[Alexandria Radio Club Reflector] Info on GMSK modem boards for use with DStar

[Marshall DeBerry]

After tonight's net, several members expressed an interest in obtaining more
information on GMSK modem boards that can be utilized with DStar.  I am
including below a variety of links to various vendors, as well as two Yahoo
groups that may be of interest.

73

  Marshall

 

 

 

MoenComm GMSK board set:

 

http://www.moencomm.com/

 

Satoshi web site:

 

http://www.d-star.asia/index.html.en

 

W9ARP Hot Spot info:

 

http://www.w9arp.com/hotspot/

 

Jim Moen DStar info:

 

http://www.k6jm.com/dstar.htm

 

Two Yahoo groups of interest: GMSK-DV-node and GMSK-DV-modem

 

http://groups.yahoo.com/group/gmsk_dv_node/  and
http://groups.yahoo.com/group/gmsk_dv_modem/

 

 

 

Some info from GMSK-DV-Node Yahoo group on using narrow FM mode:

 

 

From: gmsk_dv_node at yahoogroups.com [mailto:gmsk_dv_node at yahoogroups.com] On
Behalf Of D Star
Sent: Monday, September 19, 2011 10:01 AM
To: gmsk_dv_node at yahoogroups.com
Subject: RE: [gmsk_dv_node] D-STAR Deviation

 

GMSK for D-STAR indeed uses about 1.2kHz of RF deviation.

 

The amount of AF audio drive to get this deviation depends on the radio, and
can

range from as little as 100mV to as much as 2.5V to get to maximum
deviation;

this largely depends on how we're tapping into the modulator circuit.

 

A Kenwood TM-D700 indeed is one of those radios that requires a lot of audio

drive.  I have to full-tilt the TX trimpot to get it to that level.

 

If a radio has "Narrow FM" (NFM, NAR) mode available, always use that, as it

will change the various circuits to output less wide signals, which usually
also

require less audio drive as a side effect.

 

Normal-mode FM _can_ indeed be driven into overload, meaning you will output

a (much!) too wide signal. This signal usually can not (or very hard at
best) be

decoded by "real" narrow-mode systems, such as the various Icom D-STAR

radios.  Symptom: plenty of RF signal, but very bad audio... the radio''s
filters

are chopping off most of the "too wide" signal, and the resulting signal is
too

bad to be decoded..

 

--fred

 

 

  _____  

From: gmsk_dv_node at yahoogroups.com [mailto:gmsk_dv_node at yahoogroups.com] On
Behalf Of John D. Hays
Sent: Friday, September 16, 2011 6:01 PM
To: gmsk_dv_node at yahoogroups.com
Subject: [gmsk_dv_node] D-STAR Deviation

  

There has been a lot of discussion about "low audio" on GMSK boards with
various radios of late. 

 

Just remember, part of the advantage of D-STAR is narrow deviation, the
modulation mask is 6.25 Khz. which can only be achieved with low deviation
(about 1.2 KHz. I believe) so if your receiver is set to wide FM for the
typical FM repeater, the audio should sound a little "weak" --  I've found
that if you have a good ear, you can approach the right deviation by putting
an FM receiver in NFM (usually 2.5 KHz. deviation) and key up a D-STAR
transmitter (the mobiles often do better than handhelds on having correct
deviation) and listen to the audio level and tone, then transmit using your
node adapter and adjust until you get as close as possible to the same
levels.

 

We have a tendency to think "strong" or "loud" audio is always better, but
it is better to have correct levels.  If you don't have a deviation meter,
it is difficult to get it just right.  However, you can also use an
oscilloscope tapped to the discriminator output (9600 baud pin on many
radios) to compare a unknown deviation to a known deviation:

 

"Note from WA6ILQ:

Years ago a DC coupled scope on the discriminator metering point was the
standard way to measure deviation - you put the scope probe on the
discriminator test point, key a synthesized radio and adjust the scope for a
zero volt DC reading, then go up and down 5 KHz and put marks on the scope
screen with a piece of masking tape. Then key up the radio under test and
loudly yell a long, drawn out "four" (i.e. "fouuuuuuuuuur") into the
microphone (the yell was an attempt to drive it into clipping).
Non-symmetrical deviation or clipped modulation is instantly visible, and
you can easily see if you had +/-5kHz peak deviation. If the scope had 11
horizontal graticule lines (zero, +5 and -5) you can easily adjust the
vertical positioning to dead center, then tweak the DC vertical gain to line
up the +5kHz with the top line and -5kHz to the bottom line, and then do
direct readout of peak deviation." --
http://www.repeater-builder.com/projects/poor-mans-dev-meter/devmeter.html


  _____  

John D. Hays
K7VE

PO Box 1223, Edmonds, WA 98020-1223