Re: Morphy Richards DRM27024

jamie_p84@xxxxxxxxxx wrote:
On Apr 25, 10:36 pm, "DAB sounds worse than FM" <dab.is@dead> wrote:
DAB sounds worse than FM wrote:

And as for N2 being cancelled, how do you propose it would be
cancelled? You're wrong. You're trying to suggest that a wider
bandwidth noise signal could cancel a narrower bandwidth noise
signal. How the *** would that work then?

Here's a way that might convince you that you're wrong: the FFT is a
linear operation, so if you took the FFT of two noise signals with
different spectra, if they were equal you should be able to subtract
the two spectral representations and have a result of zero across
the whole spectrum. But as the noise spectra are different, if you
subtracted one from teh other you obviously wouldn't get zero, hence
the two signals cannot be the same.

You don't appear to understand what I or the others in this thread are
saying.
You originally said that FM-stereo noise (N2)


N2 is the noise on the L-R signal, and FM stereo channel reconstruction is
given by these equations (N1 is the noise on the L+R signal):

Left = 0.5 [(L+R+N1) + (L-R+N2)] = 0.5 x (2L + N1 + N2)
Left = L + (N1 + N2)/2

Right = 0.5 [(L+R+N1) - (L-R+N2)] = 0.5 x (2R + N1 - N2)
Right = R + (N1 - N2)/2


was still audible when
listening to the output of an FM stereo tuner on a single mono speaker
carrying L+R channels combined,


No, when the setting was stereo it wasn't carrying the combined L+R signal -
that's the whole point, and there's nothing surprising about what I was
hearing.

The device was a Morphy Richards 27024 DAB/DRM/FM/MW/etc portable radio that
only has one speaker, and the receiver module, which uses software-defined
radio (i.e. the radio is demodulated etc in software, not hardware, apart
from the RF front-end doing the downconversion to IF), was set to stereo in
the settings on the radio. So the software was doing all the processing for
stereo, and then the output audio signal will simply have been one channel
of the "stereo" signal. The Roberts RD20 was also incorrectly defaulting to
stereo, and this was also using one of Radioscape's new (at the time)
receiver modules.

Radioscape won't make separate receiver modules for 1 and 2 speaker radios.
They will make one module that can be used for both, and the manufacturers
will just have to connect the single speaker to one of the pins on the
module.

So when you say I was still hearing hiss when listening to L+R mono, that's
not what I said, and that's not what I was listening to. I was listening to
the stereo reconstruction of one of the two channels.


and that switching the FM tuner to
mono mode would remove or reduce this hiss, and therefore the tuner
should have been set to FM mono mode by default.


This bit is right.


We're not talking about hiss which may be audible on the FM carrier
when listening in mono mode (N1). That doesn't come into it at all.

Allow me to reiterate:
N1 is unwanted audible noise on L+R which would be present regardless
of whether you're listening in stereo or mono.
N2 is unwanted noise derived from the demodulated L-R 'difference
signal' subcarrier.
In the left channel, we are just adding L+R and L-R together, so in
terms of noise in the left channel we have N1 + N2.
In the right channel, we are subtracting L-R from L+R (subtracting
basically means inverting L-R and then adding it to L+R). So in terms
of noise in the right channel we have N1 - N2.


Correct.


So if we switch the FM radio to stereo and then recombine its output
to mono, in terms of noise we get (N1 + N2 - N2) = N1 hence no FM
stereo hiss.


No. N2 is not "stereo hiss". N2 is the noise on the L-R channel. Stereo hiss
is all to do with the fact that the SNR is lower. Here's what I wrote in
reply to Boltar about this:

"Say the amplitude on the left signal is 5.0 and the amplitude on the right
channel signal is 4.5 at an instant in time.

An FM signal is split as follows:

Mono = L+R = 5 + 4.5 = 9.5

Difference signal for stereo = L-R = 5.0 - 4.5 = 0.5

Thermal noise, which is the noise due to the RF front-end in the receiver,
will be flat, so the SNR for the L+R is far higher than the SNR for L-R:

Difference in SNR (dB) = 10 log10 ((L+R) / (L-R)) = 10 log10 (9/0.5) = 12.6
dB

i.e. the SNR for mono is far higher than the SNR for stereo, hence changing
from stereo to mono will very likely eliminate hiss on stations that are a
bit hissy on stereo."


--
Steve - www.digitalradiotech.co.uk - Digital Radio News & Info

The adoption of DAB was the most incompetent technical
decision ever made in the history of UK broadcasting:
http://www.digitalradiotech.co.uk/dab/incompetent_adoption_of_dab.htm


.