How 'primitive' could you construct an LED light bulb?

[Crossroads Inc.]

After a recent discussion with a friend of mine concerning a Steam Punk world setting, the following question arouse.

We all know these days, that old fashion Tungsten bulbs are about as inefficient as you can get.
They give off MASSIVE heat, which is a danger, and they 'burn' huge amounts of energy for the light they give off.

LED lights by contrast are about as good as anyone could ever hope for! But...
How much "science" do you need to make them?
Depending on the level of technology, could a steampunk world (Say, 1850 tech level) hope to make something like that?
1900? 1950?

Physically, an LED Is made up of, A bit of Plastic, some lead, some very thin metal wire. And of course a silicon diode.
The properties of LED were discovered in the early 1900's using primitive crystal and silicon carbide. and the first production of an LED Was in the mid 1950's.

But...
how much more would you need to make a full one "Light Bulb" that would give off useful light comparable to a Tungsten bulb?

Thoughts?

[Lord Revan]

It took quite a while to produce high quality semi-conductors for the diode in sufficient numbers and still be commercially viable, that's why LED light bulbs for commercial market are quite a new thing.

Essentially it's not that princibles of an LED weren't known before but they weren't able to produce the semi-conductors relibly enough to produce sufficient number of them to make LED light bulbs viable, I'd say 1930s is about as low as you can go with commercially viable LED light bulbs and even that's streaching it.

[SpottedKitty]

It took quite a while to produce high quality semi-conductors for the diode in sufficient numbers and still be commercially viable, that's why LED light bulbs for commercial market are quite a new thing.

Another significant spanner only recently extracted from the works — for many years, the only LEDs available were single colour, and some colours were easier to make than others. IIRC it wasn't really practical to make LED TVs until a good blue LED was developed, and good white ones (like the one I put into this room a couple of weeks ago) are even more recent.

[Sea Skimmer]

We all know these days, that old fashion Tungsten bulbs are about as inefficient as you can get.
They give off MASSIVE heat, which is a danger, and they 'burn' huge amounts of energy for the light they give off.

Why would steampunk care about efficiency? Burn more trees in our 60psi steam non compounding engines!

Also Mercury-vapor lamps appeared at a very early point, like 1892. These have around four times the efficiency of incandescent lights and absurdly long lifespans. Some other tech options existed at an early point too but fact is nobody really cared, they wanted shit cheap back then even more then we do today. The only downside of the vapor lamps is they take up to a few minutes to make full brightness, but you get a lot of light fast. LEDs actually still don't do a good job replacing this sort of lighting which is why it's still produced in the US. Need quantum dots or something.

The properties of LED were discovered in the early 1900's using primitive crystal and silicon carbide. and the first production of an LED Was in the mid 1950's.

These were all IR light emitting devices and generally omnidirectional emitters. A useful directional LED only appeared in the 1960s and multiple colors in the 1970s as I recall, certainly not in the early 1900s was anything like that possible. Blue and yellow ones required some very precise chemistry and engineering to pull off well into the IC chip revolution and its billions of then year dollars of funding. You would need a established semi conductor industry to get anywhere useful with this, which is kind of not the point of calling it steampunk?

[Elheru Aran]

Also Mercury-vapor lamps appeared at a very early point, like 1892. These have around four times the efficiency of incandescent lights and absurdly long lifespans. Some other tech options existed at an early point too but fact is nobody really cared, they wanted shit cheap back then even more then we do today. The only downside of the vapor lamps is they take up to a few minutes to make full brightness, but you get a lot of light fast. LEDs actually still don't do a good job replacing this sort of lighting which is why it's still produced in the US. Need quantum dots or something.

I want to say mercury vapor lamps may actually have been made before tungsten-filament bulbs, but I'm probably wrong on that.

They do get a lot of use in an industrial context and in things like streetlights, though-- where you need a lot of steady light, can take the time for them to heat up, and don't want to have to bother replacing them often.

[Simon_Jester]

A quick Wikipedia search suggests:

Arc lamps based on gas were being experimented with as early as 1860 but were not developed fully until the 1890s (patented in the US in 1901).

Incandescent light bulbs were being crudely prototyped quite early in the 1800s, and the first patent was issued in Britain in 1841.

[Sea Skimmer]

Yeah the original incandescent bulb idea is very old because all metals will glow and emit light with enough electrical voltage passed through them. Humans couldn't have helped but discover this once decent electrical power supplies existed to experiment with. Its very simple, it just took a while to figure out that a vacuum would slow down the reactivity of the tungsten and make it last much longer.

The first decent battery appeared in about 1800, it's not surprising that the first incandescent bulb ideas appear shortly afterwards.

I can't see anything that would stop a vapor light from being built much earlier, it just might be absurdly expensive to produce. The ballast to let it ignite is no bold technology. But without a decent electrical grid it wouldn't have any point, decent dynamos appeared in the late 1860s. Electrical stuff is high RPM tech while prior to turbines steam engines were all low RPM and just easier to build. It's one thing to build a wheel, its another thing for it to spin at 3000rpm 24/7 while hot.

Once you have that the electric arc light is an option! This would seem like Teslapunk.

[jwl]

There's a reason blue LEDs won the Nobel prize in physics you know. They are not easy to make. And white LEDs are just blue LEDs with fluorescent coatings.

[Crossroads Inc.]

Thanks for all the feed back!
Pretty much got the answers I was looking for in terms of a light source better than standard incandescents.
True LED Lights are it seems pretty advanced stuff, but better alternatives to incandescents, namely Mercury Vapor and Fluorescents are much more easy to manufacture.

Also something that made me chortle..

Need quantum dots or something.

OH MY GODS! So I am not the only one to see these ridiculous new TV commercials?
when I first heard someone use the word "Quantum" on an advert I just about lost it )

[jwl]

A quantum dot is an actual thing though. I dunno how they would solve the problem sea skimmer is talking about though, or even how it is a problem in the first place: LEDs can turn on near enough instantaneously, which is why they can be used in televisions.

[Jaepheth]

A bit of a tangent, but Induction lighting would fit Steampunk pretty well I should think.

[Beowulf]

A quantum dot is an actual thing though. I dunno how they would solve the problem sea skimmer is talking about though, or even how it is a problem in the first place: LEDs can turn on near enough instantaneously, which is why they can be used in televisions.

I think the problem is not that they take a while to turn on, but you can't get LEDs up to the same brightness and efficiency for near the same cost. Mercury Vapor lamps are fairly high efficiency, and usually used where you need a lot of light. LEDs are usually in the single digit watt range, while Hg lamps are more in the hundreds watt range.

[Sea Skimmer]

Its basically a light quality problem, vapor lights make very uniform soft light. The high efficiency LEDs are highly directional lights, which means they really don't want to be a floodlight 40 feet up on a pole or building. It would want to look like a bunch of random bright spots. You can do things to improve this, but they'll reduce the efficiency and that in turn reduces the incentive to convert as long as LEDs are really expensive. Also LEDs have very long lifespans, but actually they start getting dimmer the moment you start using them. The rated lifespans on the box are based on 70% original illumination. Indoors people tend not to notice this, but for outdoor lighting that steady reduction is a problem, out door lighting tending to not be that bright in the first place. Also LED lifespans outdoors are a bit open to question, they certainly don't like cycling temperature extremes.

Quantum Dots share some of these problems, but with quantum dots you can improve the original light quality in the design phase by using quantum dots which replicate the precise wavelengths of light visible to humans, and nothing else. Literally the wavelength is based on how big the dot is, no filtering you needed and pretty damn precise, which figures since they are nanoscale LEDs. You can replicate exact desired spectrums. With enough money this will make all lighting design problems go away, and they use even slightly lesser power then conventional LEDs.

This tech is going to get a lot more use then just screens and lights too. We've never been able to make precise wavelengths like this in any practical manner before. Its already probably in service with certain military equipment as a countermeasure, because the right combo of quantum dots can make a IR heat AND visual spectral signature of an aircraft/ICBM warhead/satellite which is identical to that of the real thing.If it isn't, it's damn close. That negates all kinds of ECCM software filtering and physical lens filtering as equips typical modern IR weapons. It also has a bunch of possible medical and scientific uses as well as low power but long range lasers for communication due to high signal clarity. And other stuff I'm sure, nothing like this was ever an option before.

The screens should be awesome on everything but absolute black levels, and at the rate they are going up in size and down in cost they are seriously on track to just overtake OLEDs completely. OLEDs are BTW superior to normal LEDs for lighting, by a lot, and could do most of whata quantum dots will. But fact remains OLEDs remain very hard to produce, and thus very expensive. Quantum dots meanwhile are now into the massive production scal

A quick check of google suggests that quantum dots have already seriously become price competitive with OLEDs, even though mass production is barely two years old.