Tech In Plain Sight: Incandescent Bulbs

While they are dying out, you can still find incandescent bulbs. While these were once totally common, they’ve been largely replaced by LEDs and other lighting technology. However, you still see a number of them in special applications or older gear. If you are above a certain age, you might be surprised that youngsters may have never seen a standard incandescent lightbulb. Even so, the new bulbs are compatible with the old ones, so — mechanically, at least — the bulbs don’t look different on the outside.

You might have learned in school that Thomas Edison invented the light bulb, but the truth is much stranger (public domain)

It has been known for a long time that passing a current through a wire creates a glow. The problem is, the wire — the filament — would burn up quickly. The answer would be a combination of the right filament material and using an evacuated bulb to prevent the filament degrading. But it took over a century to get a commercially successful lightbulb.

We were all taught in school that Thomas Edison invented the light bulb, but the truth is much more complicated. You can go back to 1761 when Ebenezer Kinnersley first caused a wire to glow. Of course, wires would quickly burn up in the air. By the early 19th century, limelight was fairly common in theaters. Limelight — also known as the Drummond light — heated a piece of calcium oxide using a gas torch — not electric, but technically incandescence. Ships at sea and forts in the U.S. Civil War used limelights to illuminate targets and, supposedly, to blind enemy troops at night. Check out the video below to see what a limelight looks like.

Sir Humphry Davy demonstrated a dim, impractical light that used a huge battery and a thin strip of platinum. More practical was Davy’s electric arc lamp, which, after being refined by others, became common in some applications.

Arc lights had issues, though. They hissed and flickered. The carbon rods emitted carbon monoxide and ultraviolet light. They were extremely harsh and bright, and the rods burned up quickly. Everyone knew a better light bulb would be a winner, but no one knew how to create it.

Getting Closer

Starting around 1835, there were many experiments and demonstrations, but none of them really caught on. A Belgian, Marcellin Jobard, was on the right track in 1838 with a lightbulb in a vacuum with a carbon filament, but nothing really came of it. He also came up with what amounts to early emojis, but that took a long time to catch on, too.

Since platinum has a high melting point, it was a popular filament candidate. In the 1840s and 1870s, many inventors used platinum or carbon with varying degrees of success. During that same time period, there were many patents and demonstrations, but none were successfully commercialized. However, a Russian named Alexander Lodygin did patent a working bulb with carbon rods in nitrogen gas.

It isn’t clear if Henry Woodward and Mathew Evans knew of the Russian patent. In 1874, they filed a Canadian patent for a similar bulb. Ultimately, they failed to commercialize it, but they sold their patent to Thomas Edison.

Edison

Edison got serious about electric lighting in 1878. He experimented with different carbonized materials and platinum but finally settled on carbon fed by platinum wires. Using carbonized threads resulted in a bulb that lasted just over 13 hours. However, he would discover that carbonized bamboo could last 1200 hours. You can see one of the oldest surviving Edison bulbs at the Port Huron Museum and in the video below.

Many people worked on the problem throughout the 1800s. Edison arrived at a practical solution and had the mechanism in place to exploit it. However, others had light bulb patents. Albon Man and William Sawyer had bulbs that didn’t last as long as Edison’s but formed the basis for the United States Electric Lighting Company. That company’s chief engineer was Hiram Maxim, a name familiar to most ham radio operators, but this particular Hiram Maxim was the famous ham radio operator’s father.

The elder Maxim is one of several people who claimed they had actually invented the incandescent light before Edison. The courts eventually decided that some of Edison’s claims were preempted by William Sawyer’s patents, but that Edison still had other valid patent claims.

Modern Types

These early bulbs had little in common with modern bulbs. The inside of the bulb had to have very little oxygen and moisture, or the filament would oxidize or burn out. Initially, mercury vapor pumps and phosphoric anhydride were used, but this added expense to bulbs. Arturo Malignani found that red phosphorus would allow for a drier vacuum with cheaper pumps. Edison was quick to buy the patent.

However, Lodygin and others were on the right track, and using a metal filament and an inert gas to replace the oxygen would be more effective. This prevents the filament from burning and also reduces the evaporation of the filament. (See the video below if you want to see the effect of air on a tungsten filament.) He invented a process for forming thin metal filaments and sold the patent to General Electric in 1902.

The truly modern bulb is the result of a 1904 invention by Sándor Just and Franjo Hanaman. They created a tungsten filament that worked better in an argon or nitrogen atmosphere. The Hungarian company Tungsram sold these, and they could practically pass for a modern clear-glass bulb.

A modern bulb has a glass envelope and a tungsten filament, although they add a few impurities to increase the filament life. The bulb contains a low pressure of a gas like argon, nitrogen, krypton, or xenon. Modern glass bulbs are either clear or coated with kaolin clay from the inside. Some bulbs have pigments to change color or different glass to produce different colors. Bulbs used for heating sometimes have special glass or even fused quartz.

Real World Considerations

Light bulbs are one of those circuit elements we pretend are perfect, but they aren’t. Tungsten filaments have a low resistance when cold, which causes a bulb to draw a lot of current when it first turns on. As the filament gets hot, the resistance goes up, and the current goes down. Oddly enough, carbon filaments have the opposite problem. They draw more power as they get hot, which also makes them sensitive to power surges, since if they get hot, they draw more current, which causes them to draw even more current, which makes them even hotter, and the cycle repeats.

In high-reliability circuits, designers often highly derate a bulb’s specifications to get a dimmer light that lasts longer. A 5% reduction in voltage will roughly double a bulb’s lifetime but also make it about 16% dimmer. Some will also pass a small current through the bulb even when it is off to keep the filament warm. This reduces the current draw and heating associated with turning on a cold filament.

The other big problem with incandescent lights is that they are relatively inefficient since most of the energy produces heat and infrared light. A typical bulb is around 5% efficient in terms of visible light, and the best halogens come in around 10%.

Of course, this inefficiency is why there’s been a move to ban incandescent bulbs in favor of LEDs, fluorescents, and other technologies. LED lights, in contrast, can reach 30-40% efficiency. Still more light than heat, but almost an order of magnitude more efficient than plain-old incandescents.

So Much More

There’s a lot more to learn about light bulbs. In 1885, the U.S. had an estimated 300,000 carbon filament bulbs. By 1914, there were 88.5 million. In 1945, the market was around 795 million. When you deal with that kind of scale there are many innovations both in the technology and the machinery used to build them. Want to see how lightbulbs were made? Check out the video below.

We’ve talked about the early lighting market and one of its pioneers, Lewis Latimer, a few years ago. We’ve looked at the checkered history many times.

Featured image: “Yellow Bulb” by [Daniel Reche]

70 thoughts on “Tech In Plain Sight: Incandescent Bulbs

  1. I just threw away (“recycled”) another couple of LED bulbs that died. So much for “Lasts 10+ years!”. Let’s compare the impact of LED lighting on Mother Earth…

    Incandescent Bulb components:
    1) Glass
    2) Metals (Aluminum, Steel, Tungsten)

    LED Bulb components:
    1) Plastic
    2) FR-4 or similar circuit board (Epoxy, Fiberglass)
    3) Electronics (capacitors, diodes (LEDs/rectifiers), doped-silicon ICs)

    Which of these two devices is less harmful/polluting to the landfill they’ll end up in?

    Turn up nuclear power (ideally, breeder reactors) and bring back the option that pollutes far less.

    1. The trash aspect of a lightbulb is negligible either way. A few ounces of trash waste amortized over many years is essentially zero. Conversely, the energy needs of an incandescent light bulb are something like 1lb of carbon dioxide per HOUR. So until some future time when electricity generation produces no carbon dioxide, energy consumption is the only waste metric that matters for lightbulbs.

        1. A hotter planet doesn’t change how many hours of sunlight a region gets. Growing seasons would remain the same, and tropical plants such as coffee and cacao wouldn’t be able to grow in regions with less sunlight. Meaning that they’ll be forced into extinction in a warmer climate.

          But good job regurgitating that climate change denialism drek.

          1. Although i get the gist of it, you are incorrect regarding the growing season remaining the same. For the nordic countries, this is not true, the warmer it gets there, the earlier the crops can be planted, and the later they can be harvested. The scandinavian and northern russian regions have sunllight practically 24/7 in June.

        2. Wow, I’ve heard some really detached from reality commentary on rare occasions here but this takes the cake.

          You are completely ignoring the reality and the science simultaneously at the most basic level

        3. Welcome to the Permian Times, we’re your hosts, Bill and Diana Postosuchus.
          The maid headline today, What’s with all that bad smelling smoke over the Siberian Traps? Bill.

          Thanks Diane, climate scientists earn that the CO² and other gases will build in the atmosphere over the next million years and may lead to catastrophic global warming and atmospheric poisoning.

          Here to counter this, our intrepid reporter, a Dimetrodon in a funny hat.

          Thanks Bill, see, CO² is basically plant food and I eat plants so more food for me!

          Millennia later

          Tonight we take a look at The Great Dying, the Permian extinction event.
          What those silly synapsids didn’t realise was that massive global warming could never have had a good outcome.
          Over that million years the traps released 12,000 Gigatonnes of stored carbon, death was unavoidable.
          It is Earth’s most severe known extinction event, with the extinction of 57% of biological families, 83% of genera, 81% of marine species and 70% of terrestrial vertebrate species. It is also the greatest known mass extinction of insects.

          Millennia later

          Scientists recorded Humanities greatest ever release of carbon last year at 40 gigatonnes, nearly 3,400 times greater rate than The Great Dying

          12,000 GT ÷ 1,000,000 years =
          12,000,000 tons per year

          40 GT ÷ 1 year =
          40,000,000,000 tons per year

          Since 1751 we’ve released
          1.5 Teratons of carbon

          12,000 GT in the Permian over a million years
          1,500 GT in our Holocene over 274 years

          What’s that averaging? 5,000 megatons a year?
          Versus the permians 12 megatons a year?
          So we’ve averaged about 400x faster.

          Saying this is all plant food is like saying a burst damn is fine because we need water to drink.

          This is a nuclear reactor Vs a hydrogen bomb.
          Rates are important.

          1. Thank you RoboJ1M. I was going to post a reply saying something decidedly not “kind and respectful” about stupid people wasting oxygen regurgitating nonsense, but you have done a much better job. Bonus points for your comment comparing the “CO2 is plant food” to water and drowning. Hopefully that’s a simple enough analogy that even the parent poster can understand, and they will refrain from wasting our time reposting such nonsense in the future.

        1. There’s a lot of people in warmer places where that’s not often wanted, and people who have fuel burning heat or heat pump heaters might rather use those. I admit it was convenient when a desk lamp could warm your hands on a cold day, but who’d like to change their bulbs every time the weather changes?

          1. Ok, so let Scandinavians have it, and the Canadians. I’m not sure what they use in siberia, whale oil? (I kid.) But are those people really mostly using electric resistive heat? Because I think it’s incredibly likely that it’d be too expensive to do that, and in fact they’re using either fuel or a heat pump. In which case, that’s the better heat source instead of light bulbs. The primary place you see a lot of electric resistive heat is somewhere with cheap electricity and generally mild winters.

          2. But are those people really mostly using electric resistive heat?

            Quite a bit, yeah. The alternatives like heating oil, or wood burning, are mostly being depreciated for air quality concerns and the rising cost of fuel. That leaves district heating and electric heating.

            A light bulb is of course less efficient than an air source heat pump, but then again you spend a couple hundred watts to light your home and a couple kilowatts to heat it, so the difference is somewhat negligible anyways. The extra heat only becomes a real problem if you need to run your AC to remove it, which rarely happens in the nordics and most homes don’t even have a cooling unit. They just sweat through the one week of summer when it’s actually hot.

      1. But a lot of energy goes into the production of them also.

        I’ve just had two new LED bulbs fail within a week. I suspect they’re a diff batch as others I’ve had have been ok, but still…

    2. If they died, that almost certainly was the electronics that went out, not the LED itself. When they haven’t been zapped, overheated, or gotten wet, LEDs lifetime is long enough that we instead define it as the point where they have lost a certain fraction of their original brightness, even though they still turn on just fine. It might be hard for a non-expert to figure out which bulbs will last and which won’t, but there’s nothing wrong with the actual technology. Anyway the reason you have to change incandescents at the rate you do is that a less efficient bulb could last longer but would cost more to energize to the same brightness, so they set a reasonable balance. The LEDs last much longer while consuming much less energy, and if they’re not complete junk they’ll represent one of the smallest electronics waste sources in the house.

      By the way, in the U.S., the government statistics say that the difference in energy consumption if we were all using incandescents today is about the same as if we were all driving EV’s. Specifically, I compared the lighting energy usage both before and after LEDs took over, converted that to per capita, and found that it would be about the same with a crossover EV going the per capita average number of miles. The effect of air conditioning versus heating with incandescents heating the house is a complicating factor; I think the extra load on the air conditioning is likely the bigger factor of the two.

  2. “A 5% reduction in voltage will roughly double a bulb’s lifetime but also make it about 16% dimmer.” Which is part of the reason the “1000 hour cartel” idea isn’t quite logical; a longer-lasting bulb of the same brightness would cost enough in extra energy consumption that changing them more often made a fair amount of sense.

    “30-40% efficiency. Still more light than heat,” Should be more heat than light if that’s the case.

    Interestingly, as far as I remember it tends to be that automotive incandescent bulbs are vacuum type, while home use bulbs are the inert gas filled as you mentioned. Most likely argon. Oh, and halogen bulbs are just incandescent bulbs except run at a higher power density made up for by the boost in lifetime from the halogen cycle that mostly returns tungsten to the filament.
    Now some really interesting ones were nernst lamps and the carbon button lamp; the first one uses a ceramic that doesn’t get oxidized away (so that it beats carbon for efficiency, although apparently not necessarily incandescent despite the greater temperature). Though the limelight is debated, I think in general the idea with them is both a bit similar – very very high temperature oxides don’t oxidize away in air and make a fairly nice color temperature. https://hackaday.com/2020/06/16/retrotechtacular-the-nernst-lamp/
    The second, the carbon button, was an odd Tesla idea that I’m not entirely sure if anyone’s put full effort into finding the limits of – not that it’s great, but it’s interesting at least. I’d like to see if a variation on it may be made a little like a carbon arc lamp, but without the second carbon electrode and the adjustment mechanism. If so, would be nice to get an (inefficient) high power high intensity discharge out of it. May be unavoidable that the power density can’t get high enough while keeping the lifetime up; there’s a reason we have short-arc xenon now.

  3. That company’s chief engineer was Hiram Maxim, a name familiar to most ham radio operators, but this particular Hiram Maxim was the famous ham radio operator’s father.

    Its interesting how insular communities can be. In my circles ‘Hiram Maxim’ is the inventor of the machine gun and no one’s heard of Hiram Percy Maxim, who is, apparently, famous among radio enthusiasts (who have probably never heard of his father;)

    1. Most amateur radio enthusiasts are aware of the family heritage, but most don’t choose to celebrate that part of their legacy. Your circles probably would be aware also of Hiram Percy’s invention of the firearm sound suppressor (silencer), which continues to present day in the form of industrial noise control.

  4. The Edison story is mostly an American thing, I think. It’s their hero.
    Here in Germany, for a long time we considered Heinrich Göbel as the inventor.
    Also, there are indications that the ancient egypts had primtive glowing lamps already.
    Thinking of the ancient Bagdad battery, it’s a possibility.
    In practice, though, the light bulb likely had many inventors throughout human history.
    The books of history are under constant contruction.
    Science itself is just a momentary snapshot and ever changing.
    So let’s relax. ^_^

  5. I’m surprised there was no mention of Joseph Swan the English physicist who is also credited with developing incandescent lighting independently from Edison. He went into partnership with Edison forming the Edison & Swan United Electric Light Company, commonly known as “Ediswan”. It’s all in Wikipedia.

    1. It’s pretty much guaranteed that a US-centric article wouldn’t mention any British inventors ;-) ! I wouldn’t be surprised if most Americans still thought Blue Flame had the World Land Speed record (in fact it’s been held by the UK since the early 1980s)!

        1. Ha! Touché, I didn’t know! From a British perspective though, we’re always trying to fight the corner for Swann, because I understand there was some kind of joint patent for incandescent light bulbs at least with regard to them being sold in the UK.

          -cheers!

  6. Appliance bulbs, such as those in an oven, are (often?) incandescent.
    For a while “Rough Service” bulbs existed as incandescent after regular incandescent bulbs were NLA.
    I tried using an LED bulb in my garage door opener (GDO), a typical rough service application, but it gave off RFI which prevented the GDO from receiving the signal from the car remotes. The current LED in the GDO does not interfere.

    1. Many of the older LED bulbs used switch mode power supplies and they tended to produce a lot of RFI. Now that we have high voltage, multi die LEDs, most bulbs use linear power supplies. That makes them RF quiet as long as they aren’t used on a dimmer.

    1. Actually, I originally drafted this with blow by blow of everyone I could find and it was so long I was going to have to do a two parter. Then I realized we had covered much of this earlier so I included a link, deleted a bunch, and put in the line about many others. A number of people mentioned are not American but I suppose everyone sees what they want to see. Regardless, if you really want to see the details of everybody who had a potential incandescent lamp invention just before and just after Edison, the other article linked from a few years ago does a great job of covering that.

  7. i’m an incandescent hold out- i have a stick pile of them- but i also have dimmers and i almost never have my bulbs on 100%- i’ve tried to lay out my lighting to able to be bright at 75%. because of how low i run them that last a really really long time. I have one lamp that’s on every evening from sunset to 10:00 and it’s over 15 years old

      1. Depends on whether you’re penny wise or a pound fool.

        I wanted to buy high CRI LED bulbs to replace my incandescent lights, but they cost so much that I’m genuinely better off just running my remaining stock and wait for the prices to drop – if they ever do. For my desk lamps and reading lights, I buy 15 or 40 Watt “appliance bulbs” that are still available, because they don’t make proper low wattage LEDs with high CRI and low color temperature – they’re all sickly pink somehow.

        1. They cost $.99 for a high CRI LED when you search enough. Somehow being so “pound wise” in your own head has held you to thinking about prices as though you were living 10 years in the past.

          1. The bare diodes you mean? Or some no-brand fakes?

            All the cheap ones I find at the stores don’t even list the CRI because it’s terrible, not controlled or quality checked in any way – that’s how they are cheap. Last year I tried to buy two 10 Watt E27 bulbs with a CRI of 97 from a store that measures and guarantees their products are genuine, but the asking price was $50 plus shipping and they were out at the moment.

            For the amount of money, I can buy about 800 kWh of electricity and burn 2×75 Watt halogen bulbs for over 5,000 hours, which is worth about two years of normal use. I think I’ll wait another two years.

            The problem is that regular consumers don’t know or care enough to demand high CRI lighting, and can’t measure their lamps to verify, so we get crap bulbs with little quality control unless we pay extra.

        2. I don’t know where you live, but Signify/Ledvance have both high efficiency medium CRI (80) or medium efficiency high CRI (96) lamps that all cost less than €10 a piece. And you’re probably able to find them cheaper. I’m unsure if the country you’re writing from runs 10 years behind, but in 2024 in western europe you can get reliable, affordable lamps that put out light nearly indistinguishable from halogen.
          Ikea’s offerings are not half bad either, and those are less than half that price.

  8. I once toured a chem lab that had a large glove box maintaining an argon atmosphere. There was a little incandescent flashlight bulb in one corner with the glass broken off.

    When I asked about it, they said “That’s hooked up to an alarm. If oxygen starts getting into the box, the filament burns through in a few minutes, and the alarm goes off.”

    Try THAT with an LED bulb.

        1. Costs less electricity, though, can give you an output other than yes/no, and can give more than one positive result in its lifetime. And there’s other things that consume oxygen. Also getters, like in tubes.

          1. The cost of electricity for a 3 Watt bulb is negligible, and it’s a simpler system. A dedicated getter would probably run up in the hundreds for the premium they charge.

          2. Oh, and the amount of heat it generates also causes the air in the box to move and circle around, for which you would need a fan that would probably consume an extra watt or two.

          3. Maybe. I have some pretty good fans used to remove heat from a wooden shed in summer; I don’t remember which ones but one listing from that brand is rated 80 cubic meters per hour at about a quarter of a watt, in the 140mm square format. But yes, light bulbs are cheap. And it’s easy enough to get small ones made for cars, if this is your purpose. It’s just not the do all, be all, end all solution.

          4. The point is that these “better” solutions are more complicated and cost more money for the special parts, so you won’t be saving any. The cost of the fan alone would probably buy you a box of 10 flashlight bulbs and the electricity to run one basically forever, so why make it difficult?

    1. Ok, how about this – use a led and a light sensor (which can be another LED if you amplify the signal enough) and one of those getter coatings that turns color when anything non-inert gets in – oxygen, water, or suchlike. If it doesn’t act like a mirror anymore, set off the alarm. (Or, for other types, you might pick colored LEDs so that whatever color it turns will absorb the light)

      1. Trouble is, how do you get the color changing getter in the box without it reacting to the atmosphere, or when you open the box to put something in? The light bulb is inert when switched off, and it consumes whatever little oxygen remains after purging without causing a false signal.

  9. When Edison was still trying to brute force the filament material to have lamps that last more than few minutes, The Italian Alessandro Cruto was already selling 1000 hrs lamps in large quantities, as completely industrialised products.

    Saying that edison invented the incandescent lamp is false and stupid as saying that Jobs invented the smartphone o Bell invented the telephone. Only Americans can believe that….

  10. Port Huron, MI – I was born, raised, and still live here. We have many monuments to Mr. Edison. We also have some of the most beautiful and unique waterfront in the world. Worth the trip in the warm months

  11. What’s with people calling people who hire a lot of STEM people to make things under a trademark and NDA inventors and engineers? I’m seeing this with Elon too.. Tesla even worked for Edison at one point..

    I’m a big fan of NiFe/”Edison” batteries but I realize Ernst Waldemar Jungner couldn’t afford pneumonia treatment in 1924..

  12. I read the article, but honestly came here for the comments as mentioning Edison in a positive manner always brings out those that want to discredit anything and everything about him. As a fan of Edison, Tesla, and Göbel, I am fine with giving credit where credit is due even in highly debated historical notions. In Edison’s case, he successfully created a light bulb that was suitable at the time and as a result, was patentable.
    I always recommend the book “Edison, The Man and His Work” published in 1926 by George S. Bryan as it can be fully trusted as an accurate biography. It isn’t affected by “modern perspectives” of historical events.

      1. A few thoughts:

        (1) Most of history’s most creative, brilliant, productive, and inventive persons had aspects of their personalities/private lives that you’d no doubt find fault with. I’ll go so far as to suggest that if your life’s accomplishments are so great that scholars one day write books about you, I guarantee they’ll be able to cite more than one person who found you insufferable. Humans are imperfect.

        (2) It is an annoying tendency of this generation to think present-day yardsticks are somehow relevant in measuring historical events. If you care enough to understand history, than you need to first understand context.

        (3) Even if Edison was a Grade-A jerk, (and I’m not necessarily conceding that) it has no bearing on the massive beneficial effect he had on civilization as a whole.

        Who cares that Edison didn’t “invent” the light bulb? Nobody did… as with most “inventions” the incandescent lamp as we know it resulted from a progression of ideas from multiple people. Edison, however, made it practical… and affordable. Before him, the idea was largely a laboratory curiosity.

        By the way…what good is an electric light bulb (no matter who you think “invented” it) if there is no infrastructure to power it? Edison laid the groundwork for centralized power generation/distribution. That involved the creation of insulation materials, fuses, switches, sockets, and even energy metering systems for billing.

        Like many, I am a fan of Tesla, but he was a notoriously bad businessman. It seems doubtful to me that any business interest would have backed his AC system if Edison hadn’t already demonstrated a viable power generation/distribution business model (albeit DC) with his Pearl Street Station.

  13. Sat goodbye to no record in centuries of health issues, and hello to point source light..

    Btw I find it amusing the only mention of it is the dumbed down blue light posts..

  14. If you made it this far down the comments,
    read what a master told me:
    What do you need to check for a cold solder ?
    A lot of current initially so the cold solder will pop.
    How do you create that initial current spike easily?
    Easy: an old bulb. Before filament heats, current spikes
    to almost a short.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.