Need to setup SDR for Broadcastify and Zello

[rpurinton]

Hi all. Currently have a whistler 1040 and 1065 and 2 laptops running Broadcastify streams for local fire/ems and police. Both connected to the Portland Maine P25 system. PD feed monitors 2 talkgroups and fire monitors about a dozen.

we are experiencing problems with simulcast delay due to issues with poor CQPSK support on the whistlers.

I want to go SDR now and have some noob questions.

can I use multiple SDR usb receivers on to handle both Broadcastify feeds on 1 PC?

One pc has 4x USB ports, do I need 1, 2, or 4 SDR receivers?

i have virtual audio cable so that should allow me to emulate multiple soundcards.

what’s a reliable receiver these days in the sub $50 range?

what’s good windows software? ZelloVOX only works on windows so I must use windows.

can I do it all for less than the cost of 2 New sds200?

 

[KD0TAZ]

I just set up a system using SDRTrunk for a P25 feed to upgrade from my hardware scanner, and a few minor bugs aside in the current version, it seems to be solid. I've had it running for the past week with no hiccups.

What I love about it is that it is all-in-one. You don't have to run 3-4 separate programs like you do with everything else. No separate decoders, no VACs, just run the main program. This is the first version with RR/BCFY integration, and it is in alpha, but as I said it definitely gets the job done easier than any of the others. Log into RR with it and it automatically configures any feeds you have registered, plus it can load the database and tags. It can also serve the new Calls platform, and you can serve streams via Icecast. For streams it records and queues each transmission (including those that are received simultaneously) for sequential playback into the stream. I'm not sure how Zello works or what you're using it for, but I'm sure it can be made to work with SDRTrunk, as it can call scripts/programs and trigger actions. For P25, there's a hoop that you have to jump through during setup, building the decoder from source (it's a licensing thing), but it's not that difficult. I can send the built file to you if you want.

I have mine running off a single Nooelec NESDR stick ($40). The site frequencies for my streams all fit within 2MHz so I don't need more than one, but you can add as many as you need for what you stream (and how busy it is). If you have say, two analog frequencies that are more than 2MHz apart, you'd need one stick for each. If you have a trunked system whose frequencies span more than 2.2MHz, you'd need two for that.

It's running on an old i3 2310 laptop (dedicated), and it handles a Broadcastify feed, a Calls source, and a private Icecast feed (you create separate lists of monitored TGs for each stream) running simultaneously just fine, although it slows other programs down so I wouldn't try it on anything less than that.. Eventually I'll get something more powerful to run it on, but that's what I had laying around and it works.

Even if you need to actually buy a NUC or something to run it on, it'll still be far less than what a single SDS200 costs lol.

 

[Seanm214]

I just set up a system using SDRTrunk for a P25 feed to upgrade from my hardware scanner, and a few minor bugs aside in the current version, it seems to be solid. I've had it running for the past week with no hiccups.

What I love about it is that it is all-in-one. You don't have to run 3-4 separate programs like you do with everything else. No separate decoders, no VACs, just run the main program. This is the first version with RR/BCFY integration, and it is in alpha, but as I said it definitely gets the job done easier than any of the others. Log into RR with it and it automatically configures any feeds you have registered, plus it can load the database and tags. It can also serve the new Calls platform, and you can serve streams via Icecast. For streams it records and queues each transmission (including those that are received simultaneously) for sequential playback into the stream. I'm not sure how Zello works or what you're using it for, but I'm sure it can be made to work with SDRTrunk, as it can call scripts/programs and trigger actions. For P25, there's a hoop that you have to jump through during setup, building the decoder from source (it's a licensing thing), but it's not that difficult. I can send the built file to you if you want.

I have mine running off a single Nooelec NESDR stick ($40). The site frequencies for my streams all fit within 2MHz so I don't need more than one, but you can add as many as you need for what you stream (and how busy it is). If you have say, two analog frequencies that are more than 2MHz apart, you'd need one stick for each. If you have a trunked system whose frequencies span more than 2.2MHz, you'd need two for that.

It's running on an old i3 2310 laptop (dedicated), and it handles a Broadcastify feed, a Calls source, and a private Icecast feed (you create separate lists of monitored TGs for each stream) running simultaneously just fine, although it slows other programs down so I wouldn't try it on anything less than that.. Eventually I'll get something more powerful to run it on, but that's what I had laying around and it works.

Even if you need to actually buy a NUC or something to run it on, it'll still be far less than what a single SDS200 costs lol.

Are you monitoring a p25 phase 2 system or a phase 1 system?

 

[KD0TAZ]

Are you monitoring a p25 phase 2 system or a phase 1 system?

Phase 1, but SDRTrunk will do both.

 

[rpurinton]

Luckily I’m also a software dev and very familiar with compiling things.

so if I understand it, if I use just 1 SDR dongle I can track a p25 system.

why would I want to use 2?

if I bought 4 dongle’s, does that mean I can dedicate one to the control channel and be able to record 3 voice channels simultaneously?

i need to be able to separate police and fire radio traffic into two separate internet streams. Is there existing software that can send one set of talk groups to one soundcard and a different set of talk groups to a different sound card?

Should I buy 1 laptop with 4 dongle’s or 2 laptops with 2 dongle’s each?

 

[slicerwizard]

You shouldn't have any trouble doing that with one laptop and two dongles. All of the decoders (Unitrunker, SDRTrunk, DSD+ FL, OP25) should be up to the task.

 

[KD0TAZ]

Luckily I’m also a software dev and very familiar with compiling things.

Ok cool. It's just a gradle build with JDK14.

so if I understand it, if I use just 1 SDR dongle I can track a p25 system.

Yes (see next answer)

why would I want to use 2?

Each dongle can "see" about 2.2MHz of spectrum at any given time. It needs to be locked on the control channel at all times, so as long as the voice channels are +/- 2.2MHz of the control channel, it can pull them off the same tuner. The ideal scenario for one system/one dongle is if the control channel and all voice channels fall within the same 2.2MHz slice (most sites will be). This is because while the slice can dynamically "shift" 2.2MHz to either side of the control channel as needed (allowing you to technically monitor 4.4MHz on a single dongle), it can't shift to both sides at the same time. So if you have a voice channel that's lower and a voice channel that's higher, simultaneous transmissions could be missed if they are more than 2.2MHz apart. It's not common for the channels of a single site to cover that much spectrum except for busy systems in metro areas with a lot of radios.

if I bought 4 dongle’s, does that mean I can dedicate one to the control channel and be able to record 3 voice channels simultaneously?

I haven't played with Unitrunker in a while, and I know that's how it USED to work (requiring a dongle for each channel), and I've never tried OP25, but I'm pretty sure it isn't the case anymore with any of the SDRs. SDRTrunk can pull in as many channels off a single dongle as it needs - as long as they're all within that 2.2MHz slice, and as long as the computer is powerful enough to decode them all simultaneously (any i5 or i7 from the past 5 years or so shouldn't even break a sweat). Mine can do 5 voice channels simultaneously (plus the control channel) off one dongle without an issue. Never been able to try more than that though because the site is not busy enough. If you are doing multiple systems or multiple sites off the same system (again, if all the channels of multiple systems or sites all fall into the same 2.2MHz slice, then it can run multiple systems off a single dongle), you would want to get one extra dongle to ensure you have a big enough spectrum to see everything (and it can also act as a failover).

i need to be able to separate police and fire radio traffic into two separate internet streams. Is there existing software that can send one set of talk groups to one soundcard and a different set of talk groups to a different sound card?

SDRTrunk can do it natively. No separate instances/VMs, no virtual audio cables, no scripting. You just create a "playlist" with all the TGs/Aliases that you want to send to each stream. It has BCFY integration built in, so once you log in, it will automatically load the credentials and configuration for each stream you have registered to your account and offer it as a streaming option that you can assign to each "playlist". As I said I am running a Calls feed, a traditional streaming feed, and a private Icecast feed (each with different TGs) off one instance of SDRTrunk. The one thing I definitely recommend for streaming is to set up a small (128 or 256MB would be fine) RAMDrive to use as the queue temp storage. Since it records calls to queue them into the stream (then deletes them), that can put a lot of wear on a hard drive or SSD.

Should I buy 1 laptop with 4 dongle’s or 2 laptops with 2 dongle’s each?

I would suggest if you are buying equipment, don't do it with a laptop. Get a NUC. You'll get more bang for your buck (and a much smaller package) that way. Then you can run it headless using VNC. Another option you have is Dell Optiplex (business) desktops. You can get powerful off-lease used machines pretty cheap on ebay, and even the SFF/USFF machines have IPMI/BMC (remote control even if power is off or the OS is locked up) and give you KVM-over-IP.

 

[rpurinton]

First off, great information there!

So it looks like 2.5mhz between the control channel and the lowest freq. I guess 2 dongle’s is the way to go. Would I set one dongle on the data channel and let the other dongle grab voice, or is there a better route?

my other requirement is to do Zello. Zello is a push to talk app that records from an audio input with voice activation itonly transmits to Zello when there is radio traffic to send. If Silence, Zello stops transmitting.

if I run multiple Zello instances on one pc they will need unique audio inputs. I have virtual audio cable pro already, do you think SDrtrunk will be able to play the audio from different playlists out 2 different virtual sound cards?

 

[rpurinton]

Just watched a tutorial for setting up 0.4 beta. When he configured the p25 settings it offered demodulator settings for Simulcast LSM and C4FM. Our system uses Simulcast CQPSK, do you know if that will work with the LSM setting or if CQPSK is supported in a newer version?

edit never mind, Iread more and found LSM is CQPSK. Cool. Also thanks for the suggestion on using a NUC. I see 4GB and 8GB NUC, knowing Java I assume 8 GB will be better especially if creating a ram disk.

is it possible to store recordings for like 72 hours with Sdrtrunk? i am sometimes asked to provide recordings in mp3 within a few days...

 

[KD0TAZ]

First off, great information there!
View attachment 85518
So it looks like 2.5mhz between the control channel and the lowest freq. I guess 2 dongle’s is the way to go. Would I set one dongle on the data channel and let the other dongle grab voice, or is there a better route?

So 2.2MHz is the "guaranteed safe" range. Basically any dongle will be able to operate in that bandwidth. The "certified stable" bandwidth for the NESDR dongle is 2.4MHz, and it can technically do up to 3.2MHz.. SDRTrunk allows you to use a max of 2.88MHz. So since you have a 2.5MHz span it still would most likely work for you. That's something you'd have to test for yourself. If it works reliably on the 2.56 or 2.88MHz settings, then run with it. Otherwise use two. And yes if you use two, you can set it to "prefer" a specific dongle for the control channel, then it will automatically use either one for tuning voice channels depending on which frequencies are in use. If you use two, definitely set it to 2.4MHz.

my other requirement is to do Zello. Zello is a push to talk app that records from an audio input with voice activation itonly transmits to Zello when there is radio traffic to send. If Silence, Zello stops transmitting.

if I run multiple Zello instances on one pc they will need unique audio inputs. I have virtual audio cable pro already, do you think SDrtrunk will be able to play the audio from different playlists out 2 different virtual sound cards?

Ok Ive been reading on Zello (I'm assuming you are using it as private streams?), and I think it can be made to work, but it won't be the way you're thinking. The reason is you can't assign talkgroups or playlists to a particular channel or soundcard (yet - it has been recently requested in the dev's support group). It just sends the next transmission to the next unused channel in round-robin fashion. If there are three or more transmissions occurring simultaneously, the third and more don't come through locally (but they DO get streamed). The way Zello works has me kind of intrigued, so I'll see if I can make it work. It can probably be done using local Icecast server mounts and a Raspberry Pi running Android as a "repeater". Zello could probably run locally on the PC and stream the Icecast mounts via VACs too but having a separate computer would eliminate a bunch of kludges to make it all automatically run on one machine.

Just watched a tutorial for setting up 0.4 beta. When he configured the p25 settings it offered demodulator settings for Simulcast LSM and C4FM. Our system uses Simulcast CQPSK, do you know if that will work with the LSM setting or if CQPSK is supported in a newer version?

edit never mind, Iread more and found LSM is CQPSK. Cool. Also thanks for the suggestion on using a NUC. I see 4GB and 8GB NUC, knowing Java I assume 8 GB will be better especially if creating a ram disk.

Yeah definitely. The laptop I'm running it on has 8GB and it shows 59% used including the 256MB RAMDrive (1.5GB is Java). Just make sure you're looking at one of the i5 or i7 versions, not Atom, Celeron, or Pentium. They come in all sorts of flavors, and I think up to 16GB. Depending on the one you buy, it may or may not come with storage, so you may need to budget for a NVMe SSD.

And 0.5.1 alpha is the one you want. 0.4.0 does not have the Broadcastify integration yet. There have been a lot of UI changes so make sure you check the manuals, as they have been updated.

is it possible to store recordings for like 72 hours with Sdrtrunk? i am sometimes asked to provide recordings in mp3 within a few days...

Absolutely. You can set up archive recording independent of the streams. A simple script could keep the archive pruned to your desired retention time.

 

[rpurinton]

thanks it sounds like SDRtrunk is the software I want to use. I just found a dell laptop on amazon for 222$ with 8GB ram, cheaper than any NUC I could find.

If I need to get 2 laptops and 4 dongle’s to get Zello to work. I would dedicate one for fire and one for police.

Thanks for your input, looking forward to the results!

 

[KD0TAZ]

What's the CPU on that?

 

[rpurinton]

• Intel Core i5-3210M @ 2.5 GHz, 8GB RAM, 320GB HDD

 

[KD0TAZ]

Check this out.. It's $300, but literally everything is better... i5-6500T (3 generations newer than the 3210M), Quad core (vs dual core), 16GB DDR4 (vs DDR3), 250GB SSD (vs HDD). It's also VESA sized - it's a thin computer that can be mounted on the back of a monitor, and it has vPro KVM-Over-IP.

Amazon.com: HP EliteDesk 800 G2 Desktop Mini Business PC, Intel Quad-Core i5-6500T up to 3.1G,16G DDR4,240G SSD,VGA,DP,Win 10 Pro 64 bit-Multi-Language Support English/Spanish (Renewed): Computers & Accessories

Amazon.com: HP EliteDesk 800 G2 Desktop Mini Business PC, Intel Quad-Core i5-6500T up to 3.1G,16G DDR4,240G SSD,VGA,DP,Win 10 Pro 64 bit-Multi-Language Support English/Spanish (Renewed): Computers & Accessories

www.amazon.com

 

[rpurinton]

Ty for the suggestion!

Last night I ordered 4 of the RTL-SDR.com v3 for $24 each

I also ordered 4 small Omni +5db SMA antennas rated for 700-2700Mhz for $7 each.


i havent ordered any computers yet. I’m going to try the existing 2 we have that have been running the streams thus far. The CPUs have always been pretty idle on those machines.
I see that Zello api integration is in the roadmap. That’s fantasy land for me because it eliminates having to run the Zello client at all and thus eliminates the necessity to run it on windows...

I don’t know if there’s a way to support alpha tags thru Zello but that would be cool too.

 

[rabrol]

Ok Ive been reading on Zello (I'm assuming you are using it as private streams?), and I think it can be made to work, but it won't be the way you're thinking. The reason is you can't assign talkgroups or playlists to a particular channel or soundcard (yet - it has been recently requested in the dev's support group). It just sends the next transmission to the next unused channel in round-robin fashion. If there are three or more transmissions occurring simultaneously, the third and more don't come through locally (but they DO get streamed). The way Zello works has me kind of intrigued, so I'll see if I can make it work. It can probably be done using local Icecast server mounts and a Raspberry Pi running Android as a "repeater". Zello could probably run locally on the PC and stream the Icecast mounts via VACs too but having a separate computer would eliminate a bunch of kludges to make it all automatically run on one machine.

I wonder if you managed to do any further experiments with a Pi running Android and feeding audio from SDRTrunk to Zello. I use Zello extensively for my feeds, and also want to be able to monitor my local P25 talkgroups on my trust Radiotone RT4.

Did you manage to get it working?

 

[boatbod]

Two RTL dongles and one Linux laptop (or RPi4) running op25 is all you need. Op25 works well with liquidsoap for the connection to broadcastify.

 

[rpurinton]

I ended up using SDrtrunk. Unfortunately Zello has not yet been integrated so my solution involves using virtual audio cable and because I need to have police and fire on separate Zello channels I needed to get to wholy separate PCs each with their own SDR dongle‘s and it’s been working great for a couple months. Hopefully Denny will get Zello integration working soon and I’ll be able to Broadcast multiple talk groups onto multiple Zello channels with a single PC

 

[rabrol]

I ended up using SDrtrunk. Unfortunately Zello has not yet been integrated so my solution involves using virtual audio cable and because I need to have police and fire on separate Zello channels I needed to get to wholy separate PCs each with their own SDR dongle‘s and it’s been working great for a couple months. Hopefully Denny will get Zello integration working soon and I’ll be able to Broadcast multiple talk groups onto multiple Zello channels with a single PC

That would certainly be ideal! While I have no objection to Broadcastify, I have various devices set up for Zello, and I want to be able to listen to the P25 on it. Is Denny actually considering a Zello implementation in the software? I know Zello don't have a Linux client (a great shame), but it would be great if he can figure something out that works within SDRTrunk!

 

[rpurinton]

Yes there’s a feature request in for Zello API support. Unknown ETA.