[Elecraft] Audio problem in WPX SSB

Mon Apr 14 01:21:31 EDT 2008

Joe,
  Thanks for the very thoughtful analysis.
  I'm very pleased to say that the problem seems to be greatly reduced with 
the latest firmware (1.81/1.63).  I can't tell if it is completely 
eliminated, but it's certainly much reduced.  Lyle, Wayne, what did you do? 
Whatever it is, it seems to have helped greatly.
  73, andy, ae6y
----- Original Message ----- 
From: "Joe Subich, W4TV" <lists at microham-usa.com>
To: "'Andy Faber'" <andrew.faber at gte.net>; "'wayne burdick'" 
<n6kr at elecraft.com>
Cc: <elecraft at mailman.qth.net>
Sent: Sunday, April 13, 2008 6:56 PM
Subject: RE: [Elecraft] Audio problem in WPX SSB


>
> Andy,
>
>>   Here is what happens:  For phone contesting, I my software
>> CQPWIN, which plays wav files for DVK usage (as do other
>> programs).  I run the output in such a way that when the dvk
>> is playing, the mic is shut off, but the audio from the computer
>> goes into the mic connector of the radio.  As soon as it stops
>> playing, it disconnects the computer and reconnects the
>> microphone.  I run it from the computer into a W2IHY iBox for
>> isolation, and the output to the radio.  The Mic is a Heil
>> pro-set.
>
>> >   With the K3, the messages play fine, but after one finishes,
>> if I use the mic, the radio starts out at lower power and takes
>> 2-4 seconds to build back up to high power.  This is the same
>> whether using the front or rear input and whether using high or
>> low gain.
>
> After working in this quite a bit over the weekend the problem
> is related to the K3 "microphone AGC."  It appears that levels
> from the microphone and DVK must be extremely well balanced -
> much more closely balanced than with any other radio or the
> stronger input will cause the microphone AGC to reduce the
> mic gain and the weaker input will be much weaker.  The 2-4
> second delay seems to be related to an extremely long ACG time
> constant (probably intended to prevent the ACG from acting like
> a clipper).
>
> The problem is made worse - at least in my case - because the
> ALC indicator provides only 5 very small, difficult to see,
> bars for adjustment and the "steps" are rather course (6 dB
> per division).  This means that the maximum "spread" between
> the two inputs can be almost 12 dB (!) even when the display
> only shows a slight flicker between the two adjacent bars
> (e.g., the mic is barely triggering the 5th bar and the DVK
> is hitting the sixth bar "hard").
>
> After some thought there are several ways (a combination of
> all is best) to help the situation:
>
> 1) reduce the time constant (AGC "hang") to about .5 second
>   and "dump" any gain reduction when the PTT is released.
>
> 2) Use more of the "ALC" display to get better resolution.
>   There are 12 "bars" in that part of the display but only
>   six are used (with the sixth indicating overdrive).  At
>   least expand the "good" range to seven (or eight) bars
>   to coincide with the right edge of the ALC bracket.
>
> 3) Make the ALC display response logarithmic (like a VU
>   meter).  Instead of 5 x 6 dB or 30 dB total ... the same
>   total range could be displayed with the first bar = -20 dB,
>   the second = -10 dB, the third = -7 dB, the fourth = -5 dB,
>   the fifth = -3 dB, the sixth -1 dB, the seventh = "0 dB",
>   the eighth +1 dB, the ninth +3 dB and the 10th +6 dB.
>   Setting/matching levels would be MUCH easier.  It's no
>   accident that professional recording and broadcasting have
>   used similar scales for decades.
>
> 4) Give the ALC meter a dual mode display - showing both short
>   duration "peak" and average (the current ALC) would help.
>   Alternatively, any "peak hold" dot above the ALC region
>   could show gain reduction by the mic AGC (and warn about
>   overdriving the input).
>
> 5) Provide a "set-up mode" that disables mic AGC - this
>   will allow balancing sources (either FP/RP/Line or
>   multiple inputs on an external mixer) without chasing
>   the effects of the mic AGC.
>
> I understand there has been some work on the mic AGC in the
> firmware currently in test and that may help.  However, this
> area needs to be more "user friendly."  As it is, I can see
> significant issues when using the K3 in a multi-operator
> setting where the relative levels need to be reset completely
> for every operator change!
>
> 73,
>
>   ... Joe, W4TV
>
>
>
>
>> -----Original Message-----
>> From: elecraft-bounces at mailman.qth.net
>> [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of Andy Faber
>> Sent: Monday, April 07, 2008 12:08 AM
>> To: nextk at elecraft.com; wayne burdick
>> Cc: elecraft at mailman.qth.net
>> Subject: [Elecraft] Audio problem in WPX SSB
>>
>>
>> Wayne,
>>   A peculiar audio situation arose using my K3 in Aruba in
>> WPX SSB.  It
>> works the same way at home, and isn't a problem I've ever had
>> with any of
>> many other radios that I've used in the same contest configuration.
>>
>>   Here is what happens:  For phone contesting, I my software
>> CQPWIN, which
>> plays wav files for DVK usage (as do other programs).  I run
>> the output in
>> such a way that when the dvk is playing, the mic is shut off,
>> but the audio
>> from the computer goes into the mic connector of the radio.
>> As soon as it
>> stops playing, it disconnects the computer and reconnects the
>> microphone.
>> I run it from the computer into a W2IHY iBox for isolation,
>> and the output
>> to the radio.  The Mic is a Heil pro-set.
>>
>>   With the K3, the messages play fine, but after one
>> finishes, if I use the
>> mic, the radio starts out at lower power and takes 2-4
>> seconds to build back
>> up to high power.  This is the same whether using the front
>> or rear input
>> and whether using high or low gain. Also, the W2IHY box has
>> to be turned up
>> to a level that normally would overdrive other radios.
>>
>>   I tried not using the W2IHY box (ie, running the audio
>> straight from
>> either my laptop or my desktop to the radio) with the same
>> result.  It also
>> doesn't matter if the radio is using PTT or VOX, or whether
>> it just remains
>> in transmit.
>>
>>   These seems like a some kind of odd impedance issue, but,
>> as I said, it's
>> one that has never shown up with Icom-Yaesu or Kenwood radios.
>>
>>   Any ideas?  I'm using 1.78/1.58 firmware, and the radio was
>> just back to
>> you for Rene to do hardware mods (serial number 76).  In all
>> other respects,
>> I loved it in the contest.
>>
>>   73, Andy, AE6Y
>>
>>
>> _______________________________________________
>> Elecraft mailing list
>> Post to: Elecraft at mailman.qth.net
>> You must be a subscriber to post to the list.
>> Subscriber Info (Addr. Change, sub, unsub etc.):
>>  http://mailman.qth.net/mailman/listinfo/elecraft
>>
>> Help: http://mailman.qth.net/subscribers.htm
>> Elecraft web page: http://www.elecraft.com
>
>