Choppy Audio on Nexedge 48 with both DSDplus and WinDSD

Status
Not open for further replies.

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
Hi All,

I am getting choppy audio on both DSDPlus and WinDSD programs monitoring Nexedge 4800 systems on 3 radio receivers. ICOM R-7000 tapped with a direct connection to the discriminator (resistor on top board), ICOM R-8500 via the rear jack, and the PSR-800 "IF out" jack.

Other protocols like DMR and P25 sound much better.

Is this due to the channel's narrow bandwidth, the discriminator tap, or the programs themselves? If it's the discriminator, is there any way to improve the circuit to filter out any potential dc bias or noise? I don't have test equipment to see if the disc audio is clean and a flat response. Would an input buffer help in the circuit?

thanks,
 

slicerwizard

Member
Joined
Sep 19, 2002
Messages
7,643
Location
Toronto, Ontario
Is this due to the channel's narrow bandwidth, the discriminator tap, or the programs themselves? If it's the discriminator, is there any way to improve the circuit to filter out any potential dc bias or noise? I don't have test equipment to see if the disc audio is clean and a flat response. Would an input buffer help in the circuit?
If you posted raw audio recordings, those questions could be answered.
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
I recorded them at 48KHz 8 bit mono adjusting volume where there are the fewest errors on the control channel.

The PSR800 sounds the best followed by ICOM R800 and then the worst is ICOM R7000. I have recordings of the CC and voice for all three below:



PSR800 Control
https://dl.dropboxusercontent.com/u/73582985/800cc.wav

PSR800 Voice
https://dl.dropboxusercontent.com/u/73582985/800voice.wav

R7000 Control
https://dl.dropboxusercontent.com/u/73582985/7000cc.wav

R7000 Voice
https://dl.dropboxusercontent.com/u/73582985/7000voice.wav

R8500 Control
https://dl.dropboxusercontent.com/u/73582985/8500cc.wav

R8500 Voice
https://dl.dropboxusercontent.com/u/73582985/8500voice.wav

Thanks for helping out!

If you posted raw audio recordings, those questions could be answered.
 

racingfan360

Member
Joined
Dec 19, 2005
Messages
1,158
jcardani - the samples sound a bit choppy, but I've heard much worse.

Have you tried forcing DSD+ to run with the -fn function to optimize/focus just on NXDN 4800 decoding?
 

mmisk

Member
Joined
Jul 4, 2003
Messages
772
Location
Ottawa Canada
How is your signal strength?

My experience is (so far) that digitized audio does not work well if the signal is weak.

I would like to know what other listeners opinion is on this.
 

racingfan360

Member
Joined
Dec 19, 2005
Messages
1,158
>I would like to know what other listeners opinion is on this.
Agreed...signal strength makes a big difference to decode quality, as does sampling rate of the sound card (the higher the better IMHO)
 

racingfan360

Member
Joined
Dec 19, 2005
Messages
1,158
Did you try -fi or -fn? It should be -fn

I've also tuned my setup using the previous version of dsd+ with dsdstune.
(dsdtune doesn't work yet with this new release of dsdplus)
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
I think that the best way would be for someone to look at the raw audio with a scope and figure if there is an issue. I believe that slicerwizard may have the capability.

Also I can dig up the sound card buffer circuit several members here and I worked on back in 2010 if it will improve things. Then it was a switchable between a low pass filter (for LTR & Passport) and unfiltered for everything else. Since the filtering is done in software, I removed all but the buffer and some external components to improve sound card decoding. I can post this again if it would be of help. It's not really that complicated any more.
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
Update:

A 10uF cap in series with the discriminator (between discriminator audio and sound card) helped a little. A 10K to 20K resistor in series after did not affect the decode positive or negative. Also tried a 5-10K resistor taken to ground but caused more errors so that was not implemented.

Can someone please analyze the sound files above with test equipment to see if there are any issues with the above scanners' discriminator waveform such as dc bias, bad audio response, etc?

thanks!
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
Hi,

Thanks to jhampton2000 and mmisk. I tried your suggestions and there is some improvement but still the choppiness continues.

I still have not received any help about technically why these audio recordings are choppy. Most frames appear error free but there is still a choppy sound in the audio occasionally.

If someone can look at the recordings with a scope and let me know if there is anything wrong with the discriminator audio it would be greatly appreciated. I can't fix this issue without assistance.

thanks,
 

DSDPlus

Member
Joined
Jul 24, 2014
Messages
377
Can someone please analyze the sound files above with test equipment to see if there are any issues with the above scanners' discriminator waveform such as dc bias, bad audio response, etc?
The recordings are 8 bit, not good for analyzing. You did not use DSD+ recording function? PSR800 audio is very low, so there is a interface problem? But all files decode ok in DSD+ here, so can't help you more.
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
Hi DSDPlus,

I did not know about the raw audio recording function in DSDPlus until after. They were recorded in regular windows app. I can re-record with DSDPlus and post again if that would be helpful.

Kind of strange that some the audio is choppy even with no errors. I wonder if it's ambient noise from wind or something in the background.

thanks for your help!
 

DSDPlus

Member
Joined
Jul 24, 2014
Messages
377
Hi DSDPlus,

I did not know about the raw audio recording function in DSDPlus until after. They were recorded in regular windows app. I can re-record with DSDPlus and post again if that would be helpful.
Repost if you wish.


Kind of strange that some the audio is choppy even with no errors. I wonder if it's ambient noise from wind or something in the background.
There is no guarantee that the audio you posted exactly matches what DSD+ sees. Only recording through DSD+ can do that.
 

jcardani

Member
Premium Subscriber
Joined
Jan 16, 2002
Messages
1,390
Location
Orlando, FL & Ocean City, NJ
Hi DSDPlus,

I finally got a chance to record the raw audio using the R command. Here is my public dropbox link with several short files. Also please excuse the colorful language on the recordings, these are inner city (Philadelphia) tow truck drivers in Philadelphia that wreck chase.

Link to filter is here:
https://www.dropbox.com/sh/6a3vm73n3olds6m/AACFsNSPELpIJYLaN4U0pnDca?dl=0


Please let me know what you find.

thanks,
joe
 
Status
Not open for further replies.
Top