locked Received Audio Level scale floor different with different sound cards


wa0zti <wa0zti@...>
 
Edited

I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.

When using my built in sound card with nothing connected to the audio input port, the wsjtx received audio level meter
sits about 5db (like the sound card has a 5db noise floor). 
When using the new USB sound card with nothing connected to the audio input port, the wsjtx received audio level 
meter sits at 40db. OK so the cheap sound card might have a hum or high noise floor. However, if I configure the PC to
listen to the mic input on the deck top speakers, I hear nothing. 
The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?

Thanks


Bill Somerville
 

On 17/08/2021 21:20, wa0zti via groups.io wrote:
I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.

When using my built in sound card with nothing connected to the audio input port, the wsjtx received audio level meter
sits about 5db (like the sound card has a 5db noise floor).
When using the new USB sound card with nothing connected to the audio input port, the wsjtx received audio level
meter sits at 40db. OK so the cheap sound card might have a hum or high noise floor. However, if I configure the PC to
listen to the mic input on the deck top speakers, I hear nothing.
The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?

Thanks
OM,

you should not really be surprised at strange levels being sampled from a high impedance source, if the input has no DC path to ground the ADC will sample random levels as any passing electrons vary the input level. So long as the sound card behaves well with a properly matched source there's nothing to worry about.

73
Bill
G4WJS.


Bob McGraw - K4TAX <rmcgraw@...>
 

Sounds as though there are different gain structures between the sound cards.  I can't imagine, if the two gain structures are identical, there is a 35 dB difference in the noise floor.  

A man with one watch always knows what time it is.  A man with 2 or more watches is never quite sure of the time. 

73
Bob, K4TAX


Martin G0HDB
 

On Tue, Aug 17, 2021 at 10:33 PM, wa0zti wrote:
I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.
[Snipped]

The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?
Have you checked the Windows settings for the new USB sound card and confirmed that the Mic gain is set to 0dB and also that there are no enhancements enabled?  I believe some sound devices can be set to provide an additional 20dB of mic gain, to accommodate low-output microphones, so you definitely don't want to have that enabled.

Also, once you've ensured that there's no Mic gain enabled for the USB sound device you might be able to reduce the audio output level from your rig (depending on the make and model of the rig) so that you get a low reading on the WSJT-X audio level 'thermometer' when there's no input whatsoever to the sound device.

--
Martin G0HDB


wa0zti <wa0zti@...>
 

Martin,

    Your comments turned up something.
    The cheap USB sound card input is in fact a mic input (not Line-in). I had checked earlier for a mic gain setting
as you suggested some mic inputs have. It did not have a mic gain setting, just a level setting. So I forgot all about
it.
     Prior to your comment, I had checked the behavior of the mic input on my internal (mother board) sound card. I found
the receiver signal floor was 10db (higher than the line-in of the same internal sound card that had a 5db floor). But more
importantly, that mic input does have a mic gain control. So after your comment, I adjusted it to see what would happen. 
When the gain is set to 0db, the WSJT-X signal floor is <5db (much like the line-in). At +40db (max), the WSJT-X signal floor is 40db ( just like the cheap USB mic input).
     I had previously measured the USB sound card mic input signal levels (gain) and did not think it was excessive. But 
clearly it is higher than the internal sound card line-in and that (plus maybe its general noise performance of the card) is 
effecting the WSJT-X signal floor. 
      Moral of the story, its best to use a line-in. If you must use a mic input, get one with a gain control.

Good suggestion, Thanks

Derek

On Thursday, August 19, 2021, 04:09:05 AM MDT, Martin G0HDB <marting0hdb@...> wrote:


On Tue, Aug 17, 2021 at 10:33 PM, wa0zti wrote:
I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.
[Snipped]


The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?
Have you checked the Windows settings for the new USB sound card and confirmed that the Mic gain is set to 0dB and also that there are no enhancements enabled?  I believe some sound devices can be set to provide an additional 20dB of mic gain, to accommodate low-output microphones, so you definitely don't want to have that enabled.

Also, once you've ensured that there's no Mic gain enabled for the USB sound device you might be able to reduce the audio output level from your rig (depending on the make and model of the rig) so that you get a low reading on the WSJT-X audio level 'thermometer' when there's no input whatsoever to the sound device.

--
Martin G0HDB




Alan G4ZFQ
 

When the gain is set to 0db, the WSJT-X signal floor is <5db (much like the line-in).
Derek

So 0dB is OK?
I've used stereo single input sound devices like that for soundcard SDR, rather more critical than WSJT, and have been satisfied by results.
Often it seems such devices are designed like that, minimum volume is equivalent to a line input.

73 Alan G4ZFQ


wa0zti <wa0zti@...>
 
Edited

 
"So 0dB is OK?"
In my case yes. But I want to be clear. In my internal sound card (Motherboard), the mic input has a gain control
AND a level control. Its the gain control that makes the difference in WSJT-X receiver level floor as shown the 
the attached screen shots. 
The USB sound card has no gain control, just the level control. If the level control is set to 0, no RX audio gets
to the software to be decoded. Yet the receive level floor in WSJT-X is still 40db.
That is just plain strange.  Maybe its just a poor sound card except I don't hear the noise when I configure the
desk top speakers to listen to the mic input.
 
Derek
 
On Thursday, August 19, 2021, 08:46:18 AM MDT, Derek Toeppen <wa0zti@...> wrote:
 
 
 
Martin,
 
    Your comments turned up something.
    The cheap USB sound card input is in fact a mic input (not Line-in). I had checked earlier for a mic gain setting
as you suggested some mic inputs have. It did not have a mic gain setting, just a level setting. So I forgot all about
it.
     Prior to your comment, I had checked the behavior of the mic input on my internal (mother board) sound card. I found
the receiver signal floor was 10db (higher than the line-in of the same internal sound card that had a 5db floor). But more
importantly, that mic input does have a mic gain control. So after your comment, I adjusted it to see what would happen. 
When the gain is set to 0db, the WSJT-X signal floor is <5db (much like the line-in). At +40db (max), the WSJT-X signal floor is 40db ( just like the cheap USB mic input).
     I had previously measured the USB sound card mic input signal levels (gain) and did not think it was excessive. But 
clearly it is higher than the internal sound card line-in and that (plus maybe its general noise performance of the card) is 
effecting the WSJT-X signal floor. 
      Moral of the story, its best to use a line-in. If you must use a mic input, get one with a gain control.
 
Good suggestion, Thanks
 
Derek
 
On Thursday, August 19, 2021, 04:09:05 AM MDT, Martin G0HDB <marting0hdb@...> wrote:
 
 
On Tue, Aug 17, 2021 at 10:33 PM, wa0zti wrote:
I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.
[Snipped]


The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?
Have you checked the Windows settings for the new USB sound card and confirmed that the Mic gain is set to 0dB and also that there are no enhancements enabled?  I believe some sound devices can be set to provide an additional 20dB of mic gain, to accommodate low-output microphones, so you definitely don't want to have that enabled.

Also, once you've ensured that there's no Mic gain enabled for the USB sound device you might be able to reduce the audio output level from your rig (depending on the make and model of the rig) so that you get a low reading on the WSJT-X audio level 'thermometer' when there's no input whatsoever to the sound device.

--
Martin G0HDB




Ron / W4MMP
 

Hi Derek,

I have the same exact situation.  Just little back ground:  I was using a SIIG USB/Audio adapter (I have no clue which audio chip it implements).  About two Windows updates ago,  something weird started occurring.  Occasionally and intermittently,  the  WSJT-X volume indicator go to 100 and get stuck there.  There would be no decodes when this happens.  I tested with Fldigi and the same thing happened.  Restarting either application clears the issue (for a while).  It appears the driver Windows installed and the SIIG chip don't get along very well. 

So, I switched to a C-Media USB/Audio adapter.  I then found the exact same thing as you have reported.  With the level set to zero, no decodes but WSJT-X reports a value of 40db.   This does not happen with the SIIG adapter.   When the SIIG adapter level is set to zero,  WSJT-X reports zero (and obviously no decodes).  Note: nothing was changed but the adapter, not even the USB port. 

73,
Ron / W4MMP
On 8/19/2021 15:00, wa0zti via groups.io wrote:

"So 0dB is OK?"
In my case yes. But I want to be clear. In my internal sound card (Motherboard), the mic input has a gain control
AND a level control. Its the gain control that makes the difference in WSJT-X receiver level floor as shown the 
the attached screen shots. 
The USB sound card has no gain control, just the level control. If the level control is set to 0, no RX audio gets
to the software to be decoded. Yet the receive level floor in WSJT-X is still 40db.
That is just plain strange.  Maybe its just a poor sound card except I don't hear the noise when I configure the
desk top speakers to listen to the mic input.

Derek

On Thursday, August 19, 2021, 08:46:18 AM MDT, Derek Toeppen <wa0zti@...> wrote:


Martin,

    Your comments turned up something.
    The cheap USB sound card input is in fact a mic input (not Line-in). I had checked earlier for a mic gain setting
as you suggested some mic inputs have. It did not have a mic gain setting, just a level setting. So I forgot all about
it.
     Prior to your comment, I had checked the behavior of the mic input on my internal (mother board) sound card. I found
the receiver signal floor was 10db (higher than the line-in of the same internal sound card that had a 5db floor). But more
importantly, that mic input does have a mic gain control. So after your comment, I adjusted it to see what would happen. 
When the gain is set to 0db, the WSJT-X signal floor is <5db (much like the line-in). At +40db (max), the WSJT-X signal floor is 40db ( just like the cheap USB mic input).
     I had previously measured the USB sound card mic input signal levels (gain) and did not think it was excessive. But 
clearly it is higher than the internal sound card line-in and that (plus maybe its general noise performance of the card) is 
effecting the WSJT-X signal floor. 
      Moral of the story, its best to use a line-in. If you must use a mic input, get one with a gain control.

Good suggestion, Thanks

Derek

On Thursday, August 19, 2021, 04:09:05 AM MDT, Martin G0HDB <marting0hdb@...> wrote:


On Tue, Aug 17, 2021 at 10:33 PM, wa0zti wrote:
I have been using wsjtx for months with the sound card built into my PC mother board. I decided I wanted to free it up
for other things so I purchased a simple USB sound card. It simply has a headphone/speaker port and a microphone
port. And I have run into something interesting with the receiver audio when I tried to use it.
[Snipped]


The 40db meter floor does not appear on the water fall display as noise and I have no trouble decoding FT-8 signals.
Any ideas why wsjtx is reacting differently to the two cards?
Have you checked the Windows settings for the new USB sound card and confirmed that the Mic gain is set to 0dB and also that there are no enhancements enabled?  I believe some sound devices can be set to provide an additional 20dB of mic gain, to accommodate low-output microphones, so you definitely don't want to have that enabled.

Also, once you've ensured that there's no Mic gain enabled for the USB sound device you might be able to reduce the audio output level from your rig (depending on the make and model of the rig) so that you get a low reading on the WSJT-X audio level 'thermometer' when there's no input whatsoever to the sound device.

--
Martin G0HDB




Alan G4ZFQ
 

On 19/08/2021 20:00, wa0zti via groups.io wrote:
The USB sound card has no gain control, just the level control.
Derek,

As Bill has said, if it works satisfactorily then do not worry. Different devices do not always match.
It may be that the best driver is not loaded, maybe just Windows generic USB sound driver.
Maybe the manufacturer of the chip has a different driver.

73 Alan G4ZFQ


Michael Black
 

I'm seeing a lot of noise on your waterfall.   I've seen problems with some computers having really bad internal noise.

Let's talk signal levels....

Speaker level is in the region of 1V to 100V
Pro Line level is in the region of 1.228V (something we don't see in our ham equipment)
Consumer Line level is in the region of 316 mV
Microphone level is in the region of 2.5mV to 23mV -- this is where the 20dB boost is needed for those microphones that are at the low end of this range.

I've seen people plug their headphone jack into a microphone -- talk about over driving the audio chain!!!! You have to turn down the rig's speaker volume to just about minimum to make this work without harmonics....all the splatter RTTY operators are used to seeing comes from a lot of the older rigs where there was no decent audio connection other than a headphone jack.

USB devices that have the gain control are likely expecting microphone levels.

We should be seeing line levels coming from the rigs.

0dB on most Windows audio driver levels means "no amplification or attenuation" and should be either mid-range or the far right of most audio drivers. 



Mike W9MDB




Martin G0HDB
 

On Thursday, August 19, 2021, 08:46:18 AM MDT, Derek Toeppen <wa0zti@...> wrote:
 
Martin,
 
    Your comments turned up something.
[Snipped]

 
Good suggestion, Thanks
 
Derek
 
Hi again Derek, here's something else for you to check...

I note that in your screenshots of the Microphone Properties you've got the microphone level set to 100%, ie. maximum, in both cases.  If you position your mouse cursor over the Microphone slider window and then right-click you should be offered the option to select a scale of either percentage or dB; you should select dB and then adjust the slider to give a reading of 0.0dB - this may or may not be at the same position as the 100% level.

On the USB sound device I use for my datamodes operating (a G4ZLP interface device), the 0.0dB point is only about a third of the way up from the bottom left-hand end of the scale; it's well below the tick-mark that's mid-way up the scale.  You might find that the 100% point at which you've currently got the level set is actually giving you quite a lot of microphone gain, which you shouldn't need (or want).

Needless to say, you should also set the microphone boost slider to the 0dB point!

--
Martin G0HDB