Baphomet said:

It's easier, simpler, cheaper and much more reliable to use just a step down

transformer. For example

http://www.electronics.50g.com/cdcac.htm uses 2

transistors, 5 resistors, 2 diodes, plus the need for a transformer in any

event. While this isn't the exact circuit you need, I think you can get an

idea of the added complexity of going the inverter route.

Yes; unless the weight of a transformer would be a consideration,

it would (at 50 or 60 cycles) probably be the cheapest, simplest

and most reliable (fewest parts involved) solution. About the

only extra 'component' I would suggest would be a fuse in the

input! At 400 hertz (e.g. aircraft) a 3 VA transformer could be

quite small.

An ordinary mains transformer with a 6 volt 0.5 amp transformer

secondary should be pretty small; also the inductance of the

transformer might provide some protection against sudden voltage

spikes (transients) that occur on the commercial mains supply.

You might be able to salvage such a transformer from something

small such as a desk calculator/adding machine, old printer or

something similar. Even an old model railway transformer or one

of those that operates residential front door chimes (although

those latter often don't like a continuous load and and can burn

out). Even some wall warts can supply 500 milliamps at around six

volts. A transformer also nicely isolates the 'load' from the

main supply; often required and good safety design.

Only other suggestion; if you really DON'T want to use a

transformer, and assuming the load being is say close to a pure

and unvarying resistance, would to put it in series with an AC

rated capacitor across the mains supply. Fuse this at around one

amp?

This will produce a leading current across the resistance portion

of the circuit.

This is a vector diagram situation; but in this case the

resistance will be small (less than 10% or so) in comparison to

the capacitive reactance that within practical limits the 'R' can

be ignored!

Have done this several times to power the filament heater of a

single tube rated at 6.3 volts 0.3 amps. (Very similar to the 6

volt at 0.5 amp requirement here!) from 230 volt mains supply.

[6.3/0.3 = resistance of tube heater, hot = 21 0hms].

Using the formula Xc (capacitive reactance) = 1/(2pi x frequency

x capacitance);

That is Xr = One divided by the total of (two pi times frequency

in cycles, times capacity in farads).

Thus at say 60 cycles a one microfarad capacitor (10 to the minus

6 farads) will work out to;

Xc Ohms = 1/(2pi x 60 x 10^-6) = 10^6/ 375 = 2667 ohms or approx.

2700 ohms. [One microfarad].

So a 6 microfarad cap will have one sixth of the reactance and

will 'pass';

At say 230 volts then; 230/450 = 0.5 amps.

At 115 volts then double the capcitance to 12 mfd for 115/225 =

0.5 amps.

It's rough and ready but it works!

Disadvantages of the series capacitor method:

1) One side of the mains supply is connected to the load circuit.

2) The size of capacitors (and possibly their weight) to

withstand these voltages and the cost if buying new), will be

greater than that of a simple cheap transformer.

2) If the capacitor ever breaks down full line voltage will

impressed on the load circuit. Pose a possible fire and safety

hazard and will most likely destroy the load. Whatever it happens

to be?

I.e. KISS. (Keep it simple, Simon). Why complicate it?