K3 Audio Response - Version 3.33 Firmware

[P.B. Christensen]

> Yes but unless the response remains that way when an 8 ohm speaker is
> there its not a very useful measurement.

The plots were meant to observe the radical changes in audio response across
various firmware.  These aren't subtle changes as one can see from the
unloaded plots.

With 330uF caps at C9/C13 and slight EQ boost (+ 3dB) on bands 1-3, I am
confident that the response will appear as shown down to ~ 35 Hz with an
8-ohm load at the headphone output.  The speaker output response plot should
be comparable. I will be glad to make this measurement if there's any doubt.

http://216.229.20.37/images/K3-V333-EQ.jpg

Paul, W9AC
 

______________________________________________________________
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

[Ignacy]

I added computer speakers to my K3 (original C9&C13). It is like having a new radio. The audio is crisp, loud and much more undesrtandable. With 3.33., I am hearing all that basses generated by audio gurus. As an extra bonus the headphones plugged in the speakers have stronger sound as well. As a negative, there is some RF pickup.
Ignacy

[Jim Brown-10]

On Mon, 21 Sep 2009 10:32:27 -0700 (PDT), Ignacy wrote:

>I am hearing all that basses generated by audio gurus.

The word "guru" implies authoritative knowledge. Those who try to turn their
ham station into a hi-fi rig are simply misguided. In a speech to a ham
convention, retired FCC enforcement chief Riley Hollingsworth stated that
such things were out of place on the ham bands and that they should simply
get a job in a broadcast station if they wanted to do that kind of thing.

Most hams who really DO know audio limit their audio bandwidth so that their
transmitter runs more efficiently and cuts through noise and QRM better. For
the same reason, they also use good dynamics processing of the same sort
that is widely used by broadcasters. Those who transmit all that bass are
wasting at least half of their transmit power!  Those that transmit extended
HF response are selfishly using more bandwidth and generating more QRM to
their fellow hams.  

>As a negative, there is some RF pickup.

Speakers with built-in amplifiers are notorious for RFI. I often tour the
aisles at audio trade shows with a THF6A VHF/UHF talkie. In ten years, I've
seen only one amplified loudspeaker that didn't pick up RF. It cost about
$1,500 and was made in Europe.

73,

Jim Brown K9YC


______________________________________________________________
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

[Grant Youngman]

>
> The word "guru" implies authoritative knowledge. Those who try to  
> turn their
> ham station into a hi-fi rig are simply misguided.

Perhaps "different set of interests" would be more appropriate.  I do  
not consider myself misguided in the least, thank you very much.

> In a speech to a ham
> convention, retired FCC enforcement chief Riley Hollingsworth stated  
> that
> such things were out of place on the ham bands and that they should  
> simply
> get a job in a broadcast station if they wanted to do that kind of  
> thing.

And that means what?

>
> Most hams who really DO know audio limit their audio bandwidth so  
> that their
> transmitter runs more efficiently and cuts through noise and QRM  
> better.

Indeed.  It may cut through better, but if you can't understand what's  
being said, what good does "cutting through" do.


> Those that transmit extended
> HF response are selfishly using more bandwidth and generating more  
> QRM to
> their fellow hams.

Well, yes -- in the middle of a crowded band or contest or whatever.  
No one sense there should be a lack of common sense about what  
constitutes good operating practice.  But "good operating practice" is  
very dependent on time and place.  Unless of course it's the guys over  
driving their amps with their politely limited bandwidth screaming CQ  
DX so they can be heard all over the band at once :-)

Grant/NQ5T
______________________________________________________________
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

[AC7AC]

CONTENTS DELETED

The author has deleted this message.

[Jim Brown-10]

On Mon, 21 Sep 2009 14:03:02 -0500, Grant Youngman wrote:

>Indeed.  It may cut through better, but if you can't understand what's  
>being said, what good does "cutting through" do.

By "cutting through" I am specifically talking about speech
intelligibility, and thanks to my background in pro audio and sound
reinforcement, it is a topic I know quite a bit about. Sound systems
that must provide speech intelligibility in difficult environments are
carefully designed to limit the low frequency response. Three reasons.
First, the lower octave bands make very little contribution to speech
intelligibility. Second, the lower octave bands are most subject to the
effects of reverberation. Third, the lower octave bands burn power that
could be better used on the spectrum that does provide intelligibility.
The only part of the above that doesn't apply directly to radio
communications is reverberation.

Note that no one is recommending excessive modulation or processing,
which certainly does degrade intelligibility. The K3 can very easily be
set up to limit the audio bandwidth, shape it to provide pre-emphasis
for the bandwidth lost due to IF filters, and provide dynamics
processing (compression). And it sounds VERY good when done properly.

I'm coming at this from the perspective of an audio professional and a
radio professional. I'm a Fellow of the Audio Engineering Society,
passed my First Phone in 1959 and began working in broadcasting soon
thereafter. For the last 20+ years, I've made my living designing sound
systems for public spaces.

73,

Jim Brown K9YC


______________________________________________________________
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

[Joe Subich, W4TV-4]

> Personally, I find that many modern SSB rigs do limit the
> high frequencies too much. For me, there's a big improvement
> in intelligibility between rolling off the highs quickly at
> 2.5 KHz and allowing a full 2.7 to 3.0 kHz through. 3.0 kHz
> was the "standard" communications upper frequency roll-off
> for many years going back through the AM days, and I find
> it's still preferable to me. A major difference today is that
> back "then" we simply rolled off the upper frequencies with
> the very simplest audio filtering - often just using bypass
> capacitors in the audio stages that tended to attenuate
> highs. The result was substantial audio energy being
> transmitted well above 3 kHz. Nowadays we have the filters to
> limit high frequencies much better.

A clean "square sided" audio response that passes 150-200 Hz
to 2800-3000 Hz (2.8 KHz bandwidth) with a "notch" between
750 and 1100 Hz and a response that rises at 3 - 6 dB per
octave between 1000 and 3000 Hz is a thing of joy to hear.

Such a response is very efficient use of bandwidth, is
easy to understand but not "harsh" and almost "natural"
sounding.  Those who boost bass below 150-200 Hz do nothing
but make their audio "muddy" - particularly if the bass boost
is followed by any compression/clipping - and difficult to
tune.

I'd very much like to see the Elecraft equalizers modified
to match 10 of the ISO 2/3 Octave standard (63, 100, 160,
250, 400, 630, 1000, 1600, 2500, 4000 Hz) frequencies.  By
having a "band" centered a 1000 Hz, it is possible to cut the
interformat band (significantly reducing background noise
in an area lacking any voice energy), to cut audio below
even the most low pitched male voices (63 Hz), provide a
smooth rising characteristic (1/1.6/2.5K) and sharply limit
the high frequency (4K) components that cause adjacent
channel interference without contributing to communication
efficiency.  In addition, control over the two groups
(100/160/250/400 and 1600/2500) provide the ability to
"balance" the high/low formats compensating for variations
between strong/weak voices.

73,

   ... Joe, W4TV
 

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Ron
> D'Eau Claire
> Sent: Monday, September 21, 2009 3:36 PM
> To: [hidden email]
> Subject: Re: [Elecraft] [K3] K3 Audio Response - Version 3.33 Firmware
>
>
> Jim, K9YC, wrote:
>
> Speakers with built-in amplifiers are notorious for RFI. I
> often tour the
> aisles at audio trade shows with a THF6A VHF/UHF talkie. In
> ten years, I've
> seen only one amplified loudspeaker that didn't pick up RF.
> It cost about
> $1,500 and was made in Europe.
>
> -----------------------
> Ha! Bet the vendors loved your "test" ;-)
>
> I've tried a couple of different "computer speakers" as well
> as conventional "bookshelf" audio system speakers on my K3
> and have found no need for using an auxiliary amplifier. The
> K3 has plenty of audio power to drive external speakers in
> almost any Hamshack.
>
> I'm currently using a pair of LabTec "computer" speakers that
> bypass the internal amplifier by simply disconnecting the
> wall wart that supplies the amplifier. Ignacy, you might try
> that and see if they work  without powering the internal amp.
>
> As Jim observes, the "ancient ones" (and most hams today)
> don't limit their audio response out of ignorance or because
> they couldn't. They knew that bass takes lots of power and
> generally interferes with intelligibility, especially in us
> guys. Not only do low frequencies hog power as Jim noted,
> they are hardly modulated when we produce speech. Our mouths
> and lips mostly modulate the higher harmonics, typically
> above 300 Hz. The fundamental tone and low-order
> power-hogging harmonics produced by our vocal cords are just
> a "drone" with little variation other than starting and stopping.
>
> Personally, I find that many modern SSB rigs do limit the
> high frequencies too much. For me, there's a big improvement
> in intelligibility between rolling off the highs quickly at
> 2.5 KHz and allowing a full 2.7 to 3.0 kHz through. 3.0 kHz
> was the "standard" communications upper frequency roll-off
> for many years going back through the AM days, and I find
> it's still preferable to me. A major difference today is that
> back "then" we simply rolled off the upper frequencies with
> the very simplest audio filtering - often just using bypass
> capacitors in the audio stages that tended to attenuate
> highs. The result was substantial audio energy being
> transmitted well above 3 kHz. Nowadays we have the filters to
> limit high frequencies much better.
>
> I understand that there's good evidence that, when digging
> for a signal in the noise, a lower upper-frequency limit can
> provide better communications (provided the lows below 300 Hz
> or so are also attenuated). It's a matter of concentrating
> the finite amount of RF into the most effective part of the
> audio spectrum. I find such "pinched" audio tiring to listen
> to for any amount of time, and losing the higher frequencies
> makes many phonemes harder to decode, perhaps because of the
> US Army's special gift to me (tinnitus from firing too many
> rounds from my M1 rifle parked next to my ear - our steel
> helmets don't come with ear protectors).
>
> Perhaps before long we'll start to see computer generated
> speech that is optimized for minimum bandwidth and maximum
> intelligibility rather than continue to use our clumsy,
> inefficient and highly variable biological speech mechanism
> called lungs, throat and mouth ;-)
>
> After all, we've largely dispensed with that personality in
> CW since most Hams have dumped their mechanical keys for
> keyers that compensate for most variations in fingers
> movements or even to a keyboard that eliminates any chance of
> human variability or inefficiency while "pounding brass".
>
> Ron AC7AC
>
>
>
>
>
>
> ______________________________________________________________
> 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

[Eric Swartz - WA6HHQ]

Guys - We've beat this topic to death. Let's end the thread, or take it
off list, for the moment.

73,
Eric   WA6HHQ
Elecraft List Moderator
---

Jim Brown wrote:

> On Mon, 21 Sep 2009 14:03:02 -0500, Grant Youngman wrote:
>
>  
>> Indeed.  It may cut through better, but if you can't understand what's  
>> being said, what good does "cutting through" do.
>>    
>
> By "cutting through" I am specifically talking about speech
> intelligibility,
>  

______________________________________________________________
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

[Grant Youngman]

On Sep 21, 2009, at 4:14 PM, Joe Subich, W4TV wrote:

>
> A clean "square sided" audio response that passes 150-200 Hz
> to 2800-3000 Hz (2.8 KHz bandwidth) with a "notch" between
> 750 and 1100 Hz and a response that rises at 3 - 6 dB per
> octave between 1000 and 3000 Hz is a thing of joy to hear.
>
> Such a response is very efficient use of bandwidth, is
> easy to understand but not "harsh" and almost "natural"
> sounding.  Those who boost bass below 150-200 Hz do nothing
> but make their audio "muddy" - particularly if the bass boost
> is followed by any compression/clipping - and difficult to
> tune.

It's "natural" if all of your friends are parrots.

This is ridiculous.  No one here is advocating 20Hz-20Khz as a regular  
option on the ham bands.  (Well, you would if you really wanted to  
sound like YOU instead of a parrot, but that's another issue).  There  
is nothing "natural" sounding about what you propose.  Unless you have  
really bad hearing, or or just so used to thinking that nothing sounds  
better than a KWM-2 that anything else doesn't work.

All of this "efficiency" stuff is smokescreen.  You sound like a human  
being or you don't.  You can understand (out of context) what the guy  
on the other end is saying or not.

No one here is advocating using 20Hz to 20Khz transmit bandwidth in  
the context of "good amateur practice. So why is there so much hard  
core insistence than we should have bad audio ALL of the time.

Grant/NQ5T

______________________________________________________________
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

[AD4C2009]

"I'm coming at this from the perspective of an audio professional and a
radio professional. I'm a Fellow of the Audio Engineering Society,
passed my First Phone in 1959 and began working in broadcasting soon
thereafter. For the last 20+ years, I've made my living designing sound
systems for public spaces"
 
 
WOW ! I am impressed,so you are the professional audio "GURU" in this group,we are honored to have you here.WOW!!
 
AD4C


"The difference between genius and stupidity is that genius has its limits". -- Albert Einstein

--- On Mon, 9/21/09, Jim Brown <[hidden email]> wrote:


From: Jim Brown <[hidden email]>
Subject: Re: [Elecraft] [K3] K3 Audio Response - Version 3.33 Firmware
To: "Elecraft List" <[hidden email]>
Date: Monday, September 21, 2009, 8:16 PM


On Mon, 21 Sep 2009 14:03:02 -0500, Grant Youngman wrote:

>Indeed.  It may cut through better, but if you can't understand what's 
>being said, what good does "cutting through" do.

By "cutting through" I am specifically talking about speech
intelligibility, and thanks to my background in pro audio and sound
reinforcement, it is a topic I know quite a bit about. Sound systems
that must provide speech intelligibility in difficult environments are
carefully designed to limit the low frequency response. Three reasons.
First, the lower octave bands make very little contribution to speech
intelligibility. Second, the lower octave bands are most subject to the
effects of reverberation. Third, the lower octave bands burn power that
could be better used on the spectrum that does provide intelligibility.
The only part of the above that doesn't apply directly to radio
communications is reverberation.

Note that no one is recommending excessive modulation or processing,
which certainly does degrade intelligibility. The K3 can very easily be
set up to limit the audio bandwidth, shape it to provide pre-emphasis
for the bandwidth lost due to IF filters, and provide dynamics
processing (compression). And it sounds VERY good when done properly.

I'm coming at this from the perspective of an audio professional and a
radio professional. I'm a Fellow of the Audio Engineering Society,
passed my First Phone in 1959 and began working in broadcasting soon
thereafter. For the last 20+ years, I've made my living designing sound
systems for public spaces.

73,

Jim Brown K9YC


______________________________________________________________
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

[Joe Subich, W4TV-4]

> It's "natural" if all of your friends are parrots.

It is far from a "parrot" and light years ahead of the muddy
base rumble from the overly boosted, compressed and clipped
50-150 Hz  that passes for "audiophile" product these days.

> This is ridiculous.  No one here is advocating 20Hz-20Khz as
> a regular option on the ham bands.

Even 20 Hz to 5 KHz is absurd on the amateur bands.  More
than 90% of all transceivers have IF filters that on their
best days will pass 3 KHz ... most are less than that.  To
transmit a signal wider than 2.8 KHz (200 to 3200 or 150 to
3150 Hz) to 3.2 KHz (150 Hz to 3350 Hz) anywhere other than
on 10 Meters for a local QSO is an absolute exercise in ego
self-gratification and intentional QRM to other users of
the band.  

Like K9YC, I have more than  little professional familiarity
with "good" audio.  I spent my entire professional career in
in the broadcast and recording industries and started out
doing more than enough live recordings, talk shows, and live
event audio to know what constitutes "good" and "excellent"
audio for all purposes.  More importantly, I know the difference
between audio designed to go to a digital recording media and
audio that needs to work through a band limited channel.  
 
> So why is there so much hard core insistence than we should
> have bad audio ALL of the time.

Nobody is insisting that we should have band audio, ever - but
both Jim and I are saying that wide bandwidth and "good audio"
are NOT the same.  With the proper choice of passband,
equalization, and judicious use of clipping/compression, one
can have good audio in a reasonable bandwidth ... AM (5-7 KHz)
or FM (12-15 KHz) broadcast audio bandwidths are not necessary
or appropriate for amateur use at HF.
 

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Grant Youngman
> Sent: Monday, September 21, 2009 5:58 PM
> To: Elecraft List
> Subject: Re: [Elecraft] [K3] K3 Audio Response - Version 3.33 Firmware
>
>
>
> On Sep 21, 2009, at 4:14 PM, Joe Subich, W4TV wrote:
> >
> > A clean "square sided" audio response that passes 150-200 Hz to
> > 2800-3000 Hz (2.8 KHz bandwidth) with a "notch" between 750
> and 1100
> > Hz and a response that rises at 3 - 6 dB per octave between
> 1000 and
> > 3000 Hz is a thing of joy to hear.
> >
> > Such a response is very efficient use of bandwidth, is
> > easy to understand but not "harsh" and almost "natural" sounding.  
> > Those who boost bass below 150-200 Hz do nothing but make
> their audio
> > "muddy" - particularly if the bass boost is followed by any
> > compression/clipping - and difficult to tune.
>
>
> It's "natural" if all of your friends are parrots.
>
> This is ridiculous.  No one here is advocating 20Hz-20Khz as
> a regular  
> option on the ham bands.  (Well, you would if you really wanted to  
> sound like YOU instead of a parrot, but that's another
> issue).  There  
> is nothing "natural" sounding about what you propose.  Unless
> you have  
> really bad hearing, or or just so used to thinking that
> nothing sounds  
> better than a KWM-2 that anything else doesn't work.
>
> All of this "efficiency" stuff is smokescreen.  You sound
> like a human  
> being or you don't.  You can understand (out of context) what
> the guy  
> on the other end is saying or not.
>
> No one here is advocating using 20Hz to 20Khz transmit bandwidth in  
> the context of "good amateur practice. So why is there so much hard  
> core insistence than we should have bad audio ALL of the time.
>
> Grant/NQ5T
>
> ______________________________________________________________
> 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

[PA3CW]

I am a CW operator for over 30 years, first professionally and now only as hobby. My measuring instruments are my ears and the operation of the K3.  I never had such a good receiver but at first disliked the audio a bit.  It was written in many postings and many of you gave a thorough technical explanation even with different 'colors'  of the noise.

Still I trusted on my ears as a judge and i can say that the new release very much improves the way of listening to CW, SSB and AM signals. The K3 now can really sound the way I like a communications receiver to  do. It is just a bit 'warmer'  sound than before that makes listening more comfortable and less tiring.

I still have no clue about the best NR setting. It would make sense to have some background for which situation which setting is developed.  Also I sometimes wonder why there is no full decoupling possibility with DSP technology, meaning what I hear in my headphone is only a clean sinus of 600 Hz, generated by the DSP based on its input .  Is this strange thinking for CW, or technically impossible? (No immediate feature request ;).

Anyway, thanks for all the good work and 3.33 is an improvement again.  Compliments for all the development work being done. I am still proud to operate a K3.


Dick PA3CW