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

6834Re: [SDRSharp] Re: Differents calibration setting 1114 or 1126

Expand Messages
  • Marion D. Kitchens
    Apr 1, 2013
      I have 4 RTL2828U dongles here.  All have a single 28.8 MHz crystal.  Two have the E4000 tuner, and the other 2 have the R820T tuner.  All calibrate the same way.
       
      Marion,  K4GOK
       
      On Mon, 1 Apr 2013 23:55:47 +0100 Brian Gregory <bdgregory@...> writes:
       

      Could it be that some RTL2832U based dongles have more than one crystal oscillator in them, perhaps depending what tuner chip is used? In which case the calibration method needed might be different or even different for different frequency ranges.

      On 1 Apr 2013 19:43, "WB8CJW" <delshoff@...> wrote:
       

      This is how I calibrate:

      1. The dongle and upconverter needs to be powered on for about 20 minutes to stabilize.
      2. I use the kal.exe (calibrate) program to make measurements using a strong GSM channel to determine the offset.  Over the past couple months I have run the kal program using a batch script file I made to collect the data, add the date + time then append the results to a text file each time it is run.  My dongle is in an enclosure and has been surprisingly stable - within 1ppm per day.
      3. Whatever method you use to determine the 28.8 MHz. dongle error to be is entered in the Frequency correction (ppm) box.
      4. Now switching over to the upconverter I tune in WWV.  Most often 10 MHz. is the strongest, I use LSB or USB and set the bandwidth to about 1 kHz. and the Filter order to 100 or so then adjust the VFO to get the lowest audio tone.  Once the tone sounds like it is close to zero-beat I reduce the bandwidth to 0 and turn up the volume to hear the "buzz" at the lower frequency.  Now dialing the 10's of Hertz then the 1's I can hear the tick-tick sound difference and set it for the minimum.  This is usually within 1 cycle as now you can hear the variation as the dongle crystal and converter crystal vary slightly.
      5. Enter the Shift value difference so the VFO now displays 10.000.000.  The shift value is initially the frequency of the converter crystal entered as a negative number then it is adjusted to correct the error so the VFO displays the right value.
      If WWV is strong enough at the time I can leave the bandwidth set to about 1 kHz. and adjust the frequency until the tone is lowest (near zero-beat) and begin to hear the "shoop-shoop" sound as it gets closer without having to crank the volume clear up.

      The reason I use the audio zero-beat method is that I find it to be more consistent.  If WWV shows up out of line with the waterfall initially it is easy to get it close visually but depending on the Zoom value being used I find that it can vary a hundred Hertz or so each time the control is used to zoom out then back in again to recheck.  This may be due to some graphic display inaccuracy or difference on my pc.  I always see the waterfall is shifted slightly from what I find to be the audio zero-beat and ignore it.

      Dale, WB8CJW

      On 4/1/2013 11:00 AM, Arnie wrote:
       
      What mode are you using to demodulate WWV for zero-beat purposes? Seems like it should be DSB, but I am interested in what those who are more knowledgeable than I recommend.


      On Sun, Mar 31, 2013 at 2:41 PM, Christopher <mpeg2video@...> wrote:
       

      Whatever...

      --- In SDRSharp@..., jdow <jdow@...> wrote:
      >
      > Wrong way to do it. That calibrates you at the WWV frequency only.
      >
      > First calibrate the PPM using a known accurate VHF or UHF signal.
      > Then calibrate at WWV frequency with the shift value. Then you will
      > be closer to calibrated over the whole HF range.
      >
      > The other way to do it is to calibrate at the lowest WWV frequency
      > you can hear and repeat at the highest. At the low frequency tweak
      > shift. At the high frequency tweak ppm. Repeat until it seems accurate
      > enough at both frequencies. (That's how you calibrate an FT847 with
      > its two tunable crystal oscillators.)
      >
      > {^_^} Joanne/W6MKU
      >
      > On 2013/03/30 12:14, Christopher wrote:
      > > When I first ran 1126 I noticed that the ppm setting was at -120. I tuned
      > > to WWV, set the bandwidth really narrow and adjusted the ppm setting until
      > > I had centered on WWV. When finished I found that the ppm setting for my
      > > FDPP was about -2.9ppm which is where it started out with 1114.
      > >
      > > Maybe the same thing happened to your configuration?
      > >
      > > 73, Chris
      > >
      > > --- In SDRSharp@..., "hamradionl" <hamradionl@> wrote:
      > >>
      > >> hi,
      > >>
      > >> Same R820T dongle
      > >> notice a large different in calibration between
      > >> rev 1114 and rev 1126
      > >>
      > >> 1114 = +14
      > >>
      > >> 1126 = -20
      > >> between 26 and 27 i see a jump on the spectrum.
      > >>
      > >> Anybody any idea ?
      >

    • Show all 31 messages in this topic