[RTTY] DOS computer for improved RTTY reception?

[Dave Greig]

Well Stated.  Thank for the information. I most of us forget about the
internal noise of the sound card and the clipping on strong signals.
Dave Greig
N3BUO
Director of Retail
http://www.philoshopia.com
http://www.opticsbestbuy.com
http://www.redsoldier.com
http://www.chosentactical.com
Office: 817-453-0831
Toll Free: 866-759-9523


On Wed, Jan 19, 2011 at 9:57 AM, Kok Chen <chen at mac.com> wrote:

> On Jan 18, 2011, at 1/18    10:04 PM, Jim W7RY wrote:
>
> > Sounds like you need some help learning to run MMTTY!  It works
> > wonderfully!
>
> There are many who are not 100% successful when they move from a TU of
> a TNC to a sound card based software modem.
>
> The following are my opinions:
>
> The first thing and probably the most important thing is to make sure
> that (1) the sound card is getting adequate sound and (2) the sound
> card never clips.
>
> There is a happy medium between these two states.  If (1) is not
> satisfied, you will never be able to copy weak signals and if (2) is
> not satisfied, you will probably only be able to copy the strongest of
> signals and at the same time get no protection from any QRM rejecting
> filter in the software.
>
> If the sound card ever clips, even slightly (like pregnancies, there
> is no such thing as "only slightly clipping"), the software filters no
> longer work as filters.
>
> The easiest way to get a "first cut" is to find a program that has a
> spectrum display.  Watch the sound card noise floor with the rig
> turned off.  The spectrum display should be showing the noise floor of
> the sound card.  Now turn on the rig.  The noise floor should rise.
> If it does not rise, you don't have enough audio feeding the sound
> card and need to do something pronto about fixing the situation.
> Monitor strong stations and make sure it does not clip.  There usually
> is some "VU Meter" in the software that warns you when that happen.  I
> don't use Windows and cannot recommend a program, but software modems
> such as cocoaModem has a spectrum display built in, that has enough
> dynamic range to help you to make such adjustments.
>
> Getting a noise floor rise of 10 dB is probably a good stating point.
> A 3 dB rise means that the noise floor of your rig is the same as the
> noise floor of the sound card.  If the rise is more than 10 dB, you
> are just wasting the dynamic range of the sound card.
>
> If you can't get a noise floor to rise by 10 dB and still keep strong
> stations from clipping the sound card, the sound card does not have
> sufficient dynamic range for your rig, and you will need some kind of
> adjustment to "ride the gain." Either a digital attenuator in the
> sound card, or an analog pot.  Some sound cards don't have a digital
> attenuator, or they are only good for 12 dB, in which case you will
> need external pots or the use of the RF attenuator of the rig.
>
> Typically even cheap sound cards today have more than 85 dB of dynamic
> range, so if you take away the 10 dB of noise floor excess, you still
> end up with 75 dB of total dynamic range.  This should be sufficient
> for many rigs, especially if you adjust the AGC of your rig properly.
> Even if the rig advertises 100 dB of dynamic range, it does not mean
> that the line output has 100 dB worth of dynamic range.
>
> Some other things to observe are things like filtering.  If the
> software already has a matched filter, do not turn on the rig's twin-
> passband filters (or whatever they are called).  The software's
> matched filter should already have optimized for the signal and any
> further change of filter shape will ruin the copy.  For that matter,
> many narrow I.F. filters have terrible group delays when operated
> anywhere even near the skirts of their filters.  Our own Jeff AC0C has
> done some experiments on group delays of roofing filters.  This is why
> there is the common observation that when there is a weak signal, it
> is better to widen the filter for better copy -- you are both letting
> in more keying sidebands and also cleaning up the group delay
> characteristics when you do that.
>
> Treat the I.F. filter as the "roofing filter" of your sound card --
> use it only to keep loud QRM from clipping the sound card.  Let the
> software filters handle all the detail filtering.  If the software is
> written by someone who understands RTTY, its filtering will be
> superior to the ones in the rig -- as long as the sound card never
> clips.
>
> Take care of those items, and a good software demodulator will wipe
> the floor with the best hardware TU/TNC/Modem (yes, the ST-8000
> included).
>
> Be sure to follow good audio practices.  Check for ground loops (since
> those can really affect the noise floor).
>
> 73
> Chen, W7AY
>
> _______________________________________________
> RTTY mailing list
> RTTY at contesting.com
> http://lists.contesting.com/mailman/listinfo/rtty
>