Chapter 2, Section 14
Part of AM Radio Facts and Trivia: The AM Broadcast Band in the USA. Updated November 21, 2026.
Walk up to most AM radio towers and look at the bottom. The steel does not touch the concrete. It stands on a block of glazed porcelain, sometimes no taller than a person, that carries the whole tower: tens or even hundreds of tons. That block is the base insulator. This page covers the styles, the one company still making them, how engineers size them, what crosses them, how they fail, and the stations that show it best.
This page builds on Inside the AM Tuning House: Base Insulators and ATUs (Chapter 2, Section 4) and Feeding an AM Tower (Chapter 2, Section 5).
Most AM towers are series fed: the steel tower itself is the antenna, and the transmitter’s radio energy (RF) is fed in at the bottom. The tower works against the buried ground radials, so it must not touch the earth. The base insulator keeps the two apart while it carries the tower’s weight. Everything above the porcelain is “hot” with RF, which is why the base sits inside a locked fence.
Insulated towers of this kind came into use in the United States in the early 1930s. Some insulators from the mid-1930s are still in service today.
1. One cone under a guyed tower. A guyed tower (one held up by guy wires) stands on a single cone or cylinder of porcelain. The porcelain is in compression, meaning it is squeezed, which porcelain handles very well. The larger units sit on a rocking plate so the tower can sway a little in wind without bending the porcelain.
2. A set under each leg of a self-supporting tower. A self-supporting tower has three or four spread legs and no guys. Wind can push one leg down while it pulls another leg up, so each leg needs an insulator that takes both: download (weight pushing down) and uplift (wind pulling up). One common design is two porcelain pieces held in a steel frame, the lower one carrying the weight and the upper one holding the leg down. Austin Insulators builds theirs with ceramic for the weight plus tension straps, tightened at the factory, for the uplift.
3. The Blaw-Knox pivot. The famous diamond-shaped Blaw-Knox towers of the 1930s, such as WSM and WLW, balance their narrow bottom point on a single insulator shaped like a ball and socket, which is why they need guy wires at their wide middle. A 1936 Lapp Insulator patent by Ralph Jenner, who also helped invent the diamond tower, shows a base of two porcelain cones set point to point, meeting at a joint that lets the tower move while the porcelain is only ever squeezed.
Drawing 2.14-1
Every source found for this page describes glazed porcelain, a fired ceramic. The ribs around the outside (sheds) make the path along the surface longer, so rain and dirt are less likely to let electricity creep across. Austin’s insulators are hollow and filled with oil. The company says the oil keeps water from condensing inside, and makes a crack easy to spot because the oil leaks out. Austin also designed a “safety sleeve” insulator meant to keep a tower standing even if the outer ceramic broke.
Austin Insulators began in 1933, when Arthur O. Austin left the Ohio Brass Company and started his own firm in Barberton, Ohio. It was later bought by Decca and moved to the Toronto area, and its managers bought it in 2000. In 2017 its owner told Radio World it was the only maker of radio insulators left in the world. It is the only maker found still selling AM tower base insulators in 2026. (Sources differ on the early history: one says Austin died in 1964 and his heirs closed the firm, the company says Decca bought it in 1969.)
Lapp Insulator of LeRoy, New York, made the giant insulators of the 1930s, including WSM’s. Lapp was still repairing broadcast base insulators around 2010, but no source shows it selling new ones today.
Austin publishes its guyed tower line. “Working load” is the most the insulator should carry in the worst wind and ice expected, not on a calm day. Every rating has a safety margin of at least 3 to 1. The voltage rating is tested with water dripping on the porcelain; the second figure is with an optional rain hood.
| Model | Ceramic pieces | Working load, pounds | About tons | Wet voltage rating, kV (standard / with rain hood) |
|---|---|---|---|---|
| A-4197-L | 1 | 40,000 | 20 | 28 / none |
| A-4722-B | 1 | 80,000 | 40 | 39 / 53 |
| A-0881 | 1 | 150,000 | 75 | 67 / 95 |
| A-0729 | 1 | 200,000 | 100 | 67 / 95 |
| A-0167 | 1 | 270,000 | 135 | 67 / 95 |
| A-3663-B | 3 | 375,000 | 188 | 71 / 99 |
| A-4447-B | 5 | 625,000 | 312 | 71 / 99 |
| A-3820-R | 8 | 1,000,000 | 500 | 71 / 99 |
| A-4544 | 12 | 1,500,000 | 750 | 71 / 99 |
| A-4598 | 21 | 2,500,000 | 1,250 | 71 / 99 |
Table 2.14-2
Austin’s self-supporting line runs from 25,000 pounds down and 17,500 pounds up to 900,000 pounds down and 600,000 pounds up. Heights, weights and prices are not published. Austin also makes exact-fit replacements for other makers’ insulators.
Weight. A guyed tower’s insulator carries the tower’s own weight, plus the downward pull of every guy wire, plus ice and wind. The guys alone add a lot: each one is tightened to about 7 to 15 percent of its breaking strength, and all of that pulls down on the base. Insulators are built with big margins. WGY’s 1936 insulator was tested until it broke at a little over 1,200,000 pounds, about three times its working load, the same 3 to 1 margin Austin uses today. The tower design standard, TIA-222, lets a tower designer count only 40 to 50 percent of an insulator’s breaking strength. Insulators take very little sideways push, only 3 to 6 percent of their weight rating, which is another reason guyed towers sit on a pivot or rocking plate.
Voltage. The taller the tower in wavelengths, the higher the RF voltage across its base. A tower one quarter wave tall (90 electrical degrees) is easy: at 50 kW it puts about 5,500 volts across the base at the highest modulation peaks. A 140 degree tower has a much higher base impedance and puts about 21,400 volts across the same porcelain. Austin recommends running at no more than 60 percent of its wet voltage rating, and less near salt air, fog, dust or insects. For short towers, weight decides which insulator to use. For tall towers, voltage starts to matter too.
Drawing 2.14-3
Capacitance. The insulator and everything bridging it acts like a small capacitor across the base of the tower. On a short tower it hardly matters. On a tall tower it can change the base impedance a great deal, so engineers count it when they design the tuning unit. In one published example, about 100 picofarads across the base of a 150 degree tower turned its 782 ohms into 623 ohms plus a large reactive part, while a 90 degree tower barely changed. No maker publishes a capacitance figure for its insulators.
The FCC rules. The FCC rules for measuring a station’s antenna resistance and power say only “at the base of the antenna.” The insulator itself shows up in the 2008 rule that lets directional stations prove their pattern with a computer model (47 CFR 73.151(c)). That rule limits the base region capacitance the model may assume to 250 picofarads unless real values are used, and lets engineers measure the feed “with its insulator short circuited.” One engineering firm tells its crews to measure right across the insulator and to write down the insulator’s model number.
The insulator separates the tower from the ground, but several things still have to get across it safely.
Drawing 2.14-4
Guy insulators. Steel guy wires are also broken up by porcelain insulators, often egg-shaped “Johnny Balls,” into pieces no longer than about one tenth to one eighth of a wavelength (about 71 feet at 1380 kHz), so the guys do not act as extra antennas and bend the pattern. After a 2015 windstorm damaged about a dozen guy insulators at KLIZ in Brainerd, Minnesota, its night pattern phases shifted about 2 degrees.
Warning signs are cracks, dirt, carbon tracking (black burned trails left by small arcs across the surface), lightning marks, an oil leak on oil-filled units, and a sudden change in the tower’s impedance or reflected power. Once water and ice get into a crack, it can grow until the insulator fails. Keep paint and wet rags off the porcelain. The best insurance on record is simple: a spare insulator in storage, and a close look at the base after every ice storm and windstorm.
Documented failures in the United States are mostly cracked or broken porcelain from wind, ice or contamination, not from the RF itself.
| Station, place, year | What happened | Sources |
|---|---|---|
| KLOE 730, Goodland, Kansas, December 2025 | Two of four porcelain base insulators broke in high wind on December 18, a third the next weekend. The tower leaned, the station went off the air, and the tower fell into a wheat field on December 27 or 28. Replacements were “not readily available,” there was no insurance, and the owner chose to shut the station down. | 2 |
| WMVA, Martinsville, Virginia, December 1994 | Ice, then high wind, on a 425 foot self-supporting tower: three of four base insulators cracked or broke. A tower crew fitted a spare that was kept in storage. | 1 |
| WIP, Philadelphia, about 2010 | A 300 foot Blaw-Knox tower twisted in a spring windstorm and fractured a base insulator. A spare went in; Lapp repaired the original in about six months. | 1 |
| KKPZ, Portland, Oregon, before 2011 | Paint overspray on two insulators was wiped with a wet rag. On the next foggy morning they steamed and held an arc until the transmitters were shut off. The cracks went below the glaze. | 1 |
| KYA, San Francisco, February 1986 | The tower fell in high wind, “possibly” because a leg insulator had been replaced using a threaded rod too short to reach its nut. | 1, hedged |
Table 2.14-5
Replacing one. A guyed tower is the easier case: the crew eases the guys and lifts the tower with a crane or jacks. A self-supporting leg cannot simply be lifted, because a gust could tip the tower once the bolts are out. At KKPZ, a tower crew built brackets that held the leg both up and down, raised it about one eighth of an inch, rotated the old insulator out and a salvaged one in, and bolted on an adaptor a structural engineer designed because the bolt pattern was different. It took several days of calm weather, after a year of watching the cracks grow. The station expected to spend “a couple of thousand dollars,” using salvaged parts; new replacements were called “very expensive.” No current prices are published.
| Station | Tower | Base insulator | Sources |
|---|---|---|---|
| WSM 650, Nashville | 808 ft Blaw-Knox diamond, 1932 | Two-section Lapp pivot insulator built for 1,200,000 pounds. The station’s engineer estimated over 300 tons sit on it, but was not certain. | 2 |
| WLW 700, Mason, Ohio | Blaw-Knox diamond | Still stands on its original 1930s insulators. Sources differ on the tower height (736 or 747 feet) and on 1933 or 1934 for its date. | 1 to 2 |
| WGY 810, Schenectady, New York | 625 ft guyed tower, 1936 and after | Its insulator was tested until it broke, at a little over 1,200,000 pounds, against a working load of about 430,000 pounds: about 3 to 1. A concrete wall went around it in 1942 against wartime gunfire. | 1 |
| KHQ 590, Spokane, Washington | 805 ft tower, early 1940s | Stood on an Austin oil-filled insulator that was heated to cut electrical leakage in damp weather. | 1 |
Table 2.14-6
The base insulator is 1930s technology that has outlived most of its makers. With one company left and many towers standing on Lapp or early Austin units no longer made, a single cracked insulator can force a hard choice: scavenge a used part, build an adaptor, rebuild the tower as a folded unipole (wires fed down the outside of a grounded tower, so the base can be bolted straight to the concrete), or go silent. A folded unipole needs no base insulator, no lighting transformer and no static choke, and lets the owner rent the tower to other users such as cellular companies. KLOE’s owner priced one after the December 2025 collapse, and chose silence instead.
Sources: Austin Insulators product pages (guyed bases, self-supporting bases, specifications, history, transformer history), retrieved November 2026; Cris Alexander, “Two Out of Four Must Go,” Radio World, August 24, 2011 (KKPZ); John Bisset, Radio World, January 10, 2012 (WMVA) and March 12, 2012 (WIP); Radio World, December 22, 2025, and RadioInsight, December 30, 2025 (KLOE); James O’Neal, “Arthur O. Austin’s Continuing Legacy,” Radio World, April 21, 2017; Paul Thurst, Engineering Radio blog, March 2010 (WGY tower) and September 15, 2011 (base voltage, lighting transformers and chokes); Jeremy Ruck, Radio World, September 15, 2016 (base impedance and capacitance); James Hatfield, NAB Engineering Handbook, 1999; Hatfield and Dawson measurement guide; Mark Persons, Radio World, March 7 and May 3, 2017 (guy insulators, arc gaps); The Broadcasters Desktop Resource, 2011, 2012 and 2015; Richard Fry, Topband list, January 16, 2014; 47 CFR 73.51, 73.54 and 73.151(c) (eCFR); FCC 17-119 (2017); TIA-222-G explainer (Tower Numerics); US Patent 2,064,121 (Lapp, 1936); oldradio.com and The BDR (WSM); Fybush.com Site of the Week, November 11, 2011 (WLW); Wikipedia (Blaw-Knox tower, list of mast collapses, folded unipole). A note on the facts: the Austin catalog figures come from one source, the company itself. The WGY, WMVA, WIP, KKPZ, KHQ and KYA stories each rest on one published account; the KLOE and WSM facts have two. The base voltages are from one published calculation, checked on this site. Sources disagree on Austin’s early history, WLW’s height and date, and the proof-test level in TIA-222-G (60 or 45 percent).
© 2026 Chris, N6JET. Free to read and share for non-commercial use with credit (Creative Commons BY-NC 4.0). Not for sale.
All AM radio stories by subject · AM radio interactive tools and simulators · AM Trivia Challenge quiz