ad: M2Ant-1

About bandpass filtering and filters to improve noise?

Discussion in 'General Technical Questions and Answers' started by KE0EYJ, Jul 24, 2017.

ad: L-HROutlet
ad: l-rl
ad: FBNews-1
ad: Left-3
ad: Subscribe
ad: L-MFJ
ad: Left-2
  1. SM0XHJ

    SM0XHJ Ham Member QRZ Page

    Sure, they have the potential to be better than any PLL based system, and in many cases they are. My comment was not on your statement, but an earlier statement that basically said the SDR's per definition won't have phase noise issues. If done right, they can have very good phase noise. If done wrong they can be terrible...
    So far we haven't seen many copies of SDR's, but I'm pretty sure when copies start to be made by people with no clue about HF design (no company/country mentioned) we will start to see some pretty bad SDR's too ;-)
     
  2. SM0AOM

    SM0AOM Ham Member QRZ Page

    There is a whole spectrum of SDR hardware realisations, and some are really bad, being done by people without any RF design knowledge. It takes a through understanding of analogue circuit design to build an SDR that will approach the design limits of the components involved.

    But, even the quite bad SDR:s and I have tested some, require that the R&S SME03 signal generators are shelved in favour of SMLR and HP3200B generators when measuring RMDR.

    SM5BSZ has done extensive tests of more SDR:s and the best require even better equipment to evaluate properly.


    73/
    Karl-Arne
    SM0AOM
     
  3. SM0XHJ

    SM0XHJ Ham Member QRZ Page

    Yes, and that brings us back to the weird and contradicting RMDR values in ARRL's and Sherwoods tests. Their blocking dynamic range measurement techniques seem to differ quite a lot, but their RMDR measurements should be, and have in the past, line up quite well. But for the IC-7300 they report 114 dB (which is probably on the border of what ARRL actually can measure) and 81 dB respectively.
    Either one of them made an error when measuring, or there is some design fault in the IC-7300 that in some circumstances impacts the RMDR.
     
  4. W6RZ

    W6RZ Premium Subscriber QRZ Page

    You're mixing up RMDR and 3rd order IMD numbers. They are two different things. Rob Sherwood does not measure RMDR, only 3rd order IMD. Here's the IC-7300 ARRL test report that show 3rd order IMD values of 77 dB to 79 dB which is in line with the Sherwood number of 81 dB.

    [​IMG]
     
  5. SM0XHJ

    SM0XHJ Ham Member QRZ Page

    Please read the first line of his table:
    "Sorted by Third-Order Dynamic Range Narrow Spaced - or- ARRL RMDR (Reciprocal Mixing Dynamic Range) if Phase Noise Limited"
    The columns for narrow and wide space dynamic range contains either 3rd order IMD or RMDR, depending on which is worst.
    So yes, it might be the 3rd order IMD that is the limit. But why would the wide spaced 3rd order IMD be that bad?
     
  6. AI3V

    AI3V Ham Member QRZ Page

    Jitter on a "clock" is exactly the same thing as phase noise on a "oscillator".

    There ain't no such thing as a free lunch.

    Rege
     
  7. W6RZ

    W6RZ Premium Subscriber QRZ Page

    I guess Sherwood is measuring RMDR. You can derive the RMDR from the LO Noise (db/Hz) entry in the Sherwood table. RMDR = LO Phase Noise - 27.
     
  8. SM0XHJ

    SM0XHJ Ham Member QRZ Page

    If all phase noise came from the LO, yes. But phase noise, or jitter, can be added in many places, even inside the FPGA.
     
    AI3V likes this.
  9. AI3V

    AI3V Ham Member QRZ Page

    The op lives in a highrise apartment in Seoul. Population 10 million.

    I live in a house in Pittsburgh, poulation, 300 thousand. I had to add preselectors to every single VHF/UHF/shf rig I own. I suspect my HF rig suffers also, but I haven't got roundtuit.

    Rege
     
  10. AI3V

    AI3V Ham Member QRZ Page

    That assumes some sort of lo that doesn't have any Spurs, just a steady rolloff of phase noise vs frequency.

    That's a big assumption.

    Rege
     

Share This Page