<img height="1" width="1" alt="" style="display:none" src="https://www.facebook.com/tr?id=311512539006924&amp;ev=NoScript">
  • Telos Alliance
  • Telos
  • Omnia
  • Axia
  • Linear Acoustics
  • twntyfive-seven
  • Minnetonka Audio

Blog Central

NRSC, Let's Do This Less Often

Posted by Kirk Harnack [TWiRT] on Jan 9, 2015 3:13:00 PM

Find me on:

TWiRT 241Every year AM stations in the US must measure their transmission for spectral purity. Many AM stations say they can’t afford this. Mark Persons presents a few ideas on NRSC measurements. Plus he and Chris Tobin discuss sleuthing some really difficult interference problems.

 

 

 

Watch the Video!

 

Read the Transcript!

Kirk: This Week in Radio Tech Episode 241 is brought to you by Axia Audio and the new Fusion AoIP Mixing Console, packed with features and capabilities refined from over a decade's worth of IP audio experience; by the Telos ProSTREAM X/2 and 9X/2 audio processing and stream-encoding software with adaptive streaming technology; and by Lawo and the new crystalCLEAR Virtual Radio Console. crystalCLEAR is the radio console with a multi-touch touchscreen interface.

Hey, every year AM stations in the US must measure their transmission for spectral purity. Many AM stations say they can't afford this. Mark Persons presents a few ideas on NRFC measurements, plus he and Chris Tobin discuss sleuthing some really difficult interference problems.

Hey, welcome in. It's time for This Week in Radio Tech, our first show of 2015. Glad you're here. Really glad you're here. We've had a lot of fun doing these shows. It's our 241st episode. I'm Kirk Harnack, your host for the show. It's the show where we talk about radio broadcasting, audio quality, RF transmitters, everything from the microphone to the light bulb at the top of the tower and sometimes even what's in the consumer or listener's home for picking us up.

So that's what the show is all about, and we've got a great show for you today because we've got a guest that's just out of this world fantastic. We'll introduce him in just a minute. First of all, let's bring in the best-dressed engineer in radio from his office somewhere in Manhattan, his lair of IP codicology. It's Chris Tobin. Hey Chris, how are you doing?

Chris: I'm well, Kirk. And yourself?

Kirk: Fabulous. Fabulous.

Chris: Excellent. Excellent. Yes, I'm looking forward to the show. It's going to be fun. Mark is going to be a fun guest, and as we've had in the past, it'll be good conversation.

Kirk: Your title on the lower third says IP Solutionist. Give us the elevator speech as to what that's about.

Chris: Well, a lot of people are trying to produce content, radio shows, TV shows, and they're trying to figure out ways to get it from point A to point B and the traditional methods are somewhat expensive and fading away. So I hope people do it with the technology similar to what we're using right now, which is Skype. Technologies made by companies like yours at Telos, Omnia, and Axia; companies like Tieline, Comrex and elsewhere; and I hope to bring the two together. So when you're producing a show and you're trying to get it to a network for distribution, I can help you do it over the Internet or private networks.

Kirk: I should point out that people should not judge IP's ability to handle live stuff by how many times they have had to push the print button in the past.

Chris: That's correct.

Kirk: That's not a measure of how well IP can work, and Skype itself is not a good measure of how IP can work. A lot of things in Skype are totally beyond the control of the user, and oftentimes Skype is being used over connections that are let's say questionable so it does the best that it can. So yeah. And we've talked on a number of shows about what IP can do for you, and I've always been excited about this as a technology. I'm glad it's gotten to the point, Chris, where consultants are needed and you are one of them.

Chris: Thank you. Yes, I actually spoke to someone this week who is producing a daily morning show, distributes it on a radio network on the regional size, and is having problems with the connection. So I asked him "What are you using?" He's using Skype. I said, "Okay, and you expect what out of that?" "It should work. Why shouldn't it work?" I said, "How much did you pay for it?" He goes "Nothing." "You get what you pay for."

I said, "If you'd like, you can spend a couple of bucks and I can guarantee you nice connectivity for your morning show that gets distributed on 75 radio stations. Are you making money with this show?" He goes "Yes." "If you want to continue making it, you want to invest a little." So Skype does work, but it has that place. There's a place and time for it, and there's a place and time it shouldn't be used.

Kirk: Yep, absolutely. Hey, and our show today, also on our show, is our guest, and that is Mark Persons. Mark has been on the show before and he's back and he's got more stuff to talk about that's going to make your socks roll and down. Hey Mark, welcome in. Glad you're here.

Mark: Well, thank you, and gosh, I'm glad that you invited me back. It's always fun being on the show with you. It's just something I've been looking forward to. So let's get on with it.

Kirk: Well, we're going to start the show out with an ad to pay for this part of the show, and I'm delighted that we have three sponsors on This Week in Radio Tech. You know, for a long time we went without any sponsors and then my friends at Telos picked up a sponsorship and now we have a couple more sponsorships as well so that is just delightful. First of all, I'd like to tell you about the folks at Axia and what they are doing for broadcasters in terms of IP audio.

You know, for years you've heard me talk about Axia and Livewire and Axia IP Audio Consoles and routing systems, and Axia has a new website. Well that's not in itself big news, but it does help you navigate the world just a bit. If you still go to axiaaudio.com, you'll get to the new telosalliance.com website. Click on Axia, and bam you're there.

On the Axia website you can get a quick look at all of the networked radio consoles that Axia's offering, starting with the Fusion AoIP Mixing Console. This is a brand new console for Axia. In fact, I'll be doing a dealer webinar for the Axia dealers in a couple of weeks and then a public webinar about a month after that. So in February watch out for a webinar about the Fusion Mixing Console.

Then there are the workhorses, the Element AoIP Mixing Console; the IQ; the Radius which is the console that we're doing the show through right now. There's a Radius Console at the GFQ Network. Plus there's the desk compact console and the rack compact console. I have one of these rack consoles myself at our stations in American Samoa in the news room. There's the power station integrated console engine; the studio engine mixing engine. It's a standalone. It's a componentized mixing engine.

And then there are also the Core 32 and the Core 16 mixing engines. These, by the way, have Ethernet switches that are built into them that you don't have to buy an Ethernet switch for each studio. You just plug the things that you need into that switch that's built in.

And then if you go on down this page, that's why I'm pointing out this page. Yes, I know you can walk through this but I really want you to know about it. The Axia xNodes, this is the new line of audio input/output devices for Axia. They come in analog, AES/EBU, and also the combo version. And coming out for you folks in TV there's an SDI version that will be available very soon. It was introduced a year ago at NAB and it's just about to start shipping.

There's also the xSwitch. You know if you don't want to buy a Cisco switch and configure it yourself, get an Axia xSwitch. It probably costs less than the Cisco switch does, and it's already configured for Axia. You just plug stuff right into it. Plus of course it's fully compatible with the other things that you do on an IP network.

So that's a great page to start with. Click on any of the things that you're interested in, maybe the new Fusion Console, or maybe you need a smaller console like the Radius or the IQ. I've got a couple of Radius consoles at my stations in American Samoa as well as the rack console. And if you've been paying attention, you know that we have a big Element Console at our stations in Greenville, Mississippi. There's over 5,000 Axia consoles on the air or in production right now. Stuff works, and it's all AoIP and very flexible. So check it out if you would.

There's also a bit of a word about AES67 on this website, so if you're interested in making things talk to each other, you can find out the state of AES67 at Axia. Thanks a lot to Axia Audio for sponsoring This Week in Radio Tech.

All right. Hey, Chris Tobin is going to handle most of the interview duties today. Mark Persons, I know that you've written a few articles here lately. Some are on your own website and some have been in Radio World or maybe both. I understand that you've had close to 160 articles in Radio World in your career, and that is fantastic. Just amazing. A wealth of information. So you and Chris kind of kick it off with whichever subject you think you want to start with, Mark.

Mark: Well thanks for that. Actually not all of them appeared in Radio World, but gosh, almost everything recently has appeared in Radio World. I enjoy doing it. It's like having a conversation. I write the way I talk. It's "This is what happened. This is what I do. This is how I do it." And I just try to be open about that. Then on my website, I also put on articles that talk about, "Oh, I ran into this kind of a problem on an FM exciter or RF amplifier." And I put the solutions on there. That's mwpersons.com. Mwpersons.com.

And then a lot of people read that, comment on it, then they say, "Oh, I see you did this, but I can't quite bring myself to do it myself. I'm a little weak in the knees about that." So they send the equipment to me anyway and I repair it then send it back and everyone's happy. It's absolutely amazing how that has worked out.

Let's see, we got an FM exciter in from the CBC in Canada and it was a Harris THE-1. Well, gosh, Harris doesn't support that anymore because they don't make that one and this one required a fairly substantial upgrade in its output section. I don't know if you're aware of this, but the output transistors for that exciter aren't available anymore. And so I worked up a solution for that. Not transistors, but a whole RF palette that went in there. And boom, the exciter is working again. I send it back to the CBC. They were happy.

We got one of those same exciters in, get this, from Hong Kong, a radio station there. They were having trouble repairing it. So gosh, they sent it halfway around the world. It was a $300 repair, $600 to ship it. I mean, I don't know where the expertise is nowadays, but it must be just in a few shops in the US here because that's kind of the way it came down.

What else is going on? Oh, I should mention something. Last episode, that was December 23rd, you were talking with Shane Tovin, a very nice young man. I've known him for his entire broadcasting career. He started out in Minnesota, by the way. But he and you, Kirk, mentioned the pink envelope. Well, what was that all about?

Well, if you haven't been in broadcasting for more than 20 years, you don't know what that is. The pink envelope was a set of passwords, so to speak, that went on the teletype. Get this. The teletype, you know? The one with the roll of paper and it hammered away automatically like a typewriter. Have you ever heard of a typewriter? Anyway, when an EBS alarm came in, you were to use that password and it came in a pink envelope. And so that's where all this happened from the last time. So now we know what pink envelopes are.

Okay, where were we? Okay, the reason why you asked me to be on the show was because I had written an article and it had to do about proposing changes to the FCC rules and those changes involve let's not measure annually NRSC, the occupied bandwidth, and RF harmonics of AM radio stations. Well why is that? Well if you're like me you know that most AM radio stations don't have an extra penny to spend and I have for years, since 1990 or so, or was it '92, when all these stations had to get measurements annually, okay? Not FMs, just the AMs, to prove that their bandwidth was right and they weren't putting out excessive RF harmonics.

To be truthful, the first year that I went around and started doing this, I found about 10% of the stations out of compliance. Most of them had problems with second or third harmonic radiation from the transmitter, so if the station was at 1500 on the dial then they appeared at 3000 and 4500 on the short wave dial. Well you can fix those things. And oh, I've got some here. Here we go. You remember these? Mica capacitors for AMs and coils?

Kirk: Yeah.

Mark: Well you put them together and you set them up for... you calculate the amount of inductance and capacitance so that it becomes equal capacitive reactance with equal inductive reactance and it winds up with in essence shorting that frequency to ground but letting other frequencies through with a little bit of extra inductance.

And then you say, "Well, what's this funny thing on top of the coil here?" Okay, it's a short. "Well what's that for?" Well, the coil was made longer than it needed to be and then I put this shorting part on here and experimentally determined the exact location of where it needed to be to be on frequency. And then this trap, it's called a trap, would go from ground to probably a 50-ohm point in the system like the output terminals of the transmitter or something like that. There are lots of choices.

Anyway, so those issues pretty much went away the first year but broadcasters still were required to measure occupied bandwidth as well. Well, okay, but here's my experience: most of those transmitters out there are solid-state and it's true on rare occasions solid-state transmitters will get into trouble and radiate some spurs, okay?

But here is my experience: a solid-state transmitter might have gone in 20 years ago, okay? So here I am out in the field with a spectrum analyzer making measurements on their transmitter for the 20th time, and for the 20th time it passed the test, and for the 20th time I've had to send the broadcaster a bill for $350 or something like that.

Now it may have made me a little bit of money along the way, but quite frankly I was feeling bad for these guys. And then on rare occasions things go wrong, but usually when those things go wrong the broadcaster knows about it anyway because something else is up. His audio has gone to heck or something like that. Okay, so they replace something in the transmitter, and bang, you're working again.

So I wrote this article saying we don't really need to do this. But guess what? I got hit hard by everybody in the business who makes spectrum analyzer measurements and makes money by it, okay? So guess what? They didn't want to see their income go away, so therefore I was the bad guy. That's how it came down.

Chris: Well of course. You got rid of the cottage industry. That's how things work. But unfortunately it is in business and in general things evolve. Things move forward. There's change, and you have to sort of understand and adapt it and go with it, otherwise if you try to hold it in place so that you can benefit from it eventually you lose out. You were just kind enough to point out to people and say, "Hey, we don't need to be doing it this way. There's another approach."

So the same people that came after you could've probably looked at that as an opportunity and said, "Huh, maybe now it's time to go to those broadcasters and educate them on what they have and don't have."

Mark: And the other part of the whole thing is that gosh, if you do it for an AM, shouldn't you be doing it for an FM too?

Chris: Yes.

Mark: It should be both or none, either one. And I have some examples on my website where FMs get in trouble for a number of reasons. The one that comes to mind is Harris MX-15 Exciter. In its power amplifier module there are electrolytic capacitors that will dry out. This is an FM now, don't forget. So when this happens, the exciter will probably put out spurs, additional signals, 200 or more kilohertz displaced from the wanted signal, from the station's frequency. So the station gets heard three places on the dial. Maybe it's 400 kilohertz. And usually those things can pass right through the transmitter, so you have no problems. It goes right on the air.

Now any manager would say, "Well, this is actually a good thing being heard on more spots on the dial," but the truth is the FCC doesn't like that at all. But it can happen, and the answer is replace some capacitors in the RF amplifier and you're fine again. So that's on my website. What else can go wrong? Oh, you have a collocated AM and FM transmitter.

Well, the AM transmitter, let's make it simple. It's at 1000 on the dial. That's one megahertz. It's very possible that if your station on 98.1, your FM station, is on 98.1, you might find two signals, one at 97.1 and one at 99.1. Gosh, again, you're heard at three spots of the dial. Now what's the usual cause for this? Chris, do you run into this? I have.

Chris: I have not personally, but I do know folks who have. I've worked with a couple of engineering firms that have told me the stories, but I'll let you continue on.

Mark: Oh, okay. All right. Well what usually happens is that you plug the composite audio in the back of the exciter and this exciter happens to have a balanced input, very nice for keeping audio clean. But the balanced input, which means that the shield on the BNC-to-BNC capable between the two, is actually not a ground potential.

Chris: No, it's floating. It's floating.

Mark: So that's where the one megahertz get introduced into the exciter, and the answer is put a strap, a braided strap probably, from that BNC input around the connector to ground at the exciter. Now you'd say, "Well gosh, I could have a ground-loop problem." Well, exciters really do belong, in my opinion, in the rack where the STL is or where the processing is. And the further that you move those two devices apart, the more chance you have of actually getting into trouble with distance and with hum and all of that kind of thing.

So my plan is whenever I've installed FM exciters and transmitters, or at transmitter sites, I put the exciter in a rack. I put it there with the STL; I put it there with the processing, whichever it happens to be, short jumper-cables between the two. Make sure the exciter is bonded to the rack ground which is connected to the transmitter ground. Make sure the processing or STL is bonded to that same ground. You don't have signal-to-noise problems, and you don't normally have those problems of the AM and the FM mixing together.

Chris: And that's a problem I've actually had at collocated sites, not only for the FM and AM, but just for audio devices in the facility where things are not properly balanced where you can get common-mode rejection taking place properly functioning. So I've seen that happen a lot.

Kirk: Mark, I'm so glad you explained that. I had the same thing happen at a colo AM/FM in a small town. And I found the problem quite just by experimenting. I had my spectrum analyzer handy, watching it, while I tried grounding various places.

I knew that it was the AM getting in through the composite cable, but I didn't know for sure the best way to try to handle it, so I just tried grounding various different things with a braid until I found "A-ha, if I ground this close to the composite input of the exciter, the outer conductor of that, to the station ground then it got better and better." I was able to get it to the point where it was below the FCC-required however low that was. I was measuring it, and it got just below that.

And in this case, I forget for whatever reason, but the exciter really needed to stay in the transmitter where it was. For whatever reason, and I agree with you, I like to bring that exciter out of the transmitter, put it near the audio processing or the rack, and then make my lengthy cable the RF cable from the exciter's output. For whatever reason that wasn't an option that day so I had to find the root cause. There you go. Thanks for explaining it.

Mark: Oh, and BE transmitters, there's plenty of rack space there so sometimes an STL winds up in that rack space in the BE transmitter and then again the cable between the two is short. So that works out.

Chris: Now Mark, you had mentioned earlier at your shop you document the work you do and you have the notes up on the website. I'm looking at the website right now, mwpersons.com tech tips. I'm looking. Warm, hard feelings, I have used MX and MS-50 exciters over the years and have had to fix them. It's always been an interesting task when you're working on those boxes. And then using the MSP-90 audio processors they were designed to work with. That whole package was fun to play with. I'm sure you've had those.

Mark: Yes, but that MSP-90 audio processing from Harris really wasn't such a good choice.

Chris: No.

Mark: I mean the Orban products had it beat hands-down. You might remember that.

Chris: Yes. I worked at a facility that was a complete turnkey and that's why we had them, but I did have fun one night on an overnight doing maintenance and putting an Optimod 8000 in its place just for fun and the FM program director wanted to know what we did on the overnight because it sounded really nice. [Laughs] I was like oh, I don't have the heart to tell him.

Mark: And it was probably an Orban 8000 which is the lowest of the line.

Chris: Yes it was.

Mark: And boy, have they changed. Has that become something different now. Or whatever it turns out to be. There are wonderful audio processors out today that beat the heck out of the MSP-90 that Harris in my opinion never should've made in the first place.

Chris: Oh, I'm sure we're not alone in that thought. I know several people have had some interesting comments. But the reason I mentioning just the website and the MX-15 and what you were talking about is your documentation of the work you've done and the sort. I'm curious, because I've done the same.

As a matter of fact, today I was working on a piece of equipment and I was sitting down taking notes and documenting everything I was doing, all the measurements and testing I was doing with it. Have you found that even though it can be tedious documenting things nowadays when you have touchscreen-type tablets and paper is something of a bygone era for some folks, do you find that in the 11th hour, the late night, thank goodness you did some documenting on a project on a particular repair five years ago that is now in the file cabinet or may be electronically transferred to the tablet and you're able to refer to it in short in the time of your trouble call or emergency calls?

Mark: What I have is a fairly complete set of instruction manuals for equipment, especially the stuff I work on, the exciters or whatever.

Chris: Sure.

Mark: And in pencil I make changes because I might be back there a year later and discover even a better solution to the problem, so I can use the eraser on the pencil. So that really works well. And I enjoy putting the information to the Internet. It's just a hobby of mine, and it makes me feel good to do that, to share the information with the world, not to keep it to myself.

Chris: Oh. Oh, so you're not worried about job security? Got it, okay. Moving right along.

Mark: No.

Chris: [Laughs] I have worked in many a place where documentation didn't exist and I was told in very indirect ways that was a form of security for the previous persons. I said, "Well, if that was security, why am I in their place? Just curious."

Kirk: Yeah.

Mark: Oh, and I also will ask the clients to send their books so that when I make a change in their exciter I can pencil the change in their books too.

Chris: Yes, when I have worked at a customer's site, on many occasions what I'll do is I look at the site and see what condition it's in. If I can tell they're not really taking good care because they have many contractors going through, what I'll do usually is take a small little plastic dog, put my changes and documents in that bag, and place it inside the device if it's not an issue. This way when the next person comes along and has to work on it, they pop it open and go "Oh, look at this." At least they have an idea of who was here last and what was going on.

Mark: Well, as you get older, you need those hints to keep reminding you what you did last time. So I do this as self-defense, actually.

Chris: Oh that's fine. No, no, I appreciate it because I can tell you I've gone to - as a matter of fact, two weeks ago I was working in a low-power TV transmitter room working with an engineer, this contract engineer for the station, and it's a hybrid so there were multiple antennas being combined into this system. I was curious. One of the dummy loads that was used for the rejects for any out-of-band material was somewhat warm and it shouldn't be according to the manufacturer's manual.

So I asked him. I said, "When was the last time you checked this? Is there a spectral chart somewhere? Something you may have printed out from a spectrum analyzer? Blah, blah, blah." Nothing. It was like "Well, no, we just turn this knob here and look for the number to go down on the display then we know it's back to normal." "Really? Okay. According to my measurements, you're about 40 watts below where you should be. But okay, we'll see where we go from there."

So I'm just using this as an example of where documentation, though it can be for some these days tedious, would've been very helpful in two ways: one, to shorten my time out. Rather than be there for nine hours, I would've been there for six. Yes I made more money, but the point is I don't like doing it, like you say, in that fashion.

And also it doesn't help the person who is hired as the contract guy to have to have someone else come out, spend so much time, then you're like really? Really, this shouldn't go this way. So I'm just trying to point out that documentation or documenting what you do, whether it's a transmitter, studio, or building a new system, is definitely worth the trouble or the time it takes.

Mark: And really you need to use a spectrum analyzer any time you install a new transmitter or any time you make changes in a transmitter setup. You need to have a spectrum analyzer out there to check and document what you just did.

Now here's a funny story about spectrum analyzers and documenting. Oh gosh, it must be ten year ago, a client was building a six kilowatt FM and it was going to be on one of two towers and the two towers were oh, a couple hundred feet apart. And on those two towers were also, get this, we're in Class C territory in Minnesota, 100 kilowatt facilities.

So here's the little guy, the little six kilowatt, coming in and being mixed with three 100 kilowatts right there on-site. Now I knew there were going to be mixing products, but I didn't know how bad, because you really can't know very well in advance. Okay, so I turned the transmitter on one day with a spectrum analyzer and looked at the results and saved the plot. I turned it off and said, "Okay, now I've got to look at this. How many dB do I need to bring all those mixing products, plural, down?" And so I went to the books and I discovered gosh, I need a three cavity filter.

So the client was good enough so that he didn't worry about the time. He figured we've got to do it right. We'll take whatever time is necessary. We'll order the parts necessary and put them in before we really go on the air. But here's how the telephone conversation went with the client when I called them up. I said, "I did the measurements, and you'll be happy to know there are no mixing products in the amateur radio bands. You're a lucky guy." And then he says, "Oh?" And I said, "Yeah, but as a minor problem, some of the products, two of them actually, wound up in the aviation band." I said, "Well, there's fewer pilots than there are hams anyway. It's not as big a deal."

Well, the phone went pretty silent at that point, and then I told him the bad news about how much I thought it was going to cost and then he said okay. But I thought it was pretty funny to hit him with it that way.

Kirk: And on that note, we are going to take a break. We're talking to Mark Persons about troubleshooting, repairs, documentation, and maybe asking the FCC to change a rule about NRSC measurements. I know that has a few engineers' knickers in a twist. Hey, you're watching This Week in Radio Tech, or listening to it, and our show is the 241st episode. Oh my goodness.

Our show is brought to you in part by a product from the Telos Alliance called the ProSTREAM. Now for years you may have heard about or remember the Omnia A-XE. That's software, it runs in a Windows environment, that does audio processing on your audio stream and then it can do encoding for web streaming. You can then send it to SHOUTcast, to your own SHOUTcast server; you can send it to a Wowza server or external replication service or content distribution networks.

Well we've got a great update to the Omnia A-XE. In fact, it's so good, but we've just renamed it. It's the ProSTREAM X/2. Now this is software. It runs in a Windows environment. I don't know how far back it'll run. It runs fine on Windows XP and Windows 7. It'll run on Windows 8 and it'll run on Windows Server. And the ProSTREAM X/2, for every license that you purchase on it, you get to bring in another audio source.

Now with that audio source, you can do lots of things with it. You can simply audio process it and send it back out again, so if you just want an audio processor this is one way to get it done. If you want to process it and then encode it, you can do that, and you can encode it multiple ways. Process it in multiple ways and encode in multiple ways. In fact, there is no limit, the license that you buy when you buy a ProSTREAM X/2, it only is good for each input. If you buy five licenses, you can bring five inputs into it. But each input you can encode in as many different ways as you'd like, and each way that you're encoding you can send to multiple servers.

So if you want to do let's say High-Efficiency AAC v2 as your encoding algorithm, you can send that out to half a dozen servers if you want to that are geographically diverse or for backup or redundancy and that kind of thing.

You can also, for confidence monitoring or for just a few streams, you can have clients connect directly to the ProSTREAM X/2 software by making that available on the public Internet. So you have lots of options there.

And one more cool option, multi-rate streaming. Now this may be a new concept to a lot of our listeners, and it's kind of new to me, but if you want to be able to serve listeners in various environments at their desktop with a PC, a laptop roaming around a coffee shop, or mobile over a 3G or 4G environment into a car or a mobile cell phone, you want to be able to stream at various different rates. And wouldn't it be great if the listener doesn't have to choose the rate that he or she wants to listen to? It happens automatically.

Well that's what multi-rate streaming is, and the ProSTREAM X/2 has got that built into it. There are standards, of course different standards. There are standards from Microsoft, Smooth Streaming; there's Apple's HLS; there's MPEG-DASH. These are built-in to ProSTREAM X/2.

Now if you want it one more better, there's another version of ProSTREAM X/2. It's the ProSTREAM 9X/2. If you want absolutely the best audio processing on earth for your stream, well the 9X/2 has Omnia 9 processing built into it. You pay extra for it, and you get the famous Leif Claesson Omnia 9 processing for your audio stream.

Let me encourage you to go to the Telos Alliance website. Under Telos Systems look for streaming and you'll find the ProSTREAM X/2 and the ProSTREAM 9X/2. I've got this running on my radio stations. Yeah, it's real similar to Omnia A-XE. It's a new, update look, new interface, and more capabilities. The 9X/2 option give you the audio processing of the Omnia 9 basically built into the software itself. It's really awesome and it's some of the best streaming you're going to find.

Oh, one more thing. The built-in metadata engine, we have a built-in scripting system that allows an ingest of just any kind of metadata about your stream and parsing it out and formatting it to go along right with your stream. It's just awesome. Check it out on the web at telosalliance.com or you can go to telos-systems.com. You'll get there. Thanks a lot to ProSTREAM, the ProSTREAM X/2, for sponsoring This Week in Radio Tech.

All right, we're talking to Mark Persons. Chris Tobin is here with us and he and Mark are having a conversation about some of the things that Mark's been up to lately. Mark, you talked about NRSC. Was there anything else on that you wanted to cover, like maybe what your proposals are exactly? If it worked out the way you'd like to see it, we wouldn't be doing tests annually I think you said. We'd do them some less often amount, and we'd do FM perhaps on the same schedule. You and I before the show were talking about if an AM and FM are together or if there's more than one FM. You know, if there are things that are just likely to be the genesis of a spectral problem, maybe we ought to check those more often. What are your thoughts on that?

Mark: Well going back to what I wrote in the article, I thought there was a loophole in the rules that allowed you to throw away whatever measurements you made on an FM two years later. Well, usually you wouldn't be inspected until after the two days and they probably never made the inspections or the spectrum analyzer measurements in the first place so I felt that should be corrected. And I also felt that if we were only going to do these measurements once, that we should do measurements and keep the documentation forever and then if there are any changes, and by the way any changes in bandwidth could be caused by something as simple as replacing an audio processor or adding an SCA, those RDS things. Under FCC rules you're supposed to make new measurements.

So, you know, not many stations are doing that so an inspector who was really up on the rules could cite the station for being incorrect. And I agree that measurements should be made more frequently. But going back to this whole thing, I also mentioned that you need to be aware of any changes in the environment. Here's what I mean. The station's been on the air for years. Everything is fine. Then a news station is built a mile away down the road, okay? And it's another FM. And guess what? You don't know it, but if you were tuning around in the right place, you'd probably find a mixing product between the two stations and it's probably FCC illegal. So you need to stay on top of these things to ensure that you're operating within the rules.

Now I should explain a little bit more about the mixing products. We talked about AM and FM mixing, but let's say you have two FMs, okay? One's on 94.1. One is on 96.1. And they're operating either at the same site or they're operating nearby, and it could be as little as six miles down the road and they could still mix to create unwanted products.

Well, what happens? The 96.1 signal comes into the 94.1 transmitter. It goes down into the transmitter, and there's what is called return loss in the output network of the active devices, tube or transistor. It comes back out, and let's say, it was 96.1 went into 94.1. It comes out on 92.1. The math is 2a minus b, two times one minus the other.

Well, for a spectrum analyzer the real easy one is gosh, you put your station signal in the middle, the 94.1. Up a couple megahertz is 96.1. You see that one. And then you look down two megahertz from the station's signal, 92.1, then oh my gosh it's only 60 dB down. Well what are the rules? Well at a kilowatt, the rules are that any mixing products have to be 73 dB down. When you get to five kilowatts it all has to be 80 dB or more down. And so you've got to be pretty clean to make all that work right. So, you know, I'm not a stickler for FCC rules but I sure would like to keep the spectrum clean. I don't want junk everywhere. And besides, the FCC rules require it.

Kirk: Yeah. In the example that you gave, Mark, where you have stations that are even separated by some distance, you said up to six miles or so, but let's say they're more conveniently a half mile apart. I've got a transmitter site, an FM, that is actually probably a quarter mile from another FM. I think one is 25 kilowatts. Ours is three kilowatt ERP. But if you did find these mixing products, what's your go-to solution? What's the first obvious solution that you would attempt to recommend?

Mark: Well if it is really two transmitters in two locations, well even if they're at the same site, almost certainly you're into some sort of a pass-band or pass-notch filter to satisfy the FCC rules and keep it clean. And by the way, as I mentioned earlier, the mix products could be up in the aviation band and then you're really in trouble. You want to keep after these things. But I mean it could appear down in a TV channel. Well, we don't have any TV channels down in six and five anymore. We should talk about that. I think that's where AM broadcasters belong on the FM dial, but that's an entirely different subject unto itself. But no, you need to keep after these things.

Kirk: Actually, I think that could be a great place to do. Why don't you and Chris chat about... I would love to see AM broadcasters get some bandwidth in an expanded FM band because I just don't have long-term faith that the AM band is going to be usable. There's not going to be any reduction in manmade noise in the AM band, so a never-ending battle of increased power and maybe going digital but noise still interferes with digital just in a different way. You and Chris, what are your thoughts about pushing to expand into an FM band?

Chris: Well, first of all, those of you looking at a video, this is what a spectrum analyzer looks like. It's on my shoulder. It's an HP-8921A.

Kirk: You're a strong man. Look at that.

Chris: It's a hefty little box, but it's a handy box, I will say that. I just figured I'd show it off because we were talking about spectrum analyzers this episode and then there was I think oscilloscopes in another episode so I thought it might not be a bad idea to show a prop. And besides, in the repair of high-voltage blocking capacities in the article in Radio World for Mark it talks about different things and in there is a Sencore capacitor tester that I haven't used in ages. I used to have one of those. It's a classic. But you don't get to see this stuff and you can't get excited, therefore you can't learn more. So that's what this is, a spectrum analyzer, HP-8921A. Very popular box.

And then for the return loss issues at the mixing pro ducts, this is a circulator. A small one, this is for land-mobile two-way radio stuff, but a similar design could be used for FM. Then the return loss, and there's the reject load there where my finger is pointing. So just to point out what Mark was saying with the return loss and transmitter, the amplifier, and things coming back down, this type of device helps to minimize it.

But as far as AM moving up to an FM spectrum like channels five and six, I wouldn't mind it. It would make a lot of sense. The question is the business model. People have invested so much currently in their AMs on the larger station side that may not be interested in making a change. I don't know. But yeah, I think it would make sense at this point to start investigating moving. You know, FM moved off of what is it, 66 megahertz back in the early days because of propagation and chaos that was ensuing from signals being heard everywhere. So why not take that same thinking and say, "Well, AM is not being subjected to so much noise, maybe it's time for them to move to a new spectrum that makes sense." Channels five and six just might be the place.

Kirk: Mark, what would you...

Mark: I brought my spectrum analyzer in while you were talking.

Chris: He's got the newer one! Eh, you've got the newer one.

Mark: Yeah. Well what happened was the old one got too heavy for me. Don't ever get old. This newer one is a lot lighter, like I don't know, ten pounds or whatever it is. And my, are they handy and my, do they work nice, but they also get you for about $15,000 too.

Chris: But you can do a lot with them.

Mark: That's the Agilent N-9340B I guess.

Chris: That's a nice box. That's a nice box.

Mark: There's a lot of them out there. I'm an AM person. I was born into AM. My dad started in AM in the 1920s and our family owned a couple of AM stations, no FM stations, along the way. It was my life. But I have to be realistic. Oh, by the way, I've also built maybe half a dozen directional along the way. But I have to be realistic, the AM band is really noisy today and as far as I'm concerned the cow got out of the barn because the FCC never handled Part 15 enforcement so things are radiating everywhere.

But in the big scheme of things, even if they were all Part 15 compliant, there's so many of them that the AM band is really noisy anyway. So my take on it is that AM should live as long as it can, as long as they're profitable, and then when it's time the owners will turn them off and there will be fewer and fewer on the band and in the process we'll wind up with hopefully AMs migrating to an FM signal. To me that makes so much sense.

And you know, modifying an FM receiver to cover a few more megahertz? That's simple. Sure, it'll be a while before there are a bunch of them out there, but it wouldn't take too long. I mean we had the same problem with the high end of the AM band. It used to go to 1500, then it was up to 1600. That somehow all worked out. Oh, gosh, it went up to 1700. That worked out too.

Chris: Yep.

Kirk: I believe this proposal has been made to the FCC but no action has been taken on it. Is there any practical way to free up enough former TV bandwidth to make this happen, or are we just smoking something?

Mark: I don't know that because I'm not a TV person, but it seems proper to me. I understand that for digital the low TV channels, two through six, really are a bad choice but FM is a really good choice in those channels. And you mentioned way back that the FM used to be on 66 megahertz or right about there. That goes back to the days of Edwin Armstrong.

And my father knew him, by the way, and built an FM station in Duluth, Minnesota which the AM was so big at that point back in the late '30s, that kind of thing, that FM just really never caught on. But the point is I think it was all politics that moved the FM up to a higher frequency, and I think it was instigated by RCA according to the book that I read Empire of the Air. Maybe you read that book too.

Chris: Yes, yes, you're right. You're absolutely right. It was political. Those days it was RCA, Sarnoff, and politics basically. Those are the three.

Mark: Yeah. An unfortunate side-effect of all of this. And Armstrong should be given all the credit because he spent I think two years developing the technology. He started out thinking that narrow-band FM would work, then he said, "No, that isn't going to work. We have to make it wide-band FM." And finally he worked his way through all these things and making receivers work and all that before he introduced it to the world. It was absolutely fantastic, noise-free radio. They were literally using FM stations as studio transmitter links or links from city-to-city so to speak, city-to-city really, to get high-quality audio from one station to another so that it could be broadcast on an AM.

Chris: That's right.

Kirk: Mark, you sent us along a number of examples of troubleshooting and some really insidiously weird things. One was this problem at 1440 and 1450 caused by another station sometimes playing the same audio. Would you like to finish up by telling about that and what the mystery was and how it got solved? Or perhaps there's another episode that you'd like to talk about.

Mark: That one is very convoluted and you almost have to read the article to understand it. You might give a link to that article.

Kirk: Sure.

Mark: There was some interference, and they couldn't figure out where it was coming. It almost sounded like Morse code, but you couldn't decode it as Morse code because it didn't make any sense. For those of us who understand Morse code, we know what a dit and a dah is and how they form the letters A, B, and C and that kind of thing.

Well, it didn't form that. It didn't follow that pattern, but it sure sounded like Morse code. It turns out there was an audio processor, I think it was an Orban, on one of the AMs that went into brief oscillation about nine kilohertz away from carrier on certain audio peaks. And the low-pass filter, because it's a ten kilohertz low-pass filter for AM, let it through and it actually got broadcast. But all it turned out to be was just a little beep or a little signal and it became a beep when it was heard on the radio at the far end. And here was the really difficult part, was the two stations involved ran pretty much the same programming. Two different cities a ways apart, I forget, 50 miles apart or whatever it was.

And so these beeps would correspond to the modulation. You'd have to read the article to truly understand it, but there was a lot of sleuthing work there. I actually wrote the article, never experienced it myself. But the article was really about engineers getting together to get their heads together and think this through: how could this be happening?

And a number of good engineers in that particular area, it was the Appleton, Wisconsin area, that kind of thing, they did some listening on the radio and thinking and finally came up with maybe that's it. And sure enough, that's what it was. But it was an aggravating problem, and of course at first it was thought well, listeners don't know what they're listening to. How could they know anything? Well anyway, the truth is listeners were telling them the truth and they really had a lot of detective work to get there and it was a great story.

 

Kirk: And the whole story is online at your website. We'll put a link for that in the show notes.

Mark: Oh, wait a minute.

Kirk: Yeah, I'm sorry.

Mark: But there's more to the story.

Kirk: Oh, there is?

Mark: I've got to re-read the story. I think there was a real beep involved now that I think about it, but guess what? The beep was at nine kilohertz. Nobody listening to the offending station heard it. Why is that? Because AM receivers start to roll off at three kilohertz. So if somebody there had had a wide-band receiver listening to the transmitter with the bad processor, they would have heard it. But guess what? No one complained there. Only the people who were listening on three kilohertz receiver where this one kilohertz beep came through. That was it.

Chris: That's classic. That's classic.

Mark: Yeah. So anyway, it was just bad capacitors in an exciter. More bad capacitors, by the way.

Chris: Oh geez.

Mark: And then that was the answer.

Kirk: That seems to be the cause of a lot of problems, bad capacitors.

Chris: Well, those electrolytics like to dry out eventually.

Kirk: Yeah. Hey, we're going to hear from our sponsor Lawo and when we come back Mark I hope you have a final comment or maybe a tip for your fellow engineers who are listening and watching This Week in Radio Tech. And Chris Tobin, maybe you have a tip, something you've come across over the holidays that you went "Huh, that's a good idea." Or "Who expected that to be a problem," and here's what to watch out for. So when we come back, let's ask each that question.

Our show is brought to you in part by the folks at Lawo, L-A-W-O, Lawo.com. They make those big, big fine German audio consoles that are in TV trucks and TV station production houses, but they also make smaller consoles for radio stations, their crystal series and more. And their new console, one that we learned about at NAB, at the 2014 NAB in Las Vegas, is the crystalCLEAR and they call it a virtual mixing console. Why virtual? Well, because the surface that you touch is indeed a computer. It's a multi-touch touchscreen, and the application, the edge-to-edge application, looks like an audio console.

Now there's a lot of cool advantages to doing it this way. The brains of the console, the part that actually does the mixing, well that sits over in a rack. It has the audio inputs and outputs on it and also the Ethernet connections to do Ravenna or AES67. That DSP engine sits away in a rack somewhere. Connected then via an Ethernet network is a computer. It's an HP computer on the back of a big, beautiful touchscreen. And this runs the app that actually looks like the audio console. It's the crystalCLEAR Virtual Radio Mixing Console from Lawo.

It's got all the things you'd expect from a modern mixing console. It's got a couple of mix buses plus a record bus. It has of course a preview bus, a queue bus if you will. It's got scene presets. You can program those and recall them for different shows. Precision PPM stereo meters, of course. It's got both what they call Euro and US operating modes as far as where the buttons are and how the faders make the actual faded channels go on and off.

There's a large time of day clock. Of course it's synchronized with NTP if you want that. It supports guests with talkback so you can talk back to your guests privately if you need to say, "Hey Bob, I need you to wrap this show up in just a minute." There's a panic button that clears all the changes an operator may have made since the last recall of a scene.

Twenty-four sources are available. You can plug in 24 things at the back of the console. Any eight can be active for any particular scene, and so it has eight faders. And you know, that's what you need for most shows. Eight faders will get you by plenty. Hey, my radio stations use eight faders on a different brand of console but eight is great for us.

All the audio stays in the engine, so the PC is for control only. If you wanted to you could keep a spare if you have multiple studios. Keep a spare PC in case somebody breaks a touchscreen. There's of course GPIO for on-air lamps and for tally and such as that. Power supply redundancy is available. There's a couple of mic inputs and some line inputs, AES/EBU inputs, and it's really all broadcast-quality stuff. The folks at Lawo know, German engineers and very good hardware. They'd like you to check this thing out.

There's a video on the website. You can watch Michael Dosch, he's the director of Virtual Radio Projects, he's showing off the features of this new console at NAB 2014. The website is lawo.com and then look under Products and Radio Consoles and look for the name crystalCLEAR. Of course I bet you if you just Google Lawo and crystalCLEAR you'll find the page as well.

Finally, the thing that is very cool about this is all the buttons on the console are context-sensitive. So if you're dealing with a mic channel and you touch an options button for the mic channel, you get options that have to do with what that channel is being used for, not other stuff that you have to ignore. So it's very contextual in the way that it operates. So for operators, they don't have to think about much. They punch a button and bam, there's the options right there. The Lawo crystalCLEAR console, and thanks to Lawo for sponsoring This Week in Radio Tech. I appreciate it very, very much. Okay guys, gentlemen. Your tip or your final story that you learned from? Mark, how about you? What've you got for us?

Mark: Oh, great. Glad you asked. At a number of visits to radio stations when I was doing NRSC measurements on AM and looking at the audio bandwidth, I would often see that the audio bandwidth wasn't ten kilohertz each side of carrier; it was 7.5. Why was that? Well, it was a talk program going on and that was the width of the satellite channel that they were using.

But my real tip of the day is if you have anything where the transmitter acts funny because you're running ten kilohertz audio into it, because there's a lot of bizwar out on the sidebands, or if you just want to be loud on the dial and you're a talk show, switch your audio processor to five kilohertz maximum. Nobody in the audience is going to hear that they've lost that audio bandwidth. They're going to think that it sounds just the same as it did before, only louder. Some recent audio processors allow you to switch back to five kilohertz worth of bandwidth. All the rest of them are ten kilohertz. So that's my tip of the day.

Kirk: Wow. And that does say something about the state of everything, and we kind of knew it was going that way, if you, the engineer, switch between five and ten kilohertz bandwidth and the radios can't tell the difference. I think that's a little sad.

Mark: Well, it's a fact of life. We have to live with it.

Kirk: Chris Tobin?

Chris: Well, my tip of the day would be if you're maintaining a transmitter facility and you've purchased some new equipment, or even old but we'll go with the new equipment, take advantage of the spare parts kit. Option, that is. And the reason being recently, and I'll be out there tomorrow at another AM 50 kilowatt site to do some work, it turns out they had to do some work on a solid-state transmitter and some of the parts we needed are a somewhat custom design and as a result because they didn't purchase the spare parts kit we had to place an order with the manufacturer and wait for them to be shipped.

It turns out I did the math. Had they bought the kit for several hundred dollars, they would've saved on the several hundred dollars that we're now spending to get the part in for an emergency repair. So it's well worth it. It may be a tough sale to your manager, but there's ways to explain the benefits of spare parts on the shelf. And I'm sure Mark you've run into this on many an occasion.

Kirk: That's a great point.

Mark: I know what you mean.

Kirk: Go ahead, Mark.

Mark: Oh no, that's it. I'm just agreeing. Exactly.

Kirk: And Chris, you brought up a point that I had not thought of but it's absolutely right. When your transmitter, or device, whatever it is, a lot of times you can get spare part kits for consoles or for transmitters. And when it was made, the parts obviously are available when it's made and they may not be available later on. So if you buy a spare parts kit and it's shipped at the same time the product was made, guess what? They're what the thing needs and they're what the manufacturer thinks is most likely to fail and most likely to need replaced.

At the place I work, at Telos, our support department has done numerous studies to see what should be in a spare parts kit. What are the things likely to fail, and what can we make available to the customer if they want to go ahead and do this? That's great advice. And I remember Nautel transmitters, it seems like some that I bought came with a modest spare parts kit whether you wanted it or not and then you could certainly buy a bigger one. In one case I think we didn't buy an extra one, but whatever it came with, my goodness we ended up using a lot of parts out of that. So that's great advice.

Chris: Well, I can tell you the reason I'm telling you this is because I can personally say back in 1995 I had a Harris MW1A solid-state transmitter with what was it, pulse-width modulation, right, PWM I think it was, or PDM technology that was being introduced. It was a dozen or so if not more aluminum heat sinks with if I remember correctly it was six transistors per, TT03 casings, and they all worked in a certain fashion. And if you got a lightning strike or a static discharge or something, an anomaly to your output, a few of these little modules would go out of tolerance. The transistors would short. You know, the usual stuff.

And it turns out that yes, it was a common transistor you could buy from the ECG Slovenia NTE parts replacement stuff of the day. Those of you who remember that will kindly smile and go "Wow, that's something I remember." And then you would put the transistor in, fire up your transmitter, and six more modules would explode. Reason being? Harris failed to tell you when you called them up and said, "Hey, can I just replace this with a resistor from a replacement," and they say yes, you have to match the transistors in pairs. Use a transistor checker or VOM and make sure the beta and everything is close to where it should be, otherwise they will not properly switch on and off and they will make all the others even more...

Kirk: Unhappy.

Chris: Fail. Unhappy I guess. I learned that in 1985 and I swore from then on forward that I would make sure I'd find some way to maintain spare parts or at least know going in that if I'm going to fix something I'm going to find out every possible knowledge of the device.

When it came to Harris products, I would call them at 3:00 in the morning and I would badger that service tech to the point where I was like "I just want to make sure that when I replace this component, the following will take place. Are you certain?" And he's like "Well..." I said, "It's a yes or no answer. If the answer is somewhere between those two words, I'm getting off the phone and will call tomorrow with somebody else." Trust me, it will save your bacon on many occasions. And Mark, I'm sure you've been down that road on at least two occasions.

Kirk: Mark Persons has been our guest here on This Week in Radio Tech. Hey Mark, we're about out of time. Tell us where folks can find the wealth of knowledge that you've put on the Internet. What's your Internet home?

Mark: Oh, it's on my website. It's been there for 19 years. It's mwpersons.com. mwpersons.com. And there are several hundred pages there. Not just a few. If you look around, you're going to find a lot of information. And sure I build custom things and I show that on there, but the wealth of information is in the tech tips section which you'll enjoy.

Kirk: Good deal. Good deal. Thanks so much. And Chris Tobin, folks can find you at an email address. What would that be?

Chris: Yes, support@ipcodecs.com. I've had many an email recently so I'm happy to hear from those folks who have written to me.

Kirk: Good deal. All right. And I'm Kirk Harnack. I can be found at the Telos Alliance where I am working to tell stories and help folks get up to speed on some of the latest technologies. Thanks a lot to Mark Persons, to Chris Tobin, and to Andrew Zarian for producing today's show. Be sure you check out other shows that are on the GFQ Network. They may not be like this one, but they're a lot of fun anyway.

So check those out at GFQNetwork.com. Tell your friends about This Week in Radio Tech. This show will be posted on the GFQ Website and also on ThisWeekInRadioTech.com within a couple of days. So thanks a lot for your participation and enjoyment and telling your friends about This Week in Radio Tech. We'll talk to you next week on This Week in Radio Tech. Bye-bye.

Topics: Broadcast Engineering