FM transmitters – Page 7 – Engineering Radio

Shipping Container transmitter site

I do not particularly like these. I know, they are relatively inexpensive, easy to come by, easy to install, etc. However, a shipping container was not designed to house a transmitter, they have certain drawbacks. These are, in no particular order:

Air conditioning. Using a traditional Bard-type equipment shelter HVAC unit requires cutting through a lot of fairly heavy gauge steel. What’s more, the steel walls are uneven, requiring a filler.
They are by necessity, fairly narrow. Arranging racks and transmitters along the length of the unit restricts access to either the front or the back of the equipment. Meeting NEC clearance requirements for electrical panels, transfer switches, and disconnects can pose problems.
They are not very tall. Mounting overhead equipment can be problematic as one does not want to drill through the top of the container. Crosswise unistrut is one solution, but it lowers the overhead considerably.
Electrical work is slightly more dangerous. Doing any kind of electrical work, troubleshooting, repairs, etc is a little more nerve-racking when everywhere around you is a metal surface at ground potential.
They are difficult to insulate against cold and heat.
The door-latching mechanisms bind, wear out or otherwise fail over time.

All of those things being said, I am now rebuilding a transmitter site in one of these shipping containers.

Inside view of shipping container transmitter site

Fortunately, the original electrical work was not bad. The transmitter is a twenty-year-old BE FM10B, which will be retained as a backup. The new transmitter is a Gates Air FAX-10. We have installed several of these Gates Air transmitters in the last two years or so and they seem to be pretty solid units. This is the second 10KW unit I have installed.

Gatesair FAX-10 transmitter in Middle Atlantic Rack

We decided to install the FAX-10 in a Middle Atlantic rack since we did not have a whole bunch of extra room for a separate transmitter rack. The 1 5/8 inch coax switch is installed in the top of the transmitter rack along with a Tunwall TRC-1 switch control unit. The other rack will have the STL and all other ancillary gear. My idea is to have nothing in between the door and the FM10B so it can be easily removed when that day comes. Something, something about planning ahead since it will be likely myself removing the FM10B.

The Gates Air FAX-10

This is the first one of these transmitters that I have installed. This particular unit is analog only, but there is lots of room left over for an HD exciter if need be.

GatesAir Flexiva FAX-10, 10.000 watt FM transmitter — GatesAir Flexiva FAX-10, 10.000-watt FM transmitter

The size of a 10 KW FM transmitter these days is pretty small, basically taking up the equivalent of one rack. This is a relatively small transmitter room, the old tube transmitter basically took up the entire room. With this unit, there is room to install a full power spare, if that was desired.

RF modules use LDMOSFET devices, each module has a power output of approximately 1,600 watts.

GatesAir Flexiva FAX10 power amp section

There are eight power amp modules and seven switching power supplies.

GatesAir FAX10 power output — GatesAir FAX10, licensed transmitter power output

One issue at this site, there is no reliable three-phase power available. There was a three-phase open delta, but man, that thing scares me a little bit. Since this is a single-phase setup, I was curious to know what the current draw on each leg was at full power. I measured with my clamp-on ammeter; 54.3 Amps at 120 volts, or 6516 watts per leg. Overall power draws 13,032 making the AC to RF efficiency 65.2%. VSWR calculates out to 1.21, which is not great. I think the antenna could use a little bit of tuning love.

Upgrading the firmware

The original V series Nautel transmitters have required a couple of firmware upgrades in some cases.

Upgrading the power module firmware, WDVT, Rutland, VT — Upgrading the PA module firmware on Nautel V-5D transmitter, WDVT, Rutland, VT

The first was for the controller to add a little bit of bias to the PAs during analog operation. The second one I have had to do is to the PA modules themselves which were to keep the power supplies from shutting off during re-transfer from Generator power to commercial power.

I have done several of these and once you get the hang of it, it only takes a few minutes to complete. Still, I remember when transmitters didn’t have firmware. The low voltage control circuits were either 120 or 240 VAC with big relays and contactors that loudly confirmed their closure before any meters began to move.

Regarding Nautel transmitters in general; the newer models are not the same rugged, reliable designs that were common in the past. We have AM ND series transmitters that have been on the air for 20 years without a single failure. The models rolling out of the factory these days often have switching power supplies fail without reason or warning and RF pallets that are fragile things. Ah well, I suppose all things are cyclical.

Nautel V-10 repair

Not exactly sure how it happened, but one of our Nautel transmitters malfunctioned! It is a pretty rare event, so I thought the exclamation point was needed. One of the PA pallets went bad and the transmitter lost 1/2 a PA module. Since the TPO for this particular station is 7 KW, they remained on the air at full power. In the interest of staying on top of things, we fixed it anyway.

Diagnostics were simple:

Fault lights on the front of the transmitter observed
Press the status button to find out faults, which were Module D failed
To to module sub-menu, find Module D and discover Q1 disabled, and Q3 shutdown.
The problem is with Q3, order a new pallet from the manufacture

Upon removing the module, I did not see the damage at first:

Nautel V series FM transmitter PA module

It is board A3, which for this particular flavor transmitter is a Nautel Part number NAPA16-B. Once I replaced the defective module with a new one, I discovered what looks like a symptom of the greater problem:

Over to the left-middle-lower section of the board, R10 and R8 are burned open. These are surface mount 2-watt, 20-ohm resistors. A glance at the schematic shows that these are part of the bias supply. A quick set of measurements with a DVM shows that Q1 seems to be intact and not shorted. Interesting…

The question is: Is it worth trying to fix this board or should I just trash it an buy a new spare?

Update: Schematic diagram as requested: