Posted by
Brian Lloyd-6 on
May 17, 2007; 4:37am
URL: http://elecraft.85.s1.nabble.com/K3-IF-AF-Dsp-questions-comparisons-tp447571p447574.html
On May 16, 2007, at 7:52 PM, Bill W5WVO wrote:
> Hi Brian,
>
> Well... The REAL standard, the one that EVERYBODY agrees on, is
> that S9 = 50 uV. How many dB an S-unit is depends on where you
> start from to get there.
I agree. OTOH, the initial description was that we were comparing an
S1 signal with an S9+40dB signal.
But you are certainly right that, if there is a standard, 50uV into
50ohms represents S9.
> Let's say that an S-unit is 6 dB and that the minimum detectable
> signal (MDS) is 0.25 uV. That's probably about where the MDS of the
> K3 is going to come in, plus or minus a few hundredths. How many dB
> between 0.25 uV and 50 uV? The formula is dB = 20log(V1/V2), so dB
> = 20log(200) = 46 dB. At 6 dB per S unit, that's 7.67 S-units below
> S9, or about S1-and-a-third. That means that a signal that you can
> just barely distinguish from the receiver's own internal noise
> floor -- i.e., essentially the amplitude of the receiver's internal
> noise floor itself -- is over S1 on the meter. That doesn't make
> intuitive sense to me.
>
> A more reasonable S-unit is about 5 dB. Why? You put the 0.25 uV
> MDS at S0 so that you get 46 dB / 5 dB/S-unit = ~9 S-units. This
> means that anything above the MDS moves the S-meter, including any
> atmospheric noise level that rises above MDS. A no-signal condition
> (i.e., the antenna is disconnected) is S0, no meter deflection --
> where it should be, IMHO. 6 dB per S-unit might have been
> somebody's standard way back when, but as you can see, it doesn't
> really make sense.
Well, like everything else, it is an arbitrary standard. To be honest
with you I have always stuck with 6dB/S-unit as the standard and S0
being the receiver's noise floor but that doesn't take into account
the receiver's bandwidth which could affect the level of the MDS.
50uV into 50 ohm representing S9 makes sense for a single carrier but
might not be valid for a more complex signal.
And I have never found a receiver that had a calibrated S-meter. I
get a kick out of guys turning on their amplifier that provides 10dB
of gain and seeing a 4 s-unit increase. Makes no sense but it sure
makes them feel good when I report it that way. (Actually I now just
tell them that it made a 1.5 s-unit difference. :-)
> If you really want a calibrated S-meter, I think you have to go
> with 5 dB, unless you want to number the scale from S1 "on the
> pin" (or actually a little below the "pin"). Maybe that's
> acceptable. Depends on how you think about it. I like 5 dB better
> because it makes better technical and intuitive sense to me.
I certainly understand. OTOH, I believe that even the FCC question
pool refers to an S-unit as being 6dB. I also did a search on the web
and found references to 6dB/s-unit. It does make some sense as each S-
unit represents a two-fold voltage change, something easily
determined when looking at a signal with a 'scope or a meter.
> My guess is that the K3 will come with some default value for an S-
> unit, but that it will also be user-settable in the menu. Based on
> a conversation I had with Wayne a couple weeks back, that default
> will NOT be 6 dB. I'm hoping I convinced him that 5 dB (as opposed
> to something lower) is a good number, by the logic given above. 5
> dB also works well because it is a modulus of 10, 15, 20, etc.,
> which gives a more pleasing symmetry to "dB over S9".
Aw, that's just marking the meter face. :-)
> In short -- Since there really isn't an existing S-meter
> calibration standard that anybody agrees on and is currently using,
> we might as well set a standard that makes some sense. I suspect
> that the K3 will be setting standards all over the place. :-)
That certainly makes sense.
73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com
_______________________________________________
Elecraft mailing list
Post to:
[hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft
Help:
http://mailman.qth.net/subscribers.htmElecraft web page:
http://www.elecraft.com