Give us a Choice with the S Meter

Jay Nation

Jim

So I'm retesting some this morning trying to eliminate any errors in measurement caused taking the previous readings on the XVTR input and I'm also eliminating the  +10 dB and +20 dB preamps.

I've got my dipole on the ANT1 jack, so this morning my 40m noise floor in a 500 Hz filter is -103 dBm, S-3.

Switching to my Cantenna dummy load, on ANT1,  I'm seeing -120 dBm, S-1.
Switching to my unterminated/open ANT2 jack, I'm seeing -118 dBm, S-1.
Switching to the 50 ohm terminated XVTR jack, I'm seeing -119 dBm, S-1.
Removing the terminator on the XVTR jack, I'm seeing -118 dBm, S-1.
Switching to RXA with the 50 ohm terminator, I'm seeing -120 dBm, S-1.
RXA without the terminator, I'm seeing -118 dBm, -S-1.

I'm not seeing much difference between the XVTR and ANT1 jacks.

To discover the effect on the measurements that a +20 dB preamp has.

My dipole on ANT1 w/+20 dB preamp, I see -100 dBm, S-4.
Switching to my Cantenna dummy load, on ANT1, w/+20 dB preamp, I see -128 dBm, S-0.
Switching to my unterminated/open ANT2 jack, w/+20 dB preamp, I see -127 dBm, S-0.
Switching to the 50 ohm terminated XVTR jack, w/+20 dB preamp, I see -135 dBm, S-0.
Removing the terminator on the XVTR jack, w/+20 dB preamp, I see -135 dBm, S-0.
Switching to RXA with the 50 ohm terminator, w/+20 dB preamp, I see -136 dBm, S-0.
RXA without the terminator, w/+20 dB preamp, I see -135 dBm, S-0.

As I expected the +20 dB preamp isn't helping 40m RX on my dipole. I usually don't even try using it on 40m and below.

   #FlexRadio IRC chat

For real-time discussions

        SDRgadgets

User created documentation.
           Volunteer!!

         73, Jay - NO5J

Bill -VA3WTB

Is you testing done on SSB phone? witch is what we have been talking about.

I can not reach those numbers on my 6500. My radio is always sampling the receive bins as it should. Without an antenna you should see an S3 or 4 from the sampling.

Bill -VA3WTB

I noticed flex has not commented much here about the request for a non SDR signal meter, but I think I can come close.
The Flex architecture is much different then a non direct sampling radio.
Where a traditional receiver estimates signal strength by calculating a value from the AGC voltage, the FLEX-6000 actually measures the signal power in each FFT bin.

In order For Flex to mimic a traditional receiver, it would not be a small project. Because the Flex works in such a different way, it does not share any of the same circuitry as a traditional radio, so were would we tap into for the S meter?

If we look at the layout schematics of a direct sampling receiver we can see the problem here.

Jay Nation

Bill

Yes, all of those measurements were done using LSB on 40m at @ 7.168 MHz and all done with a manually set 500 Hz LSB filter width. While I was working on it I noticed during one of the measurement that there was a very low level signal spike visible just emerging from the noise floor. so I moved the slice just enough to remove that from the filter passband, so that all the measurements wer done on a signal free portion of the noise floor.

    #FlexRadio IRC chat

For real-time discussions

        SDRgadgets

User created documentation.
           Volunteer!!

         73, Jay - NO5J

James Del Principe

Great follow up, Jay.... Now I will have to see if I can duplicate your numbers....   I just can't get my head around seeing S3 or so with no input because that means it is internal noise.....   Yes, it may be reading empty bins but that still adds to the noise floor and therefore defines the weakest signal that can be heard.... OK, off to the shack.       Best 73, Jim

Michael Coslo

Hi Rex, there is a fellow who wrote an analog meter applet for SSDR. It might be adaptable for your needs. I removed it from my computers, and am embarassed to say. This might be able to suit your needs www.kd0zv.com/woodbox.zip 

Michael Coslo

No you're not. It's just a learning curve. Welcome to the zoo.

KY6LA_Howard

The above graph probably explains why you are seeing S1 because the MDS is around -120dBm without the preamp.

Craig Williams

Why do we need an S meter? From what I hear everyone is always 5 by 9.

Ken - NM9P

The S-Meter was never designed to be a "relative signal strength meter" from which users get the numbers for their RST reports.   (BTW:  The RST system was created LONG before anyone had S-meters) From a very early point, it was designed to be a calibrated signal strength meter, from which users can make precision measurements when doing antenna experiments, etc.

The S-unit is a calibrated measurement.  Each S-Unit is 6 dB stronger (or weaker) than the next one.  The "standard" calibration agreed by the engineers was 50mV at the antenna = S9.  in modern terms, this means -73 db down from a milliwatt = S9.  if you work your way down from that by 6 dB steps, you get to where your MDS is, or the noise floor of your receiver, or the level of antenna/band noise at your station.  (Note the graphs posted above) 

Somewhere along the way, various manufacturers began to boost the numbers of their S-Meters, or failed to calibrate them in sync with their preamps, so that a station that by measurement is only an S9 will register S9+20 dB whenever the +20 dB preamp is engaged.

This "S-Meter bloat" lead to a misunderstanding of the true meaning of S-Meter readings, and also lead to a type of "S-meter envy" where people actually compared their bloated signal strength readings on the same station as a way of bragging that "My rig is more sensitive than yours" because of a higher S-meter reading.  (This was very prevalent on 11 meters in the 70's and 80's)

Now with the advent of SDR's and especially the Flex, we have a lab-quality, precision instrument that is calibrated, and synced with the preamps so that the S-meter reading is always true, and therefore helpful in giving actual comparitive readings when adjusting equipment, antennas, etc.

It wouldn't make any sense to me to dumb it down so that it is a lying S-meter like many of us had in the old CB days.


Ken - NM9P

Thomas NE7X

I personally would enjoy being able to select between a bar graph and the old style simulated mechanical style S-Meter on the 6400M TFT display. My ICOM IC-7700 allows me to do this. People who don't want the analog style S-meter don't need to select it. Simple

spopiela

Flex should add the old tuning eye indicator as an option too!

WQ2H - Jim Poulette

Cat's eyes, analog simulations, I knew there was a reason I signed up here. I Can't help but wonder what Tesla would say if I requested the buggy whip option. :-) 73 Jim, WQ2H

Jay Nation

Howard

Yep, I'm some where between quiet rural, and rural. We have dirt roads, but no post office delivery, We're not off the grid, but mostly off road. Except for the buzzing insects it's pretty quiet, both RF and AF. 

    #FlexRadio IRC chat

For real-time discussions

        SDRgadgets

User created documentation.
           Volunteer!!

         73, Jay - NO5J

Jay Nation

Jim
 
The S-3 reading on an open disconnected ANT input, is a consequence of the open jack. the SO-239 is acting as a very short, totally mismatched antenna. A quiet termination, is required. A 50 ohm composition resistor in a BNC terminator, is a lot quieter, even a 50 ohm Cantenna attached thru 3, 5 foot pieces of RG8X, and a couple of Ant switches, is a lot quieter than an open SO-239. A shorted SO-239 is also not a matched quiet termination. but it's still a very short antenna.   

    #FlexRadio IRC chat

For real-time discussions

        SDRgadgets

User created documentation.
           Volunteer!!

         73, Jay - NO5J

WQ2H - Jim Poulette

They're not perfect, but these do often work for a variety of situations.

73 Jim, WQ2H

Robert Hicks

Ken, fingers got loose again. Should have been 50uV not 50mV . As we used to say, the S-meter accuracy on some of the earlier RX were inversely proportional to the physical size of the meter itself! In SSB was it peak or average? Depends on the AGC time constant. When S-units originated, there was only AM, FM, and CW in general usage and only AM and FM really leant themselves to “meaningful” S-unit reporting. S-units are to my knowledge, not used anywhere but in amateur radio. Bob W5TX

Ken - NM9P

Yeah. I knew the mV wasn’t right when I read it in the return email. But by then several others had typed other comments. But then, I once had a receiver that it seemed to take about 50 mV to reach S9! It was deaf as a post! In late 1977 I had a surplus R392 that I used for CW practice while preparing for my 13 WPM General test. If I remember correctly, it had a meter that read uV and VU units, selectable from the AGC or the Audio output line. (It has been a while). I loved that receiver. It was my first general coverage “digital” receiver. But I got blisters on my thumbs changing the heavy band switch 1 MHz at a time...chunk, chunk, chunk. Going from 10 Meters to 80 was literally a pain.

Sergey KN7K

With all the readings and testing, I can say that I have some understanding of FLEX approach to S meter. However, I do not know how to evaluate the signal report of other station. Yesterday, I listened FT5000MP and Flex 6600 with switching antenna by Delta switch. Same Stations with s1-s2 on Yaesu were s6–s7 with no preamp, both at 2.1khz filters. So, I guess I will never hear a station on 6600 below S6 or so, right? Sergey, Kn7k

Bill -VA3WTB

No. don't be confused with the RF sampling in the FFT bins with RF from the antenna,  the FFT bins are related to the pass band width readings. so that is what your seeing. As Jay pointed out a 50 ohm antenna can show lower than the bin samples on the input.
A direct sampling Flex can hear well below the noise floor.

Bill -VA3WTB

The Flex can hear things many other radios just can't hear.

Ted_Spiegel_NX6C

Thank you google image.  :-)

James Whiteway

I want one!

Sergey KN7K

Did I say it cannot? Sergey, Kn7k

Sergey KN7K

Bill, I hear all this about sampling and fft.
When we make a QSO and I will ask you to give me report, what are you going to say? Are you going to ask me to wait until you switch to 500hz?

spopiela

Wouldn't that be a hoot ... The eye is not an s meter but you could switch it in to tune the Flex when you switch to the analog S meter that some of us want. ;<) It would put Icom out of business!

Stan VA7NF

All these discussions are about summing each noise floor fft bin.  Tune a single station at any band width, and a signal strength above the noise floor then the S meter will reflect that stronger signal (Math fanatics will say the noise is still there in the sum) 

James Del Principe

Stan, it does not take a fanatic to point out it is signal plus noise to noise ration....   so when giving a report it is binning noise and external noise plus signal to binning noise plus external noise ratio.  Whew!!!    Took a bit to get that out.    HI HI       73, Jim

Stan VA7NF

James, glad you got that out and I agree; that is however, a S+N/N ratio rather than signal strength (which technically is also a ratio against a fixed standard making it a fixed voltage).
My point is binning noise + external noise is a finite value, aka linear value, and when added to a log value it looses significance at over 1 S unit signal over noise, much less significance on stronger signals.
The result is the S meter will reflect (granted "almost") the actual signal strength and all this noise level discussion will disappear in the "grass".
So, unless working with -dbm noise levels and noise mitigation activity, it is meaningless to a signal report, which in a Flex 6000, is accurately measuring a real voltage presented to the receiver.
In RS(T) the S+N/N is in the 'R'eadability and the 'S'ignal is the S meter reading.  As a CW contest lover, everyone is 599 regardless of meter reading and in-the-noise 5 tries for a call sign.

Love this FRS form and the tangents that pop up on occasion.

Adam Farson

Next time you take an inter-continental flight, ask permission to tour the flight deck. If this is granted, take a look at the HF COMM panel in the CMU (comm management unit) above the pilot's and co-pilot's chairs and tell me where the S-meter is.
73, Adam VA7OJ/AB4OJ