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

Amplitude modulation

Expand Messages
  • Andreas Robinson
    Hello all, and merry christmas! I did some calculating on AM transmission about a month ago... The conclusion is, you can do it, but it is not easy and
    Message 1 of 9 , Dec 25, 2001
    • 0 Attachment
      Hello all, and merry christmas!

      I did some calculating on AM transmission about a
      month ago...

      The conclusion is, you can do it, but it is not easy
      and probably not cheap either.

      The idea is like this, moving from the frontend to the
      sound card:

      1. First you have a regular input stage, amplifying
      the signal about 150 times.
      2. The signal is filtered so that you get 96 dB
      attenuation at 9kHz. A third order Chebyshev I filter
      is enough.
      3. The signal is passed through an analog optocoupler
      (HCNR200 or IL300)
      4. The signal is amplitude modulated to 10kHz. This is
      the tricky part. If it is done badly, a lot of
      resolution is lost.

      There are two ways to do this:

      * Modulate using a switch, flipping it back and forth
      between the signal and a constant voltage at a rate of
      10kHz. Then filter the signal using a 6th order
      elliptic filter. If you don't do that, you'll get
      aliasing in the sound card.

      or

      * Modulate with a sine wave, the old fashioned way.
      Requires an experienced analog designer to get it
      right.

      5. A buffer amplifier sends the signal along to the
      sound card.

      6. The sound card samples the signal in 40kHz or
      higher.

      Strengths: 16-bit resolution possible, at least in
      theory.

      Weaknesses: More complex analog section. It requires
      either a very high filter order (total of 9) or a low
      distortion sine-wave oscillator and modulator. It also
      needs more power because it uses one analog
      opto-coupler per channel. You have to use two low
      noise power supplies (such as batteries) as well.

      /Andreas



      _____________________________________________________
      Do You Yahoo!?
      se.yahoo.com
    • Brian Gr
      EEG People, I still see tremendous problems in making soundblaster work in this application. (See my post:
      Message 2 of 9 , Dec 25, 2001
      • 0 Attachment
        EEG People,
        I still see tremendous problems in making soundblaster work in this application. (See my post: http://groups.yahoo.com/group/buildcheapeeg/message/545 for reasons one can't get more than 4 bit resolution at EEG sampling frequencies with a soundblaster in FM modulation.)
         
        AM is not much better. When we get to actual details of trying to implement this solution, one would be sampling 16 bit data stream at 44KHz (about 10Mbyte/minute of data to process for ONE channel) The solution is not scalable...how would 2 or more channels be implemented? Multiple sound cards? Multiple carrier frequencies? That would be interesting to decode in software on the PC. In comparison to the modular EEG method (10 bit AD converter with 6 channels) this soundblaster method is a path of much higher resistance. Each additional channel imposes huge extra processor requirements at the PC end. The hardware to implement this will have poor resolution (about 4bits resolution using FM, certainly not 16 bit resolution using AM) Whereas, the modularEEG is a very straightforward from hardware and software perspective, not complex, lo processor overhead, proven accuracy, much easier to isolate, works with any PC, etc. 
         
        I have a laptop (Toshiba Tecra 8100) that doesn't have a line in available without purchasing an expensive docking station. But, it has a serial port, with basically the same configuration as every other PC.  What would we do with anyone who doesn't have full duplex sound card. These cards can't have both audio in and out simultaneously. So, audio biofeedback would require purchase of an additional card.
         
        I suggest we put the soundblaster approach aside and use a standard, inexpensive approach as has been designed in the modular_EEG design.
         
        Seasons greetings! Best wishes for the new year.
        Brian.
        ----- Original Message -----
        Sent: Tuesday, December 25, 2001 6:39 AM
        Subject: [buildcheapeeg] Amplitude modulation

        Hello all, and merry christmas!

        I did some calculating on AM transmission about a
        month ago...

        The conclusion is, you can do it, but it is not easy
        and probably not cheap either.

        The idea is like this, moving from the frontend to the
        sound card:

        1. First you have a regular input stage, amplifying
        the signal about 150 times.
        2. The signal is filtered so that you get 96 dB
        attenuation at 9kHz. A third order Chebyshev I filter
        is enough.
        3. The signal is passed through an analog optocoupler
        (HCNR200 or IL300)
        4. The signal is amplitude modulated to 10kHz. This is
        the tricky part. If it is done badly, a lot of
        resolution is lost.

        There are two ways to do this:

        * Modulate using a switch, flipping it back and forth
        between the signal and a constant voltage at a rate of
        10kHz. Then filter the signal using a 6th order
        elliptic filter. If you don't do that, you'll get
        aliasing in the sound card.

        or

        * Modulate with a sine wave, the old fashioned way.
        Requires an experienced analog designer to get it
        right.

        5. A buffer amplifier sends the signal along to the
        sound card.

        6. The sound card samples the signal in 40kHz or
        higher.

        Strengths: 16-bit resolution possible, at least in
        theory.

        Weaknesses: More complex analog section. It requires
        either a very high filter order (total of 9) or a low
        distortion sine-wave oscillator and modulator. It also
        needs more power because it uses one analog
        opto-coupler per channel. You have to use two low
        noise power supplies (such as batteries) as well.

        /Andreas



        _____________________________________________________
        Do You Yahoo!?
        se.yahoo.com


        To unsubscribe from this group, send an email to:
        buildcheapeeg-unsubscribe@egroups.com



        Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
      • JimmyB_101
        Sorry for the on going distraction of bringing up soundcards ... that was my hope for a very low cost, possibly 8 bit, 2 channel, hobby unit. No solutions
        Message 3 of 9 , Dec 26, 2001
        • 0 Attachment
          Sorry for the on going distraction of bringing up soundcards
          ... that was my hope for a very low cost, possibly 8 bit,
          2 channel, "hobby" unit. No solutions have been proposed that
          seem easy to implement. (yet?)

          I'm NOT wanting to distract the group from
          getting the latest design completed and into a buildable
          form. Maybe we can revisit this topic after that happens.

          [etc.: clueless wandering through hardware sites over the
          summer left me with the impression that an on board serial
          A>D might be the easiest, cheapest, tested method for a very
          low cost unit IF drivers can be written to retreive
          data in an accurate and timely manner for non DOS Operating
          Systems. Joerge's very low budget serial unit
          from that time period is an example of what could
          be implemented in DOS.]
        • sleeper75se
          Hi Jimmy, ... The modular EEG setup is probably is as cheap as it gets. You can replace the filters, which use a lot of resistors and capacitors, with a
          Message 4 of 9 , Dec 26, 2001
          • 0 Attachment
            Hi Jimmy,

            --- In buildcheapeeg@y..., "JimmyB_101" <jiva@h...> wrote:
            > Sorry for the on going distraction of bringing up soundcards
            > ... that was my hope for a very low cost, possibly 8 bit,
            > 2 channel, "hobby" unit. No solutions have been proposed that
            > seem easy to implement. (yet?)

            The modular EEG setup is probably is as cheap as it gets. You can
            replace the filters, which use a lot of resistors and capacitors,
            with a switched-capacitor single-chip filter to lower the part count,
            but that would make it more expensive. (They are $9 each.)

            I've got some comments on other ideas that have been discussed
            recently:

            * Using a delta-sigma ADC with built in filter.

            ADS1211 does not work the way you want it to. The filter is designed
            for sampling one channel at a time, not for switching between
            channels with every sample.

            * Signal processor + oversampling:

            Not practical unless you are prepared to solder 100-pin fine-pitch
            packages that cost maybe $10-$20. Also, DSP's are very noisy and
            power hungry. A microcontroller is not fast enough either.

            So let's just stick to what we are working on.

            You might wonder what is happening right now...
            I can't speak for Joerg, but I'm working on a new analog frontend
            using low noise LT1012 opamps that hopefully works with protection
            resistors. It will be ready for prototyping early/mid January.

            > [etc.: clueless wandering through hardware sites over the
            > summer left me with the impression that an on board serial
            > A>D might be the easiest, cheapest, tested method for a very
            > low cost unit IF drivers can be written to retreive
            > data in an accurate and timely manner for non DOS Operating
            > Systems. Joerge's very low budget serial unit
            > from that time period is an example of what could
            > be implemented in DOS.]

            If we use the standard serial protocol, that won't be a
            problem. :-)

            /Andreas
          • Jim Peters
            ... I m not going to push AM as a solution, because I know almost nothing about electronics, and the modular EEG looks the way to go, with lots of advantages.
            Message 5 of 9 , Dec 27, 2001
            • 0 Attachment
              Andreas Robinson wrote:
              > I did some calculating on AM transmission about a month ago...

              I'm not going to push AM as a solution, because I know almost nothing
              about electronics, and the modular EEG looks the way to go, with lots
              of advantages.

              But I'm not sure about all the calculations below:

              > 2. The signal is filtered so that you get 96 dB
              > attenuation at 9kHz. A third order Chebyshev I filter
              > is enough.

              Why filter at 9kHz ? We're talking about sampling at 256Hz in the
              modular EEG design, so I reckon a 1kHz carrier would be fine. If
              making fancy filters is difficult, use a 4kHz carrier. This leaves
              plenty of space between 128Hz and 4kHz for the filter to operate.

              To my simple-minded way of looking at it, you get a data point for
              every carrier cycle. All you have to do is look for a maximum (with
              maybe some fiddling due to the inversions for -ve signal levels). If
              the software tracks the signal to stay in sync with the carrier, then
              accurately spotting maxima/minima the right way around should be no
              trouble. Or is this too `cheap' an approach ?

              I was imagining a sine-wave carrier, by the way.

              > 4. The signal is amplitude modulated to 10kHz. This is
              > the tricky part. If it is done badly, a lot of
              > resolution is lost.

              > * Modulate using a switch, flipping it back and forth
              > between the signal and a constant voltage at a rate of
              > 10kHz. Then filter the signal using a 6th order
              > elliptic filter. If you don't do that, you'll get
              > aliasing in the sound card.

              I'm cautious about this approach. For a slow 1Hz wave, for example,
              this means that we would be switching between a +ve voltage and 0V for
              0.5 sec, then -ve voltage and 0V for the next 0.5sec. A soundcard
              20Hz high-pass filter would make the 0V level appear to `sag' or
              `rise' over the half-wave periods. You need a balanced +ve/-ve signal
              to avoid this.

              For example, this:

              * * +V
              * *
              *** * * * *** * * * *** 0V
              * *
              * * -V
              0 0.5 1.0 sec

              Gets distorted to something like this, which is impossible to
              interpret on the software side:

              * * * * * * *
              *** *** *** 0V
              * * * * * * *

              0 0.5 1.0 sec

              So, the carrier would have to be a sine-wave, or maybe a balanced +/-
              square wave. If a fairly accurate balanced +/- wave can't be
              generated easily, or if a modulator is hard to do well, then I guess
              the whole thing is not going to work at all anyway.

              > * Modulate with a sine wave, the old fashioned way. Requires an
              > experienced analog designer to get it right.

              Okay, not me then.

              Anyway, I agree -- stick with the serial RS232 solution that is
              currently being developed. There's no point in wasting development
              effort on a soundcard solution if it will just slow things down,
              especially when the ideas are untested and there are other known
              limitations to the approach.

              Serial is also much cleaner for the software -- you don't have to
              worry about duplex audio (as someone pointed out), or soundcard mixer
              settings or any of that stuff.

              I'll not mention AM again !

              Jim

              --
              Jim Peters (_)/=\~/_(_) UazĂș
              (_) /=\ ~/_ (_)
              jim@ (_) /=\ ~/_ (_) www.
              uazu.net (_) ____ /=\ ____ ~/_ ____ (_) uazu.net
            • sleeper75se
              Hi Jim, I know this topic is a bit off track, but I think it is interesting to discuss these kind of things, so bear with me and let me explain my thinking.
              Message 6 of 9 , Dec 28, 2001
              • 0 Attachment
                Hi Jim,

                I know this topic is a bit off track, but I think it is interesting
                to discuss these kind of things, so bear with me and let me explain
                my thinking.

                > Why filter at 9kHz ? We're talking about sampling at 256Hz in the
                > modular EEG design, so I reckon a 1kHz carrier would be fine. If
                > making fancy filters is difficult, use a 4kHz carrier. This leaves
                > plenty of space between 128Hz and 4kHz for the filter to operate.

                I chose 9kHz because I hoped the demodulation would be easier to do
                with a high carrier frequency. Filtering and downsampling from 40kHz
                to 256Hz in software after demodulation is fairly cheap, I think, but
                perhaps you are right, that 1kHz would work well too.

                It's not difficult to make fancy filters (with the right synthesis-
                software), but they cost more money. Also, if you know nothing of
                filters, but need one, you can buy 8th order switched-capacitor
                lowpass filters for $9. Just add a clock-signal - the frequency
                decides what cutoff frequency you get. Can't get any easier than
                that. :-)

                > To my simple-minded way of looking at it, you get a data point for
                > every carrier cycle. All you have to do is look for a maximum (with
                > maybe some fiddling due to the inversions for -ve signal levels).
                [snip]
                > trouble. Or is this too `cheap' an approach ?

                I think it would be hard to find the actual peaks using this method.
                In most cases it will fall between two samples, so we would need some
                way of extrapolating their actual position and height.

                > > * Modulate using a switch, flipping it back and forth
                [snip]
                > I'm cautious about this approach. For a slow 1Hz wave, for example,
                > this means that we would be switching between a +ve voltage and 0V
                for
                > 0.5 sec, then -ve voltage and 0V for the next 0.5sec.

                It's actually the other way around. The slow signal is chopped up by
                a kHz square wave that is in turn filtered (by the 9th order filter)
                to be identical to sine-wave modulated signal. The square wave does
                not have to be balanced - the high pass filter in the sound card
                would do it for us.

                > I'll not mention AM again !

                Me neither. :-)
                Unless someone asks ...


                See you later,
                Andreas
              • JimmyB_101
                ... There may be some Maxim units for about half that price. (free samples in any case)
                Message 7 of 9 , Dec 29, 2001
                • 0 Attachment
                  >
                  > It's not difficult to make fancy filters (with the right synthesis-
                  > software), but they cost more money. Also, if you know nothing of
                  > filters, but need one, you can buy 8th order switched-capacitor
                  > lowpass filters for $9. Just add a clock-signal - the frequency
                  > decides what cutoff frequency you get. Can't get any easier than
                  > that. :-)
                  >

                  There may be some Maxim units for about half that price.
                  (free samples in any case)
                • Brian Gr
                  Let s stand back a bit and look at the total differential cost of these two approaches: Approach 1, Modular_eeg 6 Channel A/D converter, microprocessor, rs232
                  Message 8 of 9 , Dec 29, 2001
                  • 0 Attachment
                    Let's stand back a bit and look at the total differential cost of these two approaches:
                     
                    Approach 1, Modular_eeg
                    6 Channel A/D converter, microprocessor, rs232 isolation $15.00?
                    I'm assuming battery power on eeg side and photo isolation for both approaches.
                    Total $15.
                     
                    Approach 2, multiple AM modulators feeding sound card line in's 
                    Assume you can drive 2 channels per sound card, although I'd be concerned about many issues to get reasonable resolution.
                     
                    Per channel costs:
                    AM modulator per channel, assume some sort of balanced modulator and additional op amps $2.50?
                    filter, Maxim, as described below $4.50
                    analog isolation per channel $5.00?
                    $12 / channel.
                    Per system costs:
                    filter clock generator $2.00? 
                     
                    System dependant potential additional costs:
                    Additional sound card if existing cards aren't full duplex....$10
                    Extra sound cards for channels 3-6...$10 each
                     
                    So, to implement a 2 channel system:
                    AM modulation approach: $24 + $2 = $26
                        data rate: 10,560 Kbytes / minute, requires considerable additional processing to retrieve samples. (assume 44000 samples / sec / channel, 16 bit samples)
                    Modular EEG approach: $15
                        data rate: 168 Kbytes / minute (assume 200 samps/sec, 6 channels, 1 channel overhead (14 bytes transmitted per sample, that's about 28000 baud serial bandwidth, same format used for 1 through 6 channels)) requires minimal processing to retrieve samples.
                     
                    4 channel system:
                    AM: $48 + $2 + $10 (extra sound card) = $60
                        data rate: 21,120 Kbytes / minute (good luck)
                    ModularEEG: $15
                        data rate: 168 Kbytes / minute
                     
                    6 channel system:
                    AM: $60 + $2 + $20 (2 extra sound cards) = $82
                        data rate: 31,680 Kbytes / minute (forget it)
                    ModularEEG: $15
                        data rate: 168 Kbytes / minute
                     
                    Add $10 to each AM system if existing PC sound card is not full duplex (assuming sound feedback is desired).
                     
                    The AM system will take much more board real estate, especially when it's expanded beyond one channel.
                    The AM modulators will undoubtedly need complex calibration and will rely on sound card performance for system accuracy. These are all expensive compared to the modular EEG approach and will result in less consistency and accuracy.
                    The AM modulator approach will require much more powerful PC processor requirements (processor utilization to extract data) and this requirement will increase as channels are increased.
                     
                    I'd much rather have the PC being able to add value to the system (such as drive more graphics, L/S or magnetic feedback stimulation) instead of reading multiple sound cards and trying to determine what's
                     
                    Can anyone think of examples of where this approach (AM modulators and sound card inputs) have been used?
                     
                    Have a happy new year!
                    Brian.
                     
                    ----- Original Message -----
                    Sent: Saturday, December 29, 2001 4:30 PM
                    Subject: [buildcheapeeg] Re:(Semi)OT> Amplitude modulation


                    >
                    > It's not difficult to make fancy filters (with the right synthesis-
                    > software), but they cost more money. Also, if you know nothing of
                    > filters, but need one, you can buy 8th order switched-capacitor
                    > lowpass filters for $9. Just add a clock-signal - the frequency
                    > decides what cutoff frequency you get. Can't get any easier than
                    > that. :-)
                    >

                    There may be some Maxim units for about half that price.
                    (free samples in any case)




                    To unsubscribe from this group, send an email to:
                    buildcheapeeg-unsubscribe@egroups.com



                    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
                  • JimmyB_101
                    My ongoing distraction of the group has been an effort to come up with (concept)some sort of very low cost 2 channel unit that folks who are interested in EEG
                    Message 9 of 9 , Dec 30, 2001
                    • 0 Attachment
                      My ongoing distraction of the group has been an effort to
                      come up with (concept)some sort of very low cost 2 channel unit
                      that folks who are interested in EEG feedback might be able to
                      "get their feet wet" with. The soundcard idea surfaced because
                      I've seen what the amateur radio group has done with them,
                      replacing $100-200 hardware units with software. No practical
                      solution has been proposed yet but I'm hoping once the current
                      modular_eeg design is complete that there might be some
                      interest in returning to a search for a very low cost
                      "hobby" unit.

                      > Let's stand back a bit and look at the total differential cost of
                      these two approaches:
                      >
                      > Approach 1, Modular_eeg
                      > 6 Channel A/D converter, microprocessor, rs232 isolation $15.00?
                      > I'm assuming battery power on eeg side and photo isolation for both
                      approaches.
                      > Total $15.
                      >
                      > Approach 2, multiple AM modulators feeding sound card line in's
                      > Assume you can drive 2 channels per sound card, although I'd be
                      concerned about many issues to get reasonable resolution.
                      >

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