The Curacao Utility Company provides the house with 220/127 VAC, three

phase, four wire service. This is a good service, as a 208/120 VAC, three

phase, four wire service in USA would be considered an industrial grade

service, costing big dollars compared to typical 240/120 VAC single phase,

three wire residential service. The house is wired with both 127 VAC, one

phase and neutral plus ground, NEMA 5-15R receptacles (the ones with the two

blades parallel to each other), and 220 VAC two phase plus ground, NEMA

6-15R receptacles (the ones where the two blades are along a common plane).

The ground pins of the house outlets all have separate conductors from

neutral and phase conductors, and all terminate in the utility room.

In the utility room, the neutrals all come to a bus, the ground wires come

to a bus, and the neutral and ground busses are bonded together. From

there, neutral and ground share a conductor, separate from the three

individual phase conductors, out to the distribution transformer located

who-knows-where. At the distribution transformer, the neutral is grounded

to provide a single point potential reference for the entire entrance and

house circuit. This ground is typically established by throwing a plate or

heavy screen in the hole before the pole holding the transformer is set.

This entrance configuration was typical of entrances in USA up until about

the late 40s or early 50s. While having a single point ground at the pole

looks good on paper, it was discovered that it causes a variety of problems,

such as voltage appearing between house ground conductors and earth

potential in and around the house. This voltage is caused by neutral

current being sourced by loads in the house returning to the transformer via

a conductor shared with the ground reference. Since the neutral conductor

has finite impedance (and is often higher than you might think due to shaky

connections and long wires), the neutral current times conductor impedance

yields a voltage appearing between house ground and the earth on which the

house is built. There are two other reasons why this entrance configuration

was abandoned. First, the efficacy of the grounds at the poles are highly

dependent on soil conductivity, i.e., the higher the ground resistance, the

more likely the circuit ground reference won't match that of the earth under

the house. Second, since the pole ground is the one and only potential

reference for the system, if it fails, the house circuit can float up to

dangerously high common mode voltages. Combine that with a primary to

secondary short in the pole transformer, and you have a lethal situation,

with thousands of volts between the house wiring and the plumbing. Yeow!

The problem was addressed in USA by adopting code requirements for two

grounding electrodes, one at the pole transformer, and one at the house,

specifically to insure that the N-G bond in the utility room matched the

earth potential around the house. This cleared up the stray voltage

problems and added redundancy to the system potential reference.

When Geoff started working on the Signal Point house, he discovered a

variety of problems, such as voltage appearing between house ground

conductors and earth potential in and around the house. He noticed that he

could get zapped by holding an electric drill (grounded to house ground) and

touching a tower (local earth ground). Geoff correctly solved the problem

by connecting the N-G bond in the utility room to the massive copper RF

ground system, well buried at the house. By adding the second grounding

electrode at the house N-G bond, voltage between house ground conductors and

local earth was eliminated and the problems disappeared.

The central dilemma in the house is that the two basic voltages available,

220 and 127 VAC, don't match USA equipment designed for 120 VAC. Devices

with switching power supplies, such as computers and LCD monitors, will run

on anything, but motors and transformers are more particular. While most

120 VAC rigs and motors will probably run on 127 VAC without trouble, if The

Curacao Utility Company has a 5% over voltage, taking the outlets to 133

VAC, expect to start to clearing fuses or worse. Therefore, connecting

motors or rigs to the 127 VAC outlets would be a very bad idea.

At the moment, the motor & rig problem is being solved via a 2000 VA

transformer on the wall next to station #1. All of the 120 VAC loads in and

around the station run from this transformer. When the outlets provide 127

VAC, the transformer output voltage is 110 VAC (I measured this) with little

load on the transformer. With all of the rigs, computers and monitors

running, the transformer strains to come up with around 105 VAC. We know

that it's straining, since it needs its own fan to keep from smoking. When

The Curacao Utility Company gave us its moment of excitement when the

contest started, the outlets were providing only about 96 VAC, thus the rigs

were seeing only about 80 VAC. Rigs go absolutely mental at that voltage,

and we all participated in the result.

We had a couple of ideas over pasta and vino to help improve the situation.

One idea was to replace the 127:110 VAC 2000 VA transformer with a 127:7 VAC

transformer, connected so that its 7 VAC secondary subtracts from the 127

VAC outlet voltage, yielding 120 VAC. This would vastly improve voltage

regulation, since the transformer only must supply 140 VA of power to run

2000 VA of load (2000 VA / 120 V = 20 A, and 20 A * 7 V = 140 VA). There

are plenty of companies out there that would supply such a transformer to a

willing customer.

An idea that I had while swimming was to flip the switch on the backs of the

MPs from the 120 VAC to the 220 VAC position and plug them into the wall.

While 120 VAC is hard to make, 220 VAC is not -- it's right there. My MP

manual says that the specified input voltage is 100-125 VAC or 200-234 VAC.

If The Curacao Utility Company went 5% over voltage, to 231 VAC, the MPs

would still be running in spec. In this configuration, a 9% under voltage

would yield 200 VAC, and the MPs would still be running in spec. The

voltage regulation would be improved, the 2000 VA transformer on the wall

would run cooler, and MPs would probably run cooler as well.

I don't know much about how the amplifiers are connected, but I assume that

they are all running from the 220 VAC outlets, with their internal voltage

select straps set for 220 VAC. Most amplifiers that I've run across have

taps for 208, 220, 230 and 240 VAC inputs. Tubes with indirectly heated

cathodes are pretty tolerant of input voltage, so even if the amplifier

power supplies were strapped for 240 VAC and were running from 220 VAC, the

impact on tube life would be minimal.

That brings us to the generator discussion. When the generator was running,

N5OT noted some absolutely bizarre voltages appearing at the 220 VAC

amplifier outlets. We got curious and started looking inside boxes and

plugs associated with the temporary generator wiring. We found that the

ground conductors were not connected between the outlets and the generator,

and that the generator neutral was not connected to anything. Since the

amplifiers are directly connected to the massive RF ground system, there is

no immediate safety issue. However, this setup really doesn't follow the

rules for best practices in auxiliary generator wiring.

An aux generator is considered a separately derived voltage source. It

follows the same rules as any separately derived source. For basic safety,

the generator frame should be grounded. We did that when we set up the

generator, but it took some work. Perhaps a driven rod or sturdy conductor

to the massive RF ground system at the generator pad location would make

grounding the generator easier. To establish a potential reference for the

generator output, the neutral pin on the 240 VAC generator plug should be

connected directly to the ground pin in the same plug. As long as the

generator frame is grounded, you now have a happy, ground referenced

circuit. The reason that the generator has separate ground and neutral pins

on its output, and says on the generator output panel that the neutral is

floating, is to allow proper connection directly to a utility panel via a

transfer switch, where the N-G bond is provided in the utility panel. In

that case, you may leave the generator ground pin open and use the panel N-G

bond to ground reference the generator output. However, when a complete

temporary wiring system is used, as we were using, we have to supply the N-G

bond. This is most easily done at the 240 VAC generator plug. The ground

conductors should additionally be connected at the amplifier outlet ends of

the temporary generator wiring, and connected through the system back to the

generator. This will insure safety in case a 240 VAC load is connected that

does not have its own connection to the massive copper RF station ground.

I think that this more or less covers our brainstorming session after the

contest. If I made any mistakes or omitted any important points, please let

me know.

73,

W8WTS.