DSD+: Can't Creat Windows

[br0adband]

First thing I noted: the signal is way way too "loud" on the spectrum, and since I have no idea what SDR device you're using all I can suggest is to back off the gain significantly.

Second thing I noticed: you're off frequency by the slightest amount but with a digital decode a few ppm can make all the difference in the world between a clear decode and all those ERR events you're seeing. Set SDR# so it's using the proper stepping size (6.25 kHz) and it shouldn't be an issue as long as you've got the ppm setting for the device correct as well. Might consider adjusting the Zoom as well to a further out view but that's a personal kind of preference so ignore it if you want.

Third thing I noticed: you're allowing DSD+ to be wide open meaning it's listening for anything that comes along - it works better by defining and limiting it to whatever specific format you want so, if you add -fr to your command line it'll listen for and decode only DMR/TRBO traffic which is more efficient operation.

I'm betting that if you adjust the stepping to 6.25 kHz it'll still be slightly off frequency because of a mis-adjusted ppm which can be handled fairly easily. You can use something like the ATSC signal at 470.310 MHz to lock in your ppm as accurately as possible.

Now, if you're using an SDR device of some kind that is using a TCXO the ppm setting shouldn't be an issue but even so it doesn't hurt to absolutely verify that device is accurate using that ATSC signal at 470.310 MHz (pretty much anywhere in the US you'll pull that in fairly strong except maybe Death Valley or whatever).

 

[UPMan]

Thanks for the tips. That was w/Tuner AGC enabled. I think the problem might be that the source signal is too weak into the antenna. BTW, using the RTS2832U R820T2 TCXO.

I'm thinking that the $25 dongle simply does not have adequate RF performance for this channel at this location (and similar negative results for some P25 channels that come in clean on a scanner). I'll try some more with a DMR handy at the office, tomorrow. Could be that there is a reason that it was only $25.

 

[br0adband]

Well, RTL sticks are notoriously "cheap USB TV tuners" anyway so their costs have been the primary reason for the explosion of popularity in SDR over the past few years since someone discovered they could perform quite well as general wideband receivers. No they ain't perfect by a longshot but when everything works, they do a pretty damned good job considering the actual cost.

I'm still using "original" RTL sticks with the R820T tuner from NooElec and haven't had issues but I do intend to get the newer R820T2 versions that are slightly more sensitive and inside the aluminum enclosure (the models sold by RTL-SDR.com specifically).

Typically most people find the RTL sticks work best by disabling the RTL AGC and Tuner AGC and just using the gain slider as required. I keep mine at 32.6 dB and sometimes at 42.1 dB depending on what I'm monitoring and never had issues but then again I'm centrally located in the Las Vegas "valley" (we're surrounded on all sides by mountains and I'm smack in the middle geographically so I get everything pretty well).

So those tips I offered will still be relevant regardless: the ppm is vitally important for this stuff more than anything else even with a TCXO-based stick; I've seen people with such devices including Airspy and SDRplay that still end up with a ppm of either +/- 1 or 2 (from 0) and if they don't account for that in the SDR software settings their digital decodes are full of errors. It frustrates them to no end because everything else seems to work properly - and they're under the impression that because of the TCXO they shouldn't have to worry about the drift potential with 1-2 ppm except they actually do - but when they realize it and make that tiny adjustment to the ppm, wham, the decodes come through without any other major issues aside from reception itself.

It's just a new way of doing things, and no offense to Uniden itself - pretty sure right now you're doing all this experimenting on your own home computer setup - but considering what I can do with these "cheap USB TV tuners" I'm pretty happy so far. I'd love to get another Uniden scanner at some point depending on how things work out so, regardless of how things transpire in the future I hope you get some enjoyment from that $25 investment into SDR.

 

[SCPD]

About the TCXO sticks from RTL-SDR.com ...

In SDR#, I can set the correction value to -1 or -2. Both are off slightly - but close enough.

In the Unitrunker program, I can set the correction to -1.5 - which is dead center.

 

[br0adband]

Well learn something new every day, go figure: I didn't realize I could set them (in Unitrunker) as real numbers, interesting. I should have realized that considering that when the Drift Correction kicks in as required sometimes I note it's a real number (and not a natural/whole number). Pretty cool.

And yes, that was my meaning that the TCXO models of RTL-based hardware (any of them, not just those from RTL-SDR.com) tend to be just 1 or 2 ppm off the zero point on either side and in some situations with a digital format it can wreck the decode. I know of two systems here in the Las Vegas area, both NXDN-based with 3.125 kHz spacing (irritating but it's in use) and if the stick I have set to the Voice role is even 1 ppm off - the stick I assign has a verified and consistent ppm of 55 - DSD+ will simply not decode it. I set it to 54 or 56 and it's nothing but garbage decodes; set it for 55 as expected and voila, good clean error free decodes, talk about picky.

Thanks for mentioning that, Rick, just another awesome aspect of Unitrunker.

 

[redbeard]

Other bonus is with 1.101 you can lose the whole virtual audio cable and SDR# mess and just use FMP.

I noticed in the pic that the audio level is not very high, which could account for errors as well as not being on freq.

You should use SDR# first to determine your ppm offset. This is the first step with every dongle no matter what. You have to know your ppm to be on freq. I use the local NWS transmitter, the ATSC pilot freq mentioned is a good way too.

Once you have your ppm correction, stop using SDR# and use FMP. You won't have to worry about input and output levels because the I/Q data is shared directly via TCP/IP to DSD+. It's like a discriminator tap from a scanner (which would be nice if they started coming with from the factory, hint, hint, cough PSR800.) Obviously using SDR# lets you tune around easy but so does FMP somewhat and you don't need to be frequency agile because you usually know what freq you are trying to monitor ahead of time.

Also FYI the Tytera is a cheap wannabe rip off of Hytera and has been proven to have poor timeslot tolerance. You should expect errors from that radio. Keep that in mind when testing _any_ DMR receiver.


Sent from my iPad using Tapatalk

 

[slicerwizard]

A correction: It is the channel display doesn't appear. The audio and event windows do appear. With my settings, they are viewable.

PC settings / System / Display / Change the size of text, apps, and other items: xxx%

What is xxx set to?

 

[garys]

I'm using an Airspy connected to SDR# to feed CC.bat with excellent results.
I'm using a R820T2 connected to FMP-VC to fee VC.bat, with very good results now that someone told me that my PPM correction was way off.

FMP-VC reports gain of 49.6, but I have no idea how that is set. Step is 6.25, which seems to work well, but I would like to know how to adjust both of those parameters.

Well, RTL sticks are notoriously "cheap USB TV tuners" anyway so their costs have been the primary reason for the explosion of popularity in SDR over the past few years since someone discovered they could perform quite well as general wideband receivers. No they ain't perfect by a longshot but when everything works, they do a pretty damned good job considering the actual cost.

I'm still using "original" RTL sticks with the R820T tuner from NooElec and haven't had issues but I do intend to get the newer R820T2 versions that are slightly more sensitive and inside the aluminum enclosure (the models sold by RTL-SDR.com specifically).

Typically most people find the RTL sticks work best by disabling the RTL AGC and Tuner AGC and just using the gain slider as required. I keep mine at 32.6 dB and sometimes at 42.1 dB depending on what I'm monitoring and never had issues but then again I'm centrally located in the Las Vegas "valley" (we're surrounded on all sides by mountains and I'm smack in the middle geographically so I get everything pretty well).

So those tips I offered will still be relevant regardless: the ppm is vitally important for this stuff more than anything else even with a TCXO-based stick; I've seen people with such devices including Airspy and SDRplay that still end up with a ppm of either +/- 1 or 2 (from 0) and if they don't account for that in the SDR software settings their digital decodes are full of errors. It frustrates them to no end because everything else seems to work properly - and they're under the impression that because of the TCXO they shouldn't have to worry about the drift potential with 1-2 ppm except they actually do - but when they realize it and make that tiny adjustment to the ppm, wham, the decodes come through without any other major issues aside from reception itself.

It's just a new way of doing things, and no offense to Uniden itself - pretty sure right now you're doing all this experimenting on your own home computer setup - but considering what I can do with these "cheap USB TV tuners" I'm pretty happy so far. I'd love to get another Uniden scanner at some point depending on how things work out so, regardless of how things transpire in the future I hope you get some enjoyment from that $25 investment into SDR.

 

[br0adband]

I think we're kinda going off-topic with that kind of a question, it might be better if you created a thread for it after searching for potential answers (presuming you haven't done that already). My apologies for nudging things off-topic in the first place.

 

[UPMan]

PC settings / System / Display / Change the size of text, apps, and other items: xxx%

What is xxx set to?

Currently 250%, but I've tried multiple values above and below that with no change in application behavior.

 

[slicerwizard]

Oh well, it was worth a shot.

 

[UPMan]

Dang. Worse today than yesterday. Using FMP and can't get DSDPlus to even respond w/errors.

Screen shot is during xmit.

 

[br0adband]

I don't use FMP (haven't gotten around to it since it can't do P25 trunking and may never handle it) but iirc FMP outputs a signal to DSD+ using TCP/IP - I think it's possible to use the virtual audio cable but I'm pretty sure the design for both FMP and DSD+ working together is FMP outputs a stream of data over a TCP/IP port for DSD+ to pick up on and work with. I'm not saying it can't or won't work at all with the virtual audio cable, just the DSD+ developers found a way around the necessity of using such utilities by using the TCP/IP aspects.

I'm sure someone else that has more experience using FMP and DSD+ together for DMR monitoring will chime in on this and hopefully be able to get things working.

 

[Voyager]

I wonder if the message about "no frequency data files found" is why it isn't opening that window.

Granted, it's a long shot, but it's about as good as anything else at this point.

I also wonder if there should be quotes around the primary and secondary frequency lists.

 

[slicerwizard]

Dang. Worse today than yesterday. Using FMP and can't get DSDPlus to even respond w/errors.

Screen shot is during xmit.

You're sending a TDMA signal from a DMR portable to DSD+? You have DSD+ running in autobaud mode, which may degrade its ability to deal with the TDMA signal. Using -fr (decode only DMR) to lock DSD+ at 4800 baud may help.

Your dongle doesn't appear to be using the correct PPM value or the dongle hasn't warmed up yet? The DSD+ audio waveform window shows a positive DC bias in the demodulated audio - adjust the PPM (with 'c' and 'C') until the audio is centered around the zero line during a transmission. Let the dongle run (e.g. feed I/Q data to FMP or SDR#) for about five minutes before attempting to determine the correct PPM value. When you're off-frequency, the SDR's (e.g. FMP's) bandpass filter will chop off part of the signal, leading to distortions in the demodulated audio.

And finally, the audio waveform doesn't look that good - I'd guess you're feeding far too much RF power into the dongle. Looks like you have the RF gain maxed out at +50 dB; use 'g' and 'G' to adjust it. I would try dropping it to 0 dB when using an HT only a feet feet from the dongle. The ragged look may also be due to the aforementioned tuning error.

For reference, I'm attaching a simplex DMR waveform grab (right-clicking in the DSD+ audio window freezes it so you can inspect the audio). It was a nice clear PNG image before this site turned it into an artifact-laden JPEG - oh well. Anyhoo, clean signal reception will have the audio centered around the zero line and you should be able to tell it's a four level signal. I've highlighted a few points where the signal hits the different levels. Also, you can see how the signal's envelope fits well inside the +/-25% deviation lines - it stays flat with no wild excursions above or below those boundaries (the no-signal static on the left doesn't count, obviously)

Anyway, just some things to keep in mind.

I don't use FMP (haven't gotten around to it since it can't do P25 trunking and may never handle it) but iirc FMP outputs a signal to DSD+ using TCP/IP - I think it's possible to use the virtual audio cable but I'm pretty sure the design for both FMP and DSD+ working together is FMP outputs a stream of data over a TCP/IP port for DSD+ to pick up on and work with. I'm not saying it can't or won't work at all with the virtual audio cable, just the DSD+ developers found a way around the necessity of using such utilities by using the TCP/IP aspects.

I'm sure someone else that has more experience using FMP and DSD+ together for DMR monitoring will chime in on this and hopefully be able to get things working.

The screenshot shows that he's getting audio through VB Cable ok, but I agree, I would use the cleaner TCP linking. Just specify -o<port#> on the FMP command line and -i<port#> on the DSD+ command line, then deal with the Windows Firewall popup. A typical port number for use with FMP/DSD+ seems to be 20001, so FMP -o20001 and DSDPlus -i20001.

I wonder if the message about "no frequency data files found" is why it isn't opening that window.

Granted, it's a long shot, but it's about as good as anything else at this point.

I also wonder if there should be quotes around the primary and secondary frequency lists.

No and no.

 

[UPMan]

Woo Hoo! Thanks, slicer, I can decode myself, now.

 

[slicerwizard]

What fixed it?

 

[UPMan]

Probably PPM adjustment. Other adjustments were minor. Didn't switch from virtual cable, yet, since the audio was getting into DSD.

 

[garys]

The PPM adjustment drove me crazy for much longer than I want to admit. I finally got it a few days ago, but not without help.

I found that moving to TCP from virtual cable was really simple as well. It also seems improve signal processing.

Probably PPM adjustment. Other adjustments were minor. Didn't switch from virtual cable, yet, since the audio was getting into DSD.

 

[Voyager]

Yes, frequency accuracy (or lack of it) destroys DMR - even on the Motorola radios.