[Elecraft] fldigi KX3 question

[Tim Henrion]

> On Mar 28, 2016, at 9:56 PM, Don Wilhelm <w3fpr at embarqmail.com> wrote:
> 
> Tim,
> 
> The default bandwidths are to work well with the internal decoder where you tune to the desired signal with the VFO,
> If you are using a computer application, widen the bandwidth and let the software application do the decoding.
> 
> 73,
> Don W3FPR
> 

Hi Don,
While I understand that *some* of the data submodes are for decoding data on-board the radio, the “audio data modes” are specifically for getting audio data onto and off of the radio (i.e. no on-board decoding). When in AUDIO A data submode, the radio should be passing as much of the audio bandwidth through as the roofing filter being used would allow so that the computer software can see the entire passband.

Quoting the manual:
-----
Many audio-generated data transmissions can be heard on the bands, using PSK31, RTTY, Pactor, Olivia, MFSK, JT65 and other modes. A computer, sound card, and appropriate software are normally used. DATA A mode is provided for this purpose. Unlike SSB modes, DATA A disables compression and RX/TX EQ. Upper sideband is the default.
——

DATA A mode specifically disables compression and EQ to keep the radio from dorking with the TX/RX audio. It should be disabling audio bandwidth filtering as well, as there is no valid premise for the radio making assumptions about the audio being passed. Pretty much all digital decoding software allows multi-decode across the entire audio passband. The wideband Olivia modes can require up to 2Khz of audio bandwidth. Having the radio artificially impose audio bandwidth filtering narrower than the roofing filter bandwidth is pointless and counterproductive.

Tim Henrion
KC1EOQ

> On 3/28/2016 8:57 PM, Tim Henrion wrote:
>> 
>>> On Mar 28, 2016, at 7:46 PM, Don Wilhelm < <mailto:w3fpr at embarqmail.com>w3fpr at embarqmail.com <mailto:w3fpr at embarqmail.com>> wrote:
>>> 
>>> Note that the default bandwidth for DATA A and AFSK A data submodes is narrow on the KX3. 
>> 
>> Another one for the ‘Feature Request’ list. The “audio” data submode bandwidths are way too narrow out of the box. Does anyone have any idea why this is? I’m not sure I see much of a reason for any audio filtering at all other than the traditional 3K roofing filter. The downstream PC is going to do its own audio filtering at the DSP level before decoding the data anyway.
>> 
>> For the OP: If you like to tweak hardware, there are a number of performance improvements that can be made to the SignaLink USB hardware. Out of the box, its got a pretty high noise floor and is susceptible to noise on the USB +5 line, which directly feeds the onboard amp bias network without NO filtering. http://www.frenning.dk/OZ1PIF_HOMEPAGE/SignaLinkUSB-mods.html <http://www.frenning.dk/OZ1PIF_HOMEPAGE/SignaLinkUSB-mods.html> is a good place to start if you’re interested in better performance from your SignaLink.
>> 
>> 
>