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

Variable Dielectric Capacitor?

Expand Messages
  • Jim Miller
    I ve been thinking about the maximizing capacitor efficiency for a STL for mostly QRP but perhaps as high as 50W. One thing I haven t seen tried is keeping the
    Message 1 of 12 , Dec 9, 2005
    • 0 Attachment
      I've been thinking about the maximizing capacitor efficiency for a STL for mostly QRP but perhaps as high as 50W. One thing I haven't seen tried is keeping the plates of the capacitor simple and well secured to the ends of the loop and moving a dielectric substance between them.

      I was thinking that you could move a piece of high quality dielectric between an otherwise airspaced set of plates. Teflon that nearly filled the interstitial area would approximately double the capacitance. Polystyrene would be a bit higher. Glass with a dielectric of 5-10 would have a much larger effect.

      If the dielectric piece were accurately sized or indexed it could be moved into place relatively quickly and a smaller piece of dielectric material could be moved to accomplish the quite small (1-2 pf) variability needed for tuning without as much motion sensitivity.

      The advantage to this would be fixed, well coupled capacitor plates to maximize efficiency.

      The capacitance change to accomodate 30M operation on a 20M loop requires an approximate doubling of capacitance. 40M requires another doulbing. It seems that a variable dielectric approach with a combination of glass and teflon dielectrics might be able to span this with the glass occupying none, 1/4 or 1/2 of the area of the plates and a wafer of teflon working in the other half of the area to provide the fine tuning.

      I'm sure there's some fringing effects that come into play that affect the ratios of areas and probably some issues for folks who like to run high power but does the basic approach seem feasible?

      73

      jim ab3cv

      [Non-text portions of this message have been removed]
    • aa5tb
      Hi Jim, Your ideas for varying the capacitance of a small transmiting loop antenna are good but I ve tried many things over the years and have discovered for
      Message 2 of 12 , Dec 9, 2005
      • 0 Attachment
        Hi Jim,

        Your ideas for varying the capacitance of a small transmiting loop
        antenna are good but I've tried many things over the years and have
        discovered for myself the following regarding the capacitor design in
        small transmitting loops antennas:

        1. Even the best solid dielectrics get warm above QRP power levels.

        2. Maybe only a few watts are being lost in the dielectric but, any
        temperature change whether due to RF heating or the sun will cause the
        capacitance to change.

        3. Due to the extremely high Q of the antenna, any capacitance change
        will cause the resonant frequency to change. At 100 W I found my
        high-Q antennas with various solid dielectrics drifting many kHz after
        several seconds of CW tranmission.

        Therefore, I have found it necessary to use only air or vacuum as a
        dielectric for any power levels above a few watts. Maybe for small
        changes with a mostly air dielectric and small dielectric pieces these
        negative effects can be minimized. I don't want to discourage you as
        you may be able solve some unique problems. What ever capacitor
        design you come up with it needs to handle very high voltage AND
        current (out of phase with each other of course) even for low power
        levels.

        73,
        Steve Yates - AA5TB
      • dldorrance
        OK, nobody else bit at this so I will put in my 2 cents. Assuming transmitting loops, the voltage between capacitor plates will be quite high. Polyethylene
        Message 3 of 12 , Dec 10, 2005
        • 0 Attachment
          OK, nobody else bit at this so I will put in my 2 cents.

          Assuming transmitting loops, the voltage between capacitor plates will
          be quite high. Polyethylene will not work; you get arcing which melts
          the polyethylene. Telflon will work, as it resists about 1000
          volts/mil thickness. However I have read that the Q of a loop
          made with Teflon dielectric caps is degraded. My limited personal
          experience with Teflon caps in loops seems to confirm this. Perhaps
          there is a plastics engineer on this list who could provide a more
          cogent explanation. Plastics have differing behavior with regard to
          HF fields, and I assume this may have something to do with the
          apparent degradation.

          It is often stated that all that is needed to test a plastic's ability
          to "handle RF" is the microwave oven test. If it heats up it's no
          good; if it doesn't it will work. I suspect the whole story is a bit
          more complicated.

          Dave WA6YSO

          --- In loopantennas@yahoogroups.com, "Jim Miller" <jim@j...> wrote:
          >
          > I've been thinking about the maximizing capacitor efficiency for a
          STL for mostly QRP but perhaps as high as 50W. One thing I haven't
          seen tried is keeping the plates of the capacitor simple and well
          secured to the ends of the loop and moving a dielectric substance
          between them.
          >
          > I was thinking that you could move a piece of high quality
          dielectric between an otherwise airspaced set of plates. Teflon that
          nearly filled the interstitial area would approximately double the
          capacitance. Polystyrene would be a bit higher. Glass with a
          dielectric of 5-10 would have a much larger effect.
          >
          > If the dielectric piece were accurately sized or indexed it could be
          moved into place relatively quickly and a smaller piece of dielectric
          material could be moved to accomplish the quite small (1-2 pf)
          variability needed for tuning without as much motion sensitivity.
          >
          > The advantage to this would be fixed, well coupled capacitor plates
          to maximize efficiency.
          >
          > The capacitance change to accomodate 30M operation on a 20M loop
          requires an approximate doubling of capacitance. 40M requires another
          doulbing. It seems that a variable dielectric approach with a
          combination of glass and teflon dielectrics might be able to span this
          with the glass occupying none, 1/4 or 1/2 of the area of the plates
          and a wafer of teflon working in the other half of the area to provide
          the fine tuning.
          >
          > I'm sure there's some fringing effects that come into play that
          affect the ratios of areas and probably some issues for folks who like
          to run high power but does the basic approach seem feasible?
          >
          > 73
          >
          > jim ab3cv
          >
          > [Non-text portions of this message have been removed]
          >
        • Jim Miller
          Hi Dave Thanks for diving in. I was beginning to think I d dropped the proverbial t**d in the holiday punch bowl! ;-) I did some research on dielectrics before
          Message 4 of 12 , Dec 10, 2005
          • 0 Attachment
            Hi Dave

            Thanks for diving in. I was beginning to think I'd dropped the proverbial
            t**d in the holiday punch bowl! ;-)

            I did some research on dielectrics before posting and thats how i came up
            with glass, teflon and polystyrene as candidates. They each have sufficient
            dielectric strength to withstand the voltages involved and loss tangent
            about as low as it goes in the 100Mhz to 3Ghz range. Although polystyene has
            half the dielectric strength of teflon's 1kv/mil it appears to have a much
            lower loss tangent. I think I'll start with it as my first try as it's cheap
            and easy to work with.

            http://www.rfcafe.com/references/electrical/dielectric_constants_strengths.htm
            is the most comprehensive source of relavant material properties I could
            find.

            Loopcalc and several other calculators I've tried show around 3kv for a 1.5M
            diameter loop resonant at 40M @ 50W so a thickness of polystyrene of 10mils
            or greater should work.

            Since the "caps" in this scenario would be simple opposing copper plates
            1/8" apart and well soldered to flanges created at the ends of the soft
            copper tubing their intrinsic airspaced loss should be low. My first attempt
            will be just single set of opposing plates with a sheet of polystyrene
            gradually introduced to see how it affects the tuning.

            If this works out I've got some small stepper motors that will fit inside
            the copper tubing which could affect the motion necessary at least for a
            single band simple tuned approach.

            If nothing else it should keep the neighbors amused. 'll tell them it's some
            of my artist daughter's schulpture!

            73

            jim ab3cv
          • Jim Miller
            Steve Good points all. I ll post results as I get them. I m sure there s a reason why this isn t widespread so I m not expecting miracles! 73 jim ab3cv
            Message 5 of 12 , Dec 10, 2005
            • 0 Attachment
              Steve

              Good points all. I'll post results as I get them. I'm sure there's a reason
              why this isn't widespread so I'm not expecting miracles!

              73

              jim ab3cv
            • VFelix
              Hey Steve, why not try Polypropylene mylar caps instead, they are a better cap the electrolytic. :) Vince
              Message 6 of 12 , Dec 10, 2005
              • 0 Attachment
                Hey Steve, why not try " Polypropylene mylar caps " instead, they are a
                better cap the electrolytic. :)
                Vince



                aa5tb wrote:

                >Hi Jim,
                >
                > Your ideas for varying the capacitance of a small transmiting loop
                >antenna are good but I've tried many things over the years and have
                >discovered for myself the following regarding the capacitor design in
                >small transmitting loops antennas:
                >
                >< snip >
                >
                >
              • VFelix
                Hey Steve, why not try Polypropylene mylar caps instead, they are a much better cap then the electrolytic. Vince
                Message 7 of 12 , Dec 10, 2005
                • 0 Attachment
                  Hey Steve, why not try " Polypropylene mylar caps " instead, they are a
                  much better cap then the electrolytic.
                  Vince



                  aa5tb wrote:

                  >Hi Jim,
                  >
                  > Your ideas for varying the capacitance of a small transmiting loop
                  >antenna are good but I've tried many things over the years and have
                  >discovered for myself the following regarding the capacitor design in
                  >small transmitting loops antennas:
                  >
                  >
                  >
                  >
                • mdgolfbum
                  While poking around for sources of polystyrene I found that the ordinary and fragile clear plastic CD jewel cases are made of pure polystyrene! That was a nice
                  Message 8 of 12 , Dec 11, 2005
                  • 0 Attachment
                    While poking around for sources of polystyrene I found that the
                    ordinary and fragile clear plastic CD jewel cases are made of pure
                    polystyrene!

                    That was a nice surprise! Who doesn't have a box full of those in the
                    basement? ;-)

                    fwiw

                    jim ab3cv
                  • Jim Miller
                    it occurred to me that single sided copper clad board might be suitable for the capacitor plates as long as the substrate material was not interior to the
                    Message 9 of 12 , Dec 11, 2005
                    • 0 Attachment
                      it occurred to me that single sided copper clad board might be suitable for
                      the capacitor plates as long as the substrate material was not interior to
                      the capacitor plates and subject to the fields. i don't think it should
                      contribute to loss in that postion.

                      comments?

                      tnx

                      jim ab3cv
                    • martin.sivak
                      Hi, did anybody actually tried this back in 2005 or later? Are there any results? This thread is pretty old and seems to lack any conclusion :) I was planning
                      Message 10 of 12 , May 16, 2013
                      • 0 Attachment
                        Hi,

                        did anybody actually tried this back in 2005 or later? Are there any results? This thread is pretty old and seems to lack any conclusion :)

                        I was planning to try to do exactly this (variable dielectric capacitor) and found this thread while doing my research.

                        >
                        > it occurred to me that single sided copper clad board might be suitable for
                        > the capacitor plates as long as the substrate material was not interior to
                        > the capacitor plates and subject to the fields. i don't think it should
                        > contribute to loss in that postion.

                        Exactly my thinking, it might make one very cheap variable capacitor that can withstand both high voltage and high current.

                        My plan is to take two copper clad boards, separate them using plastic spacers (4 or 5mm) and slide polystyrene board between the copper planes.

                        According to the resources (links below) I found online, polystyrene has tangent loss of about 2 - 5 x 10^-4 and relative permitivity of about 2.5. That is actually not bad at all as Mica has "similar" loss properties.

                        http://www.rfcafe.com/references/electrical/dielectric-constants-strengths.htm
                        http://www.kayelaby.npl.co.uk/general_physics/2_6/2_6_5.html
                        http://www.eccosorb.eu/sites/default/files/files/dielectric-chart.pdf

                        Polystyrene also supposedly has much higher dielectric strength (197kV/cm) than air (10-30kV/cm) so arcing does not have to be an issue (especially for QRP or low power) as well.

                        73!

                        Martin OK7MS
                      • William Young
                        No, I haven t tried a variable dielectric capacitor, but I have tried what I called a floating rotor variable capacitor with a metal plate rotating into and
                        Message 11 of 12 , May 16, 2013
                        • 0 Attachment
                          No, I haven't tried a variable dielectric capacitor, but I have tried what I called a "floating rotor" variable capacitor with a metal plate rotating into and out of the space between two fixed capacitor plates.  The resulting change in capacitance for 180 degrees of rotation is small but very useful for fine tuning a receiver.  See QEX May/June 2002, page 31.  Bill Young WD5HOH 
                           

                          To: loopantennas@yahoogroups.com
                          From: mars@...
                          Date: Thu, 16 May 2013 12:45:09 +0000
                          Subject: [loopantennas] Re: Variable Dielectric Capacitor?

                           


                          Hi,

                          did anybody actually tried this back in 2005 or later? Are there any results? This thread is pretty old and seems to lack any conclusion :)

                          I was planning to try to do exactly this (variable dielectric capacitor) and found this thread while doing my research.

                          >
                          > it occurred to me that single sided copper clad board might be suitable for
                          > the capacitor plates as long as the substrate material was not interior to
                          > the capacitor plates and subject to the fields. i don't think it should
                          > contribute to loss in that postion.

                          Exactly my thinking, it might make one very cheap variable capacitor that can withstand both high voltage and high current.

                          My plan is to take two copper clad boards, separate them using plastic spacers (4 or 5mm) and slide polystyrene board between the copper planes.

                          According to the resources (links below) I found online, polystyrene has tangent loss of about 2 - 5 x 10^-4 and relative permitivity of about 2.5. That is actually not bad at all as Mica has "similar" loss properties.

                          http://www.rfcafe.com/references/electrical/dielectric-constants-strengths.htm
                          http://www.kayelaby.npl.co.uk/general_physics/2_6/2_6_5.html
                          http://www.eccosorb.eu/sites/default/files/files/dielectric-chart.pdf

                          Polystyrene also supposedly has much higher dielectric strength (197kV/cm) than air (10-30kV/cm) so arcing does not have to be an issue (especially for QRP or low power) as well.

                          73!

                          Martin OK7MS


                        • kevin asato
                          Should work. By theory a capacitor is nothing more than 2 opposite potential surfaces separated by a dielectric (air, fiberglass, some sort of plastic, etc).
                          Message 12 of 12 , May 17, 2013
                          • 0 Attachment
                            Should work. By theory a capacitor is nothing more than 2 opposite potential surfaces separated by a dielectric (air, fiberglass, some sort of plastic, etc). Using square plates will produce very non linear capacitance values as capacitance is affected by the area of the opposing plates. PITA to try to calculate, too! One thing to be aware with what you are planning is that you may have to account for air acting as a dielectric when you insert and remove the polystyrene between the two plates. May not be a large contribution to the overall capacitance but needs to be looked at. Not trying to talk you out of trying this. Let us know how it turns out!

                            >My plan is to take two copper clad boards, separate them using plastic spacers (4 or 5mm) and slide polystyrene board between the copper planes.

                            I took a basic electricity course seemingly eons ago where we took two sheets of aluminum foil and slightly larger pieces of wax paper. The wax paper was inserted between the foil and the 2nd sheet was laid on top of the foil of what would be the inner surface of the capacitor. The entire assembly was rolled/folded up into a tubular capacitor type form and jumper wires were connected, one to each sheet of foil. The capacitor was then connected to a low current assembly with a known value resistor and an NE-2 neon bulb. We could then see the neon bulb flash at a rate corresponding to the value of the capacitors made by the class. Since this class was geared towards elementary school children, we did not try to calculate/approximate the dielectric value of wax paper, though in retrospect, it would be a good exercise for high school physics.

                            73,
                            kevin
                            kc6pob


                          Your message has been successfully submitted and would be delivered to recipients shortly.