[Elecraft] MMTTY FSK

[Kok Chen]

On Jun 12, 2010, at 6/12    5:09 AM, Richard Ferch wrote:

> Brett Howard wrote:
>> I do see FSK as a method that is slightly less prone
>> to TX distortions.
>
> I have done some testing with a second
> receiver, and my K3's RTTY signal seems to be cleaner in AFSK than FSK
> (lower odd-order sidebands).

That is indeed correct.  AFSK can potentially produce cleaner signals  
than FSK, as long the transmit IMD is good.  Take a look at the last  
two figures on this page:

http://homepage.mac.com/chen/Technical/K3/Digital/digital.html

cocoaModem applies a filter to reduce the keying sidebands.  Except  
for having two tones, the result is similar to what Alex VE3NEA had  
described in his CW keying experiments article in QEX a couple of  
years ago.

The difference between FSK and AFSK can be even greater than what I'd  
measured if the K3 transmitter were cleaner.  Waveshaping the keying  
signal created albeit small, temporal overlap between the two RTTY  
tones, whereas an unwaveshaped FSK signal has no overlap and can be  
used even with class C amplifiers.

You can indeed see the low level IMD spikes from the in the last  
screen capture.

73
Chen, W7AY