Archive

July 2016
M T W T F S S
« Jun    
 123
45678910
11121314151617
18192021222324
25262728293031

Categories

A Linux based remote control system

We are extending LANs out to transmitter sites for many reasons; backup audio, control and monitoring, security systems, VOIP phones, etc.

I am casually (very casually) toying around with creating my own Linux based remote control system.  The ongoing Windows 10 upgrade debacle continues to not end, I can’t help but think that there are many potential clients who could use a reliable transmitter/studio remote control and monitoring system based on a stable operating system.  Hmm, sounds like a sales pitch 😉

Anyway, I have run across several Ethernet board manufactures that offer a variety of boards with 8-12 contact closures and a variety of analog and digital inputs.  Most new transmitters have some sort of web GUI which are great for transmitter control and monitoring.  As we all know, there is more than just a transmitter at any given transmitter site.  In addition to the transmitter, I would like to control and monitor things like tower lights, interface and control of coax switches, temperature monitoring, generator status, the old non-web interface backup transmitters, STL signal strength for those old 950 MHz links, etc.

Since Google is my friend (when they are not storing my search data), I came up with this: Internet-ethernet-12-channel-relay-board

That particular PC board is made in Bulgaria, which is home to this: Mount Buzludzha

What I like about these particular boards is the DRM software (DRM has, apparently, many different meanings) which will run on Linux or Windows.  There are also iOS and Andriod applications that can be used as well.  It appears that the GUI can be customized for various uses.   This seems like it is written in Java, so perhaps I could have some Java expert customize it for radio use.  It looks like up to 32 boards can be controlled by a single instance of the DRM software.  Alarm reporting would be via SNMP trap and email.

I don’t know, there is one particular cluster of stations that needs new remote control gear at almost every transmitter site.  Perhaps a little alpha testing is in order?  It could be fun…

Anyway, just a thought…

 

The Horns of a Dilemma

Alternate title: Building and ATU in a truck body tool box.

Alternate title II: I should get paid extra for this shit.

There is an AM radio station that is near death but the owners do not want it to go away.  Nor to they want to spend very much money to keep it around, thus the dilemma.  At the transmitter site, there are a multitude of problems; leaking roof, very old rusty ATU, rotting support posts and transmission line bridge, equipment racks rusting out, nothing is grounded properly, the building is full of junk, snakes and mice have moved in.  To further complicate things, the tower and transmitter building serve as an STL relay point for two of the market’s FM stations.  There is also two translators with antennas on the tower.  The ATU and tower light choke box are rusting through, which is causing arcing and broadband RF noise that is interfering with the FM station’s STL receiver.  There was a home made isocoupler for one of the translators that was allowing AM RF back into the building which was creating havoc with everything.  Because of this, the AM station is currently silent.  In short, it is a mess.

WCHN ATU

WCHN ATU

The red box on the bottom is the ATU, the plywood box on the top with the peeling yellow paint is the home made isocoupler, the tower light choke box is behind the isocoupler.

Crumbling old ATU output capacitor in series with tower

Crumbling old ATU output capacitor in series with tower

This was the capacitor that was feeding the antenna, .0041uf, 10KV 8 amps.

We started remediation on this last February, which is not optimum time for replacing rotting wooden posts.  However, we were able to clean out the building.  The leaking roof has been repaired.  I was able to find a few old racks from a Schafer Automation system to replace the rusted out original racks.  I began the process of grounding the equipment racks, the incoming transmission lines for the STL, etc.

Cool morning, Garter Snakes warming themselves on top of a Moseley DSP-6000

Cool morning, Garter Snakes warming themselves on top of a Moseley DSP-6000

Garter Snake

Garter Snake

We will have to find out how they are getting in, the plug up those holes.

Then there was the ATU and tower light choke enclosures.  Original to the 1952 sign on, they were past their serviceable days.  Since this is all being done on a budget and nobody wants to spend money on an AM station that has little or no listeners and even less revenue, we had a problem.

Then somebody suggested building an ATU in a truck body tool box.  Well…  This isn’t the Meadowlands, so if there are no other alternatives then okay, I guess.  Off to Amazon to order a tool box.  This particular unit seems fine, my only comment is on the gauge aluminum (or aluminium if you prefer), which is slightly thin for holding up all those parts.

ATU built in a truck body tool box

Fabrication shop, ATU built in a truck body tool box

Still, the box itself is nice enough and certainly better than the old one.  I was able to reuse the inductor and the Delta current meter but the old Sangamo capacitors crumbled in my hands when I removed them.  I also saved the feed through bowls, J-plugs and other parts.  I used some copper strap to run a good RF ground from the input to the ground connection.  Overall, I am pretty pleased with the finished product.  It is a little bit tight in there, but this station only runs 1 KW, so it should be fine.

Replacement ATU mounted

Replacement ATU mounted

So, new pressure treated posts installed, the box was mounted and the transmission line connected.

Replacement ATU under power.

Replacement ATU under power.

Reused Schaffer Automation racks, much better than the 1950's Gates racks

Reused Schafer Automation racks, much better than the 1950’s Gates racks

The reused racks are old, but serviceable and a big improvement over the old, rusting out racks.  I was able to bond each rack to the ground strap that used to connect to the RCA BTA-1 transmitter.  There is one more rack to install to the right of these two.  That should give us more than enough rack space for this site.

The station is back on at full power and not interfering with the FM STLs or the translators.  You can actually touch the rack and not get an RF burn!

We are also working on an air conditioner.

Other work at this site; cleaning out the building, replacing the tower light photocell, installing a ground buss bar, some STL lightning protectors, dress the transmission lines, etc.  It is a work in progress.

 

Medium Frequency ATU design

This is a topic I have covered before, but it is worth doing it again for future reference.  The previous post covered downgrading an AM transmission facilities for WGHQ, Kingston, NY.

This is part II of that process.

WGHQ transmitter site, towers 1 and 2 removed

WGHQ transmitter site, towers 1 and 2 removed

The old towers have been cut up and put in a scrap metal dumpster. They are off to China to be melted down and made into a submarine or a missile or a tank or something useful like that.

Towers scrapped

Towers scrapped

The directional array had a three towers in a straight line with a common point impedance of 60 Ohms.  Dropping two towers greatly changed the electrical characteristics of the remaining tower, therefore the existing ATU needed a bit of reworking to match the 50 Ohm transmitter output.

First step, correct a few deficiencies left over from the old array.

Vise grip tower feed

Vise grip tower feed

This vise grip RF connection has to go. The problem is where the tower erectors attempted to solder the copper tubing.  That tower base plate is pretty big and I would wager they didn’t use enough heat to make the solder connection.  They were probably working in the winter time, thus the “temporary” fix.  This tower was put up in 1993, so that temporary fix lasted 23 years.

I removed the offending tool and soldered the connection to another part of the tower with silver solder.  The smaller cross bar made a good connection point.

RF feed correctly connected to the tower

RF feed correctly connected to the tower

After soldering, I cleaned up and sprayed some grey primer on it to prevent rust forming where I scraped the paint off.

Next, I made an impedance measurement:

WGHQ tower base impedance measurement

WGHQ 920 KHz tower base impedance measurement

That junk on the upper part of the graph is coming from WHVW on 950 KHz. The tower itself looks pretty good, 77.6 Ohms resistance with 130 Ohms inductive reactance.  Since this is not a part of a directional antenna system, the ATU design is pretty straight forward.  Given that WHVW on 950 KHz is located 10.41 miles away, a low pass filter design is optimum.  A basic low pass filter T network has inductive input and output legs with a capacitive shunt leg to ground.

T network diagram

T network diagram

Each leg is used to match the 50 Ohm transmission line impedance (R1) to the 77.6 Ohm tower impedance (R2) and cancel out the 130 Ohms of inductive reactance.  This is a vector impedance problem, much like a vector force problem in physics.   Some basic arithmetic is required (always include the units):

X1, X2, X3 = √(Zin x Zout)

X1, X2, X3 = √(50Ω x 77.6Ω) or X = 62.28Ω

The value of inductance or capacitance for each leg is calculated using the basic inductance or capacitance formulas:

L (μH) = XL / 2πf(MHz)

And

C (μF) = 1 / 2πf(MHz) XC

Thus the input leg, or X1 = 62.28Ω / (6.28 x 0.92 MHz) or 10.78 μH

The Shunt leg, or X2 = 1 / (6.28 x 0.92 MHz x 62.28Ω) or .0028 μF

The output leg is a little different.  The tower has 130 Ohms of inductive reactance that needs to be cancelled out with a capacitor.  Rather than cancel out all of the inductive reactance, then add an inductive output leg, the tower reactance can be used as part of the tuning circuit.  The design calls for 62.28 Ohms inductive reactance, so 130Ω – 62.28Ω = 67.27Ω, which is the value needed to be cancelled by a capacitor:

Output leg, or X3 = 1 / (6.28 x 0.92 MHz x 67.27Ω) or .0025 μF

A little Ohm’s law is used to calculate the base current for both the day and night time operations.

Ohm's law pie chart calculator

Ohm’s law pie chart calculator

Thus the daytime base current is I = √(P/R) or I = √(1000 W/77.6Ω) or 3.58 Amps.

Night time base current is I = √(38 W/77.6Ω) or 0.70 Amps

Current handling requirements:

Base current is calculated to be 3.6 Amps at 1,000 Watts carrier power.  Allowing for 125% peak positive modulation makes it 5.7 Amps.  Having safety factor of two or 11.4 Amps output leg and 14 Amps input leg.

Voltages: 353 maximum input voltage, 439 output.

Thus, 20 amp, 10 KV parts should work well.

The designed schematic for the ATU:

WGHQ ATU Schematic diagram

WGHQ ATU Schematic diagram

Putting it all together.

Since the tower looks fairly broad at 920 KHz, we are going to attempt a nice broadband ATU to match it.  This station is currently programmed with a classic country format, and I have to tell you; those old Conway Twitty, Merle Haggard, Patsy Cline, et al., songs sound pretty good on the old AM radio.  The Subaru stock radio has HD, which also has a nice broad IF section, thus allowing all those lovely mid-high range frequencies through.

This is the existing ATU, which I believe was built by Collins in 1960:

WGHQ Tower 3 ATU

Existing WGHQ T network ATU

The ATU building is a little rough, but the ATU itself is in remarkable shape for being 56 years old.  The input leg inductor is in the center and will be reused as is. The large Jennings vacuum capacitor at the bottom is a part of the shut leg.  Its value is 2000 pF at 15 KV.  The top vacuum capacitor is series output cap, its value is 1000 pF at 15 KV.  The basic plan is to move the upper cap down in parallel with the bottom cap.  The shut leg inductor will be kept in place to tune out any access capacity.  For the output leg, I have a 2500 pF mica cap and a 10-100 pF variable cap connected in parallel.  The inductor on the output leg will be removed.

After some re-work on the ATU components, I tuned everything up.  The easiest way to do this is to disconnect the legs, measure them individually and adjust them for the desired reactance, which in this case is 62.28 ohms or thereabouts.  The output leg was measured with the tower connected since the tower reactance is a part of the tuning circuit.  The input leg was right about 10 μH.  The shunt leg turned out to be about 0.002 μF.  This is often the case, theoretical values are slightly different than field values due to stray capacitance and inductance in the connecting straps, etc.

This is the load, as measured at the output terminals on the transmitter:

WGHQ tower load as measured at the transmitter output terminals

WGHQ tower load as measured at the transmitter output terminals

Slightly asymmetric on 910 KHz, but overall pretty good. There is a fair amount of phase rotation in the transmission line due to the length from transmitter to tower (855 feet, 260.6 meter), which works out to be 0.93 wave length allowing for the 86% velocity factor of the transmission line.

Time to pack up and go home.

Downgrading an AM radio station

WGHQ in Kingston, NY has been downgraded from a 5KW DA-1 to a 1KW non-DA system.  This was done because two of the three towers in the directional antenna array dated from 1960, were in very rough condition and needed to be replaced.  The remaining tower (furthest from the transmitter building) had been replaced in 1994, is in good condition and is being kept as the non-directional radiator.

Here are a few pictures:

WGHQ 3 tower directional antenna array, Port Ewen, NY

WGHQ 3 tower directional antenna array, Port Ewen, NY

More deferred maintenance

More deferred maintenance

RF and tower light feed disconnected from tower base

RF and tower light feed disconnected from tower base

Second tower base vegetation not as bad, tower disconnected

Second tower base vegetation not as bad, tower disconnected

WGHQ transmitter and original Collins phasing cabinet

WGHQ transmitter and original Collins phasing cabinet

First tower video (sorry, I appear to have no idea what I am doing with the camera):

Second tower video, this one is better:

Towers on the ground:

I made measurements on the third tower and constructed a temporary ATU with parts on hand to get the station back on the air. They are now running 1 KW day, 38 watts night, as per their CP. I will be going back up to finish the job once the brush has been removed from around the existing tower and the ATU building has been repaired.  The coverage with 1 KW is not bad, actually:

Predicted coverage map from FCC website

Predicted coverage map from FCC website

The translator is on its way.

Hitchhiker’s guide

Anyone that grew up a geek in the late 70’s to early 80’s (ahem) will get the references in this video:

For those of you that are unfamiliar:

The Hitchhiker’s Guide to the Galaxy is a comedy science fiction series created by Douglas Adams. Originally a radio comedy broadcast on BBC Radio 4 in 1978, it was later adapted to other formats, including stage shows, novels, comic book adaptations, a 1981 TV series, a 1984 computer game, and 2005 feature film.

I was most familiar with the video game, which came out about the same time I bought my first computer, the beloved Apple IIc. That might have been in 1986 or so.

Anyway… It is nice to see a new generation of enthusiasts among the current Engineering students.

So long, and thanks for all the fish.

The Ubiquiti Nano-Beam

I installed one of these wireless links between two transmitter buildings recently.  The Ubiquiti gear is not my first choice, however, the client insisted that we use this equipment likely because of its inexpensive nature (less than $65.00 per unit).  My overall impression is so-so.  They are fairly easy to set up; the AirOS is intuitive and easy to navigate around.  I had to upgrade the firmware, change the default user name and pass word, assign IP addresses, subnet mask, gateway information, SSIDs, security parameters, etc.  All of that was very easy to figure out.  My grip is this; it seems the hardware is a bit plastic-y (e.g. cheep).  I know some of the Ubiquiti models are better than others.  I hear good things about the airFiber units but they still don’t compare to the Cambium/Canopy gear.

For this installation, I used the shielded Ubiquiti “Tough Cable” with the shielded Ubiquiti RJ-45 connectors and Ubiquiti Ethernet Surge Protectors.  When making the Ethernet cables up, I made sure the shield drain wire was connected to the metal body on the RJ-45 connector.  I tested everything with my trusty Fluke Microscanner cable verifier which also shows continuity for the shield.  I am still not completely confident that the out door units will survive a lightning strike on the 898 foot (273.7 meter) guyed tower nearby.  Time will tell.

The system has a wireless path length of about 200 meters plus another 60 meters or so of Ethernet cable.  Latency when pinging the gateway across the entire network is about 3 to 4 ms (laptop>switch>nanobeam<->nanobeam>switch>gateway).  The network is being used for remote control/monitoring of a transmitters and backup audio via Comrex Bric link II IP CODECs.

screen shot; Nano Beam Air OS

screen shot; Nano Beam Air OS v7.2.2

On the plus side, the 802.11ac link is very fast; 650+ Mbps unwashed link speed is pretty impressive.  Strip off the wireless LAN headers and that likely translates to greater than 500 Mbps goodput.  Also, the inexpensive nature of these units means that we can keep a few spares on hand in case something does suffer catastrophic damage due to a storm.  The AirOS v.7 is pretty cool with the RF constellation and other useful tools like airView (spectrum analyser with water fall display), discover, ping, site survey, speed test, trace route and cable test.

After installing the updated firmware, which fixes a major security flaw with the web interface, the link was established with three mouse clicks.  After that, I ran speed tests back and forth for several minutes.  Basically, the speed on the LAN is reduced because of the 100 Mbps switch.  Even so, that should be more than enough to handle the traffic on this segment of the network.

Picking your feet in Poughkeepsie?

Why yes, I have done that and it can be quite entertaining.  The FCC has graced our humble presence and apparently cracked down hard on one of those pie-rite types:

The New York Office received information that an unlicensed broadcast radio station on 87.9 MHz was allegedly operating in Poughkeepsie, New York. On January 14, 2016, agents from this office confirmed by direction finding techniques that radio signals on frequency 87.9 MHz were emanating from the basement of El Patron Restaurant, located at 411 Main Street, Poughkeepsie, New York. The Commission’s records show that no license was issued for operation of a broadcast station on 87.9 MHz at this location in Poughkeepsie, New York.

More here.

Hot Snot! One down and several to go? What is also interesting is the frequency; 87.9 MHz. That falls outside of the frequency range of the NY State anti-pirate law passed in 2011. That well crafted bit of useless legislation only covers 88-108 MHz.

If you are wondering about the title, it is from the movie “The French Connection:”

This is how I imagine those rough enforcement bureau types interrogating a busted pirate. Well, at least back in the days of Alexander Zimny and Judah Mansbach anyway.

That video is hilarious, back in the day.

Decommissioning transmitters

I was at a transmitter site a few days ago scrapping a Continental 814-R1 transmitter and started thinking (always a dangerous thing) about how many of these units I have decommissioned over the years.  It turns out, quite a few:

Make/Model Year new* Year removed Station Disposition
GE BT25A 1948 1994 WPTR Donated/scrapped
Gates BC5P 1960 2004 WWLO Donated
Harris MW5A 1982 2000 WLNA Scrapped
Gates BC1T 1961 2001 WLNA Donated
Harris FM20H3 1972 2001 WYJB Scrapped
RCA BT1AR 1960 2001 WROW Donated
Harris BC1G 1972 2001 WDFL Abandoned
Harris FM20H3 1971 2005 WHUD Scrapped
BE FM30A 1988 2005 WHUD Cannibalized
Harris FM5G 1972 2008 WSPK Scrapped
Mc Martin BF3.5K 1976 2011 WCTW Scrapped
RCA BTF-10ES 1978 2011 WRKI Scrapped
Gates BC1T 1964 2011 WINE Scrapped
Continental 315F-R1 1985 2013 WVMT Donated
Collins 813F 1975 2014 WKXZ Scrapped
RCA BTA1AR 1965 2014 WCHN Scrapped
Collins 813F2 1978 2015 WKXZ Scrapped
Collins 830D-1A 1968 2014 WKXZ Scrapped
Harris FM20H3 1972 2013 WYJB Scrapped
Harris BC5HA 1973 2013 WROW Scrapped
Harris FM10H 1971 2013 WMHT-FM Scrapped
Harris FM2.5H3 1973 2015 WEXT Scrapped
Mc Martin BF3.5K 1972 2014 WSRK Scrapped
CCA FM5000G 1980 2015 WTBD Scrapped
RCA BTF1E 1972 2016 WZOZ Scrapped
QEI 695T3.5 1996 2015 WBPM Scrapped
QEI 695T5 1996 2015 WBPM Scrapped
Harris HT3.5 1997 2015 WUPE-FM Scrapped
Harris Z5CD 1997 2015 WXPK Cannibalized
Energy Onix SSA1000 2000 2015 WDHI Cannibalized
Harris MW1 1982 2016 WPUT Abandoned
Mc Martin BF1K 1982 2016 WSUL Scrapped
Mc Martin BF3.5K 1982 2016 WSUL Scrapped
Continental 814R1 1980 2016 WDBY Scrapped

*In some cases the “Year New” is a guess based on when the station went on the air.  Before you write me and say “But model XYZ transmitter wasn’t made until 19XX, I did not look at every name plate and write all the information down as I did this.

Like everything else, there is a process to this.

RCA BTA-10U AM transmitter

RCA BTA-10U AM transmitter

First of all, if the transmitter was made before 1978, the possibility of PCB capacitors and transformers exists. In the case of the GE BT25A, massive amounts of PCBs needed to be disposed of properly. According to current federal laws, ownership of PCBs and PCB contaminated items cannot be transferred. Thus, the transformer casings were cleaned out and taken to Buffalo to be buried in a PCB certified landfill.   Otherwise, most other transmitters, such as the RCA BTA-10, may have a few PCB capacitors in them and perhaps the modulation transformer.  Those items can be disposed of by calling an authorized environmental disposal company like Clean Harbors.

The rest of the transmitter is stripped of any useful parts.  Things like vacuum variable capacitors, rectifier stacks, blower motors (if they are in good condition), HV power supply contactors, unique tuning parts, whole control and metering boards, tube sockets, etc.

The remaining carcase is then disassembled and hauled off.  I got a guy that will do this for relatively little money.  He takes the transmitter back to his warehouse and cuts it up, sorts all of the various metals out, then takes it to the scrap yard.  This includes things like cutting all of the windings off of transformers and power supply chokes, sorting out the brass and copper tuning parts, etc.

Remember kids:

μ = 1/1,000,000, 10-6 or 0.000001

Microaggression: If the worst thing that happened to you is you suffered through one millionth of an aggression, you are having a pretty good day.

The 16 channel bi-directional STL

As a part of our studio build out in Walton, we had to install a high capacity STL system between the studio and transmitter site. Basically, there are five radio stations associated with this studio and the satellite dish and receivers are going to be located at the transmitter site.

The audio over IP gear is getting really sophisticated and better yet, more reliable.  For this application, we are using a Cambium networks (Motorola Canopy) PTP-250 radio set and a pair of Wheatstone IP88 blades on either site.  Since there is quite a bit of networked gear at the transmitter site, the IP88’s will live on their own VLAN.  The PTP-250’s will pass spanning tree protocol, rapid spanning tree protocol, 802.1Q and other layer two traffic.

The Wheatsone IP88A blades are the heart of the system.  Not only do they pass 16 channels of audio, we can also pass 8 logic closures bi-directionally.  This is key because we are shipping satellite audio and contact closures back from the transmitter site.  The IP88A set up is fairly easy, once the IP address is entered.  The web GUI is used for the rest of the configurations including making the connections between units.

Pair of Wheatstone IP88A AoIP interfaces

Pair of Wheatstone IP88A AoIP interfaces

The switches are managed units.  The switchports need to be set up via command line to pass VLAN traffic.  There is an appendix in the IP88 manual that outlines how to do this with various manages switches.  This is the most important step for drop out free audio.  The switchports that connect to the two radios are set up as trunk ports using either VTP or 802.1Q.

Cambium PTP-250 5.8 GHz out door units

Cambium PTP-250 5.8 GHz out door units

The PTP-250 radios were already on hand, new in box.  They are built really well and look like they should not break in a year or so.  These particular units are connectorized, therefore an external antenna was needed.  There are many such antennas, this system ended up with a RF Engineering & Energy 5150-5850 MHz dual polarized parabolic dish with RADOMES.  RADOMES are necessary to prevent ice or snow build up in the winter.

RF Engineering & Energy 5150-5850 MHz dual polarized parabolic dish with LMR400 jumpers

RF Engineering & Energy 5150-5850 MHz dual polarized parabolic dish with LMR400 jumpers

STL link dish installed

STL link dish installed

1 1/2 inch EMT going from TOC to roof

1 1/2 inch EMT going from TOC to roof

Since the path is only 3.37 miles (5.43 kilometers), I set them up with a 40 MHz wide channel.  This is a rural, small town setting.  When I looked at the 5.8 GHz band on a spectrum analyser, it looks fairly uncongested.  These are MIMO single or dual payload selectable.  I will try them as single payload units, since the path is short and the band uncongested.  This should keep the throughput high.

Studio to transmitter site LAN extension

Studio to transmitter site LAN extension

The PTP-250’s use POE injectors in mounted in the rack rooms.  CAT5e shielded cable with the proper connectors properly applied is a must for lighting protection.  The PTP-250 units came with Cambium PTP-LPU lightning protectors.  I also installed Polyphaser AL-L8XM-MA type N surge suppressors on each RF port of each PTP-250.