[Elecraft] K3 - AFSK anamoly? No replies, try again

Thu Feb 18 17:52:29 EST 2016

Are you *absolutely sure* you are using the 400 Hz IF filter?  By ear,
it sounds as if you have a wider IF filter that is centered on the
carrier and allowing the image to reach the DSP.

Since your display shows FL4, is it possible that you have all your
filters reversed?

73,

    ... Joe, W4TV

On 2/18/2016 11:56 AM, Mike Murray wrote:
> After doing some more testing in last weekends RTTY contest, it appears
> that I made an erroneous statement as to what I was hearing and where.
> What I found is that as I tune down in frequency and find a signal that
> will decode, if I then tune down another 170 Hz I'll hear the inverted
> signal.  Still using dual passband DSP filter at 400 Hz and 400 Hz, 8 pole
> roofing filter as before.  I have a brief video on uTube showing the effect
> at:
>
> *http://tinyurl.com/hxykq9c <http://tinyurl.com/hxykq9c>*
>
> Anyone have additional thoughts as to what's causing this or where I should
> look next?
>
> Mike - W0AG
>
> On Thu, Jan 7, 2016 at 9:41 PM, Wes (N7WS) <wes at triconet.org> wrote:
>
>> Assuming that when the OP says "tuning" he is turning the VFO knob, then
>> here's my take.  All of the worry about roofing filters (gad how I hate
>> that term) is for nothing. The crystal filters in conjunction with
>> so-called hardware AGC are there to protect the delicate 2nd mixer from
>> overload in the presence of strong signals. (Conventional wisdom says
>> "protect the DSP" but the mixer is way weaker than the DSP. )
>>
>> Absent that condition, the demodulation BW is set in DSP and non-hardware
>> AGC is developed in that same bandwidth.  Depending upon the strength of
>> the desired signal and AGC threshold and slope, some amount of AGC will be
>> developed when the desired signal is within the DSP BW.  Hence there is
>> some amount of gain reduction in play.  When you tune to the opposite
>> sideband, that AGC gain reduction goes away, raising the amplitude of the
>> unwanted sideband with respect to the desired signal.  In other words, you
>> cannot measure opposite sideband rejection unless 1) the gain remains
>> constant or 2) some amount of known attenuation is included and accounted
>> for in the calculation.  This is pretty much what Lyle was talking about
>> earlier.
>>
>> One more thing.  If we understand that a superhetrodyne receiver is
>> nothing more than a narrow BPF that can be tuned over the spectrum then
>> that "roofing filter" that overlays the DSP filter tunes right along with
>> it.   The belief that somehow when you tune "below zero beat" the crystal
>> filter stays put while the DSP passband moves is silly.
>>
>> Wes  N7WS
>>
>>
>>
>>
>> On 1/7/2016 3:24 PM, Mike Murray wrote:
>>
>>> Joe, et.al.,
>>>
>>> I finally got back to the shack to check the setup.  It looks like I was
>>> actually using a 1.8 kHz SSB roofing filter in conjunction with the 500 Hz
>>> dual PB DSP filter.  I also checked the filter offsets which were all at
>>> 0.0 (all 8 pole) and I was using AFSK A mode.  Is the effect I'm seeing
>>> due
>>> to use of the 1.8 filter or do I have more research to do?
>>>
>>> Thanks for everyone's input and patience - seems like the learning curve
>>> is
>>> getting worse with age.
>>>
>>> Mike - W0AG
>>>
>>> On Wed, Jan 6, 2016 at 6:14 PM, Joe Subich, W4TV<lists at subich.com>
>>> wrote:
>>>
>>> Since I'm new to RTTY, I probably didn't describe my question in
>>>>
>>>>> correct terminology - I mainly only operate CW. I'll try again....
>>>>>
>>>>> And I'll ask my questions again ... what *ROOFING FILTER* and what DSP
>>>> bandwidth are you using?
>>>>
>>>> As Eric said in reply to your posting, this sounds like the offset is
>>>> incorrect on the roofing filter and allowing you to tune "past" zero
>>>> beat.  This is particularly true if you are using the SSB filter and
>>>> a wide DSP setting instead of the more typical CW filter (500 Hz)
>>>> centered on the mark/space tones (915/1085 Hz in your case).
>>>>
>>>> Note: RTTY is typically lower sideband with MARK being the lower audio
>>>> tone and space being the higher audio tone because of the "reversal"
>>>> that occurs in the audio to RF translation.  In the K3 AFSK A is LSB
>>>> but DATA A is USB ... if you are using them interchangeably that may
>>>> also be causing you some confusion.
>>>>
>>>> 73,
>>>>
>>>>     ... Joe, W4TV
>>>>
>>>>
>>>> On 1/6/2016 6:47 PM, Mike Murray wrote:
>>>>
>>>> Bob, Joe, Bill & Mike,
>>>>>
>>>>> Since I'm new to RTTY, I probably didn't describe my question in correct
>>>>> terminology - I mainly only operate CW.  I'll try again....
>>>>>
>>>>> I'm using DATA A/AFSK A, setup for low tones (915/170) and the RTTY
>>>>> dual-PB
>>>>> filter.  To illustrate what I'm hearing/seeing, imagine tuning down from
>>>>> 14.090 until you hear and can decode an RTTY signal.  Then continue
>>>>> tuning
>>>>> down 915 Hz (to what I probably erroneously referred to as zero beat)
>>>>> and
>>>>> then down another 915 Hz, I can now hear the same signal but inverted
>>>>> (slightly weaker, but definitely there).  It is not decodeable, but it
>>>>> is
>>>>> definitely there.  I assumed (yeah, I know) that the opposite sideband
>>>>> should be suppresed, but I think that's what I'm hearing.  Any advice
>>>>> will
>>>>> be appreciated.
>>>>>
>>>>> Mike - W0AG
>>>>>
>>>>> ______________________________________________________________
>>>>>
>>>>
>>>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:Elecraft at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to w0agmike at gmail.com
>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to lists at subich.com
>