On 06/18/2020 07:27 AM, Jeff Long wrote:
Ah, it's a Python heir block. You could reduce the quadrature rate
further so that the quadrature rate is closer to 200 kS/s and/or put a
200 kHz LPF in front of the block. There's internal filtering for the
various components of the baseband signal, but no internal filtering
before the PLL. It could also be just that FM stereo is more sensitive
to noise (try GQRX stereo on a station that sounds good in mono).
FM stereo is notoriously sensitive to multi-path, which can make a
signal that sounds really good in mono sound like crap in stereo.
On Thu, Jun 18, 2020 at 6:41 AM Jeff Long <willco...@gmail.com
<mailto:willco...@gmail.com>> wrote:
If there's no filter inside "WBFM Receive PLL", place one before
it. The example in gr-uhd has a 400 kHz filter there.
On Tue, Jun 16, 2020 at 8:46 PM Barry Duggan <ba...@dcsmail.net
<mailto:ba...@dcsmail.net>> wrote:
Hi,
I have a broadcast FM (mono) receiver which works well, with
good audio
and clean traces in the Time Domain. See
https://wiki.gnuradio.org/index.php/File:USRP_FM_fg.png
Then I made a stereo version. See
https://wiki.gnuradio.org/index.php/File:USRP_FM_stereo_fg.png
Using
it, the sound has a distortion but does not have any audio
underruns.
The traces in the Time Domain have a lot of "fuzz" (noise?).
Do I need a filter? If so, would it come before or after the WBFM
Receive PLL block?
Once I have a good design, I will use it as an Example Flowgraph.
Thanks for your help.
--
Barry Duggan KV4FV
