K3 APF

classic Classic list List threaded Threaded
34 messages Options
12
Reply | Threaded
Open this post in threaded view
|

K3 APF

N5GE
Outstanding!

I have been spotting and working weak signal DX on 40m tonight.  I am able to
copy and work stations with S-3 signals on 40m here in TX with the APF On.

I worked ZL2MS and was able to give him a 549 report.  We were both running
100w.  His antenna a GP and mine a half sloper.  His report to me was 349, so I
believe my advantage was the APF.  It cuts right through the noise on 40m here
in Arlington, TX.

The filter in the rig is the 250Hz CW filter.

Good Job Y'All

Oh, and I am finding these weak signals with my new P3 ;o)

73,

Tom Childers
Radio Amateur N5GE
Licensed since 1976
QCWA Member 35102

______________________________________________________________
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
Amateur Radio Operator N5GE
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Bill W4ZV
For my own curiosity I did some measurements of the latest APF. They agree with
measurements of the first release with the exception that the filter peak is now
zero beat instead of +10 Hz:

Zero beat = 7040.021
Flat passband = 021-020 (both 0.0 to -0.1 dB)
-1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
-6 dB passband = 037-009 (-5.8 and -6.0 dB)

Flat = 2 Hz BW (at zero beat)
-1 dB = 8 Hz BW
-6 dB = 28 Hz BW

I didn't measure the -6 dB BW carefully on the first pass since I was more
interested in the -1 dB BW, but this looks similar and agrees with Lyle's 30 Hz
design goal. As mentioned previously, the -1 dB BW is important when trying to
detect signals below the noise floor since the human ear can detect this
difference in marginal conditions.

I also checked the gain (APF vs not) which I didn't do before...+9.1 dB which is
very close to Lyle's design goal of +9 dB.

Nice job Elecraft!

73, Bill

Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Joe Subich, W4TV-4

I made a similar set of measurements ... using the XG-2 and
the AFV/dBV capability ...

I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
the  1 dB points are   8 Hz wide
the  6 dB points are  31 Hz wide
the 20 dB points are 165 Hz wide
the 30 dB points are 345 Hz wide
gain is right at 9 dB.

The measurements were made on 40 Meters with the XG-2 set to
1 uV and the K3 attenuator activated yielding a -108 dBm test
signal.

I did not make similar measurements with the original alpha
test version but this version seems subjectively less tight
than the original ... there is certainly less ringing with
this one but signals did not seem to "pop" like they did on
the earlier version when I tried it last night on 160/80/40.

73,

    ... Joe, W4TV


On 11/13/2010 1:24 PM, Bill W4ZV wrote:

>
> For my own curiosity I did some measurements of the latest APF. They agree
> with
> measurements of the first release with the exception that the filter peak is
> now
> zero beat instead of +10 Hz:
>
> Zero beat = 7040.021
> Flat passband = 021-020 (both 0.0 to -0.1 dB)
> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>
> Flat = 2 Hz BW (at zero beat)
> -1 dB = 8 Hz BW
> -6 dB = 28 Hz BW
>
> I didn't measure the -6 dB BW carefully on the first pass since I was more
> interested in the -1 dB BW, but this looks similar and agrees with Lyle's 30
> Hz
> design goal. As mentioned previously, the -1 dB BW is important when trying
> to
> detect signals below the noise floor since the human ear can detect this
> difference in marginal conditions.
>
> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
> which is
> very close to Lyle's design goal of +9 dB.
>
> Nice job Elecraft!
>
> 73, Bill
>
>
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Bill VanAlstyne W5WVO
Joe wrote:

> I did not make similar measurements with the original alpha
> test version but this version seems subjectively less tight
> than the original ... there is certainly less ringing with
> this one but signals did not seem to "pop" like they did on
> the earlier version when I tried it last night on 160/80/40.

Joe, I agree with your subjective observation. No, the signals don't "pop"
as much -- but there is less ringing. This is a trade-off, of course, and
the precise balance of peak gain versus ringing is always going to be a
matter of personal preference.

I like this less-ringy version better, I think. Haven't had time to play
with it a lot yet, but I will. I'm sure we will see many other opinions. So
far, I think the current version is very smooth and adequately "peaky".

(Opinion subject to revision upon more extensive usage.)

Bill W5WVO


-----Original Message-----
From: Joe Subich, W4TV
Sent: Saturday, November 13, 2010 20:35
To: [hidden email]
Subject: Re: [Elecraft] K3 APF


I made a similar set of measurements ... using the XG-2 and
the AFV/dBV capability ...

I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
the  1 dB points are   8 Hz wide
the  6 dB points are  31 Hz wide
the 20 dB points are 165 Hz wide
the 30 dB points are 345 Hz wide
gain is right at 9 dB.

The measurements were made on 40 Meters with the XG-2 set to
1 uV and the K3 attenuator activated yielding a -108 dBm test
signal.

I did not make similar measurements with the original alpha
test version but this version seems subjectively less tight
than the original ... there is certainly less ringing with
this one but signals did not seem to "pop" like they did on
the earlier version when I tried it last night on 160/80/40.

73,

    ... Joe, W4TV


On 11/13/2010 1:24 PM, Bill W4ZV wrote:

>
> For my own curiosity I did some measurements of the latest APF. They agree
> with
> measurements of the first release with the exception that the filter peak
> is
> now
> zero beat instead of +10 Hz:
>
> Zero beat = 7040.021
> Flat passband = 021-020 (both 0.0 to -0.1 dB)
> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>
> Flat = 2 Hz BW (at zero beat)
> -1 dB = 8 Hz BW
> -6 dB = 28 Hz BW
>
> I didn't measure the -6 dB BW carefully on the first pass since I was more
> interested in the -1 dB BW, but this looks similar and agrees with Lyle's
> 30
> Hz
> design goal. As mentioned previously, the -1 dB BW is important when
> trying
> to
> detect signals below the noise floor since the human ear can detect this
> difference in marginal conditions.
>
> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
> which is
> very close to Lyle's design goal of +9 dB.
>
> Nice job Elecraft!
>
> 73, Bill
>
>
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

wayne burdick
Administrator
The APF algorithm is unchanged. Lyle is going to verify.

Wayne

----
http://www.elecraft.com

On Nov 13, 2010, at 1:19 PM, "Bill VanAlstyne W5WVO" <[hidden email]> wrote:

> Joe wrote:
>
>> I did not make similar measurements with the original alpha
>> test version but this version seems subjectively less tight
>> than the original ... there is certainly less ringing with
>> this one but signals did not seem to "pop" like they did on
>> the earlier version when I tried it last night on 160/80/40.
>
> Joe, I agree with your subjective observation. No, the signals don't "pop"
> as much -- but there is less ringing. This is a trade-off, of course, and
> the precise balance of peak gain versus ringing is always going to be a
> matter of personal preference.
>
> I like this less-ringy version better, I think. Haven't had time to play
> with it a lot yet, but I will. I'm sure we will see many other opinions. So
> far, I think the current version is very smooth and adequately "peaky".
>
> (Opinion subject to revision upon more extensive usage.)
>
> Bill W5WVO
>
>
> -----Original Message-----
> From: Joe Subich, W4TV
> Sent: Saturday, November 13, 2010 20:35
> To: [hidden email]
> Subject: Re: [Elecraft] K3 APF
>
>
> I made a similar set of measurements ... using the XG-2 and
> the AFV/dBV capability ...
>
> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
> the  1 dB points are   8 Hz wide
> the  6 dB points are  31 Hz wide
> the 20 dB points are 165 Hz wide
> the 30 dB points are 345 Hz wide
> gain is right at 9 dB.
>
> The measurements were made on 40 Meters with the XG-2 set to
> 1 uV and the K3 attenuator activated yielding a -108 dBm test
> signal.
>
> I did not make similar measurements with the original alpha
> test version but this version seems subjectively less tight
> than the original ... there is certainly less ringing with
> this one but signals did not seem to "pop" like they did on
> the earlier version when I tried it last night on 160/80/40.
>
> 73,
>
>    ... Joe, W4TV
>
>
> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>
>> For my own curiosity I did some measurements of the latest APF. They agree
>> with
>> measurements of the first release with the exception that the filter peak
>> is
>> now
>> zero beat instead of +10 Hz:
>>
>> Zero beat = 7040.021
>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>
>> Flat = 2 Hz BW (at zero beat)
>> -1 dB = 8 Hz BW
>> -6 dB = 28 Hz BW
>>
>> I didn't measure the -6 dB BW carefully on the first pass since I was more
>> interested in the -1 dB BW, but this looks similar and agrees with Lyle's
>> 30
>> Hz
>> design goal. As mentioned previously, the -1 dB BW is important when
>> trying
>> to
>> detect signals below the noise floor since the human ear can detect this
>> difference in marginal conditions.
>>
>> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
>> which is
>> very close to Lyle's design goal of +9 dB.
>>
>> Nice job Elecraft!
>>
>> 73, Bill
>>
>>
> ______________________________________________________________
> 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
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

KK7P

> The APF algorithm is unchanged. Lyle is going to verify.

They are identical.

73,

Lyle KK7P
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF (alpha 4.21)

K9ZTV
In reply to this post by Joe Subich, W4TV-4
I agree with Joe.

My first impression was a "quieter" APF (if that's the right word).  The
"peaking" is certainly there, and signals that are not copyable at 50
cycles are definitely copyable with the new APF activated (provided
there is a copyable signal at all).  But the peaked signal doesn't take
your breath away like it did on the earlier version.  Frankly, I prefer
this newer "tamed" (if you will) iteration.  It certainly has less
ringing.  You can't have both "blasting" and "no ringing" at the same
time.  There has to be a sweet spot and I believe version 4.21 nails it.

I think there was some early confusion about what exactly is "frozen"
when the APF is turned on.  It is the SHIFT that is locked, not the
WIDTH.  The WIDTH control is still fully functional in APF mode with the
bandwidth still being clearly indicated both numerically (0.05 and up)
and by the normal width-varying graphic.  When the APF is turned on, the
normal DUAL PB icon we have grown used to that incorporates
upward-pointing winglets is still displayed, but in APF mode the
winglets are positioned at the extreme ends of the line and are fixed.  
They do not change position when the WIDTH control is rotated (as they
do when in normal DUAL PB mode).  The middle-bars between the winglets
that show passband width DO change (as they always have) to visually
indicate the change of bandwidth.  Fixing the winglets and locating them
at the far ends of the graphic display is a great idea and gives further
feedback that you are definitely in APF mode and not DUAL-PB mode.  I
doubt anyone will confuse the two because (as in the earlier version)
you have to enable the APF mode in menu item CONFIG: DUAL PB.

The SSB problem in version 4.18 that required higher settings of CMP and
ALC has been fixed.  My MH2 microphone delivers the recommended 5-7 bars
of ALC and 2-3 bars of CMP at MIC=16 and CMP=13.  These are the nominal
settings I have always run when circumstances force me to pull the mic
out of the drawer.  For some reason the boys on the Missouri SSB Traffic
Net get real agitated when I check-in using a paddle.

Great job, Lyle, et al.

73,

Kent Trimble, K9ZTV
SN 21







On 11/13/2010 2:35 PM, Joe Subich, W4TV wrote:

> I made a similar set of measurements ... using the XG-2 and
> the AFV/dBV capability ...
>
> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
> the  1 dB points are   8 Hz wide
> the  6 dB points are  31 Hz wide
> the 20 dB points are 165 Hz wide
> the 30 dB points are 345 Hz wide
> gain is right at 9 dB.
>
> The measurements were made on 40 Meters with the XG-2 set to
> 1 uV and the K3 attenuator activated yielding a -108 dBm test
> signal.
>
> I did not make similar measurements with the original alpha
> test version but this version seems subjectively less tight
> than the original ... there is certainly less ringing with
> this one but signals did not seem to "pop" like they did on
> the earlier version when I tried it last night on 160/80/40.
>
> 73,
>
>      ... Joe, W4TV
>
>
> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>    
>> For my own curiosity I did some measurements of the latest APF. They agree
>> with
>> measurements of the first release with the exception that the filter peak is
>> now
>> zero beat instead of +10 Hz:
>>
>> Zero beat = 7040.021
>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>
>> Flat = 2 Hz BW (at zero beat)
>> -1 dB = 8 Hz BW
>> -6 dB = 28 Hz BW
>>
>> I didn't measure the -6 dB BW carefully on the first pass since I was more
>> interested in the -1 dB BW, but this looks similar and agrees with Lyle's 30
>> Hz
>> design goal. As mentioned previously, the -1 dB BW is important when trying
>> to
>> detect signals below the noise floor since the human ear can detect this
>> difference in marginal conditions.
>>
>> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
>> which is
>> very close to Lyle's design goal of +9 dB.
>>
>> Nice job Elecraft!
>>
>> 73, Bill
>>
>>
>>      
> ______________________________________________________________
> 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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF (alpha 4.21)

Jim Cox
I agree, I like the newer version as the ringing is much less now.
Jim K4JAF


----- Original Message -----
From: "K9ZTV" <[hidden email]>
Cc: <[hidden email]>
Sent: Saturday, November 13, 2010 4:29 PM
Subject: Re: [Elecraft] K3 APF (alpha 4.21)


>I agree with Joe.
>
> My first impression was a "quieter" APF (if that's the right word).  The
> "peaking" is certainly there, and signals that are not copyable at 50
> cycles are definitely copyable with the new APF activated (provided
> there is a copyable signal at all).  But the peaked signal doesn't take
> your breath away like it did on the earlier version.  Frankly, I prefer
> this newer "tamed" (if you will) iteration.  It certainly has less
> ringing.  You can't have both "blasting" and "no ringing" at the same
> time.  There has to be a sweet spot and I believe version 4.21 nails it.
>
> I think there was some early confusion about what exactly is "frozen"
> when the APF is turned on.  It is the SHIFT that is locked, not the
> WIDTH.  The WIDTH control is still fully functional in APF mode with the
> bandwidth still being clearly indicated both numerically (0.05 and up)
> and by the normal width-varying graphic.  When the APF is turned on, the
> normal DUAL PB icon we have grown used to that incorporates
> upward-pointing winglets is still displayed, but in APF mode the
> winglets are positioned at the extreme ends of the line and are fixed.
> They do not change position when the WIDTH control is rotated (as they
> do when in normal DUAL PB mode).  The middle-bars between the winglets
> that show passband width DO change (as they always have) to visually
> indicate the change of bandwidth.  Fixing the winglets and locating them
> at the far ends of the graphic display is a great idea and gives further
> feedback that you are definitely in APF mode and not DUAL-PB mode.  I
> doubt anyone will confuse the two because (as in the earlier version)
> you have to enable the APF mode in menu item CONFIG: DUAL PB.
>
> The SSB problem in version 4.18 that required higher settings of CMP and
> ALC has been fixed.  My MH2 microphone delivers the recommended 5-7 bars
> of ALC and 2-3 bars of CMP at MIC=16 and CMP=13.  These are the nominal
> settings I have always run when circumstances force me to pull the mic
> out of the drawer.  For some reason the boys on the Missouri SSB Traffic
> Net get real agitated when I check-in using a paddle.
>
> Great job, Lyle, et al.
>
> 73,
>
> Kent Trimble, K9ZTV
> SN 21
>
>
>
>
>
>
>
> On 11/13/2010 2:35 PM, Joe Subich, W4TV wrote:
>> I made a similar set of measurements ... using the XG-2 and
>> the AFV/dBV capability ...
>>
>> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
>> the  1 dB points are   8 Hz wide
>> the  6 dB points are  31 Hz wide
>> the 20 dB points are 165 Hz wide
>> the 30 dB points are 345 Hz wide
>> gain is right at 9 dB.
>>
>> The measurements were made on 40 Meters with the XG-2 set to
>> 1 uV and the K3 attenuator activated yielding a -108 dBm test
>> signal.
>>
>> I did not make similar measurements with the original alpha
>> test version but this version seems subjectively less tight
>> than the original ... there is certainly less ringing with
>> this one but signals did not seem to "pop" like they did on
>> the earlier version when I tried it last night on 160/80/40.
>>
>> 73,
>>
>>      ... Joe, W4TV
>>
>>
>> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>
>>> For my own curiosity I did some measurements of the latest APF. They
>>> agree
>>> with
>>> measurements of the first release with the exception that the filter
>>> peak is
>>> now
>>> zero beat instead of +10 Hz:
>>>
>>> Zero beat = 7040.021
>>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>>
>>> Flat = 2 Hz BW (at zero beat)
>>> -1 dB = 8 Hz BW
>>> -6 dB = 28 Hz BW
>>>
>>> I didn't measure the -6 dB BW carefully on the first pass since I was
>>> more
>>> interested in the -1 dB BW, but this looks similar and agrees with
>>> Lyle's 30
>>> Hz
>>> design goal. As mentioned previously, the -1 dB BW is important when
>>> trying
>>> to
>>> detect signals below the noise floor since the human ear can detect this
>>> difference in marginal conditions.
>>>
>>> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
>>> which is
>>> very close to Lyle's design goal of +9 dB.
>>>
>>> Nice job Elecraft!
>>>
>>> 73, Bill
>>>
>>>
>>>
>> ______________________________________________________________
>> 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 

______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Bill W4ZV
In reply to this post by KK7P
KK7P wrote
> The APF algorithm is unchanged. Lyle is going to verify.

They are identical.
Maybe not quite identical.  The new algorithm is zero beat and the initial one peaked at +10 Hz above zero beat.  It could be that guys were not tuning to the actual peak so the residual ringing could have been several dB stronger than the signal (relatively) making one think there was more ringing.  Now the zero beat signal may appear stronger than the residual ringing (on noise) since it is better centered.  BTW the filter response measures about -4 dB at a 10 Hz offset, which is quite noticeable on weak signals.

I personally don't notice any difference but I had been peaking each signal using VFO-FINE in 1 Hz steps (which is how I found the +10 Hz offset in the initial version).  

73,  Bill





Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

The Smiths
In reply to this post by Bill VanAlstyne W5WVO

I'll count this as another vote in favor of adding the Variable Q setting... At least a choice of 3 perhaps Wide, Med and Narrow.  Even if one perceives the APF as "less ringy" and comments on it being a good thing, that means that they are happy to know that the Q got widened out a little, and things seem to sound "better".
 

> From: [hidden email]
> To: [hidden email]; [hidden email]
> Date: Sat, 13 Nov 2010 21:19:46 +0000
> Subject: Re: [Elecraft] K3 APF
>
> Joe wrote:
>
> > I did not make similar measurements with the original alpha
> > test version but this version seems subjectively less tight
> > than the original ... there is certainly less ringing with
> > this one but signals did not seem to "pop" like they did on
> > the earlier version when I tried it last night on 160/80/40.
>
> Joe, I agree with your subjective observation. No, the signals don't "pop"
> as much -- but there is less ringing. This is a trade-off, of course, and
> the precise balance of peak gain versus ringing is always going to be a
> matter of personal preference.
>
> I like this less-ringy version better, I think. Haven't had time to play
> with it a lot yet, but I will. I'm sure we will see many other opinions. So
> far, I think the current version is very smooth and adequately "peaky".
>
> (Opinion subject to revision upon more extensive usage.)
>
> Bill W5WVO
>
>
> -----Original Message-----
> From: Joe Subich, W4TV
> Sent: Saturday, November 13, 2010 20:35
> To: [hidden email]
> Subject: Re: [Elecraft] K3 APF
>
>
> I made a similar set of measurements ... using the XG-2 and
> the AFV/dBV capability ...
>
> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
> the 1 dB points are 8 Hz wide
> the 6 dB points are 31 Hz wide
> the 20 dB points are 165 Hz wide
> the 30 dB points are 345 Hz wide
> gain is right at 9 dB.
>
> The measurements were made on 40 Meters with the XG-2 set to
> 1 uV and the K3 attenuator activated yielding a -108 dBm test
> signal.
>
> I did not make similar measurements with the original alpha
> test version but this version seems subjectively less tight
> than the original ... there is certainly less ringing with
> this one but signals did not seem to "pop" like they did on
> the earlier version when I tried it last night on 160/80/40.
>
> 73,
>
> ... Joe, W4TV
>
>
> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
> >
> > For my own curiosity I did some measurements of the latest APF. They agree
> > with
> > measurements of the first release with the exception that the filter peak
> > is
> > now
> > zero beat instead of +10 Hz:
> >
> > Zero beat = 7040.021
> > Flat passband = 021-020 (both 0.0 to -0.1 dB)
> > -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
> > -6 dB passband = 037-009 (-5.8 and -6.0 dB)
> >
> > Flat = 2 Hz BW (at zero beat)
> > -1 dB = 8 Hz BW
> > -6 dB = 28 Hz BW
> >
> > I didn't measure the -6 dB BW carefully on the first pass since I was more
> > interested in the -1 dB BW, but this looks similar and agrees with Lyle's
> > 30
> > Hz
> > design goal. As mentioned previously, the -1 dB BW is important when
> > trying
> > to
> > detect signals below the noise floor since the human ear can detect this
> > difference in marginal conditions.
> >
> > I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
> > which is
> > very close to Lyle's design goal of +9 dB.
> >
> > Nice job Elecraft!
> >
> > 73, Bill
> >
> >
> ______________________________________________________________
> 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
     
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

The Smiths
In reply to this post by Bill W4ZV

Are you sure that the problem isn't your REF CAL settings are off from where you think they are?
 

> Date: Sat, 13 Nov 2010 15:02:43 -0800
> From: [hidden email]
> To: [hidden email]
> Subject: Re: [Elecraft] K3 APF
>
>
>
> KK7P wrote:
> >
> >
> >> The APF algorithm is unchanged. Lyle is going to verify.
> >
> > They are identical.
> >
>
> Maybe not quite identical. The new algorithm is zero beat and the initial
> one peaked at +10 Hz above zero beat. It could be that guys were not tuning
> to the actual peak so the residual ringing could have been several dB
> stronger than the signal (relatively) making one think there was more
> ringing. Now the zero beat signal may appear stronger than the residual
> ringing (on noise) since it is better centered. BTW the filter response
> measures about -4 dB at a 10 Hz offset, which is quite noticeable on weak
> signals.
>
> I personally don't notice any difference but I had been peaking each signal
> using VFO-FINE in 1 Hz steps (which is how I found the +10 Hz offset in the
> initial version).
>
> 73, Bill
>
>
>
>
>
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/K3-APF-tp5735159p5736422.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> 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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Bill W4ZV
The Smiths wrote
Are you sure that the problem isn't your REF CAL settings are off from where you think they are?
Quite sure...Lyle verified the +10 Hz offset and corrected it in the latest version.

73,  Bill
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Joe Subich, W4TV-4
In reply to this post by The Smiths

>> I'll count this as another vote in favor of adding the Variable Q
>> setting...

Not on your life.  The lack of ringing on the newer version seems to
go along with a general decline in effectiveness.  If anything I'd
prefer to see higher Q.

Just for grins I reloaded 4.16 to make the same measurements using the
XG-2 as I made on 4.21.  Here is the comparison:

    BW     4.21   4.16
  -------------------------
    0 dB    1     2   Hz
   -1 dB    8   9   Hz
   -6 dB   31  31   Hz
  -10 dB   52  49   Hz
  -20 dB  165 162   Hz
  -30 dB  345 351   Hz
    Gain  9.0 9.1   dB

Unlike W4ZV, I found only a 3 Hz offset in 4.16 (the peak response
was 3 Hz above zero beat - or the indicated spot/shift frequency).
Even though the test results were generally the same within the
measurement tolerances, I still feel the 4.16 version was more effective
in on air listening.

These measurements were generated with an XG-2 set for 1 uV with the
K3 attenuator engaged for an effective signal level of -118 dBm.

73,

    ... Joe, W4TV

On 11/13/2010 8:07 PM, The Smiths wrote:

>
> I'll count this as another vote in favor of adding the Variable Q setting... At least a choice of 3 perhaps Wide, Med and Narrow.  Even if one perceives the APF as "less ringy" and comments on it being a good thing, that means that they are happy to know that the Q got widened out a little, and things seem to sound "better".
>
>> From: [hidden email]
>> To: [hidden email]; [hidden email]
>> Date: Sat, 13 Nov 2010 21:19:46 +0000
>> Subject: Re: [Elecraft] K3 APF
>>
>> Joe wrote:
>>
>>> I did not make similar measurements with the original alpha
>>> test version but this version seems subjectively less tight
>>> than the original ... there is certainly less ringing with
>>> this one but signals did not seem to "pop" like they did on
>>> the earlier version when I tried it last night on 160/80/40.
>>
>> Joe, I agree with your subjective observation. No, the signals don't "pop"
>> as much -- but there is less ringing. This is a trade-off, of course, and
>> the precise balance of peak gain versus ringing is always going to be a
>> matter of personal preference.
>>
>> I like this less-ringy version better, I think. Haven't had time to play
>> with it a lot yet, but I will. I'm sure we will see many other opinions. So
>> far, I think the current version is very smooth and adequately "peaky".
>>
>> (Opinion subject to revision upon more extensive usage.)
>>
>> Bill W5WVO
>>
>>
>> -----Original Message-----
>> From: Joe Subich, W4TV
>> Sent: Saturday, November 13, 2010 20:35
>> To: [hidden email]
>> Subject: Re: [Elecraft] K3 APF
>>
>>
>> I made a similar set of measurements ... using the XG-2 and
>> the AFV/dBV capability ...
>>
>> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
>> the 1 dB points are 8 Hz wide
>> the 6 dB points are 31 Hz wide
>> the 20 dB points are 165 Hz wide
>> the 30 dB points are 345 Hz wide
>> gain is right at 9 dB.
>>
>> The measurements were made on 40 Meters with the XG-2 set to
>> 1 uV and the K3 attenuator activated yielding a -108 dBm test
>> signal.
>>
>> I did not make similar measurements with the original alpha
>> test version but this version seems subjectively less tight
>> than the original ... there is certainly less ringing with
>> this one but signals did not seem to "pop" like they did on
>> the earlier version when I tried it last night on 160/80/40.
>>
>> 73,
>>
>> ... Joe, W4TV
>>
>>
>> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>>
>>> For my own curiosity I did some measurements of the latest APF. They agree
>>> with
>>> measurements of the first release with the exception that the filter peak
>>> is
>>> now
>>> zero beat instead of +10 Hz:
>>>
>>> Zero beat = 7040.021
>>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>>
>>> Flat = 2 Hz BW (at zero beat)
>>> -1 dB = 8 Hz BW
>>> -6 dB = 28 Hz BW
>>>
>>> I didn't measure the -6 dB BW carefully on the first pass since I was more
>>> interested in the -1 dB BW, but this looks similar and agrees with Lyle's
>>> 30
>>> Hz
>>> design goal. As mentioned previously, the -1 dB BW is important when
>>> trying
>>> to
>>> detect signals below the noise floor since the human ear can detect this
>>> difference in marginal conditions.
>>>
>>> I also checked the gain (APF vs not) which I didn't do before...+9.1 dB
>>> which is
>>> very close to Lyle's design goal of +9 dB.
>>>
>>> Nice job Elecraft!
>>>
>>> 73, Bill
>>>
>>>
>> ______________________________________________________________
>> 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
>    
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

n7ws
In reply to this post by wayne burdick
Probably why it sounds the same to me.

Wes

--- On Sat, 11/13/10, Wayne Burdick <[hidden email]> wrote:
The APF algorithm is unchanged. Lyle is going to verify.

Wayne

----



     
______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

wayne burdick
Administrator
In reply to this post by Joe Subich, W4TV-4
Joe,

If Lyle is using the same algorithm as before, and your results are as  
indicated below (virtually a tie), I don't understand how you could be  
hearing a "lack of ringing" in the new revision. I simply cannot hear  
any difference myself, and I'm extremely picky. We have nearly 100  
people testing, and only two or three have perceived a difference, so  
given the statistical evidence, I'd suggest that background noise  
conditions are the variable here, not the firmware. Lyle has also  
completely reviewed the DSP implementation -- no change.

But since the customers are always right, I'm still in a quandry  :)

tnx
Wayne

On Nov 13, 2010, at 6:27 PM, Joe Subich, W4TV wrote:

>
>>> I'll count this as another vote in favor of adding the Variable Q
>>> setting...
>
> Not on your life.  The lack of ringing on the newer version seems to
> go along with a general decline in effectiveness.  If anything I'd
> prefer to see higher Q.
>
> Just for grins I reloaded 4.16 to make the same measurements using the
> XG-2 as I made on 4.21.  Here is the comparison:
>
>    BW     4.21   4.16
>  -------------------------
>    0 dB    1     2   Hz
>   -1 dB    8   9   Hz
>   -6 dB   31  31   Hz
>  -10 dB   52  49   Hz
>  -20 dB  165 162   Hz
>  -30 dB  345 351   Hz
>    Gain  9.0 9.1   dB
>
> Unlike W4ZV, I found only a 3 Hz offset in 4.16 (the peak response
> was 3 Hz above zero beat - or the indicated spot/shift frequency).
> Even though the test results were generally the same within the
> measurement tolerances, I still feel the 4.16 version was more  
> effective
> in on air listening.
>
> These measurements were generated with an XG-2 set for 1 uV with the
> K3 attenuator engaged for an effective signal level of -118 dBm.
>
> 73,
>
>    ... Joe, W4TV
>
> On 11/13/2010 8:07 PM, The Smiths wrote:
>>
>> I'll count this as another vote in favor of adding the Variable Q  
>> setting... At least a choice of 3 perhaps Wide, Med and Narrow.  
>> Even if one perceives the APF as "less ringy" and comments on it  
>> being a good thing, that means that they are happy to know that the  
>> Q got widened out a little, and things seem to sound "better".
>>
>>> From: [hidden email]
>>> To: [hidden email]; [hidden email]
>>> Date: Sat, 13 Nov 2010 21:19:46 +0000
>>> Subject: Re: [Elecraft] K3 APF
>>>
>>> Joe wrote:
>>>
>>>> I did not make similar measurements with the original alpha
>>>> test version but this version seems subjectively less tight
>>>> than the original ... there is certainly less ringing with
>>>> this one but signals did not seem to "pop" like they did on
>>>> the earlier version when I tried it last night on 160/80/40.
>>>
>>> Joe, I agree with your subjective observation. No, the signals  
>>> don't "pop"
>>> as much -- but there is less ringing. This is a trade-off, of  
>>> course, and
>>> the precise balance of peak gain versus ringing is always going to  
>>> be a
>>> matter of personal preference.
>>>
>>> I like this less-ringy version better, I think. Haven't had time  
>>> to play
>>> with it a lot yet, but I will. I'm sure we will see many other  
>>> opinions. So
>>> far, I think the current version is very smooth and adequately  
>>> "peaky".
>>>
>>> (Opinion subject to revision upon more extensive usage.)
>>>
>>> Bill W5WVO
>>>
>>>
>>> -----Original Message-----
>>> From: Joe Subich, W4TV
>>> Sent: Saturday, November 13, 2010 20:35
>>> To: [hidden email]
>>> Subject: Re: [Elecraft] K3 APF
>>>
>>>
>>> I made a similar set of measurements ... using the XG-2 and
>>> the AFV/dBV capability ...
>>>
>>> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
>>> the 1 dB points are 8 Hz wide
>>> the 6 dB points are 31 Hz wide
>>> the 20 dB points are 165 Hz wide
>>> the 30 dB points are 345 Hz wide
>>> gain is right at 9 dB.
>>>
>>> The measurements were made on 40 Meters with the XG-2 set to
>>> 1 uV and the K3 attenuator activated yielding a -108 dBm test
>>> signal.
>>>
>>> I did not make similar measurements with the original alpha
>>> test version but this version seems subjectively less tight
>>> than the original ... there is certainly less ringing with
>>> this one but signals did not seem to "pop" like they did on
>>> the earlier version when I tried it last night on 160/80/40.
>>>
>>> 73,
>>>
>>> ... Joe, W4TV
>>>
>>>
>>> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>>>
>>>> For my own curiosity I did some measurements of the latest APF.  
>>>> They agree
>>>> with
>>>> measurements of the first release with the exception that the  
>>>> filter peak
>>>> is
>>>> now
>>>> zero beat instead of +10 Hz:
>>>>
>>>> Zero beat = 7040.021
>>>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>>>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>>>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>>>
>>>> Flat = 2 Hz BW (at zero beat)
>>>> -1 dB = 8 Hz BW
>>>> -6 dB = 28 Hz BW
>>>>
>>>> I didn't measure the -6 dB BW carefully on the first pass since I  
>>>> was more
>>>> interested in the -1 dB BW, but this looks similar and agrees  
>>>> with Lyle's
>>>> 30
>>>> Hz
>>>> design goal. As mentioned previously, the -1 dB BW is important  
>>>> when
>>>> trying
>>>> to
>>>> detect signals below the noise floor since the human ear can  
>>>> detect this
>>>> difference in marginal conditions.
>>>>
>>>> I also checked the gain (APF vs not) which I didn't do before...
>>>> +9.1 dB
>>>> which is
>>>> very close to Lyle's design goal of +9 dB.
>>>>
>>>> Nice job Elecraft!
>>>>
>>>> 73, Bill
>>>>
>>>>
>>> ______________________________________________________________
>>> 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
>>    
> ______________________________________________________________
> 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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Bill W4ZV
wayne burdick wrote
Joe,

If Lyle is using the same algorithm as before, and your results are as  
indicated below (virtually a tie), I don't understand how you could be  
hearing a "lack of ringing" in the new revision. I simply cannot hear  
any difference myself, and I'm extremely picky. We have nearly 100  
people testing, and only two or three have perceived a difference, so  
given the statistical evidence, I'd suggest that background noise  
conditions are the variable here, not the firmware. Lyle has also  
completely reviewed the DSP implementation -- no change.
I also suggested the background noise issue to Joe.  The bands have been relatively quieter since the latest FW was released, and white noise (i.e. galactic) does not cause the same amount of ringing as impulse noise (i.e. lightning).  Therefore the quiescent state of APF may sound less "ringy" than before.

I listened to a very weak DS5USH on 80m this morning and the APF was very effective at bringing him out of the noise.  I also did a few measurements using the latest 5 Hz SHIFT.  The worst case attenuation (when off zerobeat by 5 Hz) is <0.5 dB which is probably less than most ears will detect.  However only ONE setting of SHIFT is optimum so you need to listen carefully when adjusting.

73,  Bill
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Joe Subich, W4TV-4
In reply to this post by wayne burdick

> I'm still in a quandry :)

I am too <G>.  The measured results are rather clear but I
miss the "magic" results I saw with 4.16 ...

The "feeling" I get with 4.21 is that the Q has been reduced
compared to 4.16 but since the quantitative data shows that's
obviously not the case, I'll just need to continue working
with it to see if I can regain the magic.  Maybe it is just
conditions ... noise levels have been a lot lower the last
couple days but there haven't been as many weak signals either.

73,

    ... Joe, W4TV

On 11/13/2010 11:48 PM, Wayne Burdick wrote:

> Joe,
>
> If Lyle is using the same algorithm as before, and your results are as
> indicated below (virtually a tie), I don't understand how you could be
> hearing a "lack of ringing" in the new revision. I simply cannot hear
> any difference myself, and I'm extremely picky. We have nearly 100
> people testing, and only two or three have perceived a difference, so
> given the statistical evidence, I'd suggest that background noise
> conditions are the variable here, not the firmware. Lyle has also
> completely reviewed the DSP implementation -- no change.
>
> But since the customers are always right, I'm still in a quandry :)
>
> tnx
> Wayne
>
> On Nov 13, 2010, at 6:27 PM, Joe Subich, W4TV wrote:
>
>>
>>>> I'll count this as another vote in favor of adding the Variable Q
>>>> setting...
>>
>> Not on your life. The lack of ringing on the newer version seems to
>> go along with a general decline in effectiveness. If anything I'd
>> prefer to see higher Q.
>>
>> Just for grins I reloaded 4.16 to make the same measurements using the
>> XG-2 as I made on 4.21. Here is the comparison:
>>
>> BW 4.21 4.16
>> -------------------------
>> 0 dB 1 2 Hz
>> -1 dB 8 9 Hz
>> -6 dB 31 31 Hz
>> -10 dB 52 49 Hz
>> -20 dB 165 162 Hz
>> -30 dB 345 351 Hz
>> Gain 9.0 9.1 dB
>>
>> Unlike W4ZV, I found only a 3 Hz offset in 4.16 (the peak response
>> was 3 Hz above zero beat - or the indicated spot/shift frequency).
>> Even though the test results were generally the same within the
>> measurement tolerances, I still feel the 4.16 version was more effective
>> in on air listening.
>>
>> These measurements were generated with an XG-2 set for 1 uV with the
>> K3 attenuator engaged for an effective signal level of -118 dBm.
>>
>> 73,
>>
>> ... Joe, W4TV
>>
>> On 11/13/2010 8:07 PM, The Smiths wrote:
>>>
>>> I'll count this as another vote in favor of adding the Variable Q
>>> setting... At least a choice of 3 perhaps Wide, Med and Narrow. Even
>>> if one perceives the APF as "less ringy" and comments on it being a
>>> good thing, that means that they are happy to know that the Q got
>>> widened out a little, and things seem to sound "better".
>>>
>>>> From: [hidden email]
>>>> To: [hidden email]; [hidden email]
>>>> Date: Sat, 13 Nov 2010 21:19:46 +0000
>>>> Subject: Re: [Elecraft] K3 APF
>>>>
>>>> Joe wrote:
>>>>
>>>>> I did not make similar measurements with the original alpha
>>>>> test version but this version seems subjectively less tight
>>>>> than the original ... there is certainly less ringing with
>>>>> this one but signals did not seem to "pop" like they did on
>>>>> the earlier version when I tried it last night on 160/80/40.
>>>>
>>>> Joe, I agree with your subjective observation. No, the signals don't
>>>> "pop"
>>>> as much -- but there is less ringing. This is a trade-off, of
>>>> course, and
>>>> the precise balance of peak gain versus ringing is always going to be a
>>>> matter of personal preference.
>>>>
>>>> I like this less-ringy version better, I think. Haven't had time to
>>>> play
>>>> with it a lot yet, but I will. I'm sure we will see many other
>>>> opinions. So
>>>> far, I think the current version is very smooth and adequately "peaky".
>>>>
>>>> (Opinion subject to revision upon more extensive usage.)
>>>>
>>>> Bill W5WVO
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Joe Subich, W4TV
>>>> Sent: Saturday, November 13, 2010 20:35
>>>> To: [hidden email]
>>>> Subject: Re: [Elecraft] K3 APF
>>>>
>>>>
>>>> I made a similar set of measurements ... using the XG-2 and
>>>> the AFV/dBV capability ...
>>>>
>>>> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
>>>> the 1 dB points are 8 Hz wide
>>>> the 6 dB points are 31 Hz wide
>>>> the 20 dB points are 165 Hz wide
>>>> the 30 dB points are 345 Hz wide
>>>> gain is right at 9 dB.
>>>>
>>>> The measurements were made on 40 Meters with the XG-2 set to
>>>> 1 uV and the K3 attenuator activated yielding a -108 dBm test
>>>> signal.
>>>>
>>>> I did not make similar measurements with the original alpha
>>>> test version but this version seems subjectively less tight
>>>> than the original ... there is certainly less ringing with
>>>> this one but signals did not seem to "pop" like they did on
>>>> the earlier version when I tried it last night on 160/80/40.
>>>>
>>>> 73,
>>>>
>>>> ... Joe, W4TV
>>>>
>>>>
>>>> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>>>>
>>>>> For my own curiosity I did some measurements of the latest APF.
>>>>> They agree
>>>>> with
>>>>> measurements of the first release with the exception that the
>>>>> filter peak
>>>>> is
>>>>> now
>>>>> zero beat instead of +10 Hz:
>>>>>
>>>>> Zero beat = 7040.021
>>>>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>>>>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>>>>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>>>>
>>>>> Flat = 2 Hz BW (at zero beat)
>>>>> -1 dB = 8 Hz BW
>>>>> -6 dB = 28 Hz BW
>>>>>
>>>>> I didn't measure the -6 dB BW carefully on the first pass since I
>>>>> was more
>>>>> interested in the -1 dB BW, but this looks similar and agrees with
>>>>> Lyle's
>>>>> 30
>>>>> Hz
>>>>> design goal. As mentioned previously, the -1 dB BW is important when
>>>>> trying
>>>>> to
>>>>> detect signals below the noise floor since the human ear can detect
>>>>> this
>>>>> difference in marginal conditions.
>>>>>
>>>>> I also checked the gain (APF vs not) which I didn't do
>>>>> before...+9.1 dB
>>>>> which is
>>>>> very close to Lyle's design goal of +9 dB.
>>>>>
>>>>> Nice job Elecraft!
>>>>>
>>>>> 73, Bill
>>>>>
>>>>>
>>>> ______________________________________________________________
>>>> 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
>>>
>> ______________________________________________________________
>> 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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

Steve Ellington
The magic is only good for one shot Joe.
Same thing happened when I saw the Grand Canyon for the second time.

Steve

----- Original Message -----
From: "Joe Subich, W4TV" <[hidden email]>
To: "Wayne Burdick" <[hidden email]>
Cc: "Elecraft Reflector" <[hidden email]>
Sent: Sunday, November 14, 2010 9:02 AM
Subject: Re: [Elecraft] K3 APF


>
>> I'm still in a quandry :)
>
> I am too <G>.  The measured results are rather clear but I
> miss the "magic" results I saw with 4.16 ...
>
> The "feeling" I get with 4.21 is that the Q has been reduced
> compared to 4.16 but since the quantitative data shows that's
> obviously not the case, I'll just need to continue working
> with it to see if I can regain the magic.  Maybe it is just
> conditions ... noise levels have been a lot lower the last
> couple days but there haven't been as many weak signals either.
>
> 73,
>
>    ... Joe, W4TV
>
> On 11/13/2010 11:48 PM, Wayne Burdick wrote:
>> Joe,
>>
>> If Lyle is using the same algorithm as before, and your results are as
>> indicated below (virtually a tie), I don't understand how you could be
>> hearing a "lack of ringing" in the new revision. I simply cannot hear
>> any difference myself, and I'm extremely picky. We have nearly 100
>> people testing, and only two or three have perceived a difference, so
>> given the statistical evidence, I'd suggest that background noise
>> conditions are the variable here, not the firmware. Lyle has also
>> completely reviewed the DSP implementation -- no change.
>>
>> But since the customers are always right, I'm still in a quandry :)
>>
>> tnx
>> Wayne
>>
>> On Nov 13, 2010, at 6:27 PM, Joe Subich, W4TV wrote:
>>
>>>
>>>>> I'll count this as another vote in favor of adding the Variable Q
>>>>> setting...
>>>
>>> Not on your life. The lack of ringing on the newer version seems to
>>> go along with a general decline in effectiveness. If anything I'd
>>> prefer to see higher Q.
>>>
>>> Just for grins I reloaded 4.16 to make the same measurements using the
>>> XG-2 as I made on 4.21. Here is the comparison:
>>>
>>> BW 4.21 4.16
>>> -------------------------
>>> 0 dB 1 2 Hz
>>> -1 dB 8 9 Hz
>>> -6 dB 31 31 Hz
>>> -10 dB 52 49 Hz
>>> -20 dB 165 162 Hz
>>> -30 dB 345 351 Hz
>>> Gain 9.0 9.1 dB
>>>
>>> Unlike W4ZV, I found only a 3 Hz offset in 4.16 (the peak response
>>> was 3 Hz above zero beat - or the indicated spot/shift frequency).
>>> Even though the test results were generally the same within the
>>> measurement tolerances, I still feel the 4.16 version was more effective
>>> in on air listening.
>>>
>>> These measurements were generated with an XG-2 set for 1 uV with the
>>> K3 attenuator engaged for an effective signal level of -118 dBm.
>>>
>>> 73,
>>>
>>> ... Joe, W4TV
>>>
>>> On 11/13/2010 8:07 PM, The Smiths wrote:
>>>>
>>>> I'll count this as another vote in favor of adding the Variable Q
>>>> setting... At least a choice of 3 perhaps Wide, Med and Narrow. Even
>>>> if one perceives the APF as "less ringy" and comments on it being a
>>>> good thing, that means that they are happy to know that the Q got
>>>> widened out a little, and things seem to sound "better".
>>>>
>>>>> From: [hidden email]
>>>>> To: [hidden email]; [hidden email]
>>>>> Date: Sat, 13 Nov 2010 21:19:46 +0000
>>>>> Subject: Re: [Elecraft] K3 APF
>>>>>
>>>>> Joe wrote:
>>>>>
>>>>>> I did not make similar measurements with the original alpha
>>>>>> test version but this version seems subjectively less tight
>>>>>> than the original ... there is certainly less ringing with
>>>>>> this one but signals did not seem to "pop" like they did on
>>>>>> the earlier version when I tried it last night on 160/80/40.
>>>>>
>>>>> Joe, I agree with your subjective observation. No, the signals don't
>>>>> "pop"
>>>>> as much -- but there is less ringing. This is a trade-off, of
>>>>> course, and
>>>>> the precise balance of peak gain versus ringing is always going to be
>>>>> a
>>>>> matter of personal preference.
>>>>>
>>>>> I like this less-ringy version better, I think. Haven't had time to
>>>>> play
>>>>> with it a lot yet, but I will. I'm sure we will see many other
>>>>> opinions. So
>>>>> far, I think the current version is very smooth and adequately
>>>>> "peaky".
>>>>>
>>>>> (Opinion subject to revision upon more extensive usage.)
>>>>>
>>>>> Bill W5WVO
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: Joe Subich, W4TV
>>>>> Sent: Saturday, November 13, 2010 20:35
>>>>> To: [hidden email]
>>>>> Subject: Re: [Elecraft] K3 APF
>>>>>
>>>>>
>>>>> I made a similar set of measurements ... using the XG-2 and
>>>>> the AFV/dBV capability ...
>>>>>
>>>>> I measure the peak at 1 Hz wide (e.g. -.1dB +/- 1 Hz)
>>>>> the 1 dB points are 8 Hz wide
>>>>> the 6 dB points are 31 Hz wide
>>>>> the 20 dB points are 165 Hz wide
>>>>> the 30 dB points are 345 Hz wide
>>>>> gain is right at 9 dB.
>>>>>
>>>>> The measurements were made on 40 Meters with the XG-2 set to
>>>>> 1 uV and the K3 attenuator activated yielding a -108 dBm test
>>>>> signal.
>>>>>
>>>>> I did not make similar measurements with the original alpha
>>>>> test version but this version seems subjectively less tight
>>>>> than the original ... there is certainly less ringing with
>>>>> this one but signals did not seem to "pop" like they did on
>>>>> the earlier version when I tried it last night on 160/80/40.
>>>>>
>>>>> 73,
>>>>>
>>>>> ... Joe, W4TV
>>>>>
>>>>>
>>>>> On 11/13/2010 1:24 PM, Bill W4ZV wrote:
>>>>>>
>>>>>> For my own curiosity I did some measurements of the latest APF.
>>>>>> They agree
>>>>>> with
>>>>>> measurements of the first release with the exception that the
>>>>>> filter peak
>>>>>> is
>>>>>> now
>>>>>> zero beat instead of +10 Hz:
>>>>>>
>>>>>> Zero beat = 7040.021
>>>>>> Flat passband = 021-020 (both 0.0 to -0.1 dB)
>>>>>> -1 dB passband = 026-018 (-1.2 dB and -0.8 dB)
>>>>>> -6 dB passband = 037-009 (-5.8 and -6.0 dB)
>>>>>>
>>>>>> Flat = 2 Hz BW (at zero beat)
>>>>>> -1 dB = 8 Hz BW
>>>>>> -6 dB = 28 Hz BW
>>>>>>
>>>>>> I didn't measure the -6 dB BW carefully on the first pass since I
>>>>>> was more
>>>>>> interested in the -1 dB BW, but this looks similar and agrees with
>>>>>> Lyle's
>>>>>> 30
>>>>>> Hz
>>>>>> design goal. As mentioned previously, the -1 dB BW is important when
>>>>>> trying
>>>>>> to
>>>>>> detect signals below the noise floor since the human ear can detect
>>>>>> this
>>>>>> difference in marginal conditions.
>>>>>>
>>>>>> I also checked the gain (APF vs not) which I didn't do
>>>>>> before...+9.1 dB
>>>>>> which is
>>>>>> very close to Lyle's design goal of +9 dB.
>>>>>>
>>>>>> Nice job Elecraft!
>>>>>>
>>>>>> 73, Bill
>>>>>>
>>>>>>
>>>>> ______________________________________________________________
>>>>> 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
>>>>
>>> ______________________________________________________________
>>> 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 

______________________________________________________________
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
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

AC7AC
CONTENTS DELETED
The author has deleted this message.
Reply | Threaded
Open this post in threaded view
|

Re: K3 APF

4X4DK-AMI SHAMI
In reply to this post by N5GE
Wayne

 

I also think that the background noise characteristic and intensity are the
major factors that influence

 the performance of the APF, which seems to work very well.

 

Ami  

 

4X4DK

______________________________________________________________
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
12