like an old wives' tale.
I think in this scenario he just means capacitive coupling between closely spaced wires.

There is a culture of experimenters neglecting or being afraid to correct each other on the forums.  That leads to stagnation, people don't learn because nobody corrects them and it becomes a vicious circle.
Indeed, there is.  It is pathological and unscientific.
When I make a mistake people send me apologetic private messages with their valid objections instead of just dishing it out on an open forum.

Edit:
I noticed in the subsequent message that Jack's "influence of a wire to a neighboring wire." is explained by him as the Amperean induction of two parallel wires.

verpies

  • Hero Member
  • *****
  • Posts: 3473
Re: Lenzless resonant transformer
« Reply #172 on: March 07, 2014, 06:36:20 PM »
I normally set my FG to the max. amplitude, which is 20V pp for the one i used yesterday.
Rigol specifies 20V only for loads greater than 10kΩ. For heavier loads the 20V amplitude cannot be maintained.  Rigol will display the real output amplitude if the Load's impedance is set accurately in its Utility menu.

The other FG with 30V pp i use if for some reason this 20V pp is not enough,
That old SG outputs not only higher voltage but it has a built-in low impedance amplifier that prevents its output voltage from dropping as much as Rigol's output for heavy loads.

If you had that EL2009 amplifier working, then its output voltage would be even less prone to dropping under heavy loading. 
Its output impedance is an astonishing 1Ω at 90MHz !!!

P.S.
If you'd measured the VP-P amplitude of these SigGens with a 10Ω 100Ω and 1kΩ resistors as loads, then you'd get an exact picture how their output voltage sags under load and we could even calculate exact output impedances of these SigGens.




 

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #173 on: March 08, 2014, 12:21:32 AM »

Yes, by blocking I meant minimum input current.


Can you verify that input is not affected if you disconnect the output ?
Now can you test the same setup using the other signal gen that gives 30 volt signal ? The one on the vid used 5 volts, right ? At least this is seen on the scope.


If there would have been bulb in the L3 coil to 'measure' input then that bulb would have been more lit than the output bulb ? If so, then there is still some difference compared to my setup.


From this test I now I remember what happened when I first tested this. I had 1000 nf cap in the output and I was about to quit testing, 'just one more sweep and then I dump these cores in the ocean' and as sweep was about to end at 19 kHz my output bulb started to light up. Not very bright though, so I added second 1000 nF cap and then I got better result as frequency dropped to 11 kHz. I could not add more because then my L3 started to leak as the frequency went down and not very easy to find resonance in L3 using caps for me.


So now I am thinking that one way to increase output power is to add more caps in the output side and at the same time keep L3 in the resonance as you already did. The second way could be to increase input voltage, I am not sure but now you can test this easily with your current 18 kHz setup.


If these two options give positive results then next step is to see how increasing the resonant F by spreading turns, using less turns, using thicker wire etc. affects this. Then bring it down by adding more caps in the output. But lets not think about this at the moment, better to keep variables as few as possible for now.

Ok,  answering these questions:

Quote
Can you verify that input is not affected if you disconnect the output ?

Disconnecting the output i understand as disconnecting the caps and bulb from the seondaries.
The change is in the current (which flatlines now) and the power which is much less. The resonance frequency of L3 stays around 18KHz

Quote
Now can you test the same setup using the other signal gen that gives 30 volt signal ? The one on the vid used 5 volts, right ? At least this is seen on the scope.

I did the same test with the other SG at 30V, but the 5 V you saw on the scope yesterday was a RMS value wich equals 14v PP.
The rms value now was 8.45V which equals 24V pp.  input Power increased and output bulb was brighter.
Installing an input bulb showed a slight detune from the 18KHz in L3, but at resonance input bulb was never brighter then output bulb.

Video here: https://www.youtube.com/watch?v=fIy0SSh77SM&feature=youtu.be


At this point i stopped as i think we first should agree on how to connect the both secondaries together.
According to MileHigh, now (high resonance point) the both secondaries are wrongly connected.


Facts:

when all coils are open (nothing connected), the both secondaries measure each:

713mH
1.2 Ohm
Shorting L3 does not make any difference.

When both secondaries are paralleled the way i had yesterday (high resonance frequence around 10KHz) they measure:

260uH   as in micro henries!!   
0.6 Ohm
Shorting L3 decreases inductance to 180uH

When both secondaries are parelleled the other way (low resonance frequency around 175Hz) they measure:

713mH!!  SHOULD THIS NOT BE HALF?
0.6 Ohm
Shorting L3 does not make any difference.

Video here:  https://www.youtube.com/watch?v=p03S3qmDBzU&feature=youtu.be   

So my question is:

what is the correct way to hook up the both secondaries?


Regards Itsu
« Last Edit: March 08, 2014, 10:58:41 AM by itsu »

verpies

  • Hero Member
  • *****
  • Posts: 3473
Re: Lenzless resonant transformer
« Reply #174 on: March 08, 2014, 02:01:57 AM »
713mH!!  SHOULD THIS NOT BE HALF?
It should be so only if you are connecting two separate inductors in parallel (on two separate cores). 
Note, that winding direction is not an issue for completely separate inductors.

However, if there is any flux sharing between two inductors and/or any mutual inductance, then their relative winding direction becomes an issue and formulas for Mutually Coupled Inductors in Parallel should be used instead.

Shorting L3 does not make any difference.
Of course, because L3 affects only the leakage flux that closes outside of the core through the air (bar core saturation).

When L1 & L2 are connected in parallel then the leakage flux is minimal, however if L1 & L2 are connected in antiparallel then their fluxes oppose ("fight") each other and it is much "easier' for them to close outside of the core (through the air ) than go through the core. 
This situation creates a large leakage flux outside of the core ...and this leakage flux can be easily affected by L3.

what is the correct way to hook up the both secondaries?
"Correct" to maximize what parameter?

verpies

  • Hero Member
  • *****
  • Posts: 3473
Re: Lenzless resonant transformer
« Reply #175 on: March 08, 2014, 02:52:33 AM »
I wrote this before but I will repeat it again:
Mutual magnetic coupling between two transformer windings, "consumes" the free inductance of the secondary winding, that otherwise would be available to form an LC Tank with a capacitor connected to this secondary winding.

In an extreme case (when the the mutual coupling coefficient = 1) the entire inductance of the secondary winding becomes "consumed" (converted to an EMF source) and there is no free inductance remaining in it to form an LC Tank with a capacitor connected to this winding.
The frequency response of such secondary circuit is the same as of a lone capacitor (no resonant peaks) - Itsu has witnessed this response when his windings were wound to maximize the mutual coupling coefficient.

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Lenzless resonant transformer
« Reply #176 on: March 08, 2014, 06:13:10 AM »
Itsu:

Thanks and I am not surprised that you also tried crossing the wires.  I realize you can't put everything in your clips.  So you appear to get a much lower resonant frequency but the amplitude is very low.  Something is going on, perhaps your investigations will figure it it.  Of course it's hard to visualize all the nuances associated with these things in your head.  I can only suggest that there is still something funny going on because the the L1 and L2 are seeing clockwise and counter-clockwise flux at the same instant from L3, while at the same time they are magnetically connected by the toroid.  So if L1 is driving a load, the counter-flux induced in L1 also flows into L2, etc, etc.

Verpies:

Quote
Mutual magnetic coupling between two transformer windings, "consumes" the free inductance of the secondary winding, that otherwise would be available to form an LC Tank with a capacitor connected to this secondary winding.

Indeed, that means my posting #165 is bogus.  To be honest I had some nagging doubts about it.  I realize that when the capacitor is discharging energy and sending it back to the source, it sees a real load.  The toroid couples the AC power back to the source.  So I suppose the instantaneous impedance the capacitor sees is the output impedance of the signal generator (typically a 50-ohm resistor) plus the impedance caused by the instantaneous output voltage of the signal generator signal.  And of course there is a transformer turns ratio to factor in also.

So in this case the L1 or L2 secondary is not an inductance, it's just part of the "gear system" that either transmits AC power forward into a target load, or it transmits AC power "backwards" into the signal generator source, which also looks like a load to the capacitor.  I use the term "gear system" because a great analogy for a transformer is a set of meshed gears.

In my opinion, at this point what's really happening is Itsu is exploring the phasor relationships in the power flow as he tries different configurations.  There is no pot of gold at the end of the rainbow, but it's still interesting.  There are a couple of ways to do the math behind the way the phasors act and react.  I did it in anther lifetime!  lol

MileHigh

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Lenzless resonant transformer
« Reply #177 on: March 08, 2014, 08:13:17 AM »
Itsu:

I watched clip 16 and I figured it out, but I am too tired to write it up.  Still lots of fighting going on.  If it ain't flux fighting, then it's EMF fighting.  There is your big clue.

MileHigh

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #178 on: March 08, 2014, 12:33:15 PM »
It should be so only if you are connecting two separate inductors in parallel (on two separate cores). 
Note, that winding direction is not an issue for completely separate inductors.

However, if there is any flux sharing between two inductors and/or any mutual inductance, then their relative winding direction becomes an issue and formulas for Mutually Coupled Inductors in Parallel should be used instead.
Of course, because L3 affects only the leakage flux that closes outside of the core through the air (bar core saturation).

Ok, great link, it explains exactly what is happening in this situation:

parallel aiding    LT = L1 = L2 = M  meaning 713mH whether they are paralleled or not, L stays the same

parallel opposing  LT = ( L ± M ) ÷ 2 meaning zero with perfect components, and in my real world 260uH (this was how my tests were done yesterday)
(This time, if the two inductances are equal in value and the magnetic coupling is perfect between them,
the equivalent inductance and also the self-induced emf across the inductors will be zero as the two
inductors cancel each other out. )

Quote
"Correct" to maximize what parameter?

good question, i leave that to Jack, but my feeling is telling me that we should go for the parallel aiding = 713mH setup,
i mean why should we want to cancel out anything?

Thanks,  regards Itsu

« Last Edit: March 08, 2014, 07:00:06 PM by itsu »

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Lenzless resonant transformer
« Reply #179 on: March 08, 2014, 12:37:07 PM »
Itsu:

I watched clip 16 and I figured it out, but I am too tired to write it up.  Still lots of fighting going on.  If it ain't flux fighting, then it's EMF fighting.  There is your big clue.

MileHigh

Hmmm,   in "parallel opposing" yes, both flux fighting and EMF fighting going on as mentioned earlier, but when in "parallel aiding", all should be fine and dandy, right?

Regards Itsu