Hi all, I have been lurking for a while, and something salem said caught my eye.
"It only serves to reiterate the need for properly trained and
equipped service shops to support this type of equipment on
the infrastructure side. If the shop doesn't have a secure capable
monitor and doesn't know what an "eye" is, this could be a very
good indicator of where some of the poor audio problems are
coming from."
I have been taking care of a single site trunked system for about 6 years now, and one of the thorns in my side is the DES-XL radios on my system. I am one of those guys that don't know what a "eye" is. I do beleave that I have a DES-XL capable service monitor. The alignments that I regulary peform on my Quantar repeaters, Reference osc, Power out, Devaition, Reference Mod Comp. We only use secure between portable's, no modems are involved.
Am I missing anything?
DES-XL Sounds like &^%$$#
Moderator: Queue Moderator
The "eye" Salem was referring to is the lissajous circle you attempt to get on an oscilloscope when you are aligning DVP modems to work on a specific phone line. Since you don't have encrypted audio going to a dispatch center, it's not a concern to you.
I haven't done any Securenet work through a Quantar, but I do know the MSF5000's had to be "secure capable" to pass the encryption properly. They were set to pass 5Hz to 6KHz, as opposed to the usual 300-3000Hz, and re-clocked the data, among other things. I think all Quantars are secure capable, but am not positive on that.
You think your service monitor is DES-XL capable? Are you able to keyload it and listen to encrypted communications with it? If not, then it's not. It's an option we have on both an R2001D and R2670 in our shop.
Todd
I haven't done any Securenet work through a Quantar, but I do know the MSF5000's had to be "secure capable" to pass the encryption properly. They were set to pass 5Hz to 6KHz, as opposed to the usual 300-3000Hz, and re-clocked the data, among other things. I think all Quantars are secure capable, but am not positive on that.
You think your service monitor is DES-XL capable? Are you able to keyload it and listen to encrypted communications with it? If not, then it's not. It's an option we have on both an R2001D and R2670 in our shop.
Todd
-
RadioSouth
- Batboard $upporter
- Posts: 2884
- Joined: Mon Sep 03, 2001 4:00 pm
Thanks
Thanks for the sound file! After hearing the MP3 file, my DES-XL sounds about the same. Thanks alot, The Astro sound is impressive. Question? Do you notice any loss in coverage between Astro and Analog?
DES-XL
The "XL" uses some of the bits that were used in the plain DES, or DVP
for error correction.
The result of this is that the XL encryption is not quite as clear as the plain DES, but will perform with signals so noisy that the "clear" signals are unusable.
On the other hand plain DES requires a much better than 20 DBq signal, or they get " audio holes" until the decoder can re-sync.
The audio is better than a Nextel in unit to unit transmission.
I think that the most critical adjustment is the transmitter deviation.
Since you have apparently "transparent" stations there is not much else that you can do.
On a transparent station, the station detects the 12kilobit data stream, and switches the station to digital mode. In this mode the data is recovered, re-clocked, and repeated. If this was a wire line station, the station is keyed with the normal tones, and it switches to digital if a 12KB
signal is detected on the wire line.
One last thing, use Wide filter on the service monitor receiver, or your deviation measurement will be inaccurate. You do not need the DVP test set, or an analyzer with the secure option to make this test.
I hope that this helps. Aloha Bernie
for error correction.
The result of this is that the XL encryption is not quite as clear as the plain DES, but will perform with signals so noisy that the "clear" signals are unusable.
On the other hand plain DES requires a much better than 20 DBq signal, or they get " audio holes" until the decoder can re-sync.
The audio is better than a Nextel in unit to unit transmission.
I think that the most critical adjustment is the transmitter deviation.
Since you have apparently "transparent" stations there is not much else that you can do.
On a transparent station, the station detects the 12kilobit data stream, and switches the station to digital mode. In this mode the data is recovered, re-clocked, and repeated. If this was a wire line station, the station is keyed with the normal tones, and it switches to digital if a 12KB
signal is detected on the wire line.
One last thing, use Wide filter on the service monitor receiver, or your deviation measurement will be inaccurate. You do not need the DVP test set, or an analyzer with the secure option to make this test.
I hope that this helps. Aloha Bernie
-
Josh
- Posts: 1931
- Joined: Mon Sep 03, 2001 4:00 pm
- What radios do you own?: APX4K, XTL5K, NX5200, NX700HK
"The audio is better than a Nextel in unit to unit transmission. "
Yes, although scratchy, even the DES-XL sounded just like regular analog to me, but with scratchiness in the background.
The digital Astro audio, while distorted, was still more understandable than Nextel. You get crappy audio when you compress audio on a 12.5k channel.
-Josh
Yes, although scratchy, even the DES-XL sounded just like regular analog to me, but with scratchiness in the background.
The digital Astro audio, while distorted, was still more understandable than Nextel. You get crappy audio when you compress audio on a 12.5k channel.
-Josh
-
Jonathan KC8RYW
- Posts: 1747
- Joined: Fri Dec 28, 2001 4:00 pm
I hope I'm not the only one.. but every time I try to download http://www.r0f.com/~swilling/astrosamps.mp3 it doesn't work. It can't resolve the host name.
73 DE KC8RYW
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
-
Josh
- Posts: 1931
- Joined: Mon Sep 03, 2001 4:00 pm
- What radios do you own?: APX4K, XTL5K, NX5200, NX700HK
It worked just fine for me!!KC8RYW wrote:I hope I'm not the only one.. but every time I try to download http://www.r0f.com/~swilling/astrosamps.mp3 it doesn't work. It can't resolve the host name.
DES alignment
I have not worked on very many Quantars, but the MSF is set up using the RSS and on the alignment screen the station is keyed with a 4KHZ
Dev. square wave. ( I assume that this is a "wideband" system. if not set to the appropriate spec)
The help screen on the RSS should give you the instructions.
I hope that the Quantar RSS is the same.
Since this is a square wave, there are many additional sidebands, so you must use the "wide" filter in the monitor receiver, or you will get distorted mesurements.
I have only used the Motorola R2001 series analyzer for these measurements in the last 20 years or so, but In the very early days of DVP I used a Cushman CE-4 with satisfactory results.
I suspect that the Mod compensation is also critical, but it is also very fussy on the control channel. The cat's eye and various secure test sets are necessary for alignment of complex systems, such as those with consoles, multi-site, etc. You do not need a monitor with trunking capabilities for trunking site work either.
Another question: How do these radios sound on simplex?
Aloha, Bernie
Dev. square wave. ( I assume that this is a "wideband" system. if not set to the appropriate spec)
The help screen on the RSS should give you the instructions.
I hope that the Quantar RSS is the same.
Since this is a square wave, there are many additional sidebands, so you must use the "wide" filter in the monitor receiver, or you will get distorted mesurements.
I have only used the Motorola R2001 series analyzer for these measurements in the last 20 years or so, but In the very early days of DVP I used a Cushman CE-4 with satisfactory results.
I suspect that the Mod compensation is also critical, but it is also very fussy on the control channel. The cat's eye and various secure test sets are necessary for alignment of complex systems, such as those with consoles, multi-site, etc. You do not need a monitor with trunking capabilities for trunking site work either.
Another question: How do these radios sound on simplex?
Aloha, Bernie
-
Jonathan KC8RYW
- Posts: 1747
- Joined: Fri Dec 28, 2001 4:00 pm
I applaud you, Shaun, for a noble attempt to show the audio quality differences of analog/digital/encryption.
The microphone near the speaker technique is primitive, and contributes a lot of room noise and reverberation to the recording. Perhaps you could take the audio from the speaker mic accessory jack, and isolate it with a Radio Shack transformer to the sound card input. The goal is not to over-drive the line input on the computer, and reduce the amount of ground loop problems that might occur.
Clipping seemed to be a problem on the recording. For most sources, 6 db worth of headroom should be okay. Let us say that the absolute clip level is 0db. Therefore, the audio would normally peak around –6db, as not to go into clipping. I use CoolEdit96, since it provides a VU meter during record and playback.
MP3 makes it tough to hear the effects of the encoding, when MP3 itself does lossy encoding! Try using Linear PCM audio (aka: wave files.) Mono, 16 bit, 8kHz sampling should be fine for 2-way radio recording. 11kHz will provide better coverage of high frequency response, albeit probably audio harmonics. 8 bit audio just sounds awful, and from my experience, adds hiss to otherwise good source audio.
I do this stuff a lot (well, maybe not from an Astro Saber source, though) so it sort of second nature to me. Some of this applies to Shaun’s recording; some is just for others on the board to take note of when they do their own radio recording session.
Peace,
The microphone near the speaker technique is primitive, and contributes a lot of room noise and reverberation to the recording. Perhaps you could take the audio from the speaker mic accessory jack, and isolate it with a Radio Shack transformer to the sound card input. The goal is not to over-drive the line input on the computer, and reduce the amount of ground loop problems that might occur.
Clipping seemed to be a problem on the recording. For most sources, 6 db worth of headroom should be okay. Let us say that the absolute clip level is 0db. Therefore, the audio would normally peak around –6db, as not to go into clipping. I use CoolEdit96, since it provides a VU meter during record and playback.
MP3 makes it tough to hear the effects of the encoding, when MP3 itself does lossy encoding! Try using Linear PCM audio (aka: wave files.) Mono, 16 bit, 8kHz sampling should be fine for 2-way radio recording. 11kHz will provide better coverage of high frequency response, albeit probably audio harmonics. 8 bit audio just sounds awful, and from my experience, adds hiss to otherwise good source audio.
I do this stuff a lot (well, maybe not from an Astro Saber source, though) so it sort of second nature to me. Some of this applies to Shaun’s recording; some is just for others on the board to take note of when they do their own radio recording session.
Peace,
73 DE KC8RYW
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
-
Josh
- Posts: 1931
- Joined: Mon Sep 03, 2001 4:00 pm
- What radios do you own?: APX4K, XTL5K, NX5200, NX700HK
Well, if we want to talk like this, then it might not be a bad idea to try the recording again- at least the DVP/DES part. I'd like a good clean (well, whatever it can be) sound clip.KC8RYW wrote:I applaud you, Shaun, for a noble attempt to show the audio quality differences of analog/digital/encryption.
The microphone near the speaker technique is primitive, and contributes a lot of room noise and reverberation to the recording. Perhaps you could take the audio from the speaker mic accessory jack, and isolate it with a Radio Shack transformer to the sound card input. The goal is not to over-drive the line input on the computer, and reduce the amount of ground loop problems that might occur.
Clipping seemed to be a problem on the recording. For most sources, 6 db worth of headroom should be okay. Let us say that the absolute clip level is 0db. Therefore, the audio would normally peak around –6db, as not to go into clipping. I use CoolEdit96, since it provides a VU meter during record and playback.
MP3 makes it tough to hear the effects of the encoding, when MP3 itself does lossy encoding! Try using Linear PCM audio (aka: wave files.) Mono, 16 bit, 8kHz sampling should be fine for 2-way radio recording. 11kHz will provide better coverage of high frequency response, albeit probably audio harmonics. 8 bit audio just sounds awful, and from my experience, adds hiss to otherwise good source audio.
I do this stuff a lot (well, maybe not from an Astro Saber source, though) so it sort of second nature to me. Some of this applies to Shaun’s recording; some is just for others on the board to take note of when they do their own radio recording session.
Peace,
I don't want to look a gift horse in the mouth here, though.
-Josh
-
Jonathan KC8RYW
- Posts: 1747
- Joined: Fri Dec 28, 2001 4:00 pm
1. Take one speaker/mic. Cut off the speaker/mic head. Remove the speaker mic cable jacket. Notice wires. One pair of those wires carries speaker-level audio. .... You will need to patch it into the sound card line in. The sound card line in will probably be wanting about 1 volt RMS of audio, at max. Expierment with resisters to reduce the level. If you get a low frequency Hummmmmm (ground loop) then use the audio isolation transformer that Radio Shack sells.r0f wrote:2 things.
First, someone mail me an audio adapter then. I don't have one, and Motorola doesn't sell one as far as I know. It's IMPOSSIBLE to take a direct input off of the AS3's I have unless you can interface something to that horrible speak-mic jack.
Second, MP3 does not take away from the quality. If you use abitrate, then yes it will sound bad. I encoded at 128Kbps, which is just fine, and does not take away from the clarity of the original sample. I could see maybe 32Kbps sounding like junk, but not 128Kbps. Ask anyone on here about 128Kbps audio/songs they download off the net. 128Kbps is primo. Most people these days encode at 160 or 192Kbps though..
I'm not about to upload a 10MB .wav file for you to listen to either, so sorrymp3 is used to compress .wav's, and it works just great.
Shaun
2. About MP3... it stinks. Anyone with a good pair of ears knows it, too. 128k is not good. 320K is just okay. Any lossy compression is a compromise. I know that many people use MP3 for their audio, however, that does not mean that it is right. Why compromise when it comes to hearing your favorite music? Hard drives are the cheapest they have been for years; buy a 7200RPM Seagate 80GB ATA-100 and use 44.1K, Stereo, 16 bit Linear-PCM ripped from a CD source. Best yet, you can still use Win Amp to play it, if you want.
Shaun, I agree that 10mb worth of audio over the 'net might be a little unreal.
However, the 46 second sample that you made would only take up 718Kb, at 8Khz, Mono, 16 bit PCM. That is about the same size as the MP3 you made (723Kb.)
73 DE KC8RYW
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
-
ricciticcitembo
- Posts: 730
- Joined: Tue Feb 26, 2002 4:00 pm
I agree with Todd, Thanks Shaun!
And like a few others on here use everything
from ProTools with expensive
outboard converters to HiFi Analog
recording using only class A tube
preamps(my fav). And for the purposes of the
Securenet/clear/Astro Audio comparison, the recordings are very
representative (at least what I have to compare it to ) of the differences.
After all, we are comparing encryption of a narrow band FM voice channel not a Stradivarious Concerto.
And another thing, The Astro Speaker mics are sonically welded
shut, so tapping the audio is not healthy for it. Would need an AVA, or service cable.
I built my own SVA (because it was DOA) and I do in fact have it hooked up to my car stereo, and at home I plug it right into the console, and the biggest problem with the audio, is really not the audio itself, but the squelch tail when utilizing CSQ. It's freaken LOUD! So I avoid it like the plauge in the car, and at home Run a noise reduction and compressor/limiter when operating CSQ which with carefull adjustment, helps somewhat. PL/DPL works like a champ though.
And I don't know about the Astro,
but the Saber uses a "bridged" type of audio output where NONE of the speaker outputs are "ground" and both have like 6 volts DC on them. So tapping the Audio requires carefull methods, or you will need a new radio.
Again, the recordings are more than sufficient for the application.
And like a few others on here use everything
from ProTools with expensive
outboard converters to HiFi Analog
recording using only class A tube
preamps(my fav). And for the purposes of the
Securenet/clear/Astro Audio comparison, the recordings are very
representative (at least what I have to compare it to ) of the differences.
After all, we are comparing encryption of a narrow band FM voice channel not a Stradivarious Concerto.
And another thing, The Astro Speaker mics are sonically welded
shut, so tapping the audio is not healthy for it. Would need an AVA, or service cable.
I built my own SVA (because it was DOA) and I do in fact have it hooked up to my car stereo, and at home I plug it right into the console, and the biggest problem with the audio, is really not the audio itself, but the squelch tail when utilizing CSQ. It's freaken LOUD! So I avoid it like the plauge in the car, and at home Run a noise reduction and compressor/limiter when operating CSQ which with carefull adjustment, helps somewhat. PL/DPL works like a champ though.
And I don't know about the Astro,
but the Saber uses a "bridged" type of audio output where NONE of the speaker outputs are "ground" and both have like 6 volts DC on them. So tapping the Audio requires carefull methods, or you will need a new radio.
Again, the recordings are more than sufficient for the application.
-
Josh
- Posts: 1931
- Joined: Mon Sep 03, 2001 4:00 pm
- What radios do you own?: APX4K, XTL5K, NX5200, NX700HK
ricciticcitembo wrote:I agree with Todd, Thanks Shaun!
Again, the recordings are more than sufficient for the application.
I like it, too. The analog voice sounded realistic and clear- just like the .wav file I found from some government tests of Digital radios (XTS3000). Everything else was VERY nice. I really just wanted to hear what DVP/DES/ whatver, sounded like. I got it. I am happy.
Thanks again.
-Josh
-
Jonathan KC8RYW
- Posts: 1747
- Joined: Fri Dec 28, 2001 4:00 pm
My goal wasn't to put anyone down. 
It was simply to give pointers, along with my advice. I want everyone to suceed, and do the best at everything they in life. And, since digital audio is sort of my domain, I figured I could help out so that we could progress to the best.
Being super-critical of others work is something that I've slowly become used to, after a few years of media production and study.
I'm tired now, and need to retire for the day. I just want everyone to know that my intention was not to make Shaun feel bad or prevent future audio postings. I must say that I am sorry Shaun.
It was simply to give pointers, along with my advice. I want everyone to suceed, and do the best at everything they in life. And, since digital audio is sort of my domain, I figured I could help out so that we could progress to the best.
Being super-critical of others work is something that I've slowly become used to, after a few years of media production and study.
I'm tired now, and need to retire for the day. I just want everyone to know that my intention was not to make Shaun feel bad or prevent future audio postings. I must say that I am sorry Shaun.
73 DE KC8RYW
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
Random Motorola Part Number:
SYN1894B - V3m Sprint-branded Battery Cover
-
vks737adelaide
- Posts: 12
- Joined: Sat Mar 23, 2002 4:00 pm
KC8RYW wrote:1. Take one speaker/mic. Cut off the speaker/mic head. Remove the speaker mic cable jacket. Notice wires. One pair of those wires carries speaker-level audio. .... You will need to patch it into the sound card line in. The sound card line in will probably be wanting about 1 volt RMS of audio, at max. Expierment with resisters to reduce the level. If you get a low frequency Hummmmmm (ground loop) then use the audio isolation transformer that Radio Shack sells.r0f wrote:2 things.
Shaun
Send him the speaker/mic, I'm sure he'll re-do the DEMO.
2. About MP3... it stinks. Anyone with a good pair of ears knows it, too. 128k is not good. 320K is just okay. Any lossy compression is a compromise. I know that many people use MP3 for their audio, however, that does not mean that it is right. Why compromise when it comes to hearing your favorite music? Hard drives are the cheapest they have been for years; buy a 7200RPM Seagate 80GB ATA-100 and use 44.1K, Stereo, 16 bit Linear-PCM ripped from a CD source. Best yet, you can still use Win Amp to play it, if you want.
We're not talking about music. The issue at hand is a source with an INTENDED high end (but theres more to the story, see below) of MAYBE 3.5 khz. I defy YOU to tell the difference 128k, and anything else - including raw 44.1 PCM - with THIS source material.
3. Shaun, I agree that 10mb worth of audio over the 'net might be a little unreal. However, the 46 second sample that you made would only take up 718Kb, at 8Khz, Mono, 16 bit PCM. That is about the same size as the MP3 you made (723Kb.)
Actually, it would be 718.75Kb, since we're splitting hairs here, but you would be sampling at a slower effective rate - and it is slower because you're eliminating the compression - eliminating some of the details that were displayed in the demo. Things like minimizing the cracklies in the analog encrypted signal. Try it for yourself: load the file into Cool Edit 2000, and re-sample it to 8K, 16bit. He did a great job showing what the user hears coming out of the radio speaker - which requires the higher sampling rate to capture the nuances of the audio - speaker created harmonics, etc. This demo actually answered several user's questions as to whether their systems were acting normally. Taking the audio from the demod/decoder doesn't properly show what the user hears. I believe it's now your turn to use your radio to show what a better job you could have done.
-
jjmcrowell
- Posts: 34
- Joined: Wed Nov 21, 2001 4:00 pm
I agree, r0f, it does sound just fine.
I think certain people on the board need to just take a step back and be happy with what people have provided them, instead of making ridiculous overly-analytical comments about nearly EVERY single post (you know who you are). Sorry, I just keep seeing this recurring and can no longer bite my tongue. You did fine, keep up the good work, eh!
-119
I think certain people on the board need to just take a step back and be happy with what people have provided them, instead of making ridiculous overly-analytical comments about nearly EVERY single post (you know who you are). Sorry, I just keep seeing this recurring and can no longer bite my tongue. You did fine, keep up the good work, eh!
-119