| |
ZL3AI > APRDIG 26.11.06 01:00l 234 Lines 9583 Bytes #999 (0) @ WW
BID : 9136-ZL3AI
Read: GUEST
Subj: [APRSSIG] Vol 29 #22, 2/4
Path: DB0FHN<DB0MRW<DK0WUE<I0TVL<IW8PGT<VK4TRS<ZL2BAU
Sent: 061125/2355Z @:ZL2BAU.#79.NZL.OC #:17071 [Waimate] $:9136-ZL3AI
From: ZL3AI@ZL2BAU.#79.NZL.OC
To : APRDIG@WW
Message: 10
Date: Fri, 24 Nov 2006 15:27:16 -0800
From: <scott_at_opentrac.org>
Subject: RE: [aprssig] Point-to-point telemetry addressing
>However, neither of these appear to work. The only addressing that
>seems to work is APRS,WIDE2-2, but that only works from time to time,
APRS should not be part of your path, unless you're using a program that
takes the first entry as the destination callsign. I forget which ones do
that. Do you have an example of one of your packets after it's been
digipeated? That might tell you what's going on.
Scott
N1VG
------------------------------
Message: 11
Date: Fri, 24 Nov 2006 18:43:59 -0500
From: "Michael R. Owen" <nlsa_at_nlsa.com>
Subject: Re: [aprssig] Point-to-point telemetry addressing
An HTML attachment was scrubbed...
------------------------------
Message: 12
Date: Fri, 24 Nov 2006 15:49:57 -0800
From: "Stephen H. Smith" <wa8lmf2_at_aol.com>
Subject: Re: [aprssig] Point-to-point telemetry addressing
nlsa_at_nlsa.com wrote:
>Dear friends,
>
>I know almost nothing about packet radio or APRS so please forgive me
>if my question is stupid. My home station (W9IP-1) is near Syracuse,
>NY and I'm trying to maintain a telemetry link with our cottage
>(W9IP-2) about 35 km away.
"maintain a telemetry link" ?? Unlike conventional packet, APRS stations
don't "connect" to a specific other station. You just do a one-to-many
broadcast to any and all stations within earshot.
>The terrain is rough so a direct path doesn't work. There is a good
>APRS digipeter (K2AMB-3) near W9IP-1 and another good one (KB2FAF-10)
>near W9IP-2. The two digipeters "see" each other very well. The one
>near W9IP-2 is also an IGate, which is important because I want the
>telemetry to show up on findu's database.
>
>I would think that...
>W9IP-1's path to W9IP-2 would be APRS,K2AMB-3,KB2FAF-10,W9IP-2
>and that
>W9IP-2's path to W9IP-1 would be APRS,KB2FAF-10,K2AMB-3,W9IP-1
Why are you including your own stations at the end of the path? This just
makes your own (target) station try to digipeat the packet once more. Using
specific callsigns instead of generic WIDEn-N is a good practice for
something like this since it avoids having transmissions spreading out in
every direction needless generating more QRM. And you are correct that
the directed path from each end will be a mirror image of the path from the
other end.
I assume from your usage of APRS as the first phrase in the path that you
are using UIview, since all other APRS applications have the
"destination" (any quasi-callsign beginning in APnnnn) in a separate field
or entry from the digipeater path hops.
If the device doing the telemetry is a stand-alone TNC rather than a
program running on a PC, the path will be entered into it's hardware using
an ASCII terminal program for setup. It will be something like: "APRS via
KB2FAF-10,K2AMB-3" [Exact syntax will vary depending on the particular TNC
being used.]
--
Stephen H. Smith wa8lmf (at) aol.com
EchoLink Node: 14400 [Think bottom of the 2M band]
Home Page: http://wa8lmf.com --OR-- http://wa8lmf.net
NEW! TNC Test CD
http://wa8lmf.net/TNCtest
JavAPRS Filter Port 14580 Guide
http://wa8lmf.net/aprs/JAVaprsFilters.htm
"APRS 101" Explanation of APRS Path Selection & Digipeating
http://wa8lmf.net/DigiPaths
Updated "Rev G" APRS http://wa8lmf.net/aprs
Symbols Set for UI-View,
UIpoint and APRSplus:
------------------------------
Message: 13
Date: Fri, 24 Nov 2006 18:58:38 -0500
From: "Michael R. Owen" <nlsa_at_nlsa.com>
Subject: Re: [aprssig] Point-to-point telemetry addressing
An HTML attachment was scrubbed...
------------------------------
Message: 14
Date: Fri, 24 Nov 2006 15:58:56 -0800
From: "VE7GDH" <ve7gdh_at_rac.ca>
Subject: Re: [aprssig] Point-to-point telemetry addressing
It doesn't really matter what the path from W9IP-1 to the cottage is. If the
cottage is W9IP-2, I would try a path of WIDE2-2. However, you said that you
were using APRS,WIDE2-2. That would only work if there was a digi that
responded to the alias of APRS. I'm assuming that you're not using UI-View
at W9IP-2. If you are, the "APRS" in the "unproto address" is really the
destination. Everything after that is the path. Try changing W9IP-2 so it
uses a path of WIDE2-2. This will work with any WIDEn-N digi that hears it.
If it so happens that you don't have a reliable path to a WIDEn-N digi, but
there are home fill-in digis within earshot that respond to WIDE1-1, you
could try WIDE1-1,WIDE2-1 instead. Would that make it to the IGate? If not,
you could increase the path to WIDE1-1,WIDE2-2. Generally speaking, fixed
stations shouldn't have WIDE1-1 in their path. My suggestion is for ONLY if
using WIDE1-1 will help you get to a WIDEn-N digi that would otherwise be a
bit iffy to get to directly. Re-reading you message, you said there was a
"good" digi near W9IP-2, so avoid using WIDE1-1 unless you really need to
use it.
The other possibility would be to use a specific path from W9IP-2 instead of
WIDE2-2. You could try KB2FAF-10,K2AMB-3 for the path from the cottage. Do
not include W9IP-1 as part of the path. First, it would be wasteful of
bandwidth by adding 7 bytes to every transmission, and it wouldn't be doing
any good because you are trying to receive the telemetry at W9IP-1, not
digipeat it.
The generic WIDE2-2 could spread out in multiple directions. By channelling
it to specific digipeaters, it would only propagate via the path that you
specify. However, if either of those digis went QRT, no other digis would be
used. I would probably go with the generic WIDE2-2.
You are right. You cannot expect to receive every beacon from W9IP-2. All it
takes is a packet collision for it to fail, or a burst of noise, intermod,
etc. If W9IP-2 has good "ears" it will at least hold off transmitting when
the frequency is already in use. There is always the "hidden transmitter"
that you can't do anything about, but if you have a decent antenna at W9IP-2
and the squelch is set loose or wide open, it will have a better chance of
knowing if the frequency is already in use. The danger of a loose squelch is
that if anything changes (i.e. something causes it to go open) it could end
up never transmitting. You are better off running with the squelch wide open
with the TNC set for software carrier detect.
Unless you are using UI-View at W9IP-2, get rid of the "APRS" in the unproto
path.
73 es cul - Keith VE7GDH
--
"I may be lost, but I know exactly where I am!"
------------------------------
Message: 15
Date: Fri, 24 Nov 2006 19:15:01 -0500
From: Steve Dimse <steve_at_dimse.com>
Subject: Re: [aprssig] Point-to-point telemetry addressing
What jumps out at me is the packet is really long. Do you really need to
add "Telemetry from East Homer, NY on 11/23/2006 at 11:11 PM" to every
packet? This doubles the packet length and greatly increases the chance for
collision. Though time delays do occur, they are quite rare, and the time a
packet is received at findU should be adequate for your needs. findU is
synced to a level 1 time server, whereas since your cottage does not have
internet, the computer is probably not synced at all (there appears to be a
consistant 80 second offset now) so findU's time is probably much more
accurate even including propagation delays.
Looking at the packets, you seem to get about 70-80% through, with a
shorter packet you will have even greater return.
You should also occasionally send a position packet so you will appear on
maps!
Steve K4HG
------------------------------
Message: 16
Date: Fri, 24 Nov 2006 16:23:35 -0800
From: "Stephen H. Smith" <wa8lmf2_at_aol.com>
Subject: Re: [aprssig] Point-to-point telemetry addressing
nlsa_at_nlsa.com wrote:
>Stephen -
>
>Yeah, poor choice of words. I just want to report temperatures at the
>cottage back to home, then occasionally control the cottage's
>thermostat from home. Messages exchanged maybe once every 30 minutes,
>so it's not a connection in the traditional sense. Nevertheless, I
>want to make sure that the one-to-many broadcast includes the two
>digipeters that I mentioned so that home and cottage will be
>included. Maybe I'm using the wrong tool and APRS isn't what I need?
>
>MRO
It sounds as though APRS should be able to do this. How are you
controlling the thermostat? KPC3 TNCs have single-bit open-collector
switches controllable by messages to the TNC address but TNC2s don't.
For a TNC2 or clone, the path setting of interest will be the command
"UNPROTO" typed into the TNC by an ASCII terminal or as an init string
during program startup, which determines the path used on non-connected
beaconing. This will be something like:
"UNPROTO APRS VIA K2AMB-3,KB2FAF-10 " (For the home station)
and
"UNPROTO APRS VIA KB2FAF-10,K2AMB-3" (For the cottage station)
One slight possibility is that one or the other of the digipeater operators
involved overlooked placing the digi's own callsign into the alias list of
callsigns to digipeat on, along with WIDEn-N. If so, the digi may never
respond to, and digipeat on, it's own callsign.
Another possibility is that one or ther other of the digipeater(s) uses
aliases instead of actual callsigns. Do you see the digis using their own
callsigns in the path in beacons? I.e. is the "callsign" that they send
from some sort of alias like "Syracuse" with the real callsign only in a
beacon comment field? Or do they actually use their own callsigns?
------------------------------
Read previous mail | Read next mail
| |
