3 technical questions

[k6dgw]

I understand what direct sampling is and how it works.  Why is it
preferable [or whatever other positive term you might want to use] to
the superhet design of the K3(s)?  I'm curious what sampling the entire
spectrum from 0.1 MHz to 54 MHz buys me?

If I understand the "Superhet" option for the K4 correctly
[problematical at best], it provides RF and mixer stages to an IF [~8
MHz??] which is then directly sampled by the base K4.  Since that IF
bandwidth can be adjusted by xtal filters and will be MUCH narrower than
0.1 - 54 MHz it will also be far less susceptible to overload from
strong signals not in that passband.  Is this what's going on?

For reasons I cannot remember now, I thought the K3(s) oversampled the
15 KHz 2nd IF with a delta-sigma 1-bit ADC and decimated to the
equivalent of a 22-bit word.  Some of the thredlets here have now caused
me to reconsider that belief.  Have I got it wrong?

73,
Fred ["Skip"] K6DGW
Sparks NV DM09dn
Washoe County
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Wes Stewart-2]

I'm sticking my neck out here but I think one driver is cost and that includes
assembly and tuning. The K3S for example has some hand soldered parts and if
mine is a representative sample, some selected-at-test parts, with attendant QC
issues.

Wes  N7WS

On 5/29/2019 4:01 PM, Fred Jensen wrote:
> I understand what direct sampling is and how it works.  Why is it preferable
> [or whatever other positive term you might want to use] to the superhet design
> of the K3(s)?  I'm curious what sampling the entire spectrum from 0.1 MHz to
> 54 MHz buys me?
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Bill Frantz]

Hi Skip. Let me take a stab at some of these, and thanks to
Alan, Jeff, and W2XJ for furthering my technical education about ADCs.

(1) With direct sampling, if you are operating in a contest, you
can keep an eye on other bands, like 10M in the current
conditions, while operating on bands which have propagation.
It's just another window on the screen. A superhet will severely
narrow the bandwidth in the IF stages, so you need a separate
superhet for each band (ala the 2nd receiver in the K3).

(2) As far as I can tell, the K4HD should perform as well as a
K3, or perhaps even better. Using the formula Alan posted, if
I'm tuning a 4KHz FT8 band on a K4 with its 122 MHz sample rate,
I will get a 44.8 dB improvement in S/N from the over sampling.
I would get 47.6 dB for a 2.1 KHz SSB bandwidth and 60.8 dB for
a 100 Hz CW bandwidth. (These are ranges I normally use in my
operations.) Add the 77 to 85 dB range for a real 16 bit ADC
with real support circuitry (these numbers are admittedly a
WAG), and you get some pretty respectable performance.

You also win big with the K4HD in the same way as with the K3.
The IF filters greatly suppress signals not in the passband,
such as that CW station in the tent next to your RTTY station,
(which describes the WVARA QRP field day setup).

(3) The details of the K3 ADC are beyond my current level of
knowledge, but the documentation lists it as 24 bits, as does
W2XJ's post.

73 Bill AE6JV

On 5/29/19 at 4:01 PM, [hidden email] (Fred Jensen) wrote:

>I understand what direct sampling is and how it works.  Why is
>it preferable [or whatever other positive term you might want
>to use] to the superhet design of the K3(s)?  I'm curious what
>sampling the entire spectrum from 0.1 MHz to 54 MHz buys me?
>
>If I understand the "Superhet" option for the K4 correctly
>[problematical at best], it provides RF and mixer stages to an
>IF [~8 MHz??] which is then directly sampled by the base K4. 
>Since that IF bandwidth can be adjusted by xtal filters and
>will be MUCH narrower than 0.1 - 54 MHz it will also be far
>less susceptible to overload from strong signals not in that
>passband.  Is this what's going on?
>
>For reasons I cannot remember now, I thought the K3(s)
>oversampled the 15 KHz 2nd IF with a delta-sigma 1-bit ADC and
>decimated to the equivalent of a 22-bit word.  Some of the
>thredlets here have now caused me to reconsider that belief. 
>Have I got it wrong?

-----------------------------------------------------------------------
Bill Frantz        | Privacy is dead, get over    | Periwinkle
[phone]      | it.                          | 16345
Englewood Ave
www.pwpconsult.com |              - Scott McNealy | Los Gatos,
CA 95032

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

The K4 is always doing direct sampling for panadapter purposes.

If the HDR module is turned off or is not present, the K4 also does direct sampling for demodulation.

If the HDR module is turned on, two additional ADCs are used, right at the outputs of the crystal filters (oversimplifying a bit). The original two ADCs are still being used for the panadapter.

Wayne
N6KR

> On May 29, 2019, at 5:18 PM, Bill Frantz <[hidden email]> wrote:
>
> Hi Skip. Let me take a stab at some of these, and thanks to Alan, Jeff, and W2XJ for furthering my technical education about ADCs.
>
> (1) With direct sampling, if you are operating in a contest, you can keep an eye on other bands, like 10M in the current conditions, while operating on bands which have propagation. It's just another window on the screen. A superhet will severely narrow the bandwidth in the IF stages, so you need a separate superhet for each band (ala the 2nd receiver in the K3).
>
> (2) As far as I can tell, the K4HD should perform as well as a K3, or perhaps even better. Using the formula Alan posted, if I'm tuning a 4KHz FT8 band on a K4 with its 122 MHz sample rate, I will get a 44.8 dB improvement in S/N from the over sampling. I would get 47.6 dB for a 2.1 KHz SSB bandwidth and 60.8 dB for a 100 Hz CW bandwidth. (These are ranges I normally use in my operations.) Add the 77 to 85 dB range for a real 16 bit ADC with real support circuitry (these numbers are admittedly a WAG), and you get some pretty respectable performance.
>
> You also win big with the K4HD in the same way as with the K3. The IF filters greatly suppress signals not in the passband, such as that CW station in the tent next to your RTTY station, (which describes the WVARA QRP field day setup).
>
> (3) The details of the K3 ADC are beyond my current level of knowledge, but the documentation lists it as 24 bits, as does W2XJ's post.
>
> 73 Bill AE6JV
>
> On 5/29/19 at 4:01 PM, [hidden email] (Fred Jensen) wrote:
>
>> I understand what direct sampling is and how it works.  Why is it preferable [or whatever other positive term you might want to use] to the superhet design of the K3(s)?  I'm curious what sampling the entire spectrum from 0.1 MHz to 54 MHz buys me?
>>
>> If I understand the "Superhet" option for the K4 correctly [problematical at best], it provides RF and mixer stages to an IF [~8 MHz??] which is then directly sampled by the base K4.  Since that IF bandwidth can be adjusted by xtal filters and will be MUCH narrower than 0.1 - 54 MHz it will also be far less susceptible to overload from strong signals not in that passband.  Is this what's going on?
>>
>> For reasons I cannot remember now, I thought the K3(s) oversampled the 15 KHz 2nd IF with a delta-sigma 1-bit ADC and decimated to the equivalent of a 22-bit word.  Some of the thredlets here have now caused me to reconsider that belief.  Have I got it wrong?
>
> -----------------------------------------------------------------------
> Bill Frantz        | Privacy is dead, get over    | Periwinkle
> [phone]      | it.                          | 16345 Englewood Ave
> www.pwpconsult.com |              - Scott McNealy | Los Gatos, CA 95032
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Bill Frantz]

Thanks for this explanation. Now I'm sure the K4HD will be at
least as good as the K3S and probably better.

Watching Elecraft is like watching a really good magician. With
both, you expect extraordinary things. However, with Elecraft, I
like to know how it is done. :-)

73 Bill AE6JV

On 5/29/19 at 6:09 PM, [hidden email] (Wayne Burdick) wrote:

>The K4 is always doing direct sampling for panadapter purposes.
>If the HDR module is turned off or is not present, the K4 also
>does direct sampling for demodulation.
>If the HDR module is turned on, two additional ADCs are used,
>right at the outputs of the crystal filters (oversimplifying a
>bit). The original two ADCs are still being used for the panadapter.
-----------------------------------------------------------------------
Bill Frantz        |Security, like correctness, is| Periwinkle
[phone]      |not an add-on feature. - Attr-| 16345
Englewood Ave
www.pwpconsult.com |ibuted to Andrew Tanenbaum    | Los Gatos,
CA 95032

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Edward R Cole]

Wayne, etal:

I am doing a similar thing with my K3 to view as panadapter.  I use
the LP-Pan on 8.215 IF output of my K3 and that goes to the A/D in my
Delta44 soundcard to provide a digital stream to sw in my
computer.  The 1st IF of the K3 is ahead of the roofing filters so
fairly wideband.  The LP-Pan is in fact a SDR which outputs IQ
baseband to the soundcard.  So with the K3 it is a single-conversion SDR.

LP-Pan passes nearly 400-KHz of RF bandwidth which wider than most
soundcards.  The Delta44 is 96-KHz wide soundcard.

Using a special eme sw all signals inside approx 90-KHz wide band are
simultaneously decoded and displayed showing virtually all the
calling stations.  You could also say it displays as a 90-KHz panadapter.

This sw has the ability to take two IQ streams so one can use the 1st
IF of the subRx in the K3 simultaneously.
With two antennas orthogonally polarized (Hpol and Vpol) the sw can
actually solve for the real polarization angle and maximize the signal.

So how does this relate to the K4?  I'm guessing the A/D processing
the panadpater display might also provide digital IQ output to a
computer which could perform similar wideband decoding of digital
signals.  I guess the question would be if the IQ stream is available
for export.

73, Ed - KL7UW



From: Wayne Burdick <[hidden email]>
To: Bill Frantz <[hidden email]>
Cc: Elecraft Reflector <[hidden email]>
Subject: Re: [Elecraft] 3 technical questions
Message-ID: <[hidden email]>
Content-Type: text/plain; charset=us-ascii

The K4 is always doing direct sampling for panadapter purposes.

If the HDR module is turned off or is not present, the K4 also does
direct sampling for demodulation.

If the HDR module is turned on, two additional ADCs are used, right
at the outputs of the crystal filters (oversimplifying a bit). The
original two ADCs are still being used for the panadapter.

Wayne
N6KR


73, Ed - KL7UW
   http://www.kl7uw.com
Dubus-NA Business mail:
   [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Edward H Russell]

Wayne,

 

The direct sampling ADC and the superhet ADC will probably have different
latencies, with the superhet ADC being shorter. Also the DSP filters will
have more latency than the crystal filters. Any information yet on typical
latency with direct sampling as compared to superhet?

 

Thanks,

73 Ed w2rf

 

 

 

 

 

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

Yes, the IQ stream will be available. On a K4D or K4HD it may be possible to have two IQ streams from different bands. We're still characterizing the available bandwidth.

73,
Wayne
N6KR

> On May 30, 2019, at 2:09 AM, Edward R Cole <[hidden email]> wrote:
>
> Wayne, etal:
>
> I am doing a similar thing with my K3 to view as panadapter.  I use the LP-Pan on 8.215 IF output of my K3 and that goes to the A/D in my Delta44 soundcard to provide a digital stream to sw in my computer.  The 1st IF of the K3 is ahead of the roofing filters so fairly wideband.  The LP-Pan is in fact a SDR which outputs IQ baseband to the soundcard.  So with the K3 it is a single-conversion SDR.
>
> LP-Pan passes nearly 400-KHz of RF bandwidth which wider than most soundcards.  The Delta44 is 96-KHz wide soundcard.
>
> Using a special eme sw all signals inside approx 90-KHz wide band are simultaneously decoded and displayed showing virtually all the calling stations.  You could also say it displays as a 90-KHz panadapter.
>
> This sw has the ability to take two IQ streams so one can use the 1st IF of the subRx in the K3 simultaneously.
> With two antennas orthogonally polarized (Hpol and Vpol) the sw can actually solve for the real polarization angle and maximize the signal.
>
> So how does this relate to the K4?  I'm guessing the A/D processing the panadpater display might also provide digital IQ output to a computer which could perform similar wideband decoding of digital signals.  I guess the question would be if the IQ stream is available for export.
>
> 73, Ed - KL7UW
>
>
>
> From: Wayne Burdick <[hidden email]>
> To: Bill Frantz <[hidden email]>
> Cc: Elecraft Reflector <[hidden email]>
> Subject: Re: [Elecraft] 3 technical questions
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset=us-ascii
>
> The K4 is always doing direct sampling for panadapter purposes.
>
> If the HDR module is turned off or is not present, the K4 also does direct sampling for demodulation.
>
> If the HDR module is turned on, two additional ADCs are used, right at the outputs of the crystal filters (oversimplifying a bit). The original two ADCs are still being used for the panadapter.
>
> Wayne
> N6KR
>
>
> 73, Ed - KL7UW
>  http://www.kl7uw.com
> Dubus-NA Business mail:
>  [hidden email]
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

Since we use the same DSP code for demodulation of both the direct-sampling and superhet receive paths, latencies should be very similar.

Wayne
N6KR



> On May 30, 2019, at 3:03 AM, E.H. Russell <[hidden email]> wrote:
>
> Wayne,
>  
> The direct sampling ADC and the superhet ADC will probably have different latencies, with the superhet ADC being shorter. Also the DSP filters will have more latency than the crystal filters. Any information yet on typical latency with direct sampling as compared to superhet?
>  
> Thanks,
> 73 Ed w2rf

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[DC-3]

All:

I just had the synthesizer on my K3 stop working.  It is the old style
synthesizer with a backplate (not necessary).  Does anyone who upgraded
to the new K3S synthesizers have  a a old style with or without
backplate that they would sell me?  I know for sure it is the synthesizer.


Thank you & 73's


Richard

K6VV


> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html