Joining the svxlink crowd...

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Joining the svxlink crowd...

Gullik Webjorn
Hello,

I have just started to look at svxlink, so please bear with me.
Intended usage is multi rx repeater. A few questions:

Q1: What would you recommend, Squelch via pins, or via svxlink squ
detection? We want voter, so
we are assuming svxlink squ, or else no real SNR?

Q2: If using svxlink squ, are we supposed to use an unsqelched radio? (
so to measure noise )

73 de gullik

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Rob Janssen
On Fri, Jan 20, 2017 at 03:12:40PM +0100, Gullik Webjorn wrote:
> Hello,
>
> I have just started to look at svxlink, so please bear with me.
> Intended usage is multi rx repeater. A few questions:
>
> Q1: What would you recommend, Squelch via pins, or via svxlink squ
> detection? We want voter, so
> we are assuming svxlink squ, or else no real SNR?

Definately svxlink detection via SNR!
Advantage: you can login remotely and change the parameters, vs
having to go to the site and turn the SQL knob.
 
> Q2: If using svxlink squ, are we supposed to use an unsqelched radio? (
> so to measure noise )

Yes, you need unfiltered audio "direct from the discriminator" (e.g.
a "9600 baud packet" output or one of the many mods to tap this audio)

When you use multiple rx with the Voter you will need this anyway.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Gullik Webjorn
Ok, I guessed right then, and we cannot probably use chineese

handhelds as receivers, I do not think they can run with squelch open.  
( squ = 0 ???)

Q3: If I use remotetrx + svxlink-server, will dtmf decoding happen in
remotetrx and

sent out of band to server, or they be passeed as audio and decoded in
server?

(This would enable / disable ability to control remotetrx separately.)

73 de SM6FBD

On 2017-01-20 16:24, Rob Janssen wrote:

> On Fri, Jan 20, 2017 at 03:12:40PM +0100, Gullik Webjorn wrote:
>> Hello,
>>
>> I have just started to look at svxlink, so please bear with me.
>> Intended usage is multi rx repeater. A few questions:
>>
>> Q1: What would you recommend, Squelch via pins, or via svxlink squ
>> detection? We want voter, so
>> we are assuming svxlink squ, or else no real SNR?
> Definately svxlink detection via SNR!
> Advantage: you can login remotely and change the parameters, vs
> having to go to the site and turn the SQL knob.
>  
>> Q2: If using svxlink squ, are we supposed to use an unsqelched radio? (
>> so to measure noise )
> Yes, you need unfiltered audio "direct from the discriminator" (e.g.
> a "9600 baud packet" output or one of the many mods to tap this audio)
>
> When you use multiple rx with the Voter you will need this anyway.
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> Svxlink-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/svxlink-devel


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Rob Janssen
Gullik Webjorn wrote:

> Ok, I guessed right then, and we cannot probably use chineese
>
> handhelds as receivers, I do not think they can run with squelch open.
> ( squ = 0 ???)
>
> Q3: If I use remotetrx + svxlink-server, will dtmf decoding happen in
> remotetrx and
>
> sent out of band to server, or they be passeed as audio and decoded in
> server?
>
> (This would enable / disable ability to control remotetrx separately.)
>
> 73 de SM6FBD
>

When you are looking for cheap low-performance receivers, I recommend RTL-DVB
USB sticks instead of Chinese handhelds.   They are fully functional in svxlink as
receivers and even can receive on multiple nearby channels, receive SSB, etc.

We use them for alternate receive channels in our repeater system and they work
better than first expected.  But you should put a bandpass filter in front of them.
(this is required for most Chinese handhelds just as well)

For better receive performance, you should look for surplus commercial mobile radio
transceivers like the Condor 46 or similar.  Those are easy to modify and provide much
better receive performance (sensitivity and dynamic range) than both the Chinese handheld
and the RTL stick.

All squelch/SNR/CTCSS/tone/dtmf decoding is done in the receiver software and passed
out-of-band to the central logic.
One of our group members has made a mod to allow remote control over the remoterxes
so you can temporarily disable a receiver that is malfunctioning or is being jammed.
This mod is awaiting integration into the main version.

Rob

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Catalin Tanasie
Hi,
the problem with (Chinese) portable is that they get hot quickly so you need
to ventilate them.
There is also a problem with siglev detector, it depends much on audio level
on this.
You can work with squelch open and use PL. On most devices there is a COS
signal that changes when you have the correct tone.
Even on portables you can find this signal powering the audio amplifier.
In our network we have almost 30 remotetrx-es working mostly on Motorola
mobiles/portables and some chinese portables. But we use hardware squelch.
Also we managed to implement that mod and it works flawlesly.
Congratulations to one that had the idea.
73!
Catalin YO7GQZ

-----Original Message-----
From: Rob Janssen
Sent: Friday, January 20, 2017 11:26 PM
To: Discussions about development issues
Subject: Re: [Svxlink-devel] Joining the svxlink crowd...

Gullik Webjorn wrote:

> Ok, I guessed right then, and we cannot probably use chineese
>
> handhelds as receivers, I do not think they can run with squelch open.
> ( squ = 0 ???)
>
> Q3: If I use remotetrx + svxlink-server, will dtmf decoding happen in
> remotetrx and
>
> sent out of band to server, or they be passeed as audio and decoded in
> server?
>
> (This would enable / disable ability to control remotetrx separately.)
>
> 73 de SM6FBD
>

When you are looking for cheap low-performance receivers, I recommend
RTL-DVB
USB sticks instead of Chinese handhelds.   They are fully functional in
svxlink as
receivers and even can receive on multiple nearby channels, receive SSB,
etc.

We use them for alternate receive channels in our repeater system and they
work
better than first expected.  But you should put a bandpass filter in front
of them.
(this is required for most Chinese handhelds just as well)

For better receive performance, you should look for surplus commercial
mobile radio
transceivers like the Condor 46 or similar.  Those are easy to modify and
provide much
better receive performance (sensitivity and dynamic range) than both the
Chinese handheld
and the RTL stick.

All squelch/SNR/CTCSS/tone/dtmf decoding is done in the receiver software
and passed
out-of-band to the central logic.
One of our group members has made a mod to allow remote control over the
remoterxes
so you can temporarily disable a receiver that is malfunctioning or is being
jammed.
This mod is awaiting integration into the main version.

Rob

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel 


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Gullik Webjorn
In reply to this post by Rob Janssen
Tnx Guys,

Conclusion:

Signal processing is done in remotetrx, which is good and flexible. If
we have good QTH's for the receivers, we do not really need a
highperformance receiver , suffice to say good sensitivity and good
demodulation ( low distorsion ), and preferably that all receivers are
THE SAME TYPE, to reduce tuning to setting the same level everywhere. If
this is done properly receiver switching will not be audible.

As for chineese xceivers, my PX-777 is the most sensitive 2 m receiver I
have ever seen. Old and battered up, it beats most commercial gear by
3-4 dB. But, we need unsquelched operation.

Buying a lot of 10 ensures equal characteristics....

The filter recommendation is very apropriate. However, filters for 2m
are a hassle due to size and availability. Thus, a good
(interferencefree) QTH is tha thing.

I have not looked into transmitters yet, with a single transmitter
anything goes. Considering multiple transmitters of course mandates high
quality "commercial" gear, again of same make and type to produce
coherent TX coverage. Here even most mobiles will not work due to
overheating on 100% TX duty cycle they were not designed for. Only
reasonable choice IS a REPEATER (or maybe base ) TX.

First step: Multi RX one server one TX.

73 de Gullik / SM6FBD

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel
Reply | Threaded
Open this post in threaded view
|

Re: Joining the svxlink crowd...

Rob Janssen
Gullik Webjorn wrote:

> Tnx Guys,
>
> Conclusion:
>
> Signal processing is done in remotetrx, which is good and flexible. If
> we have good QTH's for the receivers, we do not really need a
> highperformance receiver , suffice to say good sensitivity and good
> demodulation ( low distorsion ), and preferably that all receivers are
> THE SAME TYPE, to reduce tuning to setting the same level everywhere. If
> this is done properly receiver switching will not be audible.

Unless you are in the middle of nowhere, on a "good QTH" (we normally operate our
receivers on broadcast transmit towers and highrise buildings) your first concern
is not the sensitivity but the performance in the presence of strong signals on other
(nearby) frequencies.  On 70cm this may be even more of an issue than on 2m,
depending on the local usage of frequencies for commercial mobile users.

With the RTL-SDR the equality of the receivers is perfect, as all the parameters
like audio level, frequency characteristic of the modulation, etc are defined by
software,. not by hardware.   So indeed when switching there is no audio change
at all.  However, with conventional analog receivers a similar result is achievable
when you tune the system well.

On the other hand, the dynamic range of these receivers is absolutely terrible.
A strong signal on a nearby frequency makes them go deaf.

>
> As for chineese xceivers, my PX-777 is the most sensitive 2 m receiver I
> have ever seen. Old and battered up, it beats most commercial gear by
> 3-4 dB. But, we need unsquelched operation.

I don't think such a receiver will be able to handle the environment of a 2m repeater
with duplex operation (transmitter on the same site).
For this, you need a good dynamic range but also very good performance of the
frequency synthesizer.  Any noise output from the synthesizer at 600 kHz offset
will be mixed with the local transmitter and raise the noise level on the received
signal.  Those "toy receivers" do not have the specs that a surplus mobile radio has.

They also use SDR with much too few bits in the A/D conversion like the RTL-SDR
which measn the dynamic range is very small.  When using them on the rubber ducky
they work well, connected to a roof mounted antenna you already see problems,
and on a high site it will be terrible.

Personally I have good results with the Baofeng UV-B5 which performs notably better
than the widely known Baofeng UV-5R.  Others also report good results with the
Baofeng GT-3 Mk2.  But compared to a surplus mobile radio or repeater base
station it all is rubbish.

>
> Buying a lot of 10 ensures equal characteristics....

The "equal characteristics" problem is not so critical that you need to buy receivers
from the same lot.  We use different receiver models and sometimes a little difference
is audible, but for an amateur voice repeater it is perfectly OK.

>
> The filter recommendation is very apropriate. However, filters for 2m
> are a hassle due to size and availability. Thus, a good
> (interferencefree) QTH is tha thing.

Well, when you don't put a receiver on your main transmitter location, of course
this helps a lot.  on 70cm, where we need to operate a 1.6 MHz split in the Netherlands,
we actually have this situation on our pi2nos repeater (no tx and rx on the same antenna).
But there still are the signals from other people transmitting at the same site.

>
> I have not looked into transmitters yet, with a single transmitter
> anything goes. Considering multiple transmitters of course mandates high
> quality "commercial" gear, again of same make and type to produce
> coherent TX coverage. Here even most mobiles will not work due to
> overheating on 100% TX duty cycle they were not designed for. Only
> reasonable choice IS a REPEATER (or maybe base ) TX.
>
> First step: Multi RX one server one TX.
>

Yes but please understand: while it is clear that it is very important to transmit
a clean signal to be able to receive at the same site, it actually is equally important
to have a clean signal inside your receiver (local oscillator) for keeping the noise down
and avoid reciprocal mixing with the local transmitter or other signals, and to limit
the signals incoming to the receiver for the same reason.

We have had lots of interesting effects in our repeater systems that all boil down to
unwanted mixing of signals and showing up on the receiver input. This includes
hearing people that transmit on another nearby repeater, hearing transmissions
from other services (including a broadcast from a local church), insensitive receiver
due to all kinds of intermodulation, etc.   You really need good filters.

Rob

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Svxlink-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/svxlink-devel