Broadbanding an AM antenna

Many articles have been written on the topic and it is still a black art to some.  Making a Medium Frequency (MF) antenna that has enough bandwidth to pass 10 KHz audio can be challenging, to say the least.  The VSWR out to +/- 15 KHz carrier needs to be kept at a minimum and the power needs to be evenly distributed between the two sidebands.  This can become problematic with complex Directional Arrays or towers that are tall or short for their operating frequency.

When we were working on the WFAS-AM tower in White Plains, NY, it became apparent to me that something was not right.  The tower is skirted and now holds the antenna for W232AL, a 250 watt translator broadcasting the WPLJ HD-2 channel.  After installing the FM antenna, some tuning of the AM antenna was required and this is the graph of the resistance and reactance curves:

WFAS 1230 KHZ, ATU output resistance and reactance
WFAS 1230 KHZ, ATU output resistance and reactance

This looked very similar to the resistance and reactance curves before the FM antenna work was done.  Red line is resistance, the blue line is reactance.  I think it had been like this for a long time.  While it is not terrible, it is not that good either.  As alluded to in a previous post, some re-working of the ATU was needed.  After some trial and error, this is the circuit that we ended up with:

WFAS 1230 KHZ White Plains, NY ATU schematic
WFAS 1230 KHZ White Plains, NY ATU schematic

Not quite what I expected, however, it was designed with the parts on hand, excepting the vacuum variable output capacitor, which was donated by me.  That part was key in making the proper adjustments.

After my redesign and tune up of the ATU, this the resistance and reactance curves at the input terminal of the ATU:

WFAS 1230 KHz resistance and reactance after ATU modification
WFAS 1230 KHz resistance and reactance after ATU modification

The graphs have a slightly different format, but you get the idea.  The red line is resistance, the blue line is reactance and the green line is overall impedance.  The resistance is symmetrical about the carrier as is the reactance.  Truth be told, I think there is a little more that can be had here, but for now, there is no reason to go any further.  I made the initial measurements at the input of the ATU and confirmed them again at the output terminals of the transmitter.  When we turned the transmitter back on, I noticed that the modulation index had dropped by about 15 percent.  I think the reflected power was getting back into the RF sample and fooling the mod monitor.  I also noticed that the high end in particular sounded much nicer.

WFAS 1230 KHz, White Plains, NY ATU
WFAS 1230 KHz, White Plains, NY ATU

The ATU building is a little cramped and it is hard to get a good picture.  The vacuum variable capacitors were salvaged from a scrapped AM transmitter years ago.  The tower is 202 degrees tall, which is also a factor.  It will be interesting to see what seasonal changes there are with snow cover, mud, etc.

Overall, this was a fun project.

9 thoughts on “Broadbanding an AM antenna”

  1. Nice work! There’s always another theoretical fraction to pick up somewhere, but on a real world tower and ATU, that’s as good as it gets.

  2. Mike, Age and neglect have taken their toll, that is for sure. I think at least half of the AM stations need some sort of attention, but there is little interest from the current owners on improving their plants. Just a small part of the bigger problem

  3. I’m really enjoying your articles on AM antenna systems. Right now, I’m in the middle of a capital project to rehabilitate the antenna tuning units at one of the stations under me. The station is non directional day, and directional at night using 5 towers. The tuning sheds were built in 1950, and were no longer weather tight. The original Gates ATU shelves were a playground for mice, and were physically falling apart. We’ve replaced all 5 buildings, and each ATU is being disassembled, cleaned, and reassembled with new hardware. For the most part, they are still the original Gates components, although they were reconfigured in the 1970’s and again in the 1990’s.

  4. Jon;
    I did the same project some years back with an array built by Andrew in the 1940’s.
    One of the doghouses was 45 degrees sideways and heading for the swamp, Cracked base insulators, missing guy insulators, mistuned, overheating ATU’s, shorted obstruction light wiring… you name it, it was shot.
    Everything got disassembled, refurb’d or replaced, one part at a time.
    Those are fun projects, IMHO. When you aren’t pulling your hair out. But the end result should be good for another 50 years!

    Mike, For every one of these we clean up there are probably 10 more that are in similar or worse condition. Unfortunately many will probably just keep getting worse until Mother Nature reclaims the tower steel.

  5. Looks like a fairly new Phasetek install.
    Have one just like yours skirted tower & all (it’s a Phasetek/Knott). But it’s 5 kw along with 3 shorter towers for DA night.
    Would LOVE to have someone dig into it. But alas it’s not a money maker any longer (format “moved” to a simulcast FM) and not a priority.

  6. Bill, actually the ATU was made by Continental Electronics in 1986. This is a class C AM, but the 2.5 Mv contour has at least 1.5 million population. Back when there was an FM station associated with the AM, the AM was considered mostly a throw away. “Its just the AM,” was a common refrain. Now that the FM has moved further south, “Its just the AM” has taken on a totally different meaning.
    Chuck/Jon, these can be fun projects if you have the time and the owners agree to pay for the needed parts.

Leave a Reply

Your email address will not be published. Required fields are marked *