Serious question


Jerry N9AVY
 

I see stations giving out S/N  and IMD reports.  Tend to ignore them, but have always wondered if they really have any value or are just filler.   

If a station can make contact with me that's usually good enough for me.  If someone gets a 599 report from me, it's because the 599 is given to everyone and I seldom deviate from that. (No one gets offended that way, ha-ha !)

Jerry   n9avy


Paul Butzi (W7PFB)
 

I find the S/N number more useful than ‘599’, as pretty much everyone sends 599 for signal reports unless something really unusual is happening.

I try to override the default 599 and send actual signal reports but I confess I often fail to do so.

-p W7PFB
73, Don’t forget to smile and have fun!

On Dec 27, 2020, at 11:10 AM, Jerry N9AVY <n9avy@...> wrote:

I see stations giving out S/N  and IMD reports.  Tend to ignore them, but have always wondered if they really have any value or are just filler.   

If a station can make contact with me that's usually good enough for me.  If someone gets a 599 report from me, it's because the 599 is given to everyone and I seldom deviate from that. (No one gets offended that way, ha-ha !)

Jerry   n9avy


Josef 'Jeff' Sipek
 

On Sun, Dec 27, 2020 at 19:10:31 +0000, Jerry N9AVY wrote:
I see stations giving out S/N  and IMD reports.  Tend to ignore them, but
have always wondered if they really have any value or are just filler.   
I find them useful. They give me a more accurate idea of what my signal
looks like on the waterfall - and since they are generated automatically,
there is less operator interpretation.

I think the fldigi manual talks a about what the two numbers mean, but the
rule of thumb KC3FL told me when I started with fldigi, was that S/N should
be equal to the the negated IMD for "perfect" signal. For example,

- 30dB S/N & -30dB IMD is great
- 30dB S/N & -15dB IMD means that I'm a bit wider than I should be

Technically, this info is contained in the RSQ, but like many I tend to
leave it at 599 unless something warrants changing it.

If a station can make contact with me that's usually good enough for me. 
And that's fine too, after all - it is 'only' a hobby. :)

Jeff (AC1JR).


Eric KG6MZS
 

Jerry,

From what I have seen those reports are generated by programs like DM-780.  They fail to discriminate between empty rails and whatever else they may be receiving at any given time, so I think they are pretty meaningless.  For instance: right now I have DM-780 on 20m and when I put the cursor on the background with no signals, DM-780 reports -30db IMD and a 32db s/n.  Those would be good numbers on an actual signal, but utterly meaningless on background noise.


Another country heard from

73

Eric

KG6MZS

On 12/27/20 11:10 AM, Jerry N9AVY wrote:
I see stations giving out S/N  and IMD reports.  Tend to ignore them, but have always wondered if they really have any value or are just filler.   

If a station can make contact with me that's usually good enough for me.  If someone gets a 599 report from me, it's because the 599 is given to everyone and I seldom deviate from that. (No one gets offended that way, ha-ha !)

Jerry   n9avy


Jerry N9AVY
 

Eric:

Thanks for info.  Tend to ignore those numbers because there's no place in log for them , hi, hi !

Jerry

On Sunday, December 27, 2020, 07:44:10 PM CST, Eric KG6MZS <contact@...> wrote:


Jerry,

From what I have seen those reports are generated by programs like DM-780.  They fail to discriminate between empty rails and whatever else they may be receiving at any given time, so I think they are pretty meaningless.  For instance: right now I have DM-780 on 20m and when I put the cursor on the background with no signals, DM-780 reports -30db IMD and a 32db s/n.  Those would be good numbers on an actual signal, but utterly meaningless on background noise.


Another country heard from

73

Eric

KG6MZS

On 12/27/20 11:10 AM, Jerry N9AVY wrote:
I see stations giving out S/N  and IMD reports.  Tend to ignore them, but have always wondered if they really have any value or are just filler.   

If a station can make contact with me that's usually good enough for me.  If someone gets a 599 report from me, it's because the 599 is given to everyone and I seldom deviate from that. (No one gets offended that way, ha-ha !)

Jerry   n9avy


 

One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


Jerry N9AVY
 

Love it when a station gives me a 599 and keeps asking for fill....  Emoji

Jerry  n9avy

On Monday, December 28, 2020, 09:33:23 AM CST, John - KC3FL via groups.io <kc3fl@...> wrote:


One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


 

Finally figured out how to attach a picture so here is a good visual explanation of IMD

John
KC3FL


DAVE KB3RAN 1692/381
 

Does a 599 count when the overall signal is really good but qsb garbles the message?  I need to review what constitutes the RST values again.

Dave
KB3RAN
1692/381
APE Event Mgr


-----Original Message-----
From: Jerry N9AVY <n9avy@...>
To: main@070Club.groups.io <main@070Club.groups.io>
Sent: Mon, Dec 28, 2020 10:40 am
Subject: Re: [070Club] Serious question

Love it when a station gives me a 599 and keeps asking for fill....  Emoji

Jerry  n9avy

On Monday, December 28, 2020, 09:33:23 AM CST, John - KC3FL via groups.io <kc3fl@...> wrote:


One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


--
Dave H KB3RAN 1692/381


Jerry N9AVY
 

John:

As far as I can see there is no S/N generated or available on MixW 3.2.   

Jerry

On Monday, December 28, 2020, 09:44:44 AM CST, John - KC3FL via groups.io <kc3fl@...> wrote:


Finally figured out how to attach a picture so here is a good visual explanation of IMD

John
KC3FL

Attachments:


Stephen Melachrinos
 

I usually quote a signal quality "at the peaks" and then say there's QSB.

Steve
W3HF


-----Original Message-----
From: DAVE KB3RAN 1692/381 via groups.io <davehardy0101@...>
To: n9avy@... <n9avy@...>; main@070Club.groups.io <main@070Club.groups.io>
Sent: Mon, Dec 28, 2020 10:46 am
Subject: Re: [070Club] Serious question

Does a 599 count when the overall signal is really good but qsb garbles the message?  I need to review what constitutes the RST values again.

Dave
KB3RAN
1692/381
APE Event Mgr


-----Original Message-----
From: Jerry N9AVY <n9avy@...>
To: main@070Club.groups.io <main@070Club.groups.io>
Sent: Mon, Dec 28, 2020 10:40 am
Subject: Re: [070Club] Serious question

Love it when a station gives me a 599 and keeps asking for fill....  Emoji

Jerry  n9avy

On Monday, December 28, 2020, 09:33:23 AM CST, John - KC3FL via groups.io <kc3fl@...> wrote:


One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


--
Dave H KB3RAN 1692/381


Eric KG6MZS
 

John,

I'm glad you joined the conversation because, like Jerry, I tend to disregard the IMD/s:n reports as mostly meaningless.  As I mentioned before, DM780 will generate a -30db IMD and a +32db s/n on dead air.  By your criteria below, that should be a good signal.

My understanding is that an accurate IMD/s:n measurement needs to be on "empty rails" or when no text is shifting the phase.  I'm pretty sure that the software is not being that discriminating when it snags the numbers.  I'm also pretty sure that the ops handing them out aren't painstakingly writing them down and sending them manually when they see empty rails and might be getting an accurate reading.

As you know, I rely on my KK7UQ to let me know what my IMD is.  From that perspective, FT-8's automated reports are much more meaningful.  They tell me how loud my signal is in another operator's waterfall relative to all the other signals they might be receiving.  The quality of my signal is already known to me via my own feedback from the IMD meter calibrated on PSK-31

Eric

On 12/28/20 7:33 AM, John - KC3FL via groups.io wrote:

One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


Stephen Melachrinos
 

John -

You are right--RST/RSQ are subjective, and S/N and IMD are measured. But the problem is that it's not always clear what is being measured. Both are based on power measurements, but the software can't really tell the difference between true noise power and another signal--they are just measuring noise within a particular bandwidth, and only of the signal as presented as audio to the software. Here are some places where that could be a problem.

- If there is another signal right next to the signal you are copying (like 50 Hz away), your software will measure its power and attribute that to IMD.
- We usually think of S/N in terms of background noise (true QRN), but in a very crowded band the software will measure the composite signal power of all the other signals within its measurement bandwidth.
- Because the software only looks at the audio signal (after the "downconversion" performed by your SSB demod), it can't tell the difference between distortions in the other station's transmitter and ones created in your receiver.

Now in a very quiet band (low noise and few other signals), the software has an easy job, and the numbers are meaningful. The pictures you posted are examples of that. But some folks use macros that automatically extract the S/N and IMD and send it to everyone all the time. Since I can't tell what that station's waterfall looks like, I typically ignore those numbers because I don't have confidence that they are being measured accurately or appropriately.

The bottom line is that sometimes S/N and IMD are significant, but IMO it's usually too hard to figure out whether it's one of those times (unless I know the other station well). 

Steve
W3HF


-----Original Message-----
From: John - KC3FL via groups.io <kc3fl@...>
To: main@070Club.groups.io
Sent: Mon, Dec 28, 2020 10:33 am
Subject: Re: [070Club] Serious question

One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.


Eric KG6MZS
 

Now that's interesting.  I didn't know that, but it makes sense.  Thanks Stephen.  Learn something new every day.

On 12/28/20 8:14 AM, Stephen Melachrinos via groups.io wrote:
- If there is another signal right next to the signal you are copying (like 50 Hz away), your software will measure its power and attribute that to IMD.


 

Like I said, the software has a hard time displaying the correct IMD because close by signals can mess it up and the only way the software can get an accurate reading is with an idle signal.  So it is only useful if the software catches the signal at some point where there are no characters being sent and the signal is in the clear, that's why I depend on my KK7UQ IMD meter.  If I see that the computer got it wrong, I'll send it again when I see that it is correct.  I pay more attention to the S/N and adjust my power accordingly, I already know my IMD.  Guess it's just a personal choice.

John
KC3FL


Bob Motyl KK6KMU
 

John,

Thank you for your S/N and IMD post.  Greatly appreciated.  As  amateur radio operators we have a responsibility to minimize any unnecessary noise that is transmitted by our station.  If the audio levels that are going into your transmitter are too high the rigs output amplifier is driven into its nonlinear region and creates higher order harmonics (noise).  Most PSK operators have their levels properly adjusted.  Occasionally I do see someone with too much audio input, and politely recommend they reduce it.  I appreciate knowing how my station is performing.  Keep the S/N and IMD reports coming.

Bob KK6KMU


Jerry N9AVY
 

Often copy signals all bunched together like a DX pileup.  Never understand why ops don't spread out a bit....

Jerry, n9avy

, On Monday, December 28, 2020, 10:33:29 AM CST, Eric KG6MZS <contact@...> wrote:


Now that's interesting.  I didn't know that, but it makes sense.  Thanks Stephen.  Learn something new every day.

On 12/28/20 8:14 AM, Stephen Melachrinos via groups.io wrote:
- If there is another signal right next to the signal you are copying (like 50 Hz away), your software will measure its power and attribute that to IMD.


 

John, may I ask?, What software is that?
Milt
N6MG

-----Original Message-----
From: John - KC3FL via groups.io <kc3fl@...>
To: main@070Club.groups.io
Sent: Mon, Dec 28, 2020 7:44 am
Subject: Re: [070Club] Serious question

Finally figured out how to attach a picture so here is a good visual explanation of IMD

John
KC3FL

Attachments:


David, K9DWR
 

On Dec 28, 2020, at 14:48, N6MG - Milt. via groups.io <n6mg=verizon.net@groups.io> wrote:

John, may I ask?, What software is that?
That’s fldigi (you can see the name in the title bar if you look closely)



David, K9DWR
#1604 LONP #255
david@graniteweb.com


 

Eric, 
I also use the KK7UQ meter, as well as the PSK Meter.
They're both useful.
Milt.
N6MG


-----Original Message-----
From: Eric KG6MZS <contact@...>
To: main@070Club.groups.io
Sent: Mon, Dec 28, 2020 8:11 am
Subject: Re: [070Club] Serious question

John,
I'm glad you joined the conversation because, like Jerry, I tend to disregard the IMD/s:n reports as mostly meaningless.  As I mentioned before, DM780 will generate a -30db IMD and a +32db s/n on dead air.  By your criteria below, that should be a good signal.
My understanding is that an accurate IMD/s:n measurement needs to be on "empty rails" or when no text is shifting the phase.  I'm pretty sure that the software is not being that discriminating when it snags the numbers.  I'm also pretty sure that the ops handing them out aren't painstakingly writing them down and sending them manually when they see empty rails and might be getting an accurate reading.
As you know, I rely on my KK7UQ to let me know what my IMD is.  From that perspective, FT-8's automated reports are much more meaningful.  They tell me how loud my signal is in another operator's waterfall relative to all the other signals they might be receiving.  The quality of my signal is already known to me via my own feedback from the IMD meter calibrated on PSK-31
Eric
On 12/28/20 7:33 AM, John - KC3FL via groups.io wrote:

One of my favorite subjects.  An RSQ is an arbitrary number that the op makes up.  Some give a meaningful RSQ but most give the 599.  S/N and IMD are real numbers that your software generates and tells all about the signal.  The IMD without the S/N is useless because if the S/N is 15 dB and the IMD is -15 dB, that is a great  signal, however if the S/N were 30 dB and the IMD was -15 dB, that is being over driven and probably is 2 or 3 hundred Hz wide.  The closer the IMD is to the negative value of the S/N, the cleaner the signal because all the harmonics are suppressed.  Unfortunately the software has a tough time getting the IMD right as you can tell as the IMD figure usually jumps around.