Tower Talk

WSM-650 tower, Nashville, TN, 1943

Let's take another look at towers in the mediumwave band across the U.S. I did a previous series on RADIO-TIMETRAVELLER entitled Mediumwave Oddities in which one installment covered tower oddities. Oddities turned out to be fairly popular judging from the number of hits.

I've always been interested in towers, and specifically mediumwave towers. Recently I've been in touch with Dan Goldfarb, owner of the Yahoo group MW Masts, and we have been exchanging information on mediumwave towers in the U.S. Dan is attempting to document every mediumwave tower in the world, and has made great progress to that end. Be sure to check out the MW Masts group. You will find interesting discussion and informative files available for download.

As I drive around the country, it occurs to me that there is a lot of mediumwave metal up in the air. I wondered how much? A few code tweaks to my Radio Data MW program gave me the answer, supplying a comprehensive and interesting fact list of tower info. We'll get into the statistics toward the end of the article.

But first let's talk about the different types of mediumwave masts in the U.S. A tower is a tower, right? Not necessarily so. There are four different types, and actually five counting the odd Franklin antenna which I'll describe shortly. Officially, the FCC defines tower types by numbers - 0, 1, 2-9 and A-Z. Sounds like a lot of varieties, but it's really not. In practice Types 0, 1, and 2 designators are used, in which Type 2 can be one or several kinds of sectionalized antenna. So there are basically three tower varieties plus the Franklin, which the FCC lumps into the Type 2 category.

By far the most common tower out there is the FCC Type 0 (some 93%), a normal mast without any frills. It is simply a standard tower almost always insulated at its base. Stations strive to erect a tower at least one-quarter wavelength tall for the frequency at which they broadcast. At the low end of the mediumwave band this can be a tall tower - a quarter wave at 540 KHz is some 455 feet tall. At the high end of the band, in the western hemisphere at least, 1700 KHz, the quarter wave mast height would be only 144 feet tall.

The aim in all cases is to push as much RF energy across the ground as possible in order to get the best signal out to the target audience. Gone are the days when stations prided themselves in a nearly cross-country audience (during darkness hours of course). It's all about local and regional target audience now, as that's where the ratings, and thus the money, comes from.

So the object is to transmit as much signal near 0 degrees elevation (the ground) as possible, known as the "elevation angle" of the signal. The quarter wave vertical mast does a pretty good job, however by extending its length a little it gets even better. Excepting the Franklin type, a mast 5/8 wavelength tall does the best job, producing the strongest signal at the lowest angle. Only some 90 towers out of more than 7000 have heights in the range of 5/8 wavelength or taller. Few stations can afford this height, or may be prohibited from erecting such a tall tower due to various restrictions. Stations strive for at least a quarter wave for efficiency's sake.

To the rescue came an old electrical innovation called top-loading, discovered by Thomas Edison and put into first practical use in 1905 and 1906 by Fessenden and Marconi in their transatlantic tests on 82 KHz. A simple flat-top method of top loading creates the necessary capacitive effect to electrically extend the length of the mast. The resultant antenna is sometimes known as an umbrella antenna, top loaded with a "capacitance hat" looking like an umbrella. Some 550 towers (a mere 7%) use the top-loaded scheme. The top-loaded tower is the FCC's Type 1 tower.

A fine example of a mediumwave top-load "hat" is that of Australia's 3WV, 594 KHz, in Western Victoria, pictured just below.

So what is a Type 2 tower, you ask? The answer is a sectionalized tower, which is a tower having insulated sections stacked atop one another.

A scant 10 towers across the U.S. are sectionalized, an idea adapted for radio during World War I by Arthur O. Austin who already held the patent on sectionalized power transmission towers. The war ended, and Austin's sectionalized tower scheme was temporarily forgotten. Then station WHK, Cleveland, Ohio's pioneer broadcaster, erected a sectionalized tower. Though hampered by a power restriction of 1000 watts and a shorter wavelength, the station was suddenly being heard in faraway Honolulu, New Zealand, and all over the United States, achieving distances comparable to its super-power brothers.

With a sectionalized tower, high angle radiation can be minimized to give better ground wave performance than is possible with a simple vertical radiator. Calculations for a 120 degree (one third wavelength) sectionalized tower versus a standard 120 degree tower show the unwanted high-angle radiation at 60 degrees above the horizon to be nearly 2.5 times more for the standard, non-sectionalized tower. So why aren't sectionalized towers more commonly used today? Probably cost, and the fact that super long distance reception is irrelevant in today's markets. It's an idea relegated to the past as far as mediumwave broadcasting goes.

There are several variations of sectionalized towers within this type. The two main varieties are non top-loaded and top-loaded, both falling into the Type 2 designation. They may be fed at the bottom or the center of the tower depending on whether the tower is grounded at the base or not. Top-loading on a sectionalized tower has the same effect as the standard tower - increasing the effective electrical height without building a taller tower.

KNBC top-loaded Franklin tower, 1949

Now we have come to the Franklin antenna, actually a variation of the sectionalized tower and another Type 2 variety. One of the drawbacks to the efficient 5/8 wavelength tower described above is an unfortunate lobe of power pushed skyward at an approximate angle of 60 degrees. During daylight hours this hardly matters as it is absorbed in the atmosphere. At nighttime this skywave reflection can cause the station to interfere with itself out at certain critical distances within the broadcast target area.

The true Franklin antenna is composed of two half wave radiators mounted one above the other and insulated from each other. The tower base is insulated in the usual way. The difficulty is that the feed is in the middle of the tower instead of at the bottom. Feed line is run up inside the lower section, and a phasing network is used to feed both sections so that the currents in each are in phase, and the efficiency of two half-wave radiators is realized in terms of increased ground wave and decreased high angle signals. Thus, we have the Franklin antenna, a skyscraping one wavelength tall.

Several stations claim to be using Franklin antennas. There is one true Franklin antenna array in the U.S., that of California's KFBK-1530 in Sacramento. It consists of two towers, each a full wavelength in height, each fed in the middle. The claim is that 50KW daytime KFBK-1530 has the highest published millivolt per meter level of any station in the FCC AM system - an average of 3545.89 mV/m at 1 kilometer distance. It is the highest figure found in the FCC's database, at least.

Little known is that one other station, KSTP-1500, St. Paul, Minnesota, misses the perfect Franklin definition only by a hair, being just shy of a perfect full wavelength. Its single tower is comprised of two stacked 0.498 wavelength sections, making 0.996 wavelength height in all. It's computed (not FCC figure) millivolt per meter level is actually higher than KFBK's - 3618.76 mV/m. The difference is that the FCC publishes 1 kilometer field strength values for single towers based on a standard 1KW output, only 511.77 for KSTP. In actuality, 50KW KSTP-1500 is the champion.

The WHO-1040 tower (50 KW) out of Des Moines, Iowa,, a Type 2 center fed shortened "Franklin" is grounded at the base and fed at the center with only the top section excited. It is also top-loaded. It's curious omni-directional pattern pushes a mammoth 15.8 dB gain lobe out at 41 degrees elevation to the horizon, probably the highest gain lobe of any station in the U.S. The effective radiated power (ERP) in that lobe is some 2.1 million watts! There we have the high angle lobe described in paragraphs above. The single hop bounce off the ionosphere at this angle illuminates mid-America farm country with two million watts in a perfect ring around Des Moines ranging from about 150-350 miles out. Surprisingly, distant DX from WHO is not as good as it might seem, and in fact worse than the usual 50 KW monopole.

So, what does analyzing the FCC's tower data tell us? Here are the results.

Data has been compiled from the 09-26-2012 FCC database. 4744 station facilities are represented. Since the ASRN (Antenna Structure Registration Record) is not recorded in many cases, tower heights in this list are necessarily based on the FCC's figure of radiating height of the tower, which is always indicated. In almost all cases it comes very close to actual tower height, the difference being the base insulator and top lighting.

Daytime

4740 stations are on the air during daylight hours (99 listed as silent). 7176 towers are transmitting some 27,782,249 watts. How about that for an electric bill?

6666 are normal masted towers, that is, without top loading. 500 are top loaded. 10 are sectionalized.

The tallest daytime tower belongs to WKY-930, Oklahoma City, OK at 951 feet tall (290 meters). Interestingly, the older WKY tower collapsed in a freak tornado back in June of 1998, and there is video of its collapse taken from an adjacent tower cam during the storm.

There are 5 towers exceeding 750 ft. in height. They belong to WNAX-570, KMJ-580, WSM-650, WKY-930, WHO-1040.

The shortest daytime tower belongs to KJNT-1490, Jackson, WY (listed as silent at the moment), a mere 50 feet tall (15 meters). The next shortest belongs to KBZY-1490, Salem, OR, at 52.8 feet tall. It is a top-loaded affair, the top-loading upping its effective height to a perfect quarter wavelength!

In all, there are 12 towers shorter than 75 feet. They belong to KYPA-1230, KCFM-1250, WKCY-1300, KFJL-1400, WLUX-1450, WEEO-1480, WIRB-1490, KBZY-1490, KLZN-1490, KJNT-1490, KVOG-1530, WPDC-1600.

Out of 4740 daytime stations, 3554 have omni-directional patterns, or a single tower. 1214 have shaped patterns using multiple towers.

The total height of all towers broadcasting during daylight hours is 1,775,871 feet (541,286 meters), or 336 miles (541 kilometers) of metal antenna structure up in the air!

Again, the highest daytime published millivolt per meter level goes to KFBK-1530 of Sacramento, CA with 3545.89 mV/m. KSTP-1500's actual calculated level is higher at 3618.76 mV/m.

The lowest daytime published (and actual) millivolt per meter level goes to two-tower, 250 watt WCTA-810 of Alamo, TN with 140.82 mV/m.

Nighttime

4158 stations are on the air during nighttime hours (81 listed as silent). 7689 towers are transmitting some 10,956,214 watts. Quite a drop in power.

7,335 are normal masted towers, that is, without top loading. 525 are top loaded. 9 are sectionalized.

The tallest tower "in use" day or night goes to nightime WRDT-560, Monroe, MI at 992 feet (302 meters). An interesting situation, as daytime and nighttime services transmit from different locations. WRDT-560 daytime service of 500 watts transmits from Monroe, MI (near Detroit) on 4 "short" towers of the same height (each 410 ft.). Nighttime service (a puny 14 watts) transmits from the huge 992 ft. Detroit Metro Media Center tower in suburban Oak Park. FCC records show WRDT uses the entire length of the tower to radiate its 14 watts. Even at the operating frequency of 560 KHz, the 992 ft. tower is still just 0.565 wavelength. The media tower is obviously used for multiple purposes.

There are 6 towers exceeding 750 ft. in height. They belong to WRDT-560, WNAX-570, KMJ-580, WSM-650, WKY-930, WHO-1040.

The shortest tower belongs to the nighttime operation of WSIV-1540, E. Syracuse, NY at 31.9 feet (10 meters) which appears to be operating out of a house in a residential neighborhood. It is not top-loaded either, transmitting a signal of 57 watts into a 0.050 wavelength radiator. It appears that nighttime WSIV-1540 takes the prize for the shortest wavelength antenna too.

There are 13 towers shorter than 75 feet. They belong to KYPA-1230, KCFM-1250, WKCY-1300, KFJL-1400, WLUX-1450, WEEO-1480, WIRB-1490, KLZN-1490, KBZY-1490, KJNT-1490, KVOG-1530, WSIV-1540, WPDC-1600.

Out of 4158 nighttime stations, 2530 have omni-directional patterns, or a single tower. 1655 have shaped patterns using multiple towers.

The total height of all towers broadcasting during nighttime hours is 1,965,079 feet (598,956 meters), or 372 miles (599 kilometers) of metal antenna structure up in the air! Nighttime wins over daytime with an extra 36 miles of antenna tower.

Wrap Up

Several things are apparent from these statistics. Fewer stations (582 fewer) are on the air during nighttime hours, but they use 513 more towers than the daytime group. This is because of the need for signal directivity to avoid interference.

The total power output of the nighttime group is only 39% of the daytime group, again for the same reason.

Nighttime KFBK-1530 of Sacramento, CA this time has the absolute claim to the highest computed millivolt per meter level: 3126.79 mV/m. KSTP-1500 switches to a 3 tower array at night and loses its place.

The lowest nighttime published (and actual) millivolt per meter level goes to two-tower, 1 watt WZRK-1550 of Lake Geneva, WI with a micro field strength of only 9.11 mV/m. WZRK takes the overall prize in this category.

The next time you take a drive, look for mediumwave towers. They are an interesting subject! More from the west coast. I am headed across country soon.

A nice shot of the WSM-650 tower as it appears today.

Using Arbitron To Determine Station Format

When DXing, it's often helpful to know the broadcast format of stations on frequency for identification purposes. I recently had the case where I was DXing at a remote site on 1470 KHz up along the Canadian border. Two stations were down at the noise level, one slightly above the other. The stronger one was broadcasting in country music format, the other undetermined. I couldn't quite catch a call sign.

When I got home it was a simple matter to check the Arbitron data on the two possibilities and determine which one I heard. It turned out it was WPDM-1470, out of Potsdam, NY. The weaker station was WNYY-1470 out of Ithaca, NY. The Potsdam catch was the better of the two, and at lower power too. I would not have been able to determine this without the ability to check station formats.

I've covered the availability of station lists for mediumwave DXers in two previous articles on RADIO-TIMETRAVELLER, Mediumwave Station Reference Lists and Mediumwave DX Meets The Tablet Computer. Many of these lists don't document station formats. Those that do may be out of date, or incomplete. And that's just for US data - information for Canada is even more suspect. This is not surprising considering the frequency at which stations change formats.

Those lists that do document station formats are Lee Freshwater's AM Logbook (perhaps the most complete), Topaz Designs, and presumedly the National Radio Club's AM Radio Log (though at a steep price - currently $28.95 for non-members and out of print). radio-locator - technically not a list site, but you can generate a list from this link using search parameters - will also provide a nice custom list of stations which includes format information. It seems fairly complete.

Virtually all US and Canadian stations subscribe to Arbitron. Arbitron, of Columbia, MD, is an international media and marketing research firm serving all types of media — radio, television, cable, etc. It is Arbitron's business to measure network and local market radio audiences across the United States, survey the retail, media and product patterns of U.S. consumers, and provide measurement and analytics. This includes, of course, what formats stations are using. It is probably the most up-to-date format reference of all.

The Arbitron Radio Station Information Profiles (SIPs) contain information about all radio stations in the United States: current addresses, station names (including call letters), frequencies and formats. Many if not virtually all Canadian stations are also represented in the profiles.

So how do we use Arbitron to identify station formats? Though the Station Information Profile used to be made directly available on their web site, lately the link has been moved to a members only site, My Arbitron. However, still buried deep within the Arbitron's main web site are a couple of links which can get us our needed information.

Stations are asked to update their Arbitron reference data four times per year. This goes into a seasonal "survey". We must take an educated guess of the current season on file that we wish to extract information from. Survey codes are:

• WI (winter)
• SP (spring)
• SU (summer), noted as SU12 in the example (12 = current year, 2012)
• FA (fall)

Using the call letters of the station, compose a link for the AM broadcast service as such:

http://www1.arbitron.com/sip/displaySip.do?surveyID=SU12&band=am&callLetter=WHAM

Note the use of the surveyID=SU12 (summer of 2012), and callLetter=WHAM fields. You can also check FM stations by substituting the band=fm field.

A station search page is accessible as well, if you know the current survey season and date:

http://www1.arbitron.com/sip/reviewSip.do?srvy_id=SU12

Old surveys do not seem to be accessible. Use the current season and year, or sometimes the season just past if the current season is not yet up for inspection.

Keep trying with your link until you have success. I have found the Arbitron site to be quite busy at times. During busy periods, it also may return a "survey not found" message. Persistence is good.

The Silicon Labs 477x Chip - A New DSP DX Champ?

Silicon Labs recently released a new "high performance" consumer electronics broadcast radio receiver DSP chip, the Si4770/77 A20 (477x) series. What makes this chip special or any better than the familiar Si4734 chip used in the Tecsun PL-380, PL-310, and other consumer DSP-based receivers? I recently downloaded the spec and programming sheets on the new Si477x series and had a look. This chip set has some interesting new features.

A Sensitivity Boost

One thing was immediately apparent: the new chip has received a sensitivity boost. Where the old Si4734 chip's AM sensitivity was pegged at 25µV (27.9dBµV), the Si477x's comes in at 14µV (22.9dBµV). AM adjacent channel (±9 KHz) rejection for the new chip is 62dB, alternate channel (±18 KHz) is 62dB. AM image rejection is 72dB. Impressive figures for a chip radio.

FM sensitivity has also improved from 1.1µV (Si4734) to 0.66µV (-3.5 dBµV, Si477x). Selectivity is pegged at 65dB at ±100 KHz and 72dB at ±200 KHz frequency offset.

Bandwidth Options Galore

The old Si4734 chip had 7 programmable bandwidth options, the ability to set AM IF bandwidth at various widths from 1 - 6 KHz. Five AM bandwidths were implemented in the Tecsun PL-380: 1, 2, 3, 4, and 6 KHz. The Si477x chip enters with what Silicon Labs is calling "Dynamic AM/FM channel bandwidth control". The AM and FM IF channel bandwidths are dynamically optimized according to the on-channel RSSI (signal strength), and with the ability to further refine bandwidth by detection of the adjacent and alternate channel RSSI. Channel filter bandwidth is programmatically settable from 0 Hz to 15 KHz in 100 Hz steps for both AM and FM. Further, automatic bandwidth selection can be triggered not only by RSSI, but by signal to noise ratio. It can also be hard set to a fixed width by software. This could present some interesting possibilities if full user control is ever implemented in a consumer device.

Uh-oh, Soft Mute Again

Once again, programmable audio soft mute is implemented in the new chip. Soft-mute, a further lowering of the audio level of the received signal when it drops below a prescribed strength, is undoubtedly meant to provide a more comfortable listening experience for the casual listener and not the DXer. The idea is to relieve the listener from all that nasty low level "static" and "interference", or as Silicon Labs states: "....to attenuate the audio outputs and minimize audible noise in compromised signal conditions." Soft mute attenuation in the Si477x is available in 0-31dB steps (default is 12dB of attenuation, unfortunately twice the amount of the Si4734), and can be triggered at a pre-programmed S/N ratio (default is 8dB, the Si4734 was 3dB).

New Audio Bandpass Filters

Silicon Labs has deemed these "Hi-cut" and "Lo-cut" filters. The AM hi-cut control is employed on AM audio outputs having degradation of signal quality. Signal strength and signal-to-noise ratio thresholds can be programmatically set to activate the filters. The Hi-cut filter can be disabled by setting the Hi-cut filter  margin to the default audio bandwidth for AM. AM Lo-cut is employed on audio outputs for rejection of power-supply 50/60 Hz interference. AM Lo-cut is a high pass filter, and is enabled by default and can be disabled by programming. Note that the Lo-cut filter is not available for FM.

HD Radio

For HD Radio (IBOC) buffs, the Si4777 model of this chip series has AM/FM HD Radio capabilities. Additionally, an interesting innovation has been added called "IBOC Blend". This feature supports the ability to blend between analog and digital audio. When the bit error rate of the HD Radio digital signal falls below a predefined threshold and the digital audio fades out, the analog audio is blended in. This prevents the received audio from muting when the digital signal is lost. The audio will "blend to digital" upon reacquisition of the digital signal.

It will be interesting to see if the HD version of the Si477x chip is incorporated in any new, portable consumer electronics device. Car radio seems to be the only outlet anymore for HD Radio, which may die on the vine in the end anyway.

Improved Inductance Handling

Usable antenna inductance has expanded somewhat from 180-450µH to 180-540µH, though the documentation does show the chip capable of maintaining tuning up to a 688µH inductor. Again, like the Si4734, the chip is adaptable to an external air loop antenna if a ferrite rod is not used. An air loop antenna is supported by using a transformer to increase the effective inductance of the air loop. Using a 1:5 turn ratio inductor, the inductance is increased by 25 times and easily supports all typical AM air loop antennas which generally vary between 10 and 20µH.

Frequency Coverage Suffers

Unfortunately, shortwave and longwave band coverage are not available, unlike the Si4734. FM range is 64 MHz - 108 MHz. AM range is 520 KHz - 1710 KHz, at 9 or 10 KHz channel spacing.

Summary

The Si477x is Silicon Labs' cutting edge consumer electronics broadcast AM/FM receiver radio chip. It brings us higher sensitivity for both AM and FM. IF Bandwidth is now dynamically set dependent on signal conditions, using primary, adjacent, and alternate channel signal information. Soft mute is still in force, and is even more heavy handed when left to its default settings. New high and low audio bandpass filters shape the audio signal and are dynamically set by signal conditions. HD Radio for both AM and FM is made available in one version of the chip series. Finally, ferrite loopstick inductance range is expanded.

It is likely that consumer grade AM/FM radios using this chip will stick with default settings for bandwidth and soft mute, basically rendering the increased sensitivity and enhanced DX capability of this radio chip nearly moot. Opening up the myriad bandwidth options to manual selection, allowing soft mute to be manually controlled, and further, allowing the audio bandpass filters to be manually controlled would render this a nice DX machine for both AM and FM.

Market Watch Report

RADIO-TIMETRAVELLER Maps #3

1 Watt At Night - U.S.

1 Watt At Night - U.S.

The ultimate DX. How good (or lucky) a DXer are you? Have you ever tried for extremely low power AM stations at night?

Every night, 15 stations across the U.S. transmit with an output power of only one watt. How far does one watt travel normally, let's say under static conditions during the daytime? I'm glad you asked.

One watt into an omni-directional antenna (1 tower), over medium ground conductivity (8 mS/m), will get you out 10 miles, at which point the signal enters the "fringe" area of reception (0.15 mV/m and less). A good receiver will hear somewhat farther than that. At night, and using a directional pattern, who knows how far that signal will travel? All things are possible.

-----------------------------------------------------------
CallSign       Freq    Power Class Location                     
-----------------------------------------------------------
WCKB            780        1    D  Dunn, NC                     
WRFM            990        1    D  Muncie, IN                   
WHFB           1060        1    D  Benton Harbor-St. Jo, MI     
KLEY           1130        1    D  Wellington, KS               
WGAB           1180        1    D  Newburgh, IN                 
WSQR           1180        1    D  Sycamore, IL                 
WPGR           1510        1    D  Monroeville, PA              
WENG           1530        1    D  Englewood, FL                
WJJT           1540        1    D  Jellico, TN                  
KBOA           1540        1    D  Kennett, MO                  
WBNL           1540        1    D  Boonville, IN                
KLKC           1540        1    D  Parsons, KS                  
KDYN           1540        1    D  Ozark, AR                    
WZRK           1550        1    D  Lake Geneva, WI              
WSRY           1550        1    D  Elkton, MD

Here's some statistics for our 15 one watt stations, taken from the 5-27-12 FCC database:

All 15 are Class D.

12 are located east of St. Louis, 3 are west of St. Louis.

KLEY-1130, Wellington, Kansas, is the most westerly, leaving no 1 watt stations from there all the way to the west coast.

11 are "W" callsigns. 4 are "K" callsigns.

The 15 stations broadcast using 24 towers.

11 stations broadcast an omni-directional pattern (single tower). 4 broadcast a directional pattern (multi-tower).

1 station is on the FCC "Silent" list: WZRK-1550.

Interesting to note: 5 of the 15 are on the same frequency - 1540 KHz.

The lowest powered daytime station in the FCC's database broadcasts with 135 watts.

The most interesting broadcast pattern is that of WRFM-990, Muncie, Indiana. This station broadcasts using 6 towers. Its pattern projects generally to the north, covering Muncie and Delaware County as shown in the radio-locator graphic. WRFM-990's daytime power is 250 watts.