locked WHAT is CLIPPING - when db = RED #ApacheLabs


Niece KA1ULN <ka1uln1@...>
 

the DB column to the left on the wsjtx screen. it says "RED when clipping may occur"

what is clipping and how to fix it... how to get in the "GREEN"

thanx Niece


K7RA Tad Cook
 

Clipping happens when a transmitter is over-driven, or over modulated. On
SSB it produces a wide, distorted signal.

On Thu, Sep 29, 2022 at 12:29 AM Niece KA1ULN <ka1uln1@...> wrote:

the DB column to the left on the wsjtx screen. it says "RED when
clipping may occur"

what is clipping and how to fix it... how to get in the "GREEN"

thanx Niece





--


*Listen for K7RA/B CW beacon on 28.2833 MHz.*

*Tad Cook - K7RA*
*Seattle, WA*

*k7ra@... <k7ra@...>*


Pietro Molina
 

To avoid it, reduce volume of audio input (or the source on the radio)

Pietro I2OIM

Il giorno gio 29 set 2022 alle ore 09:29 Niece KA1ULN <ka1uln1@...>
ha scritto:

the DB column to the left on the wsjtx screen. it says "RED when
clipping may occur"

what is clipping and how to fix it... how to get in the "GREEN"

thanx Niece






Jim Brown
 

On 9/28/2022 3:37 PM, Niece KA1ULN wrote:
what is clipping and how to fix it... how to get in the "GREEN"
In the analog world, it is the condition where the signal voltage is so high that the tops and bottoms of the signal extend beyond the linear range of the electronics, so those tops and bottoms are "clipped off." Clipping causes distortion that adds a harshness to audio, splatter and key clicks at radio frequencies.

In the digital world, it's when the signal level so high that the A/D and D/A converters run out of bits to describe it. Digital levels are defined in dB relative to clip; the RED indicator is usually set a dB or fractional dB below actual clip.

Think of the strength of a 16-bit digital signal as a vertical row of 96 boxes (96 dB), the weakest signals occupying relatively few boxes near the bottom, signals that clip trying to go above the top box. A signal that exactly hits the top box is 0 dB; any weaker signal is - dB. (16-bits would ideally yield 96 dB dynamic range, but we lose some of that range due to non-linearity in the converters, and it takes some boxes to handle the waveform, so practical radio systems provide about 80 dB dynamic range (difference between weakest and strongest signals that can be handled at the same time).

To decode the weakest signals in the presence of very loud ones, we need to drive the A/D converter (radio feed to computer) so that the strongest signals are fairly near digital clip. That's why, when I'm trying to work very weak signals, I set that drive so that WITH signals, the strongest stay as close to the top of the bar graph as possible but never turn red. That DOES mean that I've got to "ride gain" on the signal coming out of the radio to the computer. In most rigs, adjusting the RF gain would accomplish that need.

73, Jim K9YC


Martin G0HDB
 

On Thu, Sep 29, 2022 at 10:23 AM, Jim Brown wrote:

[Snipped]

To decode the weakest signals in the presence of very loud ones, we need
to drive the A/D converter (radio feed to computer) so that the
strongest signals are fairly near digital clip. That's why, when I'm
trying to work very weak signals, I set that drive so that WITH signals,
the strongest stay as close to the top of the bar graph as possible but
never turn red. That DOES mean that I've got to "ride gain" on the
signal coming out of the radio to the computer. In most rigs, adjusting
the RF gain would accomplish that need.
A simpler approach that doesn't involve having to "ride the gain" is to operate the rig with AGC switched *on* (despite what the WSJT-X user guide states) and to set the audio level coming out of the rig and going into the ADC to give a level somewhere near the top of the scale on the WSJT-X 'thermometer' when receiving only band noise.

Rob Sherwood NC0B, he of Rx testing fame, has demonstrated conclusively that the WSJT-X decoder (specifically the WSPR one) can decode weak signals without degradation of their SNRs in the presence of an 'interfering' in-passband signal that's 75dB to 80dB stronger than the wanted signals. I've also verified that weak FT8 signals can be successfully and correctly decoded in the presence of an in-passband interfering signal that many 10's of dB stronger than the wanted signals.

I run my rig (IC-7610) with AGC on, with a 0.8sec time constant selected, and set the audio level going into WSJT-X to give a reading of 66dB when receiving band noise; the rig's AGC means that the level indicated on the WSJT-X 'thermometer' never exceeds 80dB, which is comfortably below the clipping threshold, even when receiving multiple very strong FT8 signals, eg. on 40m at night. Using this configuration I've regularly seen FT8 decodes within the same 15sec timeslot of -24dB and >+30dB, indicating a dynamic range of at least 54dB.

I understand that Rob NC0B sets his levels when receiving WSPR on 630m to give a band-noise reading of 68dB on the WSJT-X 'thermometer' and that Rob has had his findings and approach endorsed by Joe K1JT.

You'll find Rob's write-up of his tests and conclusions via:

nc0b.com/wsjt

--
Martin G0HDB


John K.
 

I won't define clipping, but you can get the meter out of the red zone by reducing the level of your receive comm port in windows.


Ron
 

My apologies, I directed you incorrectly in my earlier post. Clipping occurs when you have your receive audio is set too high.

To correct this issue, do the following:

Right click speaker icon in system tray>Sounds>Recording>”Levels” Tab –Configure so Receive Audio Level is about 30 dB when no signal present. It's a good idea to test this when connected to a dummy load. Make sure mic isn't muted (small icon to right of recording level slider bar.

73,

Ron - KJ5XX


Reino Talarmo
 

Right click speaker icon in system tray>Sounds>Recording>”Levels” Tab –Configure so Receive Audio Level is about 30 dB when no signal present. It's a good idea to test this when connected to a dummy load. Make sure mic isn't muted (small icon to right of recording level slider bar.
Hi Ron,
The original recommendation is to set the Receive Audio Level to 30 dB with empty band noise present. Depending on the rig, antenna and band the noise level can be higher than what you will get from a dummy load.
On the other hand current understanding considers a setting based on maximum levels close to clipping to be more suitable.
73, Reino OH3mA


Jim Brown
 

On 9/29/2022 3:35 AM, Martin G0HDB wrote:
A simpler approach that doesn't involve having to "ride the gain" is to operate the rig with AGC switched*on* (despite what the WSJT-X user guide states) and to set the audio level coming out of the rig and going into the ADC to give a level somewhere near the top of the scale on the WSJT-X 'thermometer' when receiving only band noise.
While I agree that use of AGC reduces the need to ride gain, it does NOT eliminate it in the presence of very strong signals. I use an Elecraft K3 and use slow AGC to minimize its effect on decoding. Slow is plenty fast enough.
Rob Sherwood NC0B, he of Rx testing fame, has demonstrated conclusively that the WSJT-X decoder (specifically the WSPR one) can decode weak signals without degradation of their SNRs in the presence of an 'interfering' in-passband signal that's 75dB to 80dB stronger than the wanted signals. I've also verified that weak FT8 signals can be successfully and correctly decoded in the presence of an in-passband interfering signal that many 10's of dB stronger than the wanted signals.
Retired from a long career in pro audio, I didn't need to make any measurements to predict that. It's the nature of A/D conversion. :)

73, Jim K9YC


Martin G0HDB
 

On Thu, Sep 29, 2022 at 06:44 PM, Ron wrote:


My apologies, I directed you incorrectly in my earlier post. Clipping occurs
when you have your receive audio is set too high.

To correct this issue, do the following:

Right click speaker icon in system tray>Sounds>Recording>”Levels” Tab
–Configure so Receive Audio Level is about 30 dB when no signal present.
It's a good idea to test this when connected to a dummy load. Make sure mic
isn't muted (small icon to right of recording level slider bar.
Ron:

Setting the audio level with the receiver connected to a dummy load won't be a realistic scenario because band noise that's received with an antenna connected will almost inevitably be considerably higher than the thermal noise generated by the dummy load.

I live in what I believe is a fairly typical semi-rural environment albeit with several other houses within a few 10s of metres of my antennas and even on a quiet band such as 10m when I disconnect the antenna from my receiver and connect a dummy load instead the noise level as displayed on the WSJT-X 'thermometer' drops by around 20dB. This indicates that the received ambient noise in my environment is approx 20dB higher than the thermal noise generated in a dummy load.

Anyone who finds that disconnecting their antenna from the receiver results in little or no change in the reading on the WSJT-X 'thermometer' when receiving only band noise is indeed very fortunate - their ambient noise level must be extremely low!

--
Martin G0HDB


Reino Talarmo
 

Anyone who finds that disconnecting their antenna from the receiver results in little or no change in the reading on the WSJT-X 'thermometer' when receiving only band noise is indeed very fortunate - their ambient noise level must be extremely low!
Or their antenna or receiver is not working properly at all, hi!

73, Reino OH3mA


Ron
 

<snip>

Ron:

Setting the audio level with the receiver connected to a dummy load won't be a realistic

<end snip>

Just sharing what has worked well for me and how I was taught to setup the app.

73,

Ron - KJ5XX


Michael Black
 

Or their RF gain is turned down.  The S-Meter on most rigs (non SDR rigs) is based on RF gain and is useless for anything but self-reference and is NOT a measure of signal strength at the antenna or anywhere else really.  It's really not even good for self reference.
Mathematically it works like this:
SMeter = S*GainWhere S is the signal level in volts at the antenna and that gets multiplied by the RFGain.  In most every SDR system the SMeter is correct and  measures the antenna signal only (no gain).
So since gain is involved the SMeter scale no longer means 6dB per interval.  It's either greater or less than that so makes the SMeter horribly inaccurate.
This is a well known problem.
Mike W9MDB



On Thursday, September 29, 2022 at 11:47:52 PM CDT, Reino Talarmo <reino.talarmo@...> wrote:

>Anyone who finds that disconnecting their antenna from the receiver results in little or no change in the reading on the WSJT-X 'thermometer' when receiving only band noise is indeed very fortunate - their ambient noise level must be extremely low!

Or their antenna or receiver is not working properly at all, hi!

73, Reino OH3mA