Why does my S-Meter read S5 all the time?

  • 8
  • Question
  • Updated 2 years ago
  • Answered
We get a lot of questions about why the S-Meter always reads some constant level, like S5 on a particular band, for the noise floor, without any signals present.

I borrowed a graph from the Radio Amateur's Handbook, and marked it up with S-Units. It shows the relationship between atmospheric noise at the antenna versus frequency, for several environments.



Some things to note:

1.) The lower dotted line is equivalent to the quietest places on Earth. So if you have a reasonable antenna, you won't find anyplace that has less than S2 noise floor (in a CW bandwidth) on 40 Meters. Most suburbs will be more like S4 (or S5 in a SSB bandwidth.)
2.) Total noise is proportional to bandwidth. This chart is for 500 Hz receiver bandwidth. If you are using a SSB (3000 Hz bandwidth) filter the same noise floor will read/measure one and a third S-Units higher.
3.) In the US, where cities are not as tightly packed, I find the suburbs are closer to the rural line than the residential line in this chart.

--- Graham / KE9H
Photo of Graham - KE9H

Graham - KE9H, Employee

  • 28 Posts
  • 22 Reply Likes

Posted 6 years ago

  • 8
Photo of K4EAR

K4EAR

  • 138 Posts
  • 17 Reply Likes
A + S-unit shift from the 5000 with band noise.
Mike, K4EAR
Photo of K4ELO

K4ELO

  • 97 Posts
  • 18 Reply Likes
Thanks for that good info Graham. Fortunately I am Quiet Rural (I love it!) and the noise floor on 40M is usually about S2 unless there are thunderstorms.
Photo of K4EAR

K4EAR

  • 138 Posts
  • 17 Reply Likes
I ran a few RX tests for $htes and giggles...20Meters...tightest X-Y axis' on the panadapter. Preamp "0" and display FPS @1
ANT1 terminated into a 50 ohm load, CW mode 100Hz filter. The panadapter grass is ~135dbm. The S-Meter reads +/- S-2.

In SSB mode 2.7kc BW, the noise floor is the same as with CW, and the S-Meter reads S-4.

FWIW
Photo of George Molnar, KF2T

George Molnar, KF2T, Elmer

  • 1680 Posts
  • 616 Reply Likes
Would it make sense to allow S-meter "calibration" by creating a setup menu in SSDR where the operator could select one of those ITU curves? We'd still want the absolute values to read accurately, but a "calibrated" S-meter zeroing option could be nice.
Photo of Graham - KE9H

Graham - KE9H, Employee

  • 28 Posts
  • 22 Reply Likes
Hi George:
The technical definition of S-units is directly tied to signal strength. S9 is 50 microVolts into 50 Ohms. S-units are 6 dB each, below that. So the chart was presented as a way of educating the users as to the nature of the atmospheric noise and the way it changes with frequency. If you wanted a 'sliding scale' then I suppose that is what the RST signal reporting provides. Relative 'readability' and 'signal strength' and 'tone' quality, irrespective of the number of dBm involved.

--- Graham

==
Photo of George Molnar, KF2T

George Molnar, KF2T, Elmer

  • 1680 Posts
  • 616 Reply Likes
Thanks, Graham. You are, of course, quite right. My comment was more geared towards providing a "relative" measurement to compare one's environment on a daily basis to the norms. Funny, I had originally prefaced my comments with, "I'm not sure this is a good idea..."

Thanks for ALL the hard work and attention to these details that makes the difference between a good product and an excellent one.
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9199 Posts
  • 3560 Reply Likes
One thing to always keep in mind is the FLEX-6000 is a very accurate RF spectrum analyzer. This is a lot different from older radios that estimated the signal strength from a negative feedback loop (AGC). The result was changing the RF gain changed the S-meter reading.

Personally I like having a calibrated S-meter that accurately tells me the integrated power within the RX pass band.
Photo of Wm McDermott

Wm McDermott

  • 1 Post
  • 0 Reply Likes

I'm just returning to amateur radio after decades. 

My Flex 1500 shows S-2 with a dummy load connected and on some bands as high as S-4 to S-5 depending on bandwidth.  Pre-amp is 0 dB.  The power lead from a battery pack has 5 turns on a type 43 torroid and the USB cable has two extra ferrite beads and five turns on another type 43 torroid.  I'm using the most recent software and firmware.

From the comments here, particularly those of K4EAR, these S meter readings are normal.  Is that correct?  I was thinking of returning it to the factory for a look see.

Your comments are welcome. 

Wm McDermott AB9BE


Photo of AA0KM

AA0KM

  • 356 Posts
  • 65 Reply Likes
Flex-6500 antenna not connected.

SSB 2.4 filter -10db pre-amp = s-4+

CW 50hz filter -10db pre-amp =S-1+

I tried  also using a dummy load and checked every band and they are all the same with or without dummy load.

Also used different power supplies same results.

For those reading this is (not a problem) just my recent observation.

Interesting on how this all relates.
(Edited)
Photo of Graham - KE9H

Graham - KE9H, Employee

  • 28 Posts
  • 22 Reply Likes
Wm:

If you see the noise floor rise when you connect a normal antenna to the radio, the radio is doing everything it needs to do, for you to hear everything you are going to be able to hear.

With a normal antenna, dipole, similar, or better, the FLEX-1500 should run 0 dB preamp from 160 meters to 20 Meters, then +10 dB up to 10 Meters and +20 on 6 Meters.

If you are in a real quiet location, then you might run +10 dB on 20 Meters.  Judge by the action
of the noise floor when you connect a real antenna.

From your comments, I don't think there is anything wrong with your 1500.  I suspect that you
just met your first honest S-Meter in your ham career.

73's
--- Graham / KE9H

==


Photo of Gerald - K5SDR

Gerald - K5SDR, Employee

  • 831 Posts
  • 1522 Reply Likes
As a rule of thumb, you want your antenna noise to increase your S-meter by 8-10 dB.  That means that your receiver is not adding noise to the signal and that you have the correct gain to maximize performance.  If the noise goes up more than 10 dB with the antenna, you have too much gain.  

An easy way to check is to disconnect the antenna and read the dBm meter.  Connect the antenna and read the meter again.  If the noise goes up less than 8 dB, increase preamp gain.  If it goes up much more than 10 dB, decrease preamp gain.
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 645 Reply Likes
Interesting! That procedure could be done programmatically, correct? I like that!
Photo of AA0KM

AA0KM

  • 356 Posts
  • 65 Reply Likes
Well, when your used to other radios that jump up the S-unit readings when you use pre-amp and this doesn`t it fools you on how it all works..lol

I have happend to have my antenns un hooked and saw the s-units thus the investigating of the why.

A-true S-Unit reading.
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 645 Reply Likes
Frankly I think S-units is such a completely fictitious thing. There are so many variables that affect it, location, TOD, ok, rig, antenna, gray line. It's nice that S5 or 6 or 9 actually has some basis in fact but how many uVolts are at the antenna will be different in your QTH from the guy 2 doors down.
Independent of 'faking' a signal report, which we all do, 5-9 good buddy, not to be out done by the last report, you give 5-9+10 and follow that with, could you give me your call again pls.

Seriously, I do like what Gerald said, it quantifies the relationship that should exist between the effectiveness of your antenna system to your preamp level. I am going to add that to my code and see what happens. 
Photo of N3NER

N3NER

  • 186 Posts
  • 24 Reply Likes
Not sure I understand Gerald but, maybe I'm missing something.  I checked my 6300 for the first time today on a dummy load and I always have a S-3 reading and that go up to a S-5 if I enable the RF gain.  
(Edited)
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9199 Posts
  • 3559 Reply Likes
Change the RX filter size.  The S meter is showing you the totla RF power inside of the filter.  The larger the filter, the more RF power, the higher the S-meter reading.

If you are noise limited at the antenna, enabling the preamp will make the noise floor rise (a bad thing) and the s-meter will go up.  You want the noise floor to drop when you enable the preamp, which indicates you are not noise limited and RF pre-amplification is providing an increase in the signal to noise ratio.
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1784 Posts
  • 547 Reply Likes
Hi Tim,
Care to explain this a bit more: 
 "You want the noise floor to drop when you enable the preamp, which indicates you are not noise limited and RF pre-amplification is providing an increase in the signal to noise ratio."
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1784 Posts
  • 547 Reply Likes
Thanks!
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
When searching the prior posts for S meter operation, I came upon this.  My new 6400M reads ~5 S units even without the antenna connected.  I disconnected everything from the rig but the PS & I still see ~5S units at an RF gain setting of "0".  When I bump the gain up to +32 it reads 2S & at -8db gain it reads almost 6 S units.  Normal?  - Bob
Photo of Jay / NO5J

Jay / NO5J

  • 1674 Posts
  • 263 Reply Likes
Bob
Sounds a tad high, but might be completely normal. The S meter reading seen depends on, which band is selected, the preamp selection, and the filter selected,
and also whatever local noise might be being picked up. Even with my antennas disconnected, I sometimes see some readings around an S3. Noise is always present. and your radio's good at hearing it. 
 
#Flexradio

73, Jay / NO5J

Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Must be really sensitive.  Here it is with a shorted coax cable & still S5.  I own 2 old boat anchors & 3 contemporary rigs & none do that even with the preamp on.  I have
have a relatively quiet QTH.  I talk to so many Hams with that have 2-5S units higher noise than I.  I know, not all S units are created equal.  But the proof is I hear folks they do not. - Bob



 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4015 Posts
  • 975 Reply Likes
This is normal for an SDR radio, when no antenna is connected the receive bins are still being sampled. Steve from flex wrote a big article about this if I can find it.

This is a comment from Tim He was answering someone who did not believe how the S-meter works in the flex.

The S meter isn't broken so it doesn't need fixing.  The "fix" you are asking for would actually break the S meter by assigning an arbitrary signal strength value to the meter based on user perceived signal or comparison to an inaccurate standard (traditional radios).  Let me explain.

First, there is always signal present.  Blackbody, atmospheric noise, cosmic noise and thermal noise are "signals" at RF energies and all contribute to the RF noise floor.  Even with a dummy load connected, there is still signal being digitized.  There is no such practical thing as "no signal" (unless the radio is in an RF isolation chamber at absolute zero).

The signal strength meter design in the FLEX-6000 is based on how all lab grade digital spectrum analyzers operate; by accurately integrating the total RF power in the FFT bins within the receiver's passband.  This is a very important point.

Where a traditional received estimates signal strength by calculating a value from the AGC voltage, the FLEX-6000 actually measures the signal power in each FFT bin. Since S units are a standardized unit of measure for RF power as defined by the IARU, we calibrate the S-meter based on a standard, making the FLEX-6000 very accurate in this regard.

In the second paragraph, I emphasized an important point about measuring in the FFT bins, as opposed to estimating, RF signal within the receiver's passband.  This is very easy to demonstrate.  With your dummy load connected to the antenna port, change the slice to CW mode and change the filter to 50Hz.  At this RX filter passband size, there are only a few FFT bins measuring RF power and the S meter is barely moving.  If you put your cursor on the meter bar to display dBm, it may be as low as -134 dBm.  Now the IARU defines S0 as -127 dBm.  So the S meter reading in this case is actually less than S0!
(Edited)
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Thanx for that.  I look forward to seeing the article if you find it.  I'll do another search on it, too.  I saw this old post & thought I'd piggy back on it as it seemed we had similar observations.  I have two other SDR's and they & their s/w are different than what I am experiencing with Flex.  It has only been a ~week & hope to learn more in the coming weeks.   Thanx - Bob
Photo of Jay / NO5J

Jay / NO5J

  • 1674 Posts
  • 263 Reply Likes

Bob

Keep in mind your listening for signals, the AGC-T can be adjusted so that you actually hear no noise at all even when the noise floor appears to be S-anything.
If it's noise only on that frequency, why worry, Actual signals, if they happen, will be S-more. and you can't actually work the noise anyway. The meter calibration is correct on a Flexradio, not so, on almost all other rigs. It does take some getting used to. The metering gets recalibrated every time the radio restarts.
 
#Flexradio

73, Jay / NO5J

Photo of KY6LA - Howard

KY6LA - Howard, Elmer

  • 3789 Posts
  • 1638 Reply Likes

Welcome to the Modern World of dBm.  Your Flex is a laboratory grade spectrum analyzer which reads in dBm... for those who can't live in the 21st century, S0=-127dBM and S9=-73dBm...  Hopefully soon Flex will get rid of S meters and just give use much more useful dBm digital readouts.

Most legacy radios are not laboratory grade instruments so their S-Meter reading as figments of the imagination of the circuit design.  Some read S0 with no antenna attached even though there is likely a signal present.  It's really hard to compare anything to a legacy S-Meter as they are virtually meaningless relative even to a radio sitting next to them.

Looking at the screen shot above you seem to be reading about -103dBm which is in the range of the MINIMUM PRACTICAL DISCERNABLE SIGNAL for 80M.  Likely there is no point hearing below that point as the background noise (see graph at top) will cover the signals.

Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Jay - As I said, I have other SDR's so I am familiar with AGCT.  In fact, if the IC-7300 had AGCT I'd probably still have it.  It had a horribly noisy receiver as the AGC was always pumped up it seemed.  I love AGCT and the Flex utilizes it very well!!  I love how quiet I can make it & still hear a signal.  But my question wasn't about AGCT. It was about S meter behavior, AGCT does not affect the S meter,  & I have not seen this S meter behavior on my superhet rigs other SDR's I have.  Since I do not have any Flex experience I am left to look for an explanation in the manual or search user sites like this.  I am not worried, I just want to understand & if it is normal.  We all use our radios in different ways & we have different expectations.  - Bob
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Howard - I was a little surprised when I saw the S meter on the Flex.  I was expecting a dBM display.  Maybe it is in the menu somewhere.  A friend & fellow Flexer used to talk to me about it.  I'll have to search more.  - Bob
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4015 Posts
  • 975 Reply Likes
If you hover your mouse pointer over the graph on the flag, it reads dbm
Photo of Jay / NO5J

Jay / NO5J

  • 1674 Posts
  • 263 Reply Likes
Bob
My only point was noise is something best not listened to or even watched, it may just be only noise. To see the actual dBm level in real time try using Flexmeter, otherwise,
you will have to hover over the indicator in SmartSDR.

I'd also like a visible digital dBm readout to be added someday to both the Maestro and SmartSDR displays.

With my radio, an S-0 reading would be an indication of a problem. The band noise never drops that low. So, I think your probably seeing normal behavior. 
 
#Flexradio

73, Jay / NO5J

Photo of Homer1952

Homer1952

  • 36 Posts
  • 17 Reply Likes
Bill - I have been using the 6400M in rig mode 90% of the time.  Next time I fire it up in PC mode, I'll look at that, thanks.

Jay - My little mind still expects 0 S units when the receiver input is shorted.  For 50 years it has been that way for me.  I think you are right that it is normal behavior for a Flex. - Bob
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
PS- Homer1952 is my alter ego ;-)  I forgot my get satisfaction login & had to login thru gmail.  I just want to make one more comment:  As I try to understand the 6400M I don't want all the good about the radio to get lost.  I am just trying to maximize my understanding of the rig.  I have a lot of other questions that I will leave for another day.  73 Bob W8RMV
Photo of Jay / NO5J

Jay / NO5J

  • 1674 Posts
  • 263 Reply Likes
I won't argue further, actually I'm not arguing at all. It is what it is. Maybe that's how they were calibrated in the past. Makes sense. But 50 years of accuracy work, might have achieved something. What it is now, is almost never equivalent to what it was then, or what it might someday be.  
 
#Flexradio

73, Jay / NO5J

Photo of KY6LA - Howard

KY6LA - Howard, Elmer

  • 3789 Posts
  • 1638 Reply Likes

dBm Metering is on the wish list... hopefully sooner than later


HOWEVER if you hover your mouse over the S-Meter you can read dBm now

(Edited)
Photo of Ria - N2RJ

Ria - N2RJ, Elmer

  • 2315 Posts
  • 957 Reply Likes
PowerSDR has it, so not exactly new. As Howard said - on the wish list. Upvote it if you want it. 
Photo of Ria - N2RJ

Ria - N2RJ, Elmer

  • 2315 Posts
  • 957 Reply Likes
PowerSDR has it, so not exactly new. As Howard said - on the wish list. Upvote it if you want it. 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4015 Posts
  • 975 Reply Likes
I think as you can see it if you hover over the signal bar, Flex may feel this has been answered. 
Photo of KY6LA - Howard

KY6LA - Howard, Elmer

  • 3789 Posts
  • 1638 Reply Likes
@Bill. It clearly has NOT been addressed,by Flex yet.

dBm needs to be visible 100% of the.time not only after a tortuous mouse movement.

So upvote it please.
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Paul- I tried it first with a 51R non-inductive dummy load & got the same result.  That is why I decided to try it with a dead short. - Bob
Photo of Jay / NO5J

Jay / NO5J

  • 1674 Posts
  • 263 Reply Likes
Bob

Since I don't use the RXA inputs here, I keep a 51R screwed onto it. 
When I Switch the RX ANT to RXA, the preamp set for 0 dB,and set a 500 Hz filter passband, I read a minimum -123.8 dBm +/- 2 dBm delta, using Flexmeter. That's the lowest reading I can get on 40m, it also varies 1 - 2 dBm depending on the band selected. Pretty much an S-1 on the Maestro's S meter. S-0 or "S" doesn't happen. That's with Flexmeter v1.3.0.5 set for normal response, not average or peak hold.

Your @S-5 reading does suggest something's wrong. HelpDesk Ticket time.

All of these levels are actually being measured inside the radio continously, and the numerical values sent out over the Ethernet connection, Flexmeter just reads and displays the numerical values being outputted.  

If you don't already have it ...

http://www.denzone.com/uploads/FlexMeterSetupV1.3Beta.zip 

The new beta has some nice improvements. 
 
#Flexradio

73, Jay / NO5J

(edited to fix the units used)

(Edited)
Photo of Mark  K1LSB

Mark K1LSB

  • 91 Posts
  • 28 Reply Likes

Gents,

I’ve seen a number of different explanations in this thread of what the S-meter should be measuring, such as integrated (average?) power, and statements to the effect that if the RX bandwidth changes, the S-value should also change, even if there’s nothing but background noise.

I disagree.  The S-meter should only report the peak signal strength measured within the audio passband.  If the background noise is at an S-2 level then the S-meter should read “S2” regardless of how wide my receive bandwidth is.  If the S-value is expected to vary as a function of receive bandwidth, then there would arguably come a point (stay with me here) where the RB is wide enough that the S-meter should be reading S9 +10, but the strongest signal in the RB might only be an S7!  So how would anyone know the actual strength of any signal in such a scenario?

No, the S-meter should only be reporting the peak signal strength found anywhere in the audio passband.  I don’t care how wide the passband is.  It could be 500 Hz wide, or 2.7 kHz, or 250 kHz.  If there’s nothing but noise in the passband, and the noise floor anywhere in the passband is at a steady -115 dBm, then the S-meter should never read above an S2 regardless of the width of the passband.  The strongest measured signal strength (whether it be noise or a received signal) at any single spot in the passband is what the S-meter should be reporting.  Nothing more, nothing less.

Perhaps there is some other information that some users would like to also see reported, but that’s not the S-meter’s job.

Mark

[edited to correct typos]

(Edited)
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4053 Posts
  • 979 Reply Likes
From Tim,,,
With our radios we measure the actual signal strength by integrating the digitized RF power contained within the FFT bins inside the RX filter.  This is how a digital signal generator operates.

What we are saying is there is always RF power inside the filter and it is additive.  The filter bandwidth is a dependency in the power measurement process.  If you make the RX filter wider, the S-meter reading increases.  If you make it smaller, it decreases.  The standard is to measure integrated power in a 500 Hz filter.  And it is frequency dependent too.

And you dBm to S unit numbers are a little off too for frequencies below 30 MHz

S1 = -121 dBm
S2 = -115 dBm
S3 = -109 dBm
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4053 Posts
  • 979 Reply Likes
More info from Steve at Flex


The panadapter simply measures the signal in a given bandwidth and draws what it hears. If you look at any given pixel, it represents a certain amount of bandwidth. We call this the "bin size" of the FFT that is used to produce the display. If you cut the bin size into two pieces, the amount of noise in each piece goes down by half (3dB). In PowerSDR, the bin size is generally fixed for any given setup and does not change when you zoom. This is why the resolution gets worse as you zoom in on PowerSDR -- you begin to show one bin with multiple pixels. But for SmartSDR, we knew we wanted to have a larger range of zoom and this method was no longer acceptable. So we vary the bin size across a 1000:1 range. So the noise in each bin also varies. 1000:1 is a change of 30dB so from min zoom to max zoom, the noise in a bin will lower by 30dB and you see this change in the panadapter as you zoom in and out. 

When people talk about noise floor in ham radio they are generally talking about the noise level with a 500Hz bandwidth. When the panadapter is zoomed in to the max level, the bin size today is about 5.8Hz. This is a 19dB difference in noise from where a ham would say the noise floor is to what you can see on the panadapter. This means that the panadapter can see 19dB below what most hams would call the noise floor. Your ear and brain are also able to hear below the noise floor in 500Hz because of how they work. But there are limits to how well you can hear. If you've ever worked JT65 or another long-term integrating mode, you have noticed that your computer can copy signals that you cannot hear. 

So if you ask another ham "where is your noise floor on 80 meters" and he says "S5," what has he told you? Well with most hams, you don't know because you don't know the answer to these questions: 

1. What bandwidth are you using to measure the signal? 
2. Is your S-meter calibrated? 

An S5 signal corresponds to -97dBm. And if he's getting this on sideband set to, say 2.8kHz bandwidth then the actual noise floor in 500Hz would be -97 - 10*log(2800/500) = -104dBm. There's nothing magical about 500Hz, it just happens to be the convention for measuring noise in the ham radio world. In SmartSDR, if you set the passband filter to 500Hz, the S-meter in the slice will show you the 500Hz noise floor. 

If you start at maximum zoom and begin zooming out, you can see a point where the noise reading of the panadapter equals this number. What do you think this point is? ... if you've been following along, you will realize that this is the point where the FFT bin size is 500Hz. To get a rough idea if this is right, you could measure the width of your panadapter window and divide the amount of frequency displayed by this number. It should be in the 250-1000Hz range. The answer will not be exact because we do not continuously vary the FFT bin size -- we adjust it in steps and don't tell you where the steps are or what size they are. We do what's right for what you are viewing. 

I know that was a long-winded answer, but I hope it provides some insight into noise and how it changes what you see and hear. 
Photo of Mark  K1LSB

Mark K1LSB

  • 91 Posts
  • 28 Reply Likes
A level of S9 corresponds to 50 microvolts at 50 ohm impedance.  That value (50uv) has been the de facto standard since the 1930s (the 50 ohm reference was nailed down in 1981 by the International Amateur Radio Union), per Wikipedia.  If the background noise power level is 50uv, then your S-meter should read S9 regardless of zoom level or bandwidth (assuming there's no signal, only noise).

I've recorded two videos of me running a session of SDR Console V3 Beta 2, using an SDRPlay RSP1A with a 50 ohm dummy load connected to the antenna port.  I varied the receive bandwidth everywhere between 10 Hz and 350 kHz, and the zoom between 200 Hertz and 1.4 MHz.  The S-meter is reporting the maximum voltage measured anywhere in the vertical green receiver stripe in the display (the defined receive bandwidth).

Here are the videos (the S-meter is in the upper left corner of the band scope, and the Receive Bandwidth value is in small yellow font to the left of the S-meter):

https://drive.google.com/open?id=1jSQUm_SXCRfpWiPGnADd-Ple3mk5x6My

https://drive.google.com/open?id=1KJwBKiJXdHBCaw0MpGjGg6dNYp4Upgjv

Both videos show virtually no change in S-meter reading, regardless of any bandwidth or zoom changes I make.  Which is exactly as it should be, because the noise level doesn't change despite any changes I make to the zoom or bandwidth.

Simon Brown, author of SDR Console, understands that concept well.  His S-meter doesn't care what the bandwidth or zoom level are, it only cares what the maximum measured voltage is anywhere in the defined receive bandwidth, which is exactly what the IARU Region 1 Technical Recommendation R.1 says an S-meter should be measuring.  The maximum voltage in the RB is, and should be, independent of any zoom level or bandwidth.

Carry on.
(Edited)
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4053 Posts
  • 979 Reply Likes
I think Steve explained it well...  maybe if Steve had time he could walk you through it a bit more. Simons software does not work the same as the 6000s and SSDR. So as Steve mentioned as you increase the size of the bin samples, it changes noise.
(Edited)
Photo of Mark  K1LSB

Mark K1LSB

  • 91 Posts
  • 28 Reply Likes
One more time -- the S-meter should only report the maximum voltage as measured anywhere within the receive bandwidth.  Period.  There is no "integrating the total RF power" or "mean band limited power" or "spectral density plot area" or "expectations" or "what we wish to convey" or "intuitive manner" that should be considered -- there is only the peak instantaneous voltage anywhere in the RB, regardless of how wide or narrow the RB may be.  That is the only measurement in conformance with the IARU definition of what an S-meter should be measuring, with S9 being referenced to 50 microvolts (for HF).  If the highest voltage peak at any point anywhere in the RB is 50 microvolts, then the S-meter should be reading an S9, regardless of bandwidth or frequency or zoom level or time of day or grid square or anything else.

My apologies if I've been testing the patience of some readers, as that was never my objective.  My only objective has been to clarify what an S-meter should (and should not) be reporting, per the IARU definition.  For the S-meter to report anything else is to open up a whole new realm of latitude with long-accepted standards, with no limits of imagination to constrain it.

Disclosure:  I don't own a Flex, but only because my good friend across town hasn't received his 6600M yet -- he's selling me his 6300 as soon as he gets notification that his 6600M is on the brown truck.  :)
(Edited)
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4051 Posts
  • 979 Reply Likes
As said before, Steve answered this above and is the authority all Flex. He has explained this clearly.
Photo of Mark  K1LSB

Mark K1LSB

  • 91 Posts
  • 28 Reply Likes
I read what you posted that Steve said.  With all respect to Steve (and I respect his technical knowledge greatly), he may be the authority at Flex but the IARU definition is the authority on what an S-meter should be reporting.

I've posted a link below to a thread in Simon Brown's groups.io site where he replies to a number of statements that people were making about what they were expecting an S-meter to report.  One person asked why the S-meter reading didn't change whenever he adjusted the bandwidth (I wondered at the time where on earth did he ever get the notion that it should), and another person couldn't even make up his mind what he wanted -- first he says that he thought the S-meter should report the highest value of all (he means "any") FFT lines in the RB (and he's correct on that, because he's actually referring to peak voltage), but in the next sentence he says he also thinks it should display the "sum" of all the FFT lines, which is a completely different thing.  Yet another person wanted the S-meter to report signal-to-noise ratio. Simon replied with the same thing I've been saying here all along -- this is his statement:  "Let me make this very clear: the Signal Meter (S-meter) shows the peak signal level, it is not designed to show Signal to Noise [or anything else]. What you expect is a different signal indication -- I am interested in what you want to see but the S-meter logic you see now will remain as it is."

Here's that thread in Simon's site:

https://sdr-radio.groups.io/g/main/topic/5714841#31952
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
I am cursed.  I wish I never brought this up, as I am unable to unsubscribe no mater what I do.  I have unsubsribed on the thread, on the 6400M tab & on my get satisfaction acct. (where I driven to when click on the CLICK HERE at the bottom of the email)  I unchecked all the boxes.  And still keep getting these.  After all the words about this, I still don't understand the S meter & have moved on.  Flex could have taken those useless 5 S units, eliminate them & spent the money elsewhere. (tongue in cheek.  Pls no hate mail.  I have moved on. Or at least I am trying if I can get off this thread).
-Bob
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4053 Posts
  • 979 Reply Likes
Bob your sort of like me, this is all interesting but a little hard to grasp. If you go back and read my original thread were I used Steve's explanation it may come clear. To break it down, there is a big difference between how PSDR and SSDR works in this regard. In PSDR the bin size stay much the same, so when we zoom in and out the noise stays much the same and the signal meter reflects that.
In SSDR as you zoom in and out, the bins change in size so you hear more noise as you zoom in.
Because SSDR can zoom in so tight they needed to make the bins change accordingly.
(Edited)
Photo of KY6LA - Howard

KY6LA - Howard, Elmer

  • 3790 Posts
  • 1640 Reply Likes

Rather than dealing with all this theoretical mumbo jumbo, I decided to do some experiments with a beam and a dummy load on 40M 7.074 to see exactly how the S-Meter actually worked in the real world

Obviously my dummy load (MFJ) is not a perfect Faraday Cage.


Here are the results


By VISUALLY LOOKING at the Panadapter screen I could definitely see signals peaking at the peak readings

So I guess I really do not understand what the controversy is?


Photo of Lee

Lee

  • 84 Posts
  • 19 Reply Likes
OK, Guys....forget the S meter for a moment.
Let's say )(it does) I have S5 on 40m SSb BW 2.7... IF there is a signal at S3 WILL I HEAR IT?
If not...FRS has a problem as many locations actually have a very low actual noise florr in there area at certain times of the year.
Will I copy the guy at S3 if all the lower the 6400 I have will go is S5???
Thanks,
Lee
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3765 Posts
  • 907 Reply Likes
Yes, the s meter is telling us what RF is being detected in the FFT bins. so it says s something. This is not the signal we hear when an antenna is connected. The Flex can hear below the noise floor, that is why it is so good for low signal work.
Photo of Mark  K1LSB

Mark K1LSB

  • 83 Posts
  • 27 Reply Likes
Lee, the Flex S-meter doesn't work like an S-meter should.  The published spec says, for example, that if the maximum noise strength as measured anywhere in the passband is at .79uV (microvolt), then the S-meter should read S3.  But the Flex S-meter will measure something more than that, dependent on how wide the passband is.  So if you're listening to a wide slice of S3-level noise, the S-meter on your Flex might read S5 instead of S3.  So if there's a signal in there that's actually at an S4 level, you'll be able to hear it over the S3 noise floor, even though the Flex S-meter is reporting that the noise floor is at S5!  Why Flex decided to do things like that has never been satisfactorily explained to me.  But that crazy way the Flex S-meter operates gives rise to some equally crazy claims that "the Flex can hear signals below the noise floor", which is absurd.

Carry on,

Mark
Photo of Lee

Lee

  • 84 Posts
  • 19 Reply Likes
What are you guys talking about...Antenna shorted and it hears RF...from what...The short wire inside the unit? Yes of course there will be a VERY small amount...but S5!
If it hears anything with the antenna shorted..it must from inside the radio.....I don't think it is any more sensitive than any other modern radio. A good friend of mine did A/B testing with another high end radio with many weak DX signals and the Flex using very good antenna's and the Flex while very good....was not quite as sensitive when copying weak DX, even though the selectivity was better. The only way with the antenna shorted at the radio to get the hiss to lower at all is bring down the AGCT to a very low level....Is my 6400 needing repair?

By the way the reason I got into this question of S/N ratio was lately I have been experimenting with BOG antennas to improve S/N ratio on 80 mtrs. When we switched to the BOG and the noise and S meter stayed high we said..What the heck is going on here.....
Photo of Lee

Lee

  • 84 Posts
  • 19 Reply Likes
Mark, it's not just the S meter....Why is the white noise the same with NO Antenna?
Photo of Lee

Lee

  • 84 Posts
  • 19 Reply Likes
What? Hear below the noise floor...how...Maybe below the S meter reading...but the actual atmospheric noise floor?
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3765 Posts
  • 907 Reply Likes
This is pasted from Steve explaining some of this..

The panadapter simply measures the signal in a given bandwidth and draws what it hears. If you look at any given pixel, it represents a certain amount of bandwidth. We call this the "bin size" of the FFT that is used to produce the display. If you cut the bin size into two pieces, the amount of noise in each piece goes down by half (3dB). In PowerSDR, the bin size is generally fixed for any given setup and does not change when you zoom. This is why the resolution gets worse as you zoom in on PowerSDR -- you begin to show one bin with multiple pixels. But for SmartSDR, we knew we wanted to have a larger range of zoom and this method was no longer acceptable. So we vary the bin size across a 1000:1 range. So the noise in each bin also varies. 1000:1 is a change of 30dB so from min zoom to max zoom, the noise in a bin will lower by 30dB and you see this change in the panadapter as you zoom in and out. 

When people talk about noise floor in ham radio they are generally talking about the noise level with a 500Hz bandwidth. When the panadapter is zoomed in to the max level, the bin size today is about 5.8Hz. This is a 19dB difference in noise from where a ham would say the noise floor is to what you can see on the panadapter. This means that the panadapter can see 19dB below what most hams would call the noise floor. Your ear and brain are also able to hear below the noise floor in 500Hz because of how they work. But there are limits to how well you can hear. If you've ever worked JT65 or another long-term integrating mode, you have noticed that your computer can copy signals that you cannot hear. 
Photo of Mark  K1LSB

Mark K1LSB

  • 83 Posts
  • 27 Reply Likes
Bill, that's the same post of Steve's that you copied and pasted a month ago in this same thread.  As I said then, I don't care about FFT bin size, and neither should the S-meter care about FFT bin size (at least not unless you zoom in to a microscopically small slice of spectrum).  I've stated before that the S-meter reading should have no correlation to passband width or zoom or anything else except the maximum signal (or noise) level detected anywhere in the passband.  If the maximum voltage detected anywhere in the passband is 3.2 uV then the S-meter should read an S5, regardless of whether the passband is 500 Hz wide or 500 kHz wide.  In reply to your cut-and-paste a month ago in this thread I posted links to two videos I made from inside an SDR Console session where I varied the passband width everywhere between 20 Hz and 350 kHz, and the zoom between 200 Hz and 1.4 MHz, with virtually no change to the S-meter reading.  Here are those videos again:

https://drive.google.com/open?id=1jSQUm_SXCRfpWiPGnADd-Ple3mk5x6My

https://drive.google.com/open?id=1KJwBKiJXdHBCaw0MpGjGg6dNYp4Upgjv

You either didn't listen to a word I said back then, or you're not capable of understanding, because you just keep copy-and-pasting the same stuff over and over.

Simon Brown got it right in his SDR Console software.  Wolfgang Link also has a clear understanding on his page here:

http://www.giangrandi.ch/electronics/radio/smeter.shtml

For the record, he is the author of Methods of measurement for amateur radio (ISBN-13:978-8870210781)
Photo of KY6LA - Howard

KY6LA - Howard, Elmer

  • 3731 Posts
  • 1582 Reply Likes
Wolfgang Also Says

This refers to an unmodulated carrier signal (N0N) that uses almost no bandwidth; in case of real signals using a given bandwidth, this definition may not be enough since a smaller receiver bandwidth allows a weaker minimum detectable signal, but S-points are still a good tool for comparing received signals.&

So Wolfgang agrees with flex when you measure @500Hz bandwidth
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
What I don't understand is how the Flex method is better than the old method.  What is gained by the Flex method?  It seems that the Flex signal strength gage is useless below S5.  If it is useless, then why even have it?  If it is useful, nobody has stated its value when showing a constant ~S5 no matter if you have the SO239 shorted or not.  Every other radio I have shows an S0 when I short the SO239.  In fact when I would align older radios, I had to first mechanical zero the meter, then electrically zero it & then set it to spec for a S9.  This method has worked relatively well for decades.

So I am still missing the point/value of having a Signal Strength Meter read almost S5 when the SO239 is shorted out.  And apparently, I am not the only one.  The S meter only makes sense, when it shows the signal strength above the zero signal point represented by a shorted SO239/N connector.

Bob W8RMV
(Edited)
Photo of Jay / NO5J

Jay / NO5J

  • 1671 Posts
  • 259 Reply Likes

I can't ever remember giving the "Panadapter Noise Floor" a signal report. Never heard it call CQ either.

#Flexradio

73, Jay / NO5J

Photo of Jay / NO5J

Jay / NO5J

  • 1671 Posts
  • 259 Reply Likes

I can't ever remember giving the "Panadapter Noise Floor" a signal report. Never heard it call CQ either.

#Flexradio

73, Jay / NO5J

Photo of Mark  K1LSB

Mark K1LSB

  • 83 Posts
  • 27 Reply Likes
Show me where Wolfgang specifically agrees with Flex's S-meter reading at 500 Hz bandwidth.  He doesn't.

Nor does he at all agree with Flex's position that S-meter reading varies linearly with bandwidth.
Photo of Bob W8RMV

Bob W8RMV

  • 21 Posts
  • 3 Reply Likes
Said in maybe a more impactful/descriptive way, just display a Flex S meter that starts a S5/ -97dBm, since the S meter indication below that level does not impart any useful information.

Bob W8RMV
Photo of Stan - VA7NF

Stan - VA7NF

  • 447 Posts
  • 103 Reply Likes
Really tired of the repeated comments, and this may be one too.
The 'S' meter will measure a real CW signal the same at any bandwidth.  
As with 'other' radios if there were 4 CW signals within the bandwidth it would read the sum of the signals.
With noise (antenna or other sources) it still sums the signals which on larger bins may be 1000 times small bins.  Hence a strong signal will remain the same and summed noise will steadily increase with the capture area.
Give it a rest unless all you want to do is fill this forum with wide band NOISE!
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3765 Posts
  • 907 Reply Likes
I am pasting in Steve and Tims responses to these questions so people asking will understand how the flex works in this regard. I know there are people here who think Steve does not know what he is doing, I am not one of them. From my reading and seeing how Lab signal generators work I believe the Flex engineers have it right.
Photo of Lee

Lee

  • 84 Posts
  • 19 Reply Likes
Guys, no one listened to me...I'm NOT talking about measuring real signals here....I am talking about the noise floor...I'm beginning to think Flex has a high noise floor. I had a TT Orion II a few years back...very low noise floor displayed on P adapter, S meter AND the actual band noise with no signal. 

No one here has explained why with a shorted antenna I have S5 and lots of HIS...S5 is -104 DB...that's a lot of RF, where is it coming from with the antenna shorted?? When I tried a low noise BOG antenna the band noise did not go down, Why...maybe the noise IS coming from the radio even with a shorted or 50ohm terminated (I tried both, no difference) antenna port. LAst night on 40, we had a very weak signal on Freq. the ONLY time I could hear him was when his signal was OVER S5 (the band noise on my 6400. When I could not hear him, the panadapter did NOT show any signal. When I watched carefully...I just began to hear him when I could just see his signal rise above the panadapter line.

Let's talk real here not engineering talk that may or may not be practical in operations.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3765 Posts
  • 907 Reply Likes
What I posted explains all that you asked,,read it. Steve and Tim explains why.
Photo of Mark  K1LSB

Mark K1LSB

  • 83 Posts
  • 27 Reply Likes
"As with 'other' radios if there were 4 CW signals within the bandwidth it would read the sum of the signals." -- Stan VA7NF

Wrong, Stan.  Let me say this one more time, since you apparently weren't listening the previous dozen times I said it: The only thing an S-meter should be reporting is the highest instantaneous voltage measured at any single point anywhere in the pass band.  There is no 'summing" that should be taking place.

Here's a video I made this morning of some digital activity on 40 meters, running Simon Brown's SDR Console V3.  Note the S-meter only registers the greatest single signal level found inside the pass band, regardless of how many signals are present in the pass band, and regardless of how wide the pass band is.  There is no "summing of signals".

https://drive.google.com/open?id=1ememj-2hZX6fIFEVvBr6tyk4evN8zldu

That's how an S-meter should work, and that's the only way it should work.  If there were any summing taking place, then the S-meter reading at the instant the video stops should have been well over S9+50!

Any notion of "summing of signals" is pure nonsense.


@Bill VA3WTB,

You've done nothing but repeatedly point to other people's posts while contributing no thoughts of your own.  So thanks for the "wide band NOISE" (to use Stan's term).
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3765 Posts
  • 907 Reply Likes
Stan, to be clear, I use information from the experts, both Tim and Steve are. I am not an expert and have not been involved with the engineering of Flex radio.

So yes, I plan to continue using their expert information to inform others.

This conversation is no longer open for comments or replies.