„S“ units versus SNR approximation ?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by AA7EJ, Mar 11, 2017.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: FBNews-1
ad: L-MFJ
ad: Left-3
ad: Subscribe
  1. AA7EJ

    AA7EJ Ham Member QRZ Page

    I am looking for approximate comparison between RST “S” unit and SNR.



    To (gently) steer the discussion into subject - here are suggested answers, choose any or add you own. Reasonable contributions to discussion are appreciated, BS is not.



    1. I do not know

    2. What is “S” unit ?

    3. I love my cat.

    4. I have been ham since kindergärtner … got my general ticket…20 WPM...

    5. The lowest SNR detected by RBN was 6 dB, hence anything =< then that can be assumed to be S = 1.


    6. S unit step is 6 dB , hence SNR 12 db = S2, approximately.


    7. Dipole length can be calculated by….

    8. When I was in army / navy / … auxiliary service we did not like to use either .

    9. You do no need to know. It is different when it is snowing or raining.

    10. Why do you need to know?

    11. If it is Thursday it must be Holland.

    12. SNR is defined as….(insert link)

    13. None of the above

    14. All of the above


    73 Shirley
     
  2. WR2E

    WR2E Ham Member QRZ Page

    Troll much Shirley?
     
  3. K7JEM

    K7JEM Ham Member QRZ Page

    These are two different terms. S meters measure relative signal strength. IOW, a signal that is S7 is stronger than a signal that is S5. But either signal could have an SNR that is really good, or really bad. An S7 signal will not have a great SNR when the noise is S6, but it will sound great when the noise is S0. An S1 signal might have an SNR of 10dB or more, but an S7 signal with S7 noise might not even be readable.
     
    WA7PRC likes this.
  4. K8MHZ

    K8MHZ Ham Member QRZ Page

    Shirley,

    Look at it this way. For example, you are listening to a station with a very strong signal, which would, for example, be a 9 in the RST rating. A relative way of measurement would be the S meter indication on your radio. Let's use S9 for this station.

    One time you hear the station, the noise level, as you look at your S meter, is S1. That would be a SNR of 9:1, meaning the signal is 9 times stronger than the noise.

    The next time you hear the station, it is still at S9 on your meter, but the noise floor has risen to an S7 on your meter. Now, the same S9 signal has an SNR of 9:7, which is only about 1 1/3 times stronger than the noise.

    Now, can you see the difference between 'S' and SNR?
     
  5. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    Well, there's certainly a difference. But if an "S" unit is 6 dB, and noise level is S1 while a signal is S9, that wouldn't mean the signal is 9 times stronger than the noise; it would mean the signal is 48 dB stronger than the noise, which is a ratio of 250:1, not 9:1.:)

    Of course, that assumes the S meter really indicates 6 dB per "S" unit change, which most won't.

    While S meter readings are quite arbitrary and don't mean much of anything, S+N/N can be accurately measured.
     
  6. K9STH

    K9STH Platinum Subscriber Volunteer Moderator Platinum Subscriber QRZ Page

    Way too many amateur radio operators are fixated on what an "S" meter reads and not what the signal is in relation to the noise level.

    Quite a while back, the Japanese manufacturers got into a war claiming that their equipment was more sensitive than those made by other manufacturers. However, what was really done was to change the calibration of the "S" meters. Instead of the old RMA / EIA standard of S-9 = 50 microvolts and 6 dB per S-unit, the S-9 point was reduced and the dB per S-unit was decreased. Then, the "dB over" calibration ceased to be accurate. All of which increased the "S" meter reading indicating that the receivers were more sensitive.

    Today, a significant number of operators are insulted if they are not given at least an S-9 reading and are generally used to like 20 dB over or 40 dB over readings. I have heard even 60 dB over readings being handed out.

    With the S-9 = 50 microvolts standard, then 20 dB over represents a signal level of 500 microvolts, a 40 dB over represents a signal level of 5,000 microvolts, 60 dB over represents a signal level of 50,000 microvolts, and an 80 dB over represents 500,000 microvolts. Frankly, it takes a mobile sitting in your driveway to even approach 50,000 microvolts signal level.

    My "S" meters are calibrated at the S-9 = 50 microvolts standard and are pretty accurate in the "dB over" range. There is an amateur radio operator who lives 5-blocks from me who has an SB-200. He can approach 40 dB over. Often, stations don't really make it to S-9 and yet are significantly out of the noise level so are "perfect" copy. I do have better than average antennas.

    http://nebula.wsimg.com/58d5f90d22e...50C433DB440D6B60D&disposition=0&alloworigin=1

    Some operators definitely get upset when I give then what my "S" meter actually reads. As such, I rely on the system that we used when a majority of receivers did not even have an "S" meter. That is, the cerebral if I am not having any problems copying then they are a 59.

    Weak signal VHF / UHF operators are used to getting low "S" meter readings. In fact, signals not even moving the "S" meter, that are "perfect" copy, are not unusual. Such is possible because of the very low noise factor.

    Frankly, it is not what the "S" meter reads that is important. It is the relationship of the signal level to the noise level that determines how well the information is going to be received.

    Glen, K9STH
     
  7. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I used to work a lot of meteor scatter on 6m and 2m SSB (way before the weak signal digital modes were invented) and the "S" reports for m.s. had nothing to do with signal strength, at all -- they were based on meteor burst duration. Under most conditions other than a major meteor shower or storm, everybody was literally "S2," meaning bursts less than 5 seconds long.

    So, new ops unfamiliar with m.s. would visit the shack early mornings to watch me work some scatter (which near sunrise on 6m is always possible if you have big enough hardware) and hear me giving everybody "S2." They would often say, "That guy wasn't nearly S2! He didn't even move the meter!":p
     
  8. SM0AOM

    SM0AOM Ham Member QRZ Page

    The RST system is quite overcomplicated as compared to the professional QRK and QSA systems where only five levels are used.
    In my opinion, the "S-report" could further be reduced to three levels; "weak", "average" and "strong" corresponding to SNR values of 10 dB, 25 dB and 50 dB.

    About three years ago I had a "slow winter afternoon" at the Enköping radiomonitoring station while waiting for a colleague to pick me up.
    I listened around on 80 and 40 SSB using a TCI 12 dBi log-periodic antenna beamed south, using an absolute calibrated measuring receiver, the R&S ESH3, which with some "tricks" can be set to indicate peak signal to averaged noise SNR. I compared signal reports given by stations in my general area to what the ESH3 showed.

    Many stations gave S9+20 or S9+30 dB reports, that showed less than 50 dB SNR or -70 dBm signal level. An S9 report usually was less than 40 dB SNR. People that obviously had local noise problems and had the pre-amp in sometimes gave the quite humourous "you are S9+, but please repeat your name and QTH" report to stations that showed 10-12 dB SNR.

    The bottom line is that "signal reporting" in amateur radio has become quite misunderstood and notoriously unreliable...

    73/
    Karl-Arne
    SM0AOM
     
  9. WB5YUZ

    WB5YUZ Ham Member QRZ Page

    And remember, no matter how kind you are, German children are kinder!
     
    AC8UN likes this.
  10. AA7EJ

    AA7EJ Ham Member QRZ Page

    "In my opinion, the "S-report" could further be reduced to three levels; "weak", "average" and "strong" corresponding to SNR values of 10 dB, 25 dB and 50 dB."

    So if S = 9 is a benchmark corresponding to SNR 50dB that is exactly what I was looking for - APPROXIMATE caparison between RST scale and SNR.
    Close enough for what I need.

    Now I can make some "signal strengths " estimates / comparison using SNR to however inaccurate RST "S" scale I grew up with .

    Of course RBN does not measure noise , just SNR. Better than nothing or someone doctored-up rice box S-meter reading.

    But you guys enforced my felling about letting RBN give "report" , as pointed out some operators are just reluctant to give nothing but "five nine plus". Oh well.


    Thanks.
    73 Shirley
     

Share This Page