Loading ...
Sorry, an error occurred while loading the content.

MEG resonance and MOVS

Expand Messages
  • carbonprobe
    My MEG experiments have been showing me that tuning the MEG is pretty difficult. In my pursuit to replicate Naudin s MEG vers 3.0, my best MEG set up hit a
    Message 1 of 4 , Aug 29, 2002
    • 0 Attachment
      My MEG experiments have been showing me that tuning the MEG is pretty
      difficult. In my pursuit to replicate Naudin's MEG vers 3.0, my best
      MEG set up hit a resonance frequency at 14.2 KHz with a Panasonic ERZ-
      V10D621 (620 Volt MOV) as a load on each output coil. Naudin's
      frequency was ~ 22KHz using conditioned resistors (which is my next
      step). The reason I'm not hitting Naudin's 22KHz (I'm using the same
      winding ratio as he did) could be 1) load on output coils 2)Parasitic
      Capacitance in output coils. He used conditioned resistors, and I
      think this is a very big part of it. Using MOVS as a load on the
      output coils causes current and voltage to be out of phase 90 degrees
      and this is what I've been measuring. And when you have current and
      voltage 90 degrees out of phase, power is calculated to be zero. P =
      I V cos 90 = 0. My guess is all the power is being dissipated through
      the MOV as heat because it gets very hot. I used a 68 Ohm carbon
      resistor in series with the MOV to measure output current. And the
      current through it was calculated as 5mA. In the setup I've grounded
      the output coils. The voltage between the MOV and ground is about
      1.68KVpp for each side. The voltage between ground and the 68 Ohm
      resistor is 345mVrms. The current through the MOV and resistor leads
      the voltage at the MOV by 90 degrees. So there is a capacitance that
      is messing things up. There is no significant change if the outputs
      are ungrounded, the waveform just looks dirtier and it's more
      difficult to measure the voltage across the resistor. Measurements
      were made using an HP digital scope (from work) and using 4 probes (2
      per side). Experimenting by removing the capaciatance of a probe by
      just using a a wire and resistance for a probe still yields a 90
      degree phase difference between I and V. (so probe capacitance isn't
      affecting readings).

      The output wave forms look exactly like those drawn in the MEG patent
      with the following exceptions:
      1. the output coils' current and voltage is 90 degrees out of phase.
      2. the 2 spikes per cycle in the current waveforms occur when the
      voltage wave forms are at negative and positive peaks.

      Using an 1800V MOV, the resonance frequency changes slightly to
      14.8KHz and the output voltage is much much larger. Using lower rated
      MOVS cause the waveform to look less sinusoidal (and I think the
      resonance frequency was actually lower, forgot to take notes). The
      620V MOV worked the best. When both sides were unloaded the resonance
      frequency was about 22KHz.


      If my measurements are correct it seems you cannot use MOVS as a load
      on the output coils of the MEG. This is because they have reactance,
      and you can even measure the capacitance with a multimeter. I've
      read many posts on the Web stating that you cannot use reactive loads
      on the MEG (which makes sense). So I don't know why Naudin states
      that you can use a MOV (He says the load must be non-inductive).


      I've measured the Capacitance, Inductance, and Q of different rated
      MOVS I've been using and here are the results:



      *******Using BK Precision 878 LCR Meter*******


      CAPACITANCE
      120HZ 1KHZ
      ------------------------------------------
      130V 1.9nF 1.89nF
      MOV

      620V 0.105nF 0.105nF
      MOV

      1800V 0.11nF 0.076nF
      MOV

      68 Ohm 150nF 5nF
      Carbon
      Resistor




      INDUCTANCE
      120HZ 1KHZ
      ----------------------------------------------
      130V 880H 13H
      MOV

      620V 9428H 233H
      MOV

      1800V >10000H 322H
      MOV

      68 Ohm 53uH 3uH
      Carbon
      Resistor


      Q
      120HZ 1KHZ
      -----------------------------------------------
      130V 29 62
      MOV

      620V 1 55
      MOV

      1800V 3 36
      MOV

      68 Ohm 0 0
      Carbon
      Resistor







      *******Using Fluke 189 Multimeter*******


      CAPACITANCE
      --------------------------------------------------
      130V 2nF
      MOV

      620V 0.19nF
      MOV

      1800V 0.16nF
      MOV

      68 Ohm OL (does not register)
      Carbon
      Resistor




      As can be seen, the Inductance decreases in the MOVS a large amount
      as the frequency increases. But the Capacitance hardly changes at all
      with increasing frequency. And this is really what causes the
      problems. At 14.2KHz the reactance does not reach a negligible level
      in the MOVS as I have measured my voltage and current 90 degrees out
      of phase.

      There is the possibility that the capacitance of my output coils are
      adding to the problem as I have used about 6 layers of electrical
      tape within the coils to prevent arcing. And there is about 5 layers
      of electrical tape between the core and the first layer of the output
      coils. But I can measure no Capacitance in the output coils with the
      Fluke 189 Meter just like I measured no capacitance in the carbon
      resistor. But when I use the BK LCR meter things are different:



      Using BK Precision 878 LCR Meter on the RIGHT OUTPUT COIL:


      Space
      between
      C-Cores WITH MAGNET
      ---------------------------------------------------
      No Space C at 120Hz = 198nF
      C at 1KHz = 2.8nF
      L at 120Hz = 8.6H
      L at 1KHz = 8.82H

      0.5mm C at 120Hz = 708nF
      C at 1KHz = 10.38nF
      L at 120Hz= ?
      L at 1KHz = 2.39H

      0.25 inch C at 120Hz = 2.37uF
      C at 1KHz = 35nF
      L at 120Hz = 715mH
      L at 1KHz = 709mH


      Space
      between
      C-Cores NO MAGNET
      ------------------------------------------------
      No Space C at 120Hz = 206nF
      C at 1KHz = 2.9nF
      L at 120Hz = 8.29H
      L at 1KHz = 8.48H

      0.5mm C at 120Hz = 706nF
      C at 1KHz = 10.3nF
      L at 120Hz = ?
      L at 1KHz = 2.41H

      0.25 inch C at 120Hz = 2.3uF
      C at 1KHz = 34.8nF
      L at 120Hz = 721mH
      L at 1KHz = 716mH




      The left coil had similar measurements. The above measurements were
      taken when both outcoils were reduced to 1300 windings. Input coils
      are 100 turns. Originally both output coils had 1500 windings. I
      wanted to see if taking 200 windings off each side affected the rez
      frequency. But it didn't seem to affect much, it was still around
      14.2KHz. I plan on using thin mylar tape to insulate between each
      layer on my next build to minimize capacitance. But i think using
      conditioned resistors is really going to make the difference.

      In an earlier post I mentioned that I was seeing a voltage decrease
      when I took the magnet out. In my current experiments, I see no
      change in voltage or current with the magnet in or out.

      Does anybody know where Naudin got his 100K Ohm 5 watt carbon
      resistors? The closest I can find is 2 Watt.

      Ken
    • BobW
      Carbonprobe Here is a place that claims to have the 5watt resistors: http://www.micro-ohm.com/filmprecision/rj.html That is a fine report you just posted.
      Message 2 of 4 , Aug 30, 2002
      • 0 Attachment
        Carbonprobe
         
        Here is a place that claims to have the 5watt resistors: http://www.micro-ohm.com/filmprecision/rj.html
         
        That is a fine report you just posted.  Would it be possible for you to share schematics showing the scope probe hook up points to augment the description.  I think I got it but...
         
        On your surprise to find the current and voltage out of phase in the output, I am surprised to hear that it should not be.  I am currently planning a lab test to show it for sure but at this point it seems to me the output coil is a great big coil and anytime you have AC flowing in a circuit with L and R the current should be shifted by 90 degrees.  Cyril says this is not so in a transformer.
         
        On your remarks about the MOV's causing distortion, It seems to me that any non linier load will produce a distortion in the output voltage since it does not conduct for the full cycle.
         
        In the tabular section, did you mean to say uH.  Henry seems a bit huge. :}
         
        Bob
      • mayerstan
        ... you to ... the ... the output, ... planning a lab ... output coil is ... L and R ... not so in a ... that any ... since it ... huge. :} ... Bob, The
        Message 3 of 4 , Aug 30, 2002
        • 0 Attachment
          --- In MEG_builders@y..., "BobW" <rwinches@t...> wrote:
          > Carbonprobe
          >
          > Here is a place that claims to have the 5watt resistors:
          > http://www.micro-ohm.com/filmprecision/rj.html
          >
          > That is a fine report you just posted. Would it be possible for
          you to
          > share schematics showing the scope probe hook up points to augment
          the
          > description. I think I got it but...
          >
          > On your surprise to find the current and voltage out of phase in
          the output,
          > I am surprised to hear that it should not be. I am currently
          planning a lab
          > test to show it for sure but at this point it seems to me the
          output coil is
          > a great big coil and anytime you have AC flowing in a circuit with
          L and R
          > the current should be shifted by 90 degrees. Cyril says this is
          not so in a
          > transformer.
          >
          > On your remarks about the MOV's causing distortion, It seems to me
          that any
          > non linier load will produce a distortion in the output voltage
          since it
          > does not conduct for the full cycle.
          >
          > In the tabular section, did you mean to say uH. Henry seems a bit
          huge. :}
          >
          > Bob

          ------------------------------------------------
          Bob,

          The resistors that I think carbonprobe is looking for are CARBON
          COMPOSITION type. Naudin specifies carbon composition. I had
          absolutely no luck in finding 5 watters and in reviewing Naudin's
          photos of his resistors and doing some subjective scaling, I think
          that Naudin's resistors were 2 watters.

          Also re: the high inductance measurements of the secondaries, my MEGs
          had similar measurements.

          Very best,

          Stan
        • BobW
          Stan I agree with you that the inductance of the output coil is large but Ken said he was measuring the characteristics of the MOV. Bob Also re: the high
          Message 4 of 4 , Aug 30, 2002
          • 0 Attachment
            Stan
            I agree with you that the inductance of the output coil is large but Ken said he was measuring the characteristics of the MOV.
            Bob 


            Also re: the high inductance measurements of the secondaries, my MEGs
            had similar measurements.
          Your message has been successfully submitted and would be delivered to recipients shortly.