Im thinking on buying some 220-240 volt light bulbs before the ban deepens more. Anyone know of a good quality manufacturer? I tried Aerro Tech but they only make 120 volt lamps. I found some 277 volt ones by Sylvania if all else fails, which I guess are good?   

240V lamps on 120V give off a tiny fraction of the light they produce at their rated potential.  You could always run the fixture on 240V if it could handle 240V (switch, wiring, etc.), but 240V lamps are less efficient and somewhat more shock sensitive, and the step up transformer would just increase the losses.  You could always use a step down transformer with 12V/24V/30V/34V/75V lamps (potential depends on the transformer), but then system currents and line losses are major things.

240V lamps on 120V give off a tiny fraction of the light they produce at their rated potential.  You could always run the fixture on 240V if it could handle 240V (switch, wiring, etc.), but 240V lamps are less efficient and somewhat more shock sensitive, and the step up transformer would just increase the losses.  You could always use a step down transformer with 12V/24V/30V/34V/75V lamps (potential depends on the transformer), but then system currents and line losses are major things.

I know, but they last at least 4 to 10 times longer, sometimes more. A 100 watt bulb gives off light equivalent to about a 40 watt version, but with a much much longer life. Consumption drops to, roughly 40 watts. They make exceptional night lights (the 60 and 40 watt versions) as well as fixtures that are hard to reach. Plus they are great for bench top electrical testing, no need to worry about over voltages. They make for good voltage testers when differentiating between 120/208/240 .

My concern is finding a good brand rather than something low quality. Ive found incandescent bulb quality varies significantly between makers and country of origin.

Thanks!           

... A 100 watt bulb gives off light equivalent to about a 40 watt version...

Not really, it would be an equvivalent of no more than 10W. The light output drops very significantly. But the consumption would be indeed about 30..40W and with ideal quality lamp (a lamp, that will at the rated voltage fail just for the tungsten evaporation exactly at rated 1000hours and does not suffer from any other degradation mechanism) the lifetime should be over a century.

Quote from: Keyless on April 15, 2014, 02:06:56 AM

... A 100 watt bulb gives off light equivalent to about a 40 watt version...

Not really, it would be an equvivalent of no more than 10W. The light output drops very significantly. But the consumption would be indeed about 30..40W and with ideal quality lamp (a lamp, that will at the rated voltage fail just for the tungsten evaporation exactly at rated 1000hours and does not suffer from any other degradation mechanism) the lifetime should be over a century.

Good point on the lumen output, they are really dim. I was thinking in terms of consumption but they are two different things. In the past Ive put 230 volt light bulbs (relatives smuggled them out of Europe )in various fixtures like the bathroom exhaust fans and used them as night lights. Not sure how long they lasted but it was on bulb I never had to touch Unfortunately they have gotten left behind in moves over the years. 

Im thinking Osram if I can find some.     

Another trick work better: Either a diode (1N4007,...) or a 110V SIDAC in series with the lightbulb. That mean the lamp will be powered at about 70% (for the diode) or 80% (for the SIDAC) rms voltage, so it's life will become way longer, at least 10x (that would be about the limit for standard lamps for the "other" degradation mechanisms, so it won't last longer anyway), yet you will get way more efficient setup and most important, it will work with locally available bulbs...

Another trick work better: Either a diode (1N4007,...) or a 110V SIDAC in series with the lightbulb. That mean the lamp will be powered at about 70% (for the diode) or 80% (for the SIDAC) rms voltage, so it's life will become way longer, at least 10x (that would be about the limit for standard lamps for the "other" degradation mechanisms, so it won't last longer anyway), yet you will get way more efficient setup and most important, it will work with locally available bulbs...

I can confirm that. Many years ago I used a diode with the light bulb at an inaccessible place. That bulb worked 24/7 and survived for many years.
I'm not sure about 70% - I'd expect 50% as it works just one half of the sine period - but my knowledge is limited here.
You can also expect noticeable flickering as the frequency is halved.

The rms voltage is exactly 1/sqrt(2). The reason is, you cut to half the power for the given resistance (it heats up just half of the time), so as P=(V^2)/R, P/2 = (V^2)/2/R = (V/sqrt(2))^2/R. But be aware, this math is valid just for pure resistive load...

The rms voltage is exactly 1/sqrt(2). The reason is, you cut to half the power for the given resistance (it heats up just half of the time), so as P=(V^2)/R, P/2 = (V^2)/2/R = (V/sqrt(2))^2/R. But be aware, this math is valid just for pure resistive load...

True. Although in theory the math is almost correct since when the filament cools off resistance lowers a bit so more current is drawn, but not by all that much. Usually the correction factor is ignored since its trivial in cases like this. Interestingly when one ohms a cold light bulb the resistance is very low, so low that when ohms law is applied the bulb's wattage is actually much more than the label. However when its powered the intense heat raises the filaments resistance dramatically, so you only get what the label says. Lighting engineers calculate all parameters around a hot filament. Cool science I must say . The light output is less than 1/4 even the power becomes 1/4. I forget the values but its something like a 10% power drop drop results in 20% or 30% something in light reduction. Someone who knows more about this than me can answer.    

Resistance change: That's, why I wrote "for given resistance" and not "for given lamp"...

Resistance change: That's, why I wrote "for given resistance" and not "for given lamp"...

I know. I was just adding on to what you wrote. You are correct though, not doubting you. 

Diodes put a DC bias on the mains supply, which causes problems with transformer saturation and other DC component nasties.  I just use long life lamps where I need them.  I do have a light that is used as a heater for our water pressure tank that I use 240V lamps in, but there I just need the IR and convective heating from the lamp and a bit of light to tell that the lamp is working.

For the DC bias you will need way higher power than few lightbulbs. Moreover if you put the diode polarity at random for different branches, the DC components will cancel out.

what i'd be worried more about is not as much transformer saturation (you really need a hige DC load do cause that), but overloading of neutral, and damage to earthing electrodes (in TN-C-S) by sending some DC biased current through them

Putting the diodes in random directions would solve this. However, there still is a problem in a commercial installation : A surge on the line is very likely to blow most of the diodes wired in 1 direction, in that case 1/2 the lamps on the phase will be lighting on full power and the other 1/2 all on half wave in the same direction... If the users dont care to replace the diodes, the expected DC damage can happen

For large installations i'd rather wire the lamps in series pairs of identical lamps on 240v or 415v (2 lamps in series between 2 phases), or use a big choke (few 100's W HID choke) in series with a parallel group of lamps instead

For commercial and/or really large installations you can not use the diodes or so anyway.
And regarding the surge, the diodes usually withstand that. At first it is not much problem to use sufficient rating (1kV of 1N4007) and as second, for the rare pulses above that level the diodes can handle that without damage...