K3 vs. HPSDR Mercury receiver

[Dave, G4AON]

Members of the RSGB may wish to sneak a preview of the December RadCom
article on SDR where the authors compare a K3 with a HPSDR Mercury
receiver and explain why the SDR sounds better than the K3 (noise
through a crystal filter causing phase changes).

http://www.rsgb.org/membersonly/radcom/techfeatures/sdr_1208.pdf

Before you ask, I will not pass the article to non members as it is
copyright material.

73 Dave, G4AON
K3/100 #80 and Perseus SDR.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Alan Bloom]

One way to get better-sounding audio, both on transmit and receive, is
to use a wider-bandwidth crystal filter and let the DSP do most of the
filtering.  Because the digital filter is linear-phase it will sound
much more natural than an analog non-linear-phase filter.

That's why I leave ESSB enabled all the time on transmit, with the
bandwidth set for minimum.

Al N1AL


On Fri, 2008-11-07 at 10:55, Dave G4AON wrote:

> Members of the RSGB may wish to sneak a preview of the December RadCom
> article on SDR where the authors compare a K3 with a HPSDR Mercury
> receiver and explain why the SDR sounds better than the K3 (noise
> through a crystal filter causing phase changes).
>
> http://www.rsgb.org/membersonly/radcom/techfeatures/sdr_1208.pdf
>
> Before you ask, I will not pass the article to non members as it is
> copyright material.
>
> 73 Dave, G4AON
> K3/100 #80 and Perseus SDR.
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft   
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Julian, G4ILO]

Alan Bloom wrote

One way to get better-sounding audio, both on transmit and receive, is
to use a wider-bandwidth crystal filter and let the DSP do most of the
filtering.  Because the digital filter is linear-phase it will sound
much more natural than an analog non-linear-phase filter.

That's why I leave ESSB enabled all the time on transmit, with the
bandwidth set for minimum.

If crystal filters distort the phase enough to make an audible difference on speech how the heck do we ever manage to receive PSK through them?

[Alan Bloom]

On Fri, 2008-11-07 at 13:44, Julian, G4ILO wrote:

> If crystal filters distort the phase enough to make an audible difference on
> speech how the heck do we ever manage to receive PSK through them?

Most of the phase distortion tends to occur at the band edges.  As long
as the PSK signal is significantly narrower than the crystal filter,
distortion is minimized.

Al N1AL

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[KK7P]

> If crystal filters distort the phase enough to make an audible difference on
> speech how the heck do we ever manage to receive PSK through them?

The phase distortion is most pronounced near the corner frequency of the
  crystal filter.

PSK, at least PSK31, is a very narrow mode and the chang ein delay
(phase) through the filter over such a narrow bandwidth is usually
minor, as long as you don't operate at a corners.

73,

Lyle KK7P
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Jim Brown-10]

On Fri, 7 Nov 2008 13:44:26 -0800 (PST), Julian, G4ILO wrote:

>If crystal filters distort the phase enough to make an audible difference on
>speech how the heck do we ever manage to receive PSK through them?

Phase shift is greatest where amplitude response is changing rapidly -- that
is, at the skirts. Mostly we decode PSK near the center of the passband. Also,
PSK is quite narrowband, so the phase shift over the relatively narrow
bandwidth is generally minimal.

73,

Jim K9YC


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[David Woolley (E.L)]

Dave G4AON wrote:
> SDR where the authors compare a K3 with a HPSDR Mercury receiver and
> explain why the SDR sounds better than the K3 (noise through a crystal
> filter causing phase changes).

The K3 is also an SDR!


--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Dave, G4AON]

The explanation from the article is as follows:

"When noise pulses/spikes pass through a crystal
filter, the phase response of the filter
changes, depending on the noise frequency.
However, when noise pulses/spikes pass
through an ADC with a linear response, the
phase response stays the same, because the
ADC treats them in a linear manner."

While the K3 is "SDR", it also includes crystal roofing filters which
the Mercury does not. My Perseus SDR (which also operates with a high
speed ADC for direct decoding - no crystals or mixers) does deal with
noisy signals better than my K3.

Next generation transceivers will not use crystal filters and mixers,
they are already appearing and the performance is outstanding.

73 Dave, G4AON
K3/100 #80
-----------------------------------------
 >/ SDR where the authors compare a K3 with a HPSDR Mercury receiver and
/>/ explain why the SDR sounds better than the K3 (noise through a crystal
/>/ filter causing phase changes).
/
The K3 is also an SDR!

--
David Woolley
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Joe Subich, W4TV-3]

> Next generation transceivers will not use crystal filters
> and mixers, they are already appearing and the performance
> is outstanding.

If the next generation of transceivers does in fact abandon
narrow filtering ahead of the ADC it will be a sorry step
backward.  While some of the designs like the Flex-5000 and
Peresus have excellent specs for the most part when tested
against two tone situations, they are still substandard in
dealing with multiple very strong in-band signals as would
occur on 160/80/75 meters in a contest or on 40 meters with
the multiple megapower broadcasters.

To maintain linearity, an ADC must be able to withstand the
instantaneous peak signal (vector sum - not average) from
all the signals within its passband.  For some DSR designs
that maintain bandwidths from several hundred KHz to 10s
of MHz that peak signal handling capability requirement
can be significant and represents a required dynamic range
orders of magnitude greater than even the best current
technology.  It is the multi-tone interference that is best
handled by narrow filtering ahead of the ADC (e.g., the K3
design).  

73,

   ... Joe, W4TV
 

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Dave G4AON
> Sent: Saturday, November 08, 2008 6:38 AM
> To: [hidden email]
> Subject: Re: [Elecraft] K3 vs. HPSDR Mercury receiver
>
>
> The explanation from the article is as follows:
>
> "When noise pulses/spikes pass through a crystal
> filter, the phase response of the filter
> changes, depending on the noise frequency.
> However, when noise pulses/spikes pass
> through an ADC with a linear response, the
> phase response stays the same, because the
> ADC treats them in a linear manner."
>
> While the K3 is "SDR", it also includes crystal roofing filters which
> the Mercury does not. My Perseus SDR (which also operates with a high
> speed ADC for direct decoding - no crystals or mixers) does deal with
> noisy signals better than my K3.
>
> Next generation transceivers will not use crystal filters and mixers,
> they are already appearing and the performance is outstanding.
>
> 73 Dave, G4AON
> K3/100 #80
> -----------------------------------------
>  >/ SDR where the authors compare a K3 with a HPSDR Mercury
> receiver and />/ explain why the SDR sounds better than the
> K3 (noise through a crystal />/ filter causing phase
> changes). / The K3 is also an SDR!
>
> --
> David Woolley
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft   
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill W4ZV]

Dave G4AON wrote

The explanation from the article is as follows:

"When noise pulses/spikes pass through a crystal
filter, the phase response of the filter
changes, depending on the noise frequency.
However, when noise pulses/spikes pass
through an ADC with a linear response, the
phase response stays the same, because the
ADC treats them in a linear manner."

While the K3 is "SDR", it also includes crystal roofing filters which
the Mercury does not. My Perseus SDR (which also operates with a high
speed ADC for direct decoding - no crystals or mixers) does deal with
noisy signals better than my K3.

Next generation transceivers will not use crystal filters and mixers,
they are already appearing and the performance is outstanding.

73 Dave, G4AON

Blocking Dynamic Range (BDR) is the weak link of current SDR receivers (typically 120-125 dB versus ~140 dB in the K3).  That will remain the weak link until ADCs with higher resolution that the current 24 bits is exceeded.  BDR is important in multi-transmitter environments or if you are close to a neighbor (amateur or commercial BC station) that has very high signal levels (which trigger ADC front-end protection and thus limit BDR).  

Until the BDR of current generation SDRs is improved, we will not see pure SDR rigs (without crystal filter front ends) be competitive in contest environments.  

73,  Bill  W4ZV

[N8LP]

For casual SSB operating I listen to my LP-PAN / PowerSDR panadapter which is fed from ahead of the xtal filters, and for dabbling in contests or operating CW (where latency can be an issue) I listen to the K3. I am not in Europe where things can be dicey, but I have yet to see enough total peak energy in the passband to affect the ADC of the panadapter.

73,
Larry N8LP



<quote author="Joe Subich, W4TV-3">

> Next generation transceivers will not use crystal filters
> and mixers, they are already appearing and the performance
> is outstanding.

If the next generation of transceivers does in fact abandon
narrow filtering ahead of the ADC it will be a sorry step
backward.  While some of the designs like the Flex-5000 and
Peresus have excellent specs for the most part when tested
against two tone situations, they are still substandard in
dealing with multiple very strong in-band signals as would
occur on 160/80/75 meters in a contest or on 40 meters with
the multiple megapower broadcasters.

To maintain linearity, an ADC must be able to withstand the
instantaneous peak signal (vector sum - not average) from
all the signals within its passband.  For some DSR designs
that maintain bandwidths from several hundred KHz to 10s
of MHz that peak signal handling capability requirement
can be significant and represents a required dynamic range
orders of magnitude greater than even the best current
technology.  It is the multi-tone interference that is best
handled by narrow filtering ahead of the ADC (e.g., the K3
design).  

73,

   ... Joe, W4TV
 

[David Woolley (E.L)]

Dave G4AON wrote:
> Members of the RSGB may wish to sneak a preview of the December
> RadCom article on SDR where the authors compare a K3 with a HPSDR
> Mercury receiver and explain why the SDR sounds better than the K3
> (noise through a crystal filter causing phase changes).

I've read the article and it is really comparing two types of SDR
receiver, rather than SDR and analogue.

The type they are claiming the benefit for is one in which digitisation
is performed at signal frequency.  The other type is the design used by
both SoftRock and K3, in which the signal is first mixed down before
being digitised.

The sort of noise they are talking about is impulsive noise,
specifically atmospherics due to global lightening.  A filter with
non-constant group delay will turn these from clicks to chirps.

It is possible the SoftRock will do better in this respect, because
typical sound cards have a constant group delay anti-aliasing filter.
>
> http://www.rsgb.org/membersonly/radcom/techfeatures/sdr_1208.pdf
>
> </div>


--
David Woolley
"The Elecraft list is a forum for the discussion of topics related to
Elecraft products and more general topics related ham radio"
List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm>
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Geoffrey Mackenzie-Kennedy]

David,

If you have not read it, the December 2008 issue of QST contains another
review of the Perseus.

Although I own a Perseus, which I bought for use as a piece of test
equipment, I do not own a K3 thus cannot compare their 'sound'. I can say
though that I have never been comfortable with the 'sound' produced by most
commercial amateur and military receivers which use DSP, but I do find that
the 'sound' produced by Perseus, especially when a CW, SSB or AM signal is
close to the ambient noise floor, is certainly not hard on the ears - it is
quite mellow. Both its NB and NR work well.

73,
Geoff
GM4ESD



David Woolley (E.L) wrote:

> Dave G4AON wrote:
>> Members of the RSGB may wish to sneak a preview of the December
>> RadCom article on SDR where the authors compare a K3 with a HPSDR
>> Mercury receiver and explain why the SDR sounds better than the K3
>> (noise through a crystal filter causing phase changes).
>
> I've read the article and it is really comparing two types of SDR
> receiver, rather than SDR and analogue.
>
> The type they are claiming the benefit for is one in which digitisation is
> performed at signal frequency.  The other type is the design used by both
> SoftRock and K3, in which the signal is first mixed down before being
> digitised.
>
> The sort of noise they are talking about is impulsive noise, specifically
> atmospherics due to global lightening.  A filter with non-constant group
> delay will turn these from clicks to chirps.
>
> It is possible the SoftRock will do better in this respect, because
> typical sound cards have a constant group delay anti-aliasing filter.
>>
>> http://www.rsgb.org/membersonly/radcom/techfeatures/sdr_1208.pdf
>>
>> </div>
>
>
> --
> David Woolley

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com