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

Re: What about charge amps?

Expand Messages
  • qw1209po
    Thanks, Jim. Looking forward to hearing about the gotchas when you get a chance. I found this document by searching, and though it applies to much higher
    Message 1 of 24 , Sep 1, 2011
    • 0 Attachment
      Thanks, Jim. Looking forward to hearing about the "gotchas" when you get a chance.

      I found this document by searching, and though it applies to much higher frequencies (X-Rays!), still has interesting things to say about the charge amplifier:

      http://sales.hamamatsu.com/assets/applications/SSD/charge_amp_kacc9001e01.pdf

      One of the major benefits seems to be transient response, which you've demonstrated. I'm particularly curious about S/N and Fq response.
    • Zapnspark
      Continuing -- Some random thoughts that occur to me: * correction. I said the square wave scope shot was 1 kHz. It s 2 kHz. The square wave shot indicates
      Message 2 of 24 , Sep 1, 2011
      • 0 Attachment
        Continuing --

        Some random thoughts that occur to me:

        * correction. I said the square wave scope shot was 1 kHz. It's 2 kHz.
        The square wave shot indicates that the OCCAM-3b head amp response easily extends to 20 kHz. There are OCCAM-3 simulation screen captures in the Photos > Zapnspark section.

        * Microphone, P48 powered charge amplifiers need a high open loop gain, low noise, low offset drift, 20 kHz bandwidth and low power drain. Those criteria narrow the field for the choice of op-amps. Luckily, there are recently introduced op-amps that fit those requirements. Some of the really good ones are surface mount types though.

        * There can be stability/oscillation issues if you don't watch those poles and zeros in the design (e.g. phase shifts).

        * The type of JFET front end is less critical in a charge amplifier design. It functions mostly a trans-impedance device and further, it keeps a very linear transfer characteristic (due to being in that tight feedback loop) JFET gate capacitance is less of an issue because the gate is at a virtual ground. Same with stray capacitance issues.

        * Low frequency response is extended. Sub audible sounds could be a problem. It's tough to do a standard LF roll-off filter in that tight feedback loop. Lowering the 1 G resistor value might help or - a more complex approach that is used in the ZM-47 design (Neumann style feedback introduced into the capsule polarizing supply)

        I hope that's useful information.

        Cheers.

        Jim G.
        ZAPNSPARK


        --- In micbuilders@yahoogroups.com, "Zapnspark" <zapnspark@...> wrote:
        >
        > OK.
        >
        > I'll just give this a quick try for now.
        >
        > Have a look at these scope shots of an OCCAM-3b microphone circuit:
        >
        > http://goo.gl/o167m
        >
        > I introduced a 2 kHz triangle wave into the OCCAM-3b head amplifier and the traces are from pin 2 and pin 3, respectively, of the mic. XLR connector. (The mic is connected to a MACKIE 1202-VLZ pro mixer, P48 phantom power)
        > These signal traces are shown just at the verge of clipping.
        > The peak to peak amplitude on pins 2 and 3 are 5 volts p-p or - 10 volts p-p differential. Note the linearity of the triangle waveform right up to the threshold of the clipping level.
        >
        > http://goo.gl/EBA2a
        >
        > This is the 1 kHz square wave response at pins 2 and 3 of the XLR connector. No ringing or other obvious artifacts to be seen.
        >
        > There are some pro microphones will not give such nice results with these simple waveform tests.
        >
        > It's getting late for me. I'll go into some further detail in future posts or -- even better -- how about a real mic. guru jumping in here with additional information on microphone charge amplifiers?
        >
        > Charge amplifiers ain't perfection. As with most microphone circuits, there are some drawbacks and gotchas associated with this type of circuit.
        >
        > For now,
        >
        > Cheers.
        >
        > Jim G.
        > ZAPNSPARK
        >
        > --- In micbuilders@yahoogroups.com, Robb Nichols <Robb@> wrote:
        > >
        > > Thanks, Steve, for reinforcing my questions. Though data like this
        > > would certainly be interesting, I'd be happy with any comments, however
        > > subjective in nature. I'm especially curious how sensitive this type of
        > > circuit is to variations in FETs and how rugged it is. Is linearity and
        > > dynamic range comparable to voltage-mode amps? I guess this is just
        > > paraphrasing Steve's question, but I'd also be interested in any
        > > subjective opinions.
        > >
        > > Kind regards, Robb
        > >
        > > On 8/31/2011 12:46 AM, qw1209po wrote:
        > > > Do we have hard data on how much better the OCCAM circuit is to the
        > > > "traditional" Schoeps circuit? Are there comparative S/N, Fq response,
        > > > distortion and output level figures?
        > > >
        > > > I'm in the process of putting together a few mics and at this point, I
        > > > could go either way. OCCAM is a simpler build because of the lack of
        > > > calibration and it's about the same in complexity as the Schoeps.
        > >
        >
      • LESLIE WATTS
        Jim, Makes me think of Prodigy discussion earlier this year. I think we are familiar with voltage amplifier/capsule distortion mechanism with stray or dead
        Message 3 of 24 , Sep 1, 2011
        • 0 Attachment
          Jim,

          Makes me think of Prodigy discussion earlier this year.

          I think we are familiar with voltage amplifier/capsule distortion mechanism with stray or dead capacitance. (Nelville Fletcher's curved backplates
          and reduced backplate diameter compared to diaphragm help)

          But seemingly a charge amplifier will have a distortion mechanism as gain is modulated by capsule capacitance change/ feedback capacitor.
          Pure second harmonic?

          Haven't analyzed fully...

          Les

          ----- Original Message -----
          From: Zapnspark
          To: micbuilders@yahoogroups.com
          Sent: Thursday, September 01, 2011 12:30 PM
          Subject: [micbuilders] Re: What about charge amps?



          Continuing --

          Some random thoughts that occur to me:

          * correction. I said the square wave scope shot was 1 kHz. It's 2 kHz.
          The square wave shot indicates that the OCCAM-3b head amp response easily extends to 20 kHz. There are OCCAM-3 simulation screen captures in the Photos > Zapnspark section.

          * Microphone, P48 powered charge amplifiers need a high open loop gain, low noise, low offset drift, 20 kHz bandwidth and low power drain. Those criteria narrow the field for the choice of op-amps. Luckily, there are recently introduced op-amps that fit those requirements. Some of the really good ones are surface mount types though.

          * There can be stability/oscillation issues if you don't watch those poles and zeros in the design (e.g. phase shifts).

          * The type of JFET front end is less critical in a charge amplifier design. It functions mostly a trans-impedance device and further, it keeps a very linear transfer characteristic (due to being in that tight feedback loop) JFET gate capacitance is less of an issue because the gate is at a virtual ground. Same with stray capacitance issues.

          * Low frequency response is extended. Sub audible sounds could be a problem. It's tough to do a standard LF roll-off filter in that tight feedback loop. Lowering the 1 G resistor value might help or - a more complex approach that is used in the ZM-47 design (Neumann style feedback introduced into the capsule polarizing supply)

          I hope that's useful information.

          Cheers.

          Jim G.
          ZAPNSPARK

          --- In micbuilders@yahoogroups.com, "Zapnspark" <zapnspark@...> wrote:
          >
          > OK.
          >
          > I'll just give this a quick try for now.
          >
          > Have a look at these scope shots of an OCCAM-3b microphone circuit:
          >
          > http://goo.gl/o167m
          >
          > I introduced a 2 kHz triangle wave into the OCCAM-3b head amplifier and the traces are from pin 2 and pin 3, respectively, of the mic. XLR connector. (The mic is connected to a MACKIE 1202-VLZ pro mixer, P48 phantom power)
          > These signal traces are shown just at the verge of clipping.
          > The peak to peak amplitude on pins 2 and 3 are 5 volts p-p or - 10 volts p-p differential. Note the linearity of the triangle waveform right up to the threshold of the clipping level.
          >
          > http://goo.gl/EBA2a
          >
          > This is the 1 kHz square wave response at pins 2 and 3 of the XLR connector. No ringing or other obvious artifacts to be seen.
          >
          > There are some pro microphones will not give such nice results with these simple waveform tests.
          >
          > It's getting late for me. I'll go into some further detail in future posts or -- even better -- how about a real mic. guru jumping in here with additional information on microphone charge amplifiers?
          >
          > Charge amplifiers ain't perfection. As with most microphone circuits, there are some drawbacks and gotchas associated with this type of circuit.
          >
          > For now,
          >
          > Cheers.
          >
          > Jim G.
          > ZAPNSPARK
          >
          > --- In micbuilders@yahoogroups.com, Robb Nichols <Robb@> wrote:
          > >
          > > Thanks, Steve, for reinforcing my questions. Though data like this
          > > would certainly be interesting, I'd be happy with any comments, however
          > > subjective in nature. I'm especially curious how sensitive this type of
          > > circuit is to variations in FETs and how rugged it is. Is linearity and
          > > dynamic range comparable to voltage-mode amps? I guess this is just
          > > paraphrasing Steve's question, but I'd also be interested in any
          > > subjective opinions.
          > >
          > > Kind regards, Robb
          > >
          > > On 8/31/2011 12:46 AM, qw1209po wrote:
          > > > Do we have hard data on how much better the OCCAM circuit is to the
          > > > "traditional" Schoeps circuit? Are there comparative S/N, Fq response,
          > > > distortion and output level figures?
          > > >
          > > > I'm in the process of putting together a few mics and at this point, I
          > > > could go either way. OCCAM is a simpler build because of the lack of
          > > > calibration and it's about the same in complexity as the Schoeps.
          > >
          >





          [Non-text portions of this message have been removed]
        • Zapnspark
          Leslie. I recall that Prodigy Pro discussion. At that time, I passed the second harmonic argument by a physics guru. He had become interested in the finer
          Message 4 of 24 , Sep 1, 2011
          • 0 Attachment
            Leslie.

            I recall that Prodigy Pro discussion.
            At that time, I passed the "second harmonic" argument by a physics guru. He had become interested in the finer details of how condenser microphones work.
            I'll paraphrase as best I can:
            " - any second harmonic effects of the tiny diaphragm movement is extremely small compared to the other artifacts that a condenser capsule can produce"

            I think he was saying that it would probably be inaudible.
            I dunno. My math skills are so badly rusted that I can't even begin to tackle that.
            I would be interested in your opinion if you look into this further.

            Thanks.

            Jim G.
            ZAPNSPARK


            --- In micbuilders@yahoogroups.com, "LESLIE WATTS" <leswatts@...> wrote:
            >
            > Jim,
            >
            > Makes me think of Prodigy discussion earlier this year.
            >
            > I think we are familiar with voltage amplifier/capsule distortion mechanism with stray or dead capacitance. (Nelville Fletcher's curved backplates
            > and reduced backplate diameter compared to diaphragm help)
            >
            > But seemingly a charge amplifier will have a distortion mechanism as gain is modulated by capsule capacitance change/ feedback capacitor.
            > Pure second harmonic?
            >
            > Haven't analyzed fully...
            >
            > Les
            >
            > ----- Original Message -----
            > From: Zapnspark
            > To: micbuilders@yahoogroups.com
            > Sent: Thursday, September 01, 2011 12:30 PM
            > Subject: [micbuilders] Re: What about charge amps?
            >
            >
            >
            > Continuing --
            >
            > Some random thoughts that occur to me:
            >
            > * correction. I said the square wave scope shot was 1 kHz. It's 2 kHz.
            > The square wave shot indicates that the OCCAM-3b head amp response easily extends to 20 kHz. There are OCCAM-3 simulation screen captures in the Photos > Zapnspark section.
            >
            > * Microphone, P48 powered charge amplifiers need a high open loop gain, low noise, low offset drift, 20 kHz bandwidth and low power drain. Those criteria narrow the field for the choice of op-amps. Luckily, there are recently introduced op-amps that fit those requirements. Some of the really good ones are surface mount types though.
            >
            > * There can be stability/oscillation issues if you don't watch those poles and zeros in the design (e.g. phase shifts).
            >
            > * The type of JFET front end is less critical in a charge amplifier design. It functions mostly a trans-impedance device and further, it keeps a very linear transfer characteristic (due to being in that tight feedback loop) JFET gate capacitance is less of an issue because the gate is at a virtual ground. Same with stray capacitance issues.
            >
            > * Low frequency response is extended. Sub audible sounds could be a problem. It's tough to do a standard LF roll-off filter in that tight feedback loop. Lowering the 1 G resistor value might help or - a more complex approach that is used in the ZM-47 design (Neumann style feedback introduced into the capsule polarizing supply)
            >
            > I hope that's useful information.
            >
            > Cheers.
            >
            > Jim G.
            > ZAPNSPARK
            >
            > --- In micbuilders@yahoogroups.com, "Zapnspark" <zapnspark@> wrote:
            > >
            > > OK.
            > >
            > > I'll just give this a quick try for now.
            > >
            > > Have a look at these scope shots of an OCCAM-3b microphone circuit:
            > >
            > > http://goo.gl/o167m
            > >
            > > I introduced a 2 kHz triangle wave into the OCCAM-3b head amplifier and the traces are from pin 2 and pin 3, respectively, of the mic. XLR connector. (The mic is connected to a MACKIE 1202-VLZ pro mixer, P48 phantom power)
            > > These signal traces are shown just at the verge of clipping.
            > > The peak to peak amplitude on pins 2 and 3 are 5 volts p-p or - 10 volts p-p differential. Note the linearity of the triangle waveform right up to the threshold of the clipping level.
            > >
            > > http://goo.gl/EBA2a
            > >
            > > This is the 1 kHz square wave response at pins 2 and 3 of the XLR connector. No ringing or other obvious artifacts to be seen.
            > >
            > > There are some pro microphones will not give such nice results with these simple waveform tests.
            > >
            > > It's getting late for me. I'll go into some further detail in future posts or -- even better -- how about a real mic. guru jumping in here with additional information on microphone charge amplifiers?
            > >
            > > Charge amplifiers ain't perfection. As with most microphone circuits, there are some drawbacks and gotchas associated with this type of circuit.
            > >
            > > For now,
            > >
            > > Cheers.
            > >
            > > Jim G.
            > > ZAPNSPARK
            > >
            > > --- In micbuilders@yahoogroups.com, Robb Nichols <Robb@> wrote:
            > > >
            > > > Thanks, Steve, for reinforcing my questions. Though data like this
            > > > would certainly be interesting, I'd be happy with any comments, however
            > > > subjective in nature. I'm especially curious how sensitive this type of
            > > > circuit is to variations in FETs and how rugged it is. Is linearity and
            > > > dynamic range comparable to voltage-mode amps? I guess this is just
            > > > paraphrasing Steve's question, but I'd also be interested in any
            > > > subjective opinions.
            > > >
            > > > Kind regards, Robb
            > > >
            > > > On 8/31/2011 12:46 AM, qw1209po wrote:
            > > > > Do we have hard data on how much better the OCCAM circuit is to the
            > > > > "traditional" Schoeps circuit? Are there comparative S/N, Fq response,
            > > > > distortion and output level figures?
            > > > >
            > > > > I'm in the process of putting together a few mics and at this point, I
            > > > > could go either way. OCCAM is a simpler build because of the lack of
            > > > > calibration and it's about the same in complexity as the Schoeps.
            > > >
            > >
            >
            >
            >
            >
            >
            > [Non-text portions of this message have been removed]
            >
          • LESLIE WATTS
            Well, my math skills are rustier than they used to be even though I m a math type. Separating capsule capacitance into an active (function of diaphragm
            Message 5 of 24 , Sep 1, 2011
            • 0 Attachment
              Well, my math skills are rustier than they used to be even though I'm a math type.

              Separating capsule capacitance into an active (function of diaphragm amplitude) and passive can and has led to some
              wrong hypotheses as we know.

              I'll think about it. But i'll have to do the equations. Stray capacitance effects are known well enough with voltage amplifiers.

              Until I do an analysis I can't really say whether stray capacitance is a good thing, a bad thing, or a wash with
              virtual ground feedback charge to voltage converters.

              Let alone stuff like diaphragm collapse...

              Les
              ----- Original Message -----
              From: Zapnspark
              To: micbuilders@yahoogroups.com
              Sent: Thursday, September 01, 2011 3:34 PM
              Subject: [micbuilders] Re: What about charge amps?



              Leslie.

              I recall that Prodigy Pro discussion.
              At that time, I passed the "second harmonic" argument by a physics guru. He had become interested in the finer details of how condenser microphones work.
              I'll paraphrase as best I can:
              " - any second harmonic effects of the tiny diaphragm movement is extremely small compared to the other artifacts that a condenser capsule can produce"

              I think he was saying that it would probably be inaudible.
              I dunno. My math skills are so badly rusted that I can't even begin to tackle that.
              I would be interested in your opinion if you look into this further.

              Thanks.

              Jim G.
              ZAPNSPARK

              --- In micbuilders@yahoogroups.com, "LESLIE


              [Non-text portions of this message have been removed]
            • Sergei Steshenko
              I vaguely remember that mechanical second harmonic becomes a factor at 120db SPL. Regards,   Sergei.
              Message 6 of 24 , Sep 1, 2011
              • 0 Attachment
                I vaguely remember that mechanical second harmonic
                becomes a factor at 120db SPL.

                Regards,
                  Sergei.



                >________________________________
                >From: Zapnspark <zapnspark@...>
                >To: micbuilders@yahoogroups.com
                >Sent: Thursday, September 1, 2011 10:34 PM
                >Subject: [micbuilders] Re: What about charge amps?
                >
                >

                >Leslie.
                >
                >I recall that Prodigy Pro discussion.
                >At that time, I passed the "second harmonic" argument by a physics guru. He had become interested in the finer details of how condenser microphones work.
                >I'll paraphrase as best I can:
                >" - any second harmonic effects of the tiny diaphragm movement is extremely small compared to the other artifacts that a condenser capsule can produce"
                >
                >I think he was saying that it would probably be inaudible.
                >I dunno. My math skills are so badly rusted that I can't even begin to tackle that.
                >I would be interested in your opinion if you look into this further.
                >
                >Thanks.
                >
                >Jim G.
                >ZAPNSPARK
                >
              • Bhakakhan
                ...And you are one of the best explanators in the known universe. Thanks for this one... Johannes ... [Non-text portions of this message have been removed]
                Message 7 of 24 , Sep 1, 2011
                • 0 Attachment
                  ...And you are one of the best explanators in the known universe.

                  Thanks for this one...

                  Johannes
                  On 02.09.2011 10:09, Richard Lee wrote:
                  >
                  > A charge amp is one of the Classic Low Noise circuits for a capacitative
                  > source like a condensor capsule. It get very close to the theoretical
                  > limits for noise performance. It is really a virtual earth integrator.
                  >
                  > It can do this cos no evil resistors get in the way of electrons. Eg in
                  > the Schoeps circuit, the drain (and source to some extent) resistor(s)
                  > contribute significant noise.
                  >
                  > But electronic noise is only 1 part of the total noise performance of a
                  > mike. Guru Wurcer shows in
                  >
                  > http://www.linearaudio.net/current.php?volume=Volume%201
                  >
                  > how the capsule generates noise too. Volume 2 will show the Charge Amp.
                  >
                  > Because it is a well designed OPA circuit, it will give slightly more
                  > output before overload than the Schoeps.
                  >
                  > Response is not a problem with either unless you are incompetent.
                  >
                  > But the Charge Amp is not without disadvantages. You need to twiddle the
                  > bias too. see Zapnsparks circuits.
                  >
                  > Standing back as far as I can, the big disadvantage is that you need big
                  > electrolytics on the output. The Schoeps can use nice small electrolytics
                  > (or even Golden Pinnae parts for true believers).
                  >
                  > But this is a con shared by all mike circuits using OPAs.
                  >
                  > If you want to use an OPA, power from internal 9V batteries or want line
                  > level directly from a mike, a charge amp is good IF (and its a big IF)
                  > you
                  > can get at the gate of the FET without introducing nasties due to poor
                  > layout & wiring.
                  >
                  > If the capsule only allows 20dBA noise, the complexity isn't worth it.
                  > The
                  > borderline is probably around the excellent TSB160 / 165s.
                  >
                  > The Schoeps circuit is a finely tuned compromise and though I have
                  > pretended to study it since 1980, I still get "Ah ha!" moments when I
                  > look
                  > at it more closely.
                  >
                  > Wurcer is a true LN guru and has designed some of the lowest noise
                  > OPAs in
                  > the known universe for Analog Devices. Wannabe gurus should read his
                  > stuff
                  > daily on their knees facing Mecca.
                  >
                  >



                  [Non-text portions of this message have been removed]
                • Richard Lee
                  A charge amp is one of the Classic Low Noise circuits for a capacitative source like a condensor capsule. It get very close to the theoretical limits for
                  Message 8 of 24 , Sep 2, 2011
                  • 0 Attachment
                    A charge amp is one of the Classic Low Noise circuits for a capacitative
                    source like a condensor capsule. It get very close to the theoretical
                    limits for noise performance. It is really a virtual earth integrator.

                    It can do this cos no evil resistors get in the way of electrons. Eg in
                    the Schoeps circuit, the drain (and source to some extent) resistor(s)
                    contribute significant noise.

                    But electronic noise is only 1 part of the total noise performance of a
                    mike. Guru Wurcer shows in

                    http://www.linearaudio.net/current.php?volume=Volume%201

                    how the capsule generates noise too. Volume 2 will show the Charge Amp.

                    Because it is a well designed OPA circuit, it will give slightly more
                    output before overload than the Schoeps.

                    Response is not a problem with either unless you are incompetent.

                    But the Charge Amp is not without disadvantages. You need to twiddle the
                    bias too. see Zapnsparks circuits.

                    Standing back as far as I can, the big disadvantage is that you need big
                    electrolytics on the output. The Schoeps can use nice small electrolytics
                    (or even Golden Pinnae parts for true believers).

                    But this is a con shared by all mike circuits using OPAs.

                    If you want to use an OPA, power from internal 9V batteries or want line
                    level directly from a mike, a charge amp is good IF (and its a big IF) you
                    can get at the gate of the FET without introducing nasties due to poor
                    layout & wiring.

                    If the capsule only allows 20dBA noise, the complexity isn't worth it. The
                    borderline is probably around the excellent TSB160 / 165s.

                    The Schoeps circuit is a finely tuned compromise and though I have
                    pretended to study it since 1980, I still get "Ah ha!" moments when I look
                    at it more closely.

                    Wurcer is a true LN guru and has designed some of the lowest noise OPAs in
                    the known universe for Analog Devices. Wannabe gurus should read his stuff
                    daily on their knees facing Mecca.
                  • LESLIE WATTS
                    The issue for me is drums. I regularly see 140+ dBSPL at overheads with heavy hitters. Some capsules put out VOLTS. So what do I do? Switch the attenuators on,
                    Message 9 of 24 , Sep 2, 2011
                    • 0 Attachment
                      The issue for me is drums. I regularly see 140+ dBSPL at
                      overheads with heavy hitters. Some capsules put out VOLTS. So what do I do? Switch the attenuators on, slapping a couple hundred pF of capacitance across the capsule, thus enabling the distortion mechanism in voltage amplifier circuits.

                      I have not measured the distortion though. I do hear something, even though there is no gross clipping.

                      Les
                      ----- Original Message -----
                      From: Sergei Steshenko
                      To: micbuilders@yahoogroups.com
                      Sent: Thursday, September 01, 2011 6:53 PM
                      Subject: Re: [micbuilders] Re: What about charge amps?



                      I vaguely remember that mechanical second harmonic
                      becomes a factor at 120db SPL.

                      Regards,
                      Sergei.

                      >________________________________
                      >From: Zapnspark <zapnspark@...>
                      >To: micbuilders@yahoogroups.com
                      >Sent: Thursday, September 1, 2011 10:34 PM
                      >Subject: [micbuilders] Re: What about charge amps?
                      >
                      >
                      >
                      >Leslie.
                      >
                      >I recall that Prodigy Pro discussion.
                      >At that time, I passed the "second harmonic" argument by a physics guru. He had become interested in the finer details of how condenser microphones work.
                      >I'll paraphrase as best I can:
                      >" - any second harmonic effects of the tiny diaphragm movement is extremely small compared to the other artifacts that a condenser capsule can produce"
                      >
                      >I think he was saying that it would probably be inaudible.
                      >I dunno. My math skills are so badly rusted that I can't even begin to tackle that.
                      >I would be interested in your opinion if you look into this further.
                      >
                      >Thanks.
                      >
                      >Jim G.
                      >ZAPNSPARK
                      >




                      [Non-text portions of this message have been removed]
                    • LESLIE WATTS
                      Well, I have a little bit of information. This is a thesis from my Engineering alma mater UF. http://etd.fcla.edu/UF/UFE0017526/martin_d.pdf In this case they
                      Message 10 of 24 , Sep 2, 2011
                      • 0 Attachment
                        Well, I have a little bit of information.

                        This is a thesis from my Engineering alma mater UF.

                        http://etd.fcla.edu/UF/UFE0017526/martin_d.pdf

                        In this case they designed a dual backplate mems condenser mic and compared it with voltage and charge
                        amplifiers.

                        To review a few comparisons:

                        distortion:
                        a constant charge variable voltage capsule in the simple form of a moving flat plate has no inherent distortion mechanism unless there is stray capacitance. Used with a voltage amp.

                        A constant voltage variable charge capsule of the same type
                        does have an inherent distortion mechanism, but it's fairly small. Used with a charge amp.

                        Noise:

                        CCVV noise tends to go down with stray capacitance
                        CVVC noise tends to go up with stray capacitance

                        Bandwidth:

                        CVVC can have better low frequency response.

                        Numbers to quantify are in the thesis, but are only for low sensitivity compliance controlled pressure mics. I expect they would be very different in say a low tension mid tuned
                        resistance controlled pressure gradient mic.

                        All in all their device looks pretty darned good....with the voltage amp they had something like a 20 dBSPL self noise
                        with 165 dBSPL 3% distortion level. The charge amp ones were 20dB noisier due to stray capacitance, but then the
                        diaphragm capacitance is very very small.

                        I'll have to run down and see if I can get some of these. Much better than the Knowles Sisonic I've been messing with...they have a 75kHz bandwidth but limited dynamic range. Might make a pretty good measurement mic though.

                        I was just there a little while back but the main mic guys were off somewhere so I didn't get to see them.

                        Les


                        [Non-text portions of this message have been removed]
                      • Robb N
                        Thanks to all those who weighed in on this discussion. It is all very interesting! Leslie originally brought up the issue of distortion: But seemingly a
                        Message 11 of 24 , Sep 7, 2011
                        • 0 Attachment
                          Thanks to all those who weighed in on this discussion. It is all very interesting!

                          Leslie originally brought up the issue of distortion: "But seemingly a charge amplifier will have a distortion mechanism as gain is
                          modulated by capsule capacitance change/ feedback capacitor.
                          Pure second harmonic?" I don't really understand this, but I'm wondering if it is an issue while using piezo transducers. (Remember that my application would be for hydrophones.) The capacitance in a typical piezo element, or even a bender mechanism, should remain consistent, even at high SPLs. Could I assume that this distortion mechanism might be insignificant in my application?

                          Charge amps are very often used with hydrophones in practice. But we try to build hydrophones for the recordist that connect directly to standard mass-produced audio components. Also, sea state noise swamps all but noisiest hydrophone, but we also have customers that are using them for studio and field recording applications--and occasionally even weatherproof mics. So I look at circuits a little differently from the industry standards.

                          Richard wrote, "Response is not a problem with either unless you are incompetent." That hits a little closer to home than I wish, but I'm not afraid to seek help at the right time. He goes on to write, "...a charge amp is good IF (and its a big IF) you can get at the gate of the FET without introducing nasties due to poor layout & wiring." I've read his LNPrimer document. Richard, do you care to elaborate on the "nasties" in question or what to do about them? Of course, comments from anyone are appreciated.

                          Thanks again and best regards, Robb
                        • LESLIE WATTS
                          Rob the distortion mechanism mentioned by the paper is for capacitive sensors only, and does not apply to piezo. I have used charge amps with piezo many many
                          Message 12 of 24 , Sep 7, 2011
                          • 0 Attachment
                            Rob the distortion mechanism mentioned by the paper
                            is for capacitive sensors only, and does not apply to piezo.

                            I have used charge amps with piezo many many times...every time I hook up My B&K accelerometers.
                            Works great.

                            Les
                            ----- Original Message -----
                            From: Robb N
                            To: micbuilders@yahoogroups.com
                            Sent: Wednesday, September 07, 2011 3:11 PM
                            Subject: [micbuilders] Re: What about charge amps?



                            Thanks to all those who weighed in on this discussion. It is all very interesting!

                            Leslie originally brought up the issue of distortion: "But seemingly a charge amplifier will have a distortion mechanism as gain is
                            modulated by capsule capacitance change/ feedback capacitor.
                            Pure second harmonic?" I don't really understand this, but I'm wondering if it is an issue while using piezo transducers. (Remember that my application would be for hydrophones.) The capacitance in a typical piezo element, or even a bender mechanism, should remain consistent, even at high SPLs. Could I assume that this distortion mechanism might be insignificant in my application?

                            Charge amps are very often used with hydrophones in practice. But we try to build hydrophones for the recordist that connect directly to standard mass-produced audio components. Also, sea state noise swamps all but noisiest hydrophone, but we also have customers that are using them for studio and field recording applications--and occasionally even weatherproof mics. So I look at circuits a little differently from the industry standards.

                            Richard wrote, "Response is not a problem with either unless you are incompetent." That hits a little closer to home than I wish, but I'm not afraid to seek help at the right time. He goes on to write, "...a charge amp is good IF (and its a big IF) you can get at the gate of the FET without introducing nasties due to poor layout & wiring." I've read his LNPrimer document. Richard, do you care to elaborate on the "nasties" in question or what to do about them? Of course, comments from anyone are appreciated.

                            Thanks again and best regards, Robb





                            [Non-text portions of this message have been removed]
                          • ricardo
                            ... Yes there are various distortion mechanisms for single backplate condensor mikes with slight variations with Charge Amp vs Source Follower. There s an old
                            Message 13 of 24 , Sep 8, 2011
                            • 0 Attachment
                              > Leslie originally brought up the issue of distortion: "But seemingly a charge amplifier will have a distortion mechanism as gain is modulated by capsule capacitance change/ feedback capacitor.

                              Yes there are various distortion mechanisms for single backplate condensor mikes with slight variations with Charge Amp vs Source Follower. There's an old Sennheiser AES preprint that explains this in detail. But these effects usually show themselves at well over 120dB spl. The difference between Source Follower (Linkwitz) and the normal Drain Follower are MUCH greater.

                              I've pretended to analyse the capsule effects in the past but this only leads me to ignore them for normal recording & measurement mikes.

                              > Pure second harmonic?" I don't really understand this, but I'm wondering if it is an issue while using piezo transducers.

                              These problems are much smaller with piezo.

                              > "...a charge amp is good IF (and its a big IF) you can get at the gate of the FET without introducing nasties due to poor layout & wiring." I've read his LNPrimer document. Richard, do you care to elaborate on the "nasties" in question or what to do about them?

                              As a reader of LNPrimer.doc you will know I'm paranoid about layout for LN.

                              A Charge Amp increases the number of connections to the most sensitive part of the circuit by 33%. Stray capacitance is important and you can easily introduce microphony. I'm not saying don't try a Charge Amp. But you need to take extra care over layout and mounting.

                              I want to use Charge Amps for an Ambisonic Tetrahedral mike and have difficulty squeezing 4 such circuits into the space.
                            • umashankar
                              dear richard   re: tetrahedral and charge amp   i too have been toying with this idea. i even modified a tsb 140 to connect a capacitor at the gate. but so
                              Message 14 of 24 , Sep 8, 2011
                              • 0 Attachment
                                dear richard
                                 
                                re: tetrahedral and charge amp
                                 
                                i too have been toying with this idea. i even modified a tsb 140 to connect a capacitor at the gate. but so far, the charge amplifier designs i looked at single fet, like the neumans. just about completed a tetra using 25 mm capsules, and there is enough space at the back of the capsules!
                                 
                                umashankar

                                i have published my poems. you can read (or buy) at http://stores.lulu.com/umashankar

                                From: ricardo <ricardo@...>
                                >To: micbuilders@yahoogroups.com
                                >Sent: Friday, September 9, 2011 6:34 AM
                                >Subject: [micbuilders] Re: What about charge amps?
                                >
                                >
                                >  
                                >
                                >> Leslie originally brought up the issue of distortion: "But seemingly a charge amplifier will have a distortion mechanism as gain is modulated by capsule capacitance change/ feedback capacitor.
                                >
                                >Yes there are various distortion mechanisms for single backplate condensor mikes with slight variations with Charge Amp vs Source Follower. There's an old Sennheiser AES preprint that explains this in detail. But these effects usually show themselves at well over 120dB spl. The difference between Source Follower (Linkwitz) and the normal Drain Follower are MUCH greater.
                                >
                                >I've pretended to analyse the capsule effects in the past but this only leads me to ignore them for normal recording & measurement mikes.
                                >
                                >> Pure second harmonic?" I don't really understand this, but I'm wondering if it is an issue while using piezo transducers.
                                >
                                >These problems are much smaller with piezo.
                                >
                                >> "...a charge amp is good IF (and its a big IF) you can get at the gate of the FET without introducing nasties due to poor layout & wiring." I've read his LNPrimer document. Richard, do you care to elaborate on the "nasties" in question or what to do about them?
                                >
                                >As a reader of LNPrimer.doc you will know I'm paranoid about layout for LN.
                                >
                                >A Charge Amp increases the number of connections to the most sensitive part of the circuit by 33%. Stray capacitance is important and you can easily introduce microphony. I'm not saying don't try a Charge Amp. But you need to take extra care over layout and mounting.
                                >
                                >I want to use Charge Amps for an Ambisonic Tetrahedral mike and have difficulty squeezing 4 such circuits into the space.
                                >
                                >
                                >
                                >
                                >

                                [Non-text portions of this message have been removed]
                              • LESLIE WATTS
                                ... You know Richard, I have designed a good many piezo microphones for industrial application. I think a pretty good piezo studio microphone could be made. It
                                Message 15 of 24 , Sep 9, 2011
                                • 0 Attachment
                                  >> Pure second harmonic?" I don't really understand this, but I'm wondering if it is an issue while using piezo transducers.

                                  >These problems are much smaller with piezo.

                                  You know Richard, I have designed a good many piezo microphones for industrial application. I think a pretty good piezo studio microphone could
                                  be made. It would work with similar electronics to a condenser mic of course. Perhaps a charge amp.

                                  I just wouldn't create a resistance controlled system with rubber things as in some of the ubiquitous ceramic microphones of the past.
                                  Rubber things aren't stable.

                                  Remember the Sonotones?

                                  I might do that if I ever get some time. Back to stamping AlBemet diaphragms on the press?

                                  Les




                                  [Non-text portions of this message have been removed]
                                Your message has been successfully submitted and would be delivered to recipients shortly.