December 2011
« Nov   Jan »



Old Year SWR

This time of year is when we all sit back and asess things that we did in the past 365 or so days. It is called reflection, which is just a civilian term for SWR (Standing Wave Ratio).

Thus, I though I would take a little time and make a few observations about the business, my part in it, and this blog.

1.  The business of Radio:

Let us be honest, Radio is not what is used to be.  Many times, what it used to be was somewhat of a free for all, wheeler dealer radio station owners cutting corners and making do with less than optimum equipment and staff.  And trade, lots and lots of trade. Only in large metropolitan areas did radio stations make enough money to throw it around, but sometimes not even then.  Radio was by no means a huge money making operation and therefore, those that worked in mostly it did it as a labor of love.  That may or may not have come across on the air.  By far, the funnest station I ever listened to was run from a closet, with a sound reinforcement board and the program directors CD collection.  What made it so much fun was they had nothing to loose, there were no restraints placed on the staff.  Once that on air enthusiasm translated to ratings, then to revenue, the magic was gone and they were just another radio station filling a spot on the dial.

The radio business has fully transitioned from a fun, seat of the pants entertainment operation to a mega money making corporate mentality under the control of mostly non-entertainment types.  Even those stations owned by smaller group owners are forced to rely on the tactics developed by the big two in order to stay in business.

Group owners will continue to extract money in whatever way they can until the money train runs off the rails.  Then, radio will be replaced by something less.

2.  Radio Engineering:

Engineering will continue to grow smaller, with more emphasis on computers, networking, and IT infrastructure.  The future distribution of music and program material will take the form of streaming (live events), pod casts (specialty shows) and subscription services.  Over the air free radio will become less and less relevant as younger “listeners” trend toward new media.  The idea of listeners may be archaic in lieu of “subscribers” or “users.”  Thus, in order to remain relevant, broadcast engineers are going to have to keep their skill sets current.  I would recommend to anyone getting into the business to get current with routers, routing tables, Cisco equipment and whatnot.  The cloud is coming and will rain on all those not adjusted to the new “broadcasting” reality.

3.  My part in the business:

A somewhat superannuated broadcast engineer who’s skill set lies mostly within the RF and heavy duty electrical areas, I am going back to college in January.  Cicso Network Administrator is the degree I am shooting for, for that is where the local jobs, both in and out of broadcasting will be.  Network Administrators are going to be the backbone of cloud computing, those that can configure routing tables will be desired.

That being said, I continue to be involved with larger RF projects and transmitter work.  It is fun for me, most of the time.  Having to drive two hours,one way on Christmas Eve to fix a backup transmitter, not so much, but those situations tend to be the exception, rather than the rule.

All in all, it is great fun to press the high voltage on button, not knowing if the transmitter will cycle on normally, or put on some type of display.

4.  The blog:

This little thing we have here has been fun.  I get good response to most articles.  I welcome all the comments and the off line e-mails that come my way.  My original intent, which is to provoke thought and dialog, remains unchanged.  This year, I have delved into areas not covered by the trade magazines, but do have at least some bearing on radio or radio related arts.  To that end, there have been several negative responses, which is fine.  I don’t pretend to know everything, if you know more, then by all means, speak up.  By and large, however, the majority of responses continue to be positive.

I continue to grow the overseas audience, with roughly 36% of the page views coming from non US IP addresses.  Persons from The UK, followed by Canada, Netherlands, Australia and Germany are the top five non-US readers of this blog.

So, I will continue to post about things in the coming year.  If any of you have any suggestions or requests, shoot me an email of leave a comment.

In the mean time, have a Happy New Year!

Synchronized FM signals

How effective are they at filling in or expanding coverage for FM stations?  The answer is, it depends.  Most have heard of the quadcast around New York City on 107.1 MHz formed in 1996-98.  It was well documented in Radio World and several other publications as a cleaver way to overcome the suburban rim shot problem.  Four signals on 107.1 were synchronized using GPS timing data, then fed the same program material.  They were WYNY, Braircliff Manor, NY; WWXY, Hampton Bays (Long Island), NY; WWYZ, Long Branch, NJ; and WWYY Belvidere, NJ.  These being four separate Class A FM stations, the 60 dBu contours did not overlap.  There was some mutual interference in some areas, but there were few if any reception negative zones where the signal strength is equal between stations.

In early 2003, I was a part of the disassembly of the quadcast.  In the end, it is difficult to point to any one thing that lead to the breakup.  The station’s owners, Big City Radio, had filed for bankruptcy.  I am not sure if the company ever had the correct formula for marketing and sales, given the strong suburban, but weak and lacking building penetration in Manhattan signal.  The station initially had a country format, something that arm chair quarterbacks said would not work in New York City.  After a few years, Big City had changed the format to Rumba, a Spanish/Caribbean music format, which did worse than Country.  The fact is, that it never lived up to expectations and the station were worth more separately than together.  Given the right circumstances, it could have worked.

The other synchronized FM broadcasts are those where boosters are employed.  These are a good deal more difficult to configure because the booster signal is within the main stations 60 dBu contour.  Often cases, where there is severe terrain shadowing or other limitations, a well positioned booster that is in a population center can greatly improve the signal in those areas.  This was formerly the duty of an FM translator, however, those stations seem to be taking on a life of their own, without regard for the intent of the current FCC rules.  Boosters can also be called a single frequency repeater or single frequency network (SFN).

The disadvantages of a SFN are the aforementioned negative reception areas.  To the receiver, this will create a multipath or picket fencing situation, which is objectionable to most listeners.  The advantages are, of course, better coverage in key areas, spectrum efficiency, and the ability to create a network of common frequency systems.  Think of how easy it would be if all NPR stations were all on the same frequency, for example.

The key to making a booster work is to synchronize several aspects of the RF and Audio signals:

  • RF carrier frequency
  • Stereo pilot frequency and phase
  • Audio amplitude and phase
This is best done using GPS receivers to synchronize the exciters and an AES/EBU audio path from the studio to both transmitters fed by one processor.  Once this is accomplished, a certain amount of delay can be added to the audio content on either the main or booster transmitter to move the interference zones away from heavily populated or trafficked areas.
WDBY, Patterson, NY 60 dBu contour

WDBY, Patterson, NY 60 dBu contour

This is the situation with WDBY in Patterson, NY.  The main transmitter site is located on a hill in Patterson and has a power level of 900 Watts at 610 feet (186 meters) HAAT. The main population area is Danbury, CT, to the south east, about 12 miles away.  Between the two, there are several imposing hills, which create reception issues in Danbury.  Therefore, WDBY FM1 was placed in service on the Danbury Medical Center.  The booster has a power output of 1,200 Watts, at 0 feet (0 meters) HAAT (49 meters AGL).

WDBY FM-1 signal, Danbury, CT 60 dBu contour

WDBY FM-1 signal, Danbury, CT 60 dBu contour

Therefore, the southern area of the 60 dBu contour is filled in by the booster.  The interference zone between the two transmitters is determined by the amount of delay in the audio between the two units.  If both are time the same, the interference will occur at precisely 1/2 the distance between the transmitter sites, which in this case is 10.18 KM from booster.  Looking at the population maps, it might be better to move that more toward the north, away from Danbury.

The formula for computing audio delay time is:

A-B=C where A is the distance between the transmitters and B is the distance to the interference zone from any given transmitter.  The product of that is multiplied by a constant of 3.34 to obtain the time delay in micro seconds.  Therefore, if the interference zone is desired to be further outside of Danbury, say 15 KM away, then the equation looks like this:

20.358 kM -15.0 kM = 5.358 KM

5.358 KM x 3.34 = 17.89 μS delay from the main transmitter site will put the interference zone out in the middle of nowhere, away from Danbury.  This is total delay between the two stations, therefore any difference in STL paths needs to be included in this figure.

Nautel has a good webinar on SFNs which can be found on their website: Single Frequency Networks Webinar

Nautel equipment has most of these features built into it, therefore, the implementation of a SFN using Nautel exciters and transmitters should be relatively straight forward.

The head smasher

I have worked in hundreds of transmitter sites over the years; AM, FM, TV, HF, Two way, Paging, Cellular, etc.  So many, I have lost count.  The one thing that is always annoying is equipment that is suspended from the ceiling at just the wrong height, AKA: The Head Smasher.  It does not matter if warning signs are posted, I’ve seen them marked with black and yellow caution tap, and so on.  If it is installed low enough for somebody to hit their head, contusions will result.

3 1/8 inch motorized coax switch mounted

3 1/8 inch motorized coax switch mounted

Thus, when it came to install this motorized 3 1/8 inch coax switch, there was only one way to do it.  Installing it the other way would result in a head smasher behind the backup transmitter because the ceilings are low.  The problem with this style mounting is how to get to the motor and clutch assembly for servicing.  There is but one inch of clearance between the top of the coax switch and the transmitter room’s ceiling.  If servicing is needed, the entire switch would need to be removed, resulting in lots of extra work and off air time.

3 1/8 inch motorized coax switch cover

3 1/8 inch motorized coax switch cover

So, an idea was formed.  Why not cut the switch cover in half and put some hinges on it.  The cover itself is made of aluminum.  I was able to carefully mark it out and cut it with a jig saw.  Then, I attached a set of hinges on the back side and a set of latches on the front.  It now opens like a clam shell.

3 1/8 inch coax switch cover modification

3 1/8 inch coax switch cover modification

Now, when access is needed to either the motor or clutch, the cover can be opened up and removed.  Unless the actual RF contact fingers burn out, there should be no need physically removed the switch for servicing.

3 1/8 inch coax switch cover, modified

3 1/8 inch coax switch cover, modified

Cover replaced.  This will not have to be removed very often, in fact, I have known some coax switches that never need service.  Still, having the ability to quickly get the cover off and do some basic repairs is a good thing.

Conduit fill

It may be surprising to some, but number of wires allowed in any given conduit is not “as many as can be jammed in there.” The National Electrical Code, AKA NEC or NFPA 70 gives specific guidance on the numbers of current carrying conductors allowed in any specific size and type of conduit.

This is due to the fact that current carrying conductors generate heat.  Cables enclosed in a conduit need to dissipate that heat so that the insulation on the cable doesn’t melt, which would be a bad outcome.

Conduit fill tables are found in Chapter 9 of the NEC.  There are several tables that give the number of conductors for each size and type of conduit.  Then there is the general rule of thumb that more than two cables, the maximum conduit fill is 40%.  This comes in handy when several different size conductors are being run in the same conduit.

An example of this is when several circuits are going across the room to the same general location, in this case, a row of transmitters and racks.  Instead of running individual conduits for all those units, one or two conduits from the electrical panel are run to a square wireway, then the individual circuits are broken out and wired from wireway to the individual loads.  In this case, the following equipment is being connected:

  • Harris FM25K: 100 amp 3 phase high voltage power supply (#2 THHN), 30 amp 3 phase transmitter cabinet (#10 THHN)
  • Harris FM3.5K: 70 amp split phase (#6 THHN)
  • Harris MW1A: 30 amp split phase (#10 THHN)
  • Two equipment racks: 20 amp single phase (#12 THHN)
  • Coax switch: 15 amp single phase (#14 THHN)
  • Dummy Load: 15 amp single phase (#14 THHN)
  • Antenna switch/dissipation network for AM station: 15 amp split phase (#14 THHN)
  • Convenience outlets for back wall: 20 amp single phase (#12 THHN)

Excluding grounding conductors, which will be addressed below, the total current carrying conductor count is thus:

  • #2 THHN: 3 each
  • #6 THHN: 3 each
  • #10 THHN: 7 each
  • #12 THHN: 6 each
  • #14 THHN: 6 each

Ampacities based on NEC table 310.16, THHN insulation in dry locations, maximum temperature rating is 90° C (194° F) based on ambient temperature of 30° C (86° F)

Grounding conductors for each of those circuits, based on NEC Table 250.122 (all conductors are copper):

  • 100 amp circuit: #8
  • 70 amp circuit: #8
  • 30 amp circuit: #10
  • 20 amp circuit: #12
  • 15 amp circuit: #14

The final conductor count is:

  • #2 THHN: 3 each
  • #6 THHN: 3 each
  • #8 THHN: 2 each
  • #10 THHN: 9 each
  • #12 THHN: 9 each
  • #14 THHN: 9 each

The plan is to use two 1 and 1/2 inch EMT conduits between the electrical service panel and the 4 x 4 square wireway. According to  NEC Chapter 9, Table 4, the 40% cross sectional size of this conduit is 526 mm2.  It is easier to simply use metric measurements for this.  The cross sectional wire areas are found in Chapter 9, Table 5.  Chart of various conductor sizes and areas:

Conductor Area (mm2) Total conductor Total area (mm2)
#2 THHN 74.71 3 224.13
#6 THHN 32.71 3 98.13
#8 THHN 23.61 2 47.22
#10 THHN 13.61 9 122.49
#12 THHN 8.581 9 77.229
#14 THHN 6.258 9 56.322

Thus, in order to break this up into two 1 and 1/2 inch conduits, the #2, #6 and #8 (main transmitter HV power supply, backup transmitter and grounds) are run in one conduit, the remaining circuits in the other.  The idea is that the main transmitter and backup transmitter will not be running simultaneously for long periods of time.  Those cable areas total 369.48 mm2, well within the 40% limit of 526 mm2 for 1 and 1/2 inch EMT.   The rest of the circuit’s cable areas total 256.041 mm2.  That leaves room for additional circuits in the second conduit if future needs dictate.  The extra conduit area will make pulling the wires through easy.

From the square wireway to the HV power supply, 1 and 1/4 inch conduit will carry the three #2 and one #8 ground.  1 and 1/4 inch EMT has a cross sectional area of 387 mm2, the conductors contained within will be 271 mm2.  Less room here, but still well withing the 40% limit.

Pictures will be posted when the project is done.

Moving the WRKI and WINE transmitter site

Blogging has been light due to work load being heavy, at the moment. We are engaged in moving transmitters out of this old house:

WINE 940 WRKI 95.1 former studio and transmitter site

WINE 940 WRKI 95.1 former studio and transmitter site

Into this new transmitter building:

WINE WRKI transmitter building at base of tower

WINE WRKI transmitter building at base of tower

The former building was the original studio for WRKI, 95.1 MHz, which signed on in 1957.  The co-located AM station, WINE 940 KHz, did not sign on until 1963.  As such, the building is a little worn around the edges, so to speak.  The FM transmitter has an auxiliary cooling device, for those hot summer days as the building itself is un airconditioned:

WRKI Harris FM25K transmitter, circa 1986

WRKI Harris FM25K transmitter, circa 1986

The rest of the building is in similar condition.  Ceiling tiles are falling off the ceiling and getting ground into the floor, junk is pile up in almost every corner, rodent feces, and the basement, don’t even get me started on the basement.

The basic floor plan for the new building is simple:

WRKI WINE transmitter room floor plan

WRKI WINE transmitter room floor plan

Right now, the preliminaries are being done, mounting the coax switch, running conduit, pulling wires, etc.

A few design notes:

  1. This building is much closer to the tower, which is sited on a high hill (715 feet, 218 Meters) and sticks up 500 feet (152.1 Meters) above that.  Basically it is the area lightning rod, thus special attention will need to be paid to grounding and bonding.  I decided to isolate the electrical ground in favor of the RF ground for lightning protection.  This involves putting toroids on the electrical ground conductors.
  2. The building itself is shielded with continuous steel plating, but that has been cut in a few areas to install air conditioners.  Those areas will have to be repaired and the AC units bonded to the steel plate.
  3. Back up cooling will be in the form of a large exhaust fan and intake louver.
  4. The tower itself is AM radiator for WINE.  It is 170 degrees tall, which means high RF fields at the base, therefore good RF bypassing is needed.
  5. The transmitter room itself is fairly small for what needs to go in there. careful design and placement is required.

Here are some in progress pictures:

WRKI backup transmitter, Harris FM3.5K, coax switch in the background

WRKI backup transmitter, Harris FM3.5K, coax switch in the background

The first order of business was retuning a Harris FM3.5K transmitter to function as the backup. The current backup transmitter is an RCA FM20E, which no longer runs. After the move is completed, that transmitter will likely be scrapped.

I attached super strut to the ceiling at four foot intervals. I used this strut to support the 4 port coax switch. All coax in the transmitter room is 3 1/8 inch hardline, which has a power rating of 40 KW.  Since the transmitter power output is 20 KW, this leaves a lot of head room for problems.  When working with 3 1/8 inch coax, it is important to remember to cut the inner conductor 1 1/2 to 1 3/4 inches sorter than the outer conductor, otherwise the stuff doesn’t go together right.

The 30 KW air cooled dummy load was moved up from the other building and connected to the coax switch.  This allowed the backup transmitter to be tested.

WRKI backup transmitter and dummy load

WRKI backup transmitter and dummy load

Three inch ground strap connects all the transmitters, racks, and dummy load to the station ground.

WRKI ground strap, new transmitter building

WRKI ground strap, new transmitter building

Electrical requirements are being met by a 400 Amp service backed up by a 120 KW generator.  Once the conduit work is finished and all the wires pulled, the coax to the old building can be cut and brought into the new building, then the station can go on the air with the “new” backup transmitter.

Good bye UHF RPU, we hardly knew you

As more and more things which use radio are invented, it is inevitable that the limited spectrum will be squeezed. We see this with BPL, which admittedly causes interference on the HF spectrum, however, proponents have deep pockets. Then there are the so called “whitespaces” between active TV channels coveted by broadband providers.  Not even the once sacred GPS system is immune interference by other radio systems being implemented by companies who “bought” the radio spectrum in question. Although it is quite beyond me exactly how one can buy or sell radio spectrum. I suppose next they will be selling sunlight and rain.

The next chunk of RF spectrum being repurposed is in the 451-457 MHz range.  This has already been eyed by the Department of Homeland Security for on scene data communications networking.  However, the latest interested party is the Alfred Mann Foundation, who builds bonic implants.  In an interesting twist, one of the plans for the spectrum in question is something called the MMNS (medical micropower network systems).  This network would be used to transmit commands from the patient’s spinal cord to prosthetic devices.

Many TV stations use 450-455 MHz band for IFB and cueing.  Radio stations use that same spectrum for remote broadcast and telemetry return links from transmitter sites.  RPU frequencies used to be very congested, as remote broadcasts were often an additional revenue stream for radio stations.  These days, most stations to “cellphone” remotes, e.g. the disk jockey goes out to a store or event and calls it in on his or her cellphone.  Some of the more fancy station use POTS codecs like the Comrex blue box or matrix and very few still use ISDN.  So the first question is how many broadcasters still use UHF (or even VHF) RPU gear (AKA The Marti)?

The second question is what type of damage or reaction could occur if a UHF RPU interfered with one of these MMNS devices?  Some RPU’s use fairly high power levels and directional antennas.  But, according to FCC Report and Order on ET Docket 09-36, it is a done deal:

The rules we adopt will allow these new types of MedRadio devices to access 24 megahertz of spectrum in the
413-419 MHz, 426-432 MHz, 438-444 MHz, and 451-457 MHz bands on a secondary basis.

It goes on to say:

Each year, millions of Americans, including injured U.S. soldiers, suffer from spinal cord injuries, traumatic brain injuries, strokes, and various neuromusculoskeletal disorders. The devices that we anticipate will operate under our new rules are designed to provide artificial nervous system functions for these patients.

Which is nice.  I suppose if someone is at the mall setting up the Marti for a remote and when it gets turned on, Grandpa starts break dancing, one should find another frequency.  Do you think the DJ’s or promotions people remember that?  No, me neither.

If this keeps up, eventually everything is going to interfere with everything else and nothing will work.

Higashi no kaze ame

Or “East wind rain,” which was the Japanese code words transmitted to their embassies indicated hostilities with the United States was imminent. While the Navy intercepted the first message to the embassies, outlining the various code words and instructions on what to do when or if they were used, they never actually intercepted the code word messages themselves.

This is not all that unusual, as the Japanese were using several different diplomatic, army and naval codes at the outbreak of the war.  Many different stations and frequencies were in use, and it is quite possible that the message was sent and never intercepted.  According to the NSA, many, but not all of the Japanese Codes had been cracked and were being read regularly.  The diplomatic code, known as Purple and well as a similar code, J-19 and a lesser version, JNA-20 were being intercepted and forwarded to Washington for decoding.  Only JN-25 was being intercepted and decoded at Pearl Harbor prior to December 7th.

In 1979 the NSA declassified over 2,000 intercepts.  They declassified more in 1994.  Those decrypts paint an interesting picture of the lead up and aftermath of Pearl Harbor.  A good book is “And I was There,” by Edwin Layton.  It was completed in the late 1980’s before the 1994 document dump.  Layton was indeed at Pearl Harbor on the morning of December 7th.  Layton does not come out and directly say that Washington knew of the attack, but rather that they should have known, warned Pearl Harbor but didn’t.

By the first week of December, the Japanese tipped their hand several times.

  • A bomb plot message was transmitted to the Naval Atache in Pearl on September 24th.  This message was sent in J-19 and requested information on shipping and locations within Pearl Harbor based on a grid.  No other locations around the Pacific required such detailed analysis by the Japanese Navy.  This was intercepted in Pearl but sent to Washington for decoding.  No information from this message was shared with the military commanders on Hawaii.
  • Several spies (Russian, British, Korean) and diplomats (Germany, Peru, Mexico) warned the US that the Japanese intended to attack Hawaii before the end of the year
  • The Japanese striking force did not maintain radio silence during it’s transit from Japan to Hawaii, they sent of 663 messages, many of which were duly intercepted by Hawaii and the Navy intercept stations in the Philippines and Guam.  The National Archives contains about 100 of those messages, however, the direction finding information attached to each message has been removed or remains classified.  In addition to this, most of the JN-25 messages sent between December 1-7 remain classified.
  • All PYB (long range flying boats) patrols from the Aleutains were discontinued in early December.  Their patrol area covered at least part of the northern route from Japan to Hawaii.
  • On the Evening of December 6th, Roosevelt was shown the first of 13 parts of a Japanese Diplomatic message which was a declaration of war on the US.  The US code breakers and translators finished the decrypt and had it in the hand of the President hours before the first bombs were dropped on Pearl Harbor, even before the Japanese envoys had read it.

It is also well known the Roosevelt wanted Japan to attack the US as justification for the entry of the US into WWII.  In early 1941, Roosevelt sent a gun boat up from the Philippines nosing around the Chinese coast to see if it could stir up any trouble.  The Japanese failed to take the bait on that and several other occasions.

December 7th, 1941

December 7th, 1941

There is still quite a bit of controversy as to who knew what and when.  Whether information was deliberately withheld, or not transmitted due to some concern of compromised intelligence or some bumbling bureaucracy is hard to tell.  Certainly, the powers that be in Washington knew more than they let on.  The military commanders on Hawaii took the fall and several thousand lost their lives that morning.  Many more would die in the coming months and the Japanese tide rolled over the western Pacific basin.

By the end of World War II, over 74 million people had died, most of them civilians.  History, do not repeat thyself.

More radio apps, then I'll leave this alone

Since I posted about Andoid phone apps for engineers last February, I noticed several others have picked up the thread and published articles as well. Good for you! I am glad that we can be of assistance here at Engineering Radio.

Radio World had a good article back in July on this subject.  I’m not going to link to it, just because.  I figured I’d add a few others that didn’t make the grade last go around, either because they didn’t exist, or I didn’t know about them.  I am limiting my choices to free apps for Android phones.

  1. Router passwords.  Often, very often in fact, the default password on any given router does not get changed when it is installed.  I found this app to be accurate and useful for speeding up various router tasks required in day to day radio engineering.  Things like opening ports for VNC, routing outside IP addresses to internal ones, etc.
  2. Navaile Electrical Calculator.  Great for National Electrical Code questions, wire sizes, breaker sizes, box fill, conduit fill, voltage drop, etc.
  3. RF & Microwave tool box.  Has handy calculators for filters, mismatch, return loss, etc, just in case those things need to be done by hand.
  4. GPS test.  Shows available GPS signals, gives time, location and accuracy in WGS84 datum.
  5. Shortwave Schedules.  Data base of shortwave schedules searchable by station, time, and frequency.
  6. Note pad.  Just what it says.
There are several other good ones, but these are the ones that I tend to use most often.

The Gates BFE-50C Amplifier

Found in a pile of junk in the corner of an older transmitter site, this Gates BFE-50C or otherwise known as an M5675 Amplifier. This was used as an IPA in a Gates FM 1C transmitter installed around 1960 or so.  The rest of the transmitter has long since departed, likely to the scrap yard, however, somebody though to remove this and set it aside.

Gates BFE-50C 50 watt VHF amplifier

Gates BFE-50C 50 Watt VHF amplifier

This unit is missing it’s grid tune knob.  The grid tune capacitor is still there, however.  There is also some evidence of heating on R403 and R407/408 likely due to a prolonged overdrive condition.  Otherwise, it is in good shape.

Gate BFE-50C 50 Watt VHF amplifier back

Gate BFE-50C 50 Watt VHF amplifier back

The design is pretty simple, a pair of 6146’s in push pull, three watts in nets about 50-60 watts out, according to the manual, which can be found here (.pdf).  The power supply voltages are fairly tame, 500 volts plate, 300 volts screen.  The one thing that this design does not have is any type of harmonic filtering.  When used with a larger transmitter, this makes sense because the transmitter output will have overall harmonic filters.  If this was to be used on it’s own for any reason, a good harmonic filter would need to be designed and installed.

Gates BFE-50C or M5675 50 watt VHF amplifier

Gates BFE-50C or M5675 50 Watt VHF amplifier

The schematic is straight forward.  Gates, the old Gates Radio of Parker Gates, designed good equipment.  Click on image for higher resolution.

Gates BFE50-C input section

Gates BFE50-C input section

It is a bit hard to see in this picture; the input section consists of three turns of #14 gauge wire coupled to two 4 turn sections of 14 gauge wire on either side of it.  This is matched to the grids Screen1 of the 6146’s with C401.  L412, C411 and L413 form a low pass filter.  L412 consist of one turn #14 gauge wire, L413 is five turns of #14 gauge wire.  All coils are 3/4 inch in diameter.

Gates BFE-50C output section

Gates BFE-50C output section

The output section is even simpler, using just one loop of small diameter copper tubing.  The plate tuning is accomplished by C407, loading is C406.  Power output is adjusted by varying the screen voltage using R405.

Advantages of this design:

  1. The 6146 tube is fairly rugged, at class AB the 50 to 60 watt output range is well within the plate dissipation for a push pull configuration.
  2. No special parts needed, everything can be found or fabricated by hand
  3. The 500 volt supply is fairly tame, maximum PA current should be less than 0.2 amps for 50 watt output and 50% PA efficiency.
  4. Output tuning and load allow for tuning into less than ideal loads, if required.
  5. If operated as a stand alone unit, some type of plate current meter should be used to aid tuning.  A harmonic Filter would need to be designed and built for the output.
All in all, a pretty cool little FM amp.

The BE FM20T transmitter

This is the main transmitter for WYJB in Albany, NY. The backup is the Harris FM20H3 on the right. I haven’t turned that unit on lately, but it normally makes quite a fuss the first time the Plate On button is pushed. The FM 20T on the other hand, is mellow and even tempered.

WYJB 95.5 Mhz, class B, transmitter Albany, NY

WYJB 95.5 Mhz, class B, transmitter Albany, NY

One other thing of note; The FM20T is still on its original tube.  I looked up the maintenance records for this transmitter, it was installed in December of 2000.  Eleven years later, the 4CX15000A is still cranking out 15 KW TPO, which is impressive.  I found that high power ceramic vacuum tubes actually seem to last longer when run closer to their limits than those that are running at half power.

Judicious management of filament voltage is required to achieve this type of longevity.  There is a set procedure for installing a large ceramic vacuum tube:

  1. After the tube is in the transmitter, run it a full filament voltage for at least an hour or so before turning on the plate voltage.  This allows the getter to absorb any stray gases in the tube.
  2. Once the plate voltage is applied, proper tuning should be completed a quickly as possible.  Tuning procedures vary from transmitter to transmitter, however, the general idea is to obtain the maximum power output for the least amount of plate current while keeping the PA bandwidth within acceptable limits.  Some transmitters can get narrow banded at high efficiencies, which manifests itself as higher AM noise.
  3. After the tube has been in use for 90-100 hours, the filament voltage should be reduced gradually until a drop in the transmitter output power is noticed, then increased by 0.1 volts.

This maximizes the filament life for that particular transmitter and power output.  Once the filament can no longer boil off enough electrons, the tube power output drops and it is time to replace it.

This site also has two other radio stations, WZMR, 104.9  and WAJZ 96.3 , both class A using solid state transmitters of less than 1,000 watts:

WAJZ and WZMR Energy Onix solid state transmitters

WAJZ and WZMR Energy Onix solid state transmitters

Not the prettiest sight in the world, but it does stay on the air.  There is no money to go back and neaten up this work, unfortunately.

The tower supports all three antennas.  There was some discussion of a common antenna for all three stations, however, WZMR is a directional station, thus it would require it’s own antenna.  Doing a common antenna for the other two stations was cost prohibitive, so the tower supports three two bay antennas.

WYJB, WZMR, WAJZ FM antennas, New Scotland, NY

WYJB, WZMR, WAJZ FM antennas, New Scotland, NY

The stations are all located in the New Scotland, NY tower farm.  WYJB is licensed to Albany, WZMR is licensed to Altamont and WAJZ is licensed to Voorheesville.


A pessimist sees the glass as half empty. An optimist sees the glass as half full. The engineer sees the glass as twice the size it needs to be.

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.
~1st amendment to the United States Constitution

Any society that would give up a little liberty to gain a little security will deserve neither and lose both.
~Benjamin Franklin

The individual has always had to struggle to keep from being overwhelmed by the tribe. To be your own man is hard business. If you try it, you will be lonely often, and sometimes frightened. But no price is too high to pay for the privilege of owning yourself.
~Rudyard Kipling

Everyone has the right to freedom of opinion and expression; this right includes the freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers
~Universal Declaration Of Human Rights, Article 19 was discovered, and not invented, and that these frequencies and principles were always in existence long before man was aware of them. Therefore, no one owns them. They are there as free as sunlight, which is a higher frequency form of the same energy.
~Alan Weiner

Free counters!