Trying to get DSD to work better, settings?

Status
Not open for further replies.

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
Hi all, I have SDRsharp and DSD 1.6 working with my NooElec that has the R820T tuner in it. Have it hooked up to a Hustler DCX discone up 15 feet. Been playing around with the settings and so far the best I have been able to get is about 70% decode on the strongest signals I can find. Here is a screen shot, anybody see something I should try or change? Thanks, Dale
 

Attachments

rbm

Member
Joined
Jan 25, 2005
Messages
1,395
Location
Upstate New York
Here are the settings that work for me.

Make sure you've calibrated the dongle first.
I use the procedure in this post:
http://forums.radioreference.com/software-defined-radio/285020-sdr-sharp.html#post2136932

Select 'Snap to Grid' and a step of 500 or 100 hz.
That way you can easily move up or down a little to improve decoding.

I think your bandwidth may be a little too narrow to allow for any variation.

So, I use 9500-11500 bandwidth to allow for some tuning error.
WFM seems to work much better for me than NFM.
Volume near 85%

The attached image is while decoding a distant signal that I have a lot of multi-path problems with.
The local P25 systems (power company) aren't normally that active on weekends here.

Rich


 
Last edited:

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
What kind of mode is it? And what's your PC specs
Hi, the mode was P25 and the computer is a dual core 1.6GHz with 4 GB ram running a dual boot Fedora, Windows 7 32 bit. I did try running with the waterfall turned off but did not see any difference. The computer does not seem like it is working hard, still snappy and running free.
 

Attachments

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
Here are the settings that work for me.

Make sure you've calibrated the dongle first.
I use the procedure in this post:
http://forums.radioreference.com/software-defined-radio/285020-sdr-sharp.html#post2136932

Select 'Snap to Grid' and a step of 500 or 100 hz.
That way you can easily move up or down a little to improve decoding.

I think your bandwidth may be a little too narrow to allow for any variation.

So, I use 9500-11500 bandwidth to allow for some tuning error.
WFM seems to work much better for me than NFM.
Volume near 85%

The attached image is while decoding a distant signal that I have a lot of multi-path problems with.
The local P25 systems (power company) aren't normally that active on weekends here.

Rich


Hi Rich, thanks for the info. I tried using WFM and it was by far the best improvement of anything I have tried to date. Not sure why but when I use WFM the highest I can get the inlvl is about 16 % Perhaps it would work even better if I could figure out why.
 

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
Wouldn't the direct route to better decoding be to use DSD+?
Maybe, I did try DSD+ ; however, the only thing I got it to decode was some Provoice. Does it need to be started with an option to read P25?

Another thing that I'm thinking might be part of the problem is I have not installed Unitracker yet. I have just been clicking on strong signals I find using SDRsharp. Perhaps there are different talk groups using the trunk that I clicked on and some might be encrypted???
 

reconrider8

Member
Feed Provider
Joined
Dec 26, 2007
Messages
2,575
Location
Eastern, NC
isnt your inlvl suppose to be around 5% for optimal decode? i read that somewhere and thats what i run on my desktop and i get near perfect decode everytime
 

n4yek

Member
Joined
Apr 20, 2004
Messages
2,488
Location
Newport, Tennessee
Last edited:

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
isnt your inlvl suppose to be around 5% for optimal decode? i read that somewhere and thats what i run on my desktop and i get near perfect decode everytime
Thanks for the info. I though it was suppose to be abt 30%, will give 5% a try and see what happens.
 

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
inlvl at 5% is barely any audio being feed into the program. 30% was optimal in DSD and around 40% to 45% using DSD+
I'm going to give DSD+ another try, will try 42.5% ;-)

Speaking of DSD+ I just tried starting it with the -h option and it comes up with a boatload of info. Going to try -f1 the next time I find a strong P25 signal.

Pressing - will toggle and display the current option at the top bar.
 

n4yek

Member
Joined
Apr 20, 2004
Messages
2,488
Location
Newport, Tennessee
I'm going to give DSD+ another try, will try 42.5% ;-)

Speaking of DSD+ I just tried starting it with the -h option and it comes up with a boatload of info. Going to try -f1 the next time I find a strong P25 signal.

Pressing - will toggle and display the current option at the top bar.
I was making changes when you replied to my post #10, probably need to read it again. Sorry..
 

KF4SIR

Member
Premium Subscriber
Joined
Sep 18, 2014
Messages
12
Hi all, here is an update: I now have DSD+ decoding at about 90 to 100% on strong P25 signals.

So far what has worked the best is using WFM mode, 12500 bandwidth, audio slider at the top of SDRsharp set all the way to the right (40 DB) this gave me a LVL of abt 33% when the signal is present and abt 66% when not.

I started DSD+ using the -f1 option.

Looks like my thoughts on a trunk signal carrying both open and encrypted data were correct, it gives the impression the decode is not working but DSD+ displays encrypted voice muted.

I took a look at the CPU usage and it looks like the computer is strong enough, running abt 70% usage.

Here is a screen shot of it working.

Thanks, Dale
 

Attachments

Status
Not open for further replies.
Top