Order Tray | Contact Us | Home | SIG Lists

[aprssig] FW: Voice Alert

Robert Bruninga bruninga at usna.edu
Mon Jul 21 13:14:22 UTC 2008


> I have tried the voice alert that Bob, WB4APR has 
> talked about.  100 Hz tone decode and encode.  
> I don't seem to be digipeated through the local
> digipeater.  Is it possible that the digipeater 
> is blocking my signal because of the tone?

Good question.  Can you take some time and do some very careful
tests?  It would be nice to find out what is going on.  But you
must remember one thing, and that is that a digi will not
digipeat the same packet more often that once every 30 seconds
or so.  So you have to be careful in your testing or you could
be confusing time-out issues with lack of digipeats.

One way to get more rapid testing is to pre-load two fixed
posits if using a D700.  THen you can alternate between them and
beat the 30 second timer.

If you are not getting digipeated. Stop, get a good clear area.
Confirm lack of digipeats with Volice Alert on.  Then turn it
off and try...  Switching POSITS on every try if needed.  Do
lots of repeated testing.  Then share your results.  Many-many
digipeaters out there were just slapped-up plug-n-play
installations with no care at all to audio levels, tone balance,
skew, roll off or pre/de-emphasis.  Some have even been reported
to not work with D700's because of tone inbalance.

Remember, 85% of mobiles in many areas are D7/D700's and they
are all factory balanced.  SO I would use one of them as your
benchmark.  Please see this web page on SIGNALS:
http://www.ew.usna.edu/~bruninga/sigs.html

The issue is pre-emphasis versus flat passband.  The D7 was made
flat.  The D700 allowed a little more pre/de-emphasis.  Everyone
running a TNC and speaker/mic connections are running with full
pre/de-emphasis.  And there is 6 dB difference!  And that is a
range difference of HALF all else being equal...

But the good news is that the difference is disastrous in one
direction, but tolerable in the other.  If I remember correctly,
it is OK to have the low tone higher than the high tone, but if
the high tone is higher than the low tone, even slightly, your
decoding goes south real fast.  These issues are compounded by
how TNC's are interfaced to particular radios though most
problems can be fixed with a capacitor and resistor either in a
lowpass or highpass configuration with values chosen to match
the exact impedance characteristics of your system.

I'd suggest using a speaker and listening to packets on 144.39
in the USA.  It is amazing and disgusting how different they all
sound!!!!

1) Many are overdeviated
2) Many are under deviated
3) Many have way too long TXD delays wasting bandwidth
4) Many are too emphasized, some too deemphasized
5) Many have tone inbalance, etc...

The combination of any of these plus all the same problems on
receipt can all compound to work against you.  And throw in a
100 HZ tone in the mix and it might be enough to put you over
the edge. 

But the problem is not the fact that you are running voice
alert, but that the digi is not properly adjusted probably...

If this thread continues, along the digi line, I will probably
change the SUBJECT because this is required reading by everyone
on APRS..

Thanks for raising it!
Bob, WB4APR


Also, be careful in judgement based on what -you-hear, because
what you hear is 100% affected by your receiver's audio
circuits.





More information about the aprssig mailing list