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

Re: S meter calibration -- HDSDR

Expand Messages
  • warrenallgyer
    Hi Chris I think you may be asking about S9? You may calibrate your HDSDR meter by putting a -73 dBm signal (50 mVRMS) into the antenna connector. You can use
    Message 1 of 10 , Jun 26, 2013
    • 0 Attachment
      Hi Chris

      I think you may be asking about S9? You may calibrate your HDSDR meter by putting a -73 dBm signal (50 mVRMS) into the antenna connector. You can use almost any level so long as you remember that one S unit equals 6 dB and then do the math.

      Warren Allgyer - 9V1TD

      --- In softrock40@yahoogroups.com, Chris Wilson <chris@...> wrote:
      >
      >
      >
      > 26/06/2013 11:14
      >
      > How much signal level should I inject from a calibrated sig gen to show 9dB on the HDSDR S
      > meter,to calibrate the meter please? Thanks.
      >
      > --
      > Best Regards,
      > Chris Wilson. 2E0ILY
      > mailto: chris@...
      >
    • Bip's
      Warren, 50 mV is way too much for S9. I suppose you meant 50 microVolt 73 Jean-Paul
      Message 2 of 10 , Jun 26, 2013
      • 0 Attachment
        Warren,
        50 mV is way too much for S9. I suppose you meant 50 microVolt

        73
        Jean-Paul

        --- In softrock40@yahoogroups.com, "warrenallgyer" <allgyer@...> wrote:
        >
        > Hi Chris
        >
        > I think you may be asking about S9? You may calibrate your HDSDR meter by putting a -73 dBm signal (50 mVRMS) into the antenna connector. You can use almost any level so long as you remember that one S unit equals 6 dB and then do the math.
        >
        > Warren Allgyer - 9V1TD
        >
        > --- In softrock40@yahoogroups.com, Chris Wilson <chris@> wrote:
        > >
        > >
        > >
        > > 26/06/2013 11:14
        > >
        > > How much signal level should I inject from a calibrated sig gen to show 9dB on the HDSDR S
        > > meter,to calibrate the meter please? Thanks.
        > >
        > > --
        > > Best Regards,
        > > Chris Wilson. 2E0ILY
        > > mailto: chris@
        > >
        >
      • Chris Wilson
        ... 26/06/2013 16:12 I am sure he does, and it s a typo. 50 uV is what I have used. It seems a tad high though, there are now a lot of weakish looking
        Message 3 of 10 , Jun 26, 2013
        • 0 Attachment
          >
          > Warren,
          > 50 mV is way too much for S9. I suppose you meant 50 microVolt

          > 73
          > Jean-Paul




          26/06/2013 16:12

          <LOL> I am sure he does, and it's a typo. 50 uV is what I have used.
          It seems a tad high though, there are now a lot of weakish looking signals
          on the bandscope showing 10 and 20 dB over S9 :) Perhaps I should have
          left it alone!

          --
          Best Regards,
          Chris Wilson. 2E0ILY
        • Bill Cromwell
          ... Hi, I don t really believe in S-meter calibration . It s only useful to see that one signal is stronger than another and maybe some wild notion about how
          Message 4 of 10 , Jun 26, 2013
          • 0 Attachment
            On 06/26/2013 11:15 AM, Chris Wilson wrote:
            >
            >
            >
            > >
            > > Warren,
            > > 50 mV is way too much for S9. I suppose you meant 50 microVolt
            >
            > > 73
            > > Jean-Paul
            >
            > 26/06/2013 16:12
            >
            > <LOL> I am sure he does, and it's a typo. 50 uV is what I have used.
            > It seems a tad high though, there are now a lot of weakish looking signals
            > on the bandscope showing 10 and 20 dB over S9 :) Perhaps I should have
            > left it alone!
            >
            > --
            > Best Regards,
            > Chris Wilson. 2E0ILY
            >

            Hi,

            I don't really believe in S-meter "calibration". It's only useful to see
            that one signal is stronger than another and maybe some wild notion
            about how much stronger (or weaker). Yeah..I try to set mine somewhere
            near 50 uV, too, but the inputs to receivers is rarely ever 50 ohms and
            even if it is 50 ohms or padded to 50 ohms when you diddle the S-meter
            it is NOT going to be 50 ohms when you expose it to 'the wild'. Worse,
            it will change with frequency and band and change by a LOT.

            I give out RST reports based on what I hear and NOT what the S-meter
            needle indicates. At my location I am not "plagued" with manmade noise
            even though I have some. How much varies from day to day and hour to
            hour but the noise is spread out up and down the bands so there are
            quiet areas - sometimes large quiet areas - all over the bands. I
            routinely listen to Q5 signals that do not tickle the S-meters. So would
            that be some kind of nonsense like a 509? I heard that on the air once
            or twice (grin). When I apply DSP filtering in the audio channel I copy
            signals that I can't hear without the DSP and even some of them out from
            under noises made by naughty television sets - loud and clear. I suspect
            that SDRs use a little bit of DSP, so the S-meter isn't really critical.
            Some of my radios don't even have an S-meter and it isn't missed.

            73,

            Bill KU8H
          • Ben
            ... Chris, 50 uV is correct. As noted elsewhere, to be strictly correct, the receiver impedance must be 50 ohms (a good SDR should be) and the source impedance
            Message 5 of 10 , Jun 26, 2013
            • 0 Attachment
              --- In softrock40@yahoogroups.com, Chris Wilson <chris@...> wrote:

              > <LOL> I am sure he does, and it's a typo. 50 uV is what I have used.
              > It seems a tad high though, there are now a lot of weakish looking signals
              > on the bandscope showing 10 and 20 dB over S9 :) Perhaps I should have
              > left it alone!

              Chris,

              50 uV is correct. As noted elsewhere, to be strictly correct, the receiver impedance must be 50 ohms (a good SDR should be) and the source impedance needs to be similar, or readings will tend to be off somewhat. Still, when working with an SDR, it *is* best to calibrate S9 to be equal to 50 microvolts. When you do have an SDR working well, you'll find that the entire S scale is meaningful. These are *not* like analog receivers in that regard.

              A good (and quick) reference on S-units, etc., may be found here...

              http://en.wikipedia.org/wiki/S_meter

              ...with the chart on that page in hand, run through your entire signal range with the generator and verify to your own satisfaction that the s meter is reading what it should. If it is, but you see what appear to be bizarre results from a real antenna, look into the antenna's impedance. It may vary wildly at different frequencies, particularly if it is a wideband design. If the antenna impedance is very high, the 50 ohm input of the radio forms a voltage divider highZ-tap-lowZ that results in low signal levels - not just low *readings*, but actually low signal levels. The S-meter is correct, it's telling you the actual signal level you're seeing at the input, it's just not good news.

              Cheers!

              --Ben
              AA7AS
            • EB4APL
              Chris, If your hardware is a Softrock Ensemble Rx, be aware that there are two 14dB attenuators in the two lower superbands , so your calibration will differ
              Message 6 of 10 , Jun 26, 2013
              • 0 Attachment
                Chris,

                If your hardware is a Softrock Ensemble Rx, be aware that there are two 14dB attenuators in the two lower "superbands", so your calibration will differ by the same amount when crossing the 8 MHz boundary.

                73 de Ignacio EB4APL



                On 26/06/2013 17:15, Chris Wilson wrote:
                 



                >
                > Warren,
                > 50 mV is way too much for S9. I suppose you meant 50 microVolt

                > 73
                > Jean-Paul

                26/06/2013 16:12

                <LOL> I am sure he does, and it's a typo. 50 uV is what I have used.
                It seems a tad high though, there are now a lot of weakish looking signals
                on the bandscope showing 10 and 20 dB over S9 :) Perhaps I should have
                left it alone!

                --
                Best Regards,
                Chris Wilson. 2E0ILY


              • warrenallgyer
                Yep... sure did. I fat-fingered it. Thanks for the correction. Warren Allgyer - 9V1TD
                Message 7 of 10 , Jun 27, 2013
                • 0 Attachment
                  Yep... sure did. I "fat-fingered" it. Thanks for the correction.

                  Warren Allgyer - 9V1TD


                  --- In softrock40@yahoogroups.com, "Bip's" <louijp@...> wrote:
                  >
                  > Warren,
                  > 50 mV is way too much for S9. I suppose you meant 50 microVolt
                  >
                  > 73
                  > Jean-Paul
                  >
                  > --- In softrock40@yahoogroups.com, "warrenallgyer" <allgyer@> wrote:
                  > >
                  > > Hi Chris
                  > >
                  > > I think you may be asking about S9? You may calibrate your HDSDR meter by putting a -73 dBm signal (50 mVRMS) into the antenna connector. You can use almost any level so long as you remember that one S unit equals 6 dB and then do the math.
                  > >
                  > > Warren Allgyer - 9V1TD
                  > >
                  > > --- In softrock40@yahoogroups.com, Chris Wilson <chris@> wrote:
                  > > >
                  > > >
                  > > >
                  > > > 26/06/2013 11:14
                  > > >
                  > > > How much signal level should I inject from a calibrated sig gen to show 9dB on the HDSDR S
                  > > > meter,to calibrate the meter please? Thanks.
                  > > >
                  > > > --
                  > > > Best Regards,
                  > > > Chris Wilson. 2E0ILY
                  > > > mailto: chris@
                  > > >
                  > >
                  >
                • warrenallgyer
                  In normal operating circumstances I would completely agree with you Bill. In the case of HDSDR the reason I do use the calibration is you have the ability to
                  Message 8 of 10 , Jun 27, 2013
                  • 0 Attachment
                    In normal operating circumstances I would completely agree with you Bill.

                    In the case of HDSDR the reason I do use the calibration is you have the ability to expand the scale of the spectrum display for single dB resolution. If you have calibrated it properly then the numbers on the vertical scale reflect that. Again, in normal QSO operations that does not have a lot of merit. However, in a controlled environment where the source impedance is set by the signal source or (better) a good set of attenuators, this calibration gives a very accurate indication of absolute power levels and an even better measurement of relative levels.

                    The meter itself on HDSDR seems to me to be relative useless except for a quick indication.... and I never have understood why it sits at S6 on a quiet band.... a topic for another day.

                    BTW.... to an earlier comment... I have bypassed the input attenuators in the two low ranges of my RXII just to preserve band to band calibration. It is good to check it on each band but I find mine holds calibration remarkably well from 3-30 MHz.

                    Warren Allgyer - 9V1TD
                    >
                    > Hi,
                    >
                    > I don't really believe in S-meter "calibration". It's only useful to see
                    > that one signal is stronger than another and maybe some wild notion
                    > about how much stronger (or weaker). Yeah..I try to set mine somewhere
                    > near 50 uV, too, but the inputs to receivers is rarely ever 50 ohms and
                    > even if it is 50 ohms or padded to 50 ohms when you diddle the S-meter
                    > it is NOT going to be 50 ohms when you expose it to 'the wild'. Worse,
                    > it will change with frequency and band and change by a LOT.
                    >
                    > I give out RST reports based on what I hear and NOT what the S-meter
                    > needle indicates. At my location I am not "plagued" with manmade noise
                    > even though I have some. How much varies from day to day and hour to
                    > hour but the noise is spread out up and down the bands so there are
                    > quiet areas - sometimes large quiet areas - all over the bands. I
                    > routinely listen to Q5 signals that do not tickle the S-meters. So would
                    > that be some kind of nonsense like a 509? I heard that on the air once
                    > or twice (grin). When I apply DSP filtering in the audio channel I copy
                    > signals that I can't hear without the DSP and even some of them out from
                    > under noises made by naughty television sets - loud and clear. I suspect
                    > that SDRs use a little bit of DSP, so the S-meter isn't really critical.
                    > Some of my radios don't even have an S-meter and it isn't missed.
                    >
                    > 73,
                    >
                    > Bill KU8H
                    >
                  • Bill Cromwell
                    ... Hi Warren, I have started using some of the DSP soundcard software as test bench equipment. In that environment I agree 100 per cent with your comments.
                    Message 9 of 10 , Jun 27, 2013
                    • 0 Attachment
                      On 06/27/2013 06:25 AM, warrenallgyer wrote:
                      >
                      >
                      > In normal operating circumstances I would completely agree with you Bill.
                      >
                      > In the case of HDSDR the reason I do use the calibration is you have
                      > the ability to expand the scale of the spectrum display for single dB
                      > resolution. If you have calibrated it properly then the numbers on the
                      > vertical scale reflect that. Again, in normal QSO operations that does
                      > not have a lot of merit. However, in a controlled environment where
                      > the source impedance is set by the signal source or (better) a good
                      > set of attenuators, this calibration gives a very accurate indication
                      > of absolute power levels and an even better measurement of relative
                      > levels.
                      >
                      > The meter itself on HDSDR seems to me to be relative useless except
                      > for a quick indication.... and I never have understood why it sits at
                      > S6 on a quiet band.... a topic for another day.
                      >
                      > BTW.... to an earlier comment... I have bypassed the input attenuators
                      > in the two low ranges of my RXII just to preserve band to band
                      > calibration. It is good to check it on each band but I find mine holds
                      > calibration remarkably well from 3-30 MHz.
                      >
                      > Warren Allgyer - 9V1TD
                      >

                      Hi Warren,

                      I have started using some of the DSP soundcard software as "test bench"
                      equipment. In that environment I agree 100 per cent with your comments.
                      My head is wrapped up tightly around operating radios on the air and my
                      comments will probably always reflect that. I think I said something
                      like "in the wild". And I do attempt to setup my radios with S meters to
                      that 50 uV spec - knowing that it won't really matter so very much when
                      I expose the antenna terminals to the real world of HF operating.

                      BTW - I do have a SoftRock kit waiting for some ham time allocated to
                      the bench to put together (SoftRock-40-R on 20 meters). I seem to have
                      found ways of abusing the DSP software (especially Spectran) so maybe
                      I'll be able to subvert the SDRs, too. Wondering what pleasant surprises
                      await.

                      73,

                      Bill KU8H
                    Your message has been successfully submitted and would be delivered to recipients shortly.