Incandescent lamps on 230 V DC? 1

[Skogsgurra]

What are the problems (if any) if one wants to run normal 230 V lamps from a DC voltage with 230 V tension?

Of course, the switches and fuses may be a problem, but the lamps as such? Is there any knowledge out there?

 

[ScottyUK]

While what VE1BLL says is true, for equivalent wattage lamps the lower voltage lamp carries more current and its filament is made from thicker material.

The surface area, which directly affects ion migration, to volume of metal varies in proportion to diameter, while the volume of metal varies in proportion the square of diameter.

Hypothesis: the thick wire can support material loss in the form of ion migration for longer than the thin one before the wire reaches the point of mechanical failure.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[Skogsgurra]

My original thinking was this: A voltage gradient between the more positive and the more negative end of the filament creates an electric field that moves ions (which I believe to be positive) so that they drift from the positive end towards the negative end. That field is - still in my thinking - intenser in a 230 V lamp than in a 12 V lamp. And if it is an AC field, the net movement is nil - I think.

I have got a lot of interesting answers to my first question. Som have asked why I ask - and that is also an interesting answer. The reason is that more and more people start thinking DC distribution instead of AC. Mostly to avoid the harmonics issue, but also to facilitate common UPS functions in "distribution islands" and to avoid losses and costs due to local rectification in every computer, TV, herd and other appliances in the house. It is therefore good to find out what influence DC can have on different components in an electric installation. Fuses, breakers, outlets and so on are also being discussed.

 

[VEBill]

"...DC distribution instead of AC..."

'Tom' and 'Nick' (Thomas Edison and Nikola Tesla) had a period of debate on this exact same question many years ago.

Nick won the debate, but Tom ended up much wealthier.

 

[Skogsgurra]

Yes, we know about that. But that was about hundred years ago. Not many computers and other non-linear loads at that time. And the need for UPS almost non-existent. A short blink didn't matter much.

 

[CarlPugh]

The filament will act as a diode. The negative end of the filament will conduct electrons to the conductor holding the positive end of the filament.
The electrons leaving the negative end of the filament will cool that end of the filament.
It would be an interesting experiment to operate the lamp and check the temperature at each end of the filament, and then reverse the polarity and check again.
I have been told that lamp life using DC is much less than when using AC.

 

[Warpspeed]

I agree. With dc, one end will always be at a high positive dc voltage with respect to the other end, and there will definitely be current flow across the vacuum.

This has all sorts of implications. The anode end would probably run considerably hotter, and then there is the problem ion migration.

In a thermionic diode the directly heated cathode is going to run at low voltage (typically 4 to 10v) and high current, and be a fairly massive structure compared to the skinny filament in a high voltage lamp. The anode would be even more robust.

Another probably more serious problem is what happens when the lamp finally gives up. With ac, the dreaded "plink" and sudden extremely bright flash at cold turn on, is actually an arc. The momentary half cycle fault current can be very high indeed, but at least with ac it will most likely be extinguished at the next zero crossing, (provided there is minimal series inductance).

With dc, a continuous arc might form which could possibly grow to frightening proportions at a high dc voltage with minimal series impedance, until the main fuse/breaker finally trips.

Interrupting large dc currents at high voltage is a very serious business indeed, as the fuse and circuit breaker manufacturers are well aware. What happens if a massive continuous dc arc escapes from inside the glass envelope ?

 

[jraef]

Wow, is this ever interesting. Who would have guessed that something we have all come to take for granted has so much to think about...

I have a question to add to this however. I have used, on lamps in very difficult access areas, a small "lamp saver button" that was inserted into the mogul base socket ahead of the lamp. This button was a diode, so the lamp would recieve 1/2 wave AC, which can be thought of as variable potential DC. These devices would make the lamps last 3-5 times longer than normal. I know it was true because I did my own test of sorts. I had a multi-lamp fixture that needed a 20ft ladder to get to. 1/2 of the lamps got the bulb saver, the rest did not. The ones with the lamp saver had roughly 2/3 of the lumen output (tested with a light meter), but lasted 5 years, opposed to 18 months without. If DC shortens the life, how do you explain this? Was it just because at the reduced output, the filaments ran cooler?

"Venditori de oleum-vipera non vigere excordis populi"

 

[Warpspeed]

Most definitely, even a 10% decrease in operating voltage will dramatically increase lamp life. The biggest killer is thermal shock at turn on, the cold inrush current can be up to ten times the rated running current. Lamps rarely fail when running, it is that "plink" at turn on.

In my home, every incandescent lamp is now operated from a lamp dimmer instead of the usual on/off switch. Just the act of twisting up the knob which takes maybe half a second, will all but eliminate thermal shock. If it can be run at slightly less than flat out, that helps too.

I have no actual figures, but from memory,lamp life now seems to be several times what it was years ago when I only had switches fitted. It is not the cost of the lamps, but the damned inconvenience.

 

[panelman]

when we used to use bulbs (in the days before LED lamps became so cheap) we used to run 130V bulbs on 110V supplies, i seem to remember bulb life went up by 2 or 300%

 

[OperaHouse]

Those lamp savers are a resistance that decreases is value when it heats up. This reduces thermal shock on the lamp filiment that increases in resistance (about 8 times)and prevents it from acting more like a fuse. These buttons are not diodes. While the 20% of life is a little shocking, empirically the effective lamp life is probably only shortened by about 30%. There are many other factors that shorten lamp life.

 

[Warpspeed]

Those negative temperature coefficient thermistors are an excellent idea, they are used extensively on slide projector lamps to limit inrush. Suitably specified they should work well on any lamp to extend it's life

OperaHouse, you wouldn't by any chance have a connection with the Sydney Opera House ?

 

[ScottyUK]

Hi OperaHouse,

I have seen the diode variety of 'lamp saver' too. A horrible product causing lots of even-order harmonics. Imagine a large housing estate or a commercial / industrial facility fitted with them! I guess the diode orientation is probably random so there would be some degree of cancellation. The NTC sounds a better way to do it if the additional heat can be safely got rid of.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[OperaHouse]

I have a 15W lamp that I use to illuminate that stairs so my dog can go out the dog dooe at night. That has a diode in series so it doesn't ruin the dogs night vision and the lamp lasts longer. I used to have it on a timer till I figured out the timer consumed just about as much as the lamp. I should use a cap after reading all this, it lasts about a year and a half.

The Opera House actually comes from Nantucket. There is a building there from the old whaling days called the Opera House. Time has a way of cleansing things. I think it wasn't much more than a brothel in its day. They still have sailboat races there for the Opera House Cup. And it might just have something to do with my having a 12 rank pipe organ in my garage. I'm up for a trip down under. I loved "Deckchair Danny."

 

[Warpspeed]

The Reason I asked was that I was involved with the design of the emergency lighting system at The Sydney Opera House. This was (and probably still is) the largest dc installation in the southern hemisphere.

The house lights normally run on 240v three phase mains power, but if the mains fails, all three phases are switched over to a 240v (nominal) nickel cadmium battery supply. It was what I was referring to in an earlier post on this thread.

I was employed as the electronics design engineer by the contractor that supplied the batteries, chargers, and diode voltage dropper. Unfortunately I was only involved in the battery room part of the project, and have zero knowledge of the lamps, or any of the problems that may have resulted with that side of it. Your handle suggested to me that you might have been in some way connected.