[OZAPRS] 10MHz HF Gateway -Re: New to list

Mon Sep 3 18:26:25 EST 2007

Hi Ray, hi all,
just to state once again: I am not selling any TNCs and I am not in any
way 
related to SCS.
I just had the luck to travel VK6/VK8 and V51 during past years and using 
such a DSP-TNC in practice.

I agree with many arguments you mentioned: phase distortion at 300baud fsk

is a problem and modern trx are much better than older ones, without high 
stability TXO.
However, my experience is that you easily get mislead by the digital
readout 
of syntesized radios. In most cases if your display reads 10.151,00LSB you

will NOT exactly hit the international APRS qrg and subsequently you will 
not be decoded by PK232 or MFJ or Kantronics or Soundcard,etc..
Ofcourse you can correct that by undergoing some tests with
ECHO-digipeaters 
on the band.
There are other traps in the firmware of modern TRX that might make it 
difficult for users:
 many trx can adjust a frequency offset to make the voice audio sounding 
more high to improve modulation for dx....

So to make the long story short: I accept that you think you dont need 
simultaneous decode in a 800Hz channel, but lets agree that it is a
helpful 
improvement of HF-APRS in 300Baud FSK especially for gateways.

All the above did not even adress the advantages of using RobustPacket
mode 
(RPR, OFDM) instead of 300FSK.
OFDM helps to solve phase distortion problems and multi path, since bit 
duration time is much longer(20msec vs. 3msec). RPR uses 8 carriers with 
abt. 50baud each, and efficient encoding with FEC and random interleaver.
Here is some example from Africa: During 7 weeks in V51  I transmitted my 
posit alternating 300FSK and RPR (same qrg and same number of packets at 
time, same trx, same antenna)  and had two gateway stations in ZS1 (Cape 
Town) to receive:
ZS1TX in 300FSK received and decoded 141 posits in 7 weeks,
ZS1AAZ in RobustPR received 1647 posits in same period.

So I know that good old proven technology is still doing the job, but
there 
is new technology that does it even better, and thats what amateur radio
is 
all about, isn't it?
No worries.
tnx fer reading
vy 73 de MIke
oe3mzc
ex vk3fpf

----- Original Message ----- 
From: "Ray Wells" <vk2tv at exemail.com.au>
To: "Mike Zwingl oe3mzc" <oe3mzc at oevsv.at>; "VK / ZL APRS Users" 
<ozaprs at aprs.net.au>
Sent: Friday, August 31, 2007 4:26 AM
Subject: Re: [OZAPRS] 10MHz HF Gateway -Re: New to list

> Mike Zwingl oe3mzc wrote:
>
>>Gday,
>>just one comment on HF-APRS Gateway operation:
>>in practice it is not the signal to noise ratio that decides about
decode 
>>or nodecode of an APRS beacon on 10Mhz.
>>
> I've only been on HF packet for about 8 years so I guess I've still got
a 
> lot to learn about it.
>
> The two most significant things that affect the ability to decode HF 
> packet are:
> 1. Noise
> 2. Phase distortion. Yes, even with just 200Hz shift it can be a real 
> problem. It's more of a problem than noise at my QTH.
>
> There is no doubt that a rig that drifts will cause problems, but with
any 
> modern synthesised rig the stability is excellent and unliley to be a 
> consideration. Even my 23 years old Kenwood TS430 (which I'm currently 
> using for the 30m gate) is more than satisfactory. I know the reference 
> crystal is off by a100Hz or so but that's another issue that I can
easily 
> resolve when the need to do so gets to the top of my priority list.
>
>>More important is to use a TNC that allows some frequency drift, like
the 
>>DSP-TNC from SCS.
>>The DSP-TNC is currently the only hardware that will simultaneously
decode 
>>several APRS signals at same time in a 800Hz wide channel!
>>So the frequency offset of a mobile station can be plus/minus 400Hz and 
>>still have successful decode on 10.147,6 USB (10151LSB)
>>See the result at VK6UZ-4 and VK8ZAB-4, Fritz and Alan are using these 
>>TNCs.
>>Most other TNCs/Soundcards require 30Hz accuracy in frequency, which is 
>>almost impossible to achive.
>>www.scs-ptc.com/datasheets/scs_datasheet_tracker_english.pdf
>>vy 73 de Mike
>>oe3mzc
>>ex VK3FPF/8
>>
> And how many are going to spend big money on another TNC when what they 
> use works fine. We all know that Pactor-II is better than 0k3 packet but

> at near $2000 in Australia for the controller it would want to be.
>
> VK6UZ and VK8ZAB are seeing lots of travellers in outback Australia, 
> that's true, but have a look at where the travellers are, and then 
> back-track to when they  were closer to stations in the eastern states
and 
> you'll see that other stations got hit more often. It's a matter of 
> distance and propagation, not frequency drift in a mobile.
>
> Far from being almost impossible to achieve, modern rigs easily maintain

> 30Hz acuracy, and better.
>
> I've had my TS430 in Central Australia on some of the most horrenous 
> tracks you can imagine - try the flint country from the vk4/vk5  border
up 
> onto the Lamamour Plateau in South Australia, or the potholes and 
> corrugations on the Birdsville Track just prior to it being graded. At
no 
> time did I detect the slightest stabilty problems, and I spent a lot of 
> time listening to  both amateur and broadcast signals  (in SSB mode)
while 
> mobile.
>
> On could be forgiven for thinking that you're selling those TNC's??
>
> Ray vk2tv 

_______________________________________________
Ozaprs mailing list
Ozaprs at aprs.net.au
http://aprs.net.au/mailman/listinfo/ozaprs