Hacker News new | past | comments | ask | show | jobs | submit login
Oxford Electric Bell (wikipedia.org)
106 points by TheAuditor on June 2, 2014 | hide | past | favorite | 60 comments

As the article mentions, the battery is likely a Zamboni pile, which has an extremely high internal resistance (i.e. can only supply very low currents) but is made for high voltage, electrostatic applications like this bell. That's why it lasted so long - modern battery chemistries have much lower internal resistance, and self-discharge would mean it being completely depleted in much less time. The battery in this bell has such high internal resistance that the self-discharge is many orders of magnitude less.

The characteristics of these batteries are different enough that you'll find lots of claims of "perpetual motion" on the Internet from people who have made them and found how long they can last when powering low-current loads like LEDs!

But this could be extremely handy for something doesn't use a lot of current, right, like an MSP430 or similar?

Why can't I buy LED lights that use these batteries and stick on things people run into in the dark?

I've always found things like this amazing. The "Centennial Light" is a lightbulb that has been in near-continuous operation for 113 years!


I had known about that, but I never had realized until re-reading that article that it now runs at 4 watts rather than the original 60 watts.

I'm still impressed totally, but I must say that that factor dampened it a bit for me. If planned obsolescence is really non-existent in that industry one wonders how long a modern low-wattage bulb built for endurance could potentially last.

A modern low-wattage bulb built for endurance could potentially last roughly forever. And during that time turn so much electricity into heat that you'd be asking for shorter-lived higher-temperature bulbs to save on energy costs.

With normal incandescent lighbulbs, electricity is about 95% of the costs, so if you doubled the lifetime of those bulbs, the most you could save would be about 2.5% of total costs, if the lower operating temperature of long-lasting bulbs didn't increase the electricity needed for the same brightness, which practically means that you probably would pay more overall.

electricity is about 95% of the costs

Really depends on the application. For difficult-to-change bulbs (in high or remote locations), longevity can be a significant factor. It's one thing to replace the bulb in a table lamp, another on a remote high tower or deep within an industrial plant in a hostile environment. Or on a spacecraft.

> A modern low-wattage bulb built for endurance

I do not think so. LED on its own sure, but most bulbs also have power converter and other parts prone to failure.

I think he meant traditional, carbon-filament lightbulb, manufactured to modern standards. If the filament used was thick enough, it would work for hundreds of years, no problem.

The wiki also lists it being shut off without specifying the reasons. "The fire department says that the bulb is at least 113 years old and has been turned off only a handful of times."

I really doubt there's anywhere in the US where power has run uninterrupted for 100+ years. Even the use of backup generators is fairly modern and those fail or run out of fuel as well.

Neat little thing, but I think back then they hand-wound the filament using much stronger and thicker (and with more resistance) wire than what we use today. Or used, considering that style of lightbulb is more or less dead. Dunno, I'm not impressed by this the same way I am by the pitch experiment or the ever-ringing bell.

I've often heard that light bulbs are not manufactured to as high standards as they could be.


But i have no idea how much truth there is to it.

I don't think it's a conspiracy, it's just cutting costs of a commodity to meet expected lifespans.

If you buy 130 volt-rated bulbs (120V is typical), they are built to a better standard and can survive exposure to poor quality electric service.

It's one of those things not taken into account by the cargo-cult conversion to CFL and LED -- the electronics in these bulbs are built for good conditions. Try running the newer bulbs for extended periods off of portable generators.

Depends on where.

Some smaller cities have municipal electricity that pretty much always works (Holyoke, MA is the one that comes to mind). Even in regional blackouts, there are a few Upstate NY towns that never go out... they can isolate themselves from the grid when necessary, and draw from the grid when they have issues.

I grew up in the boroughs of NYC, and don't recall ever having a power outage in those years. I think it flickered during a hurricane in the 80's, but never actually went out.

Where I live now (Upstate NY), forget it... even after eviscerating the street trees, any significant gust of wind has a probability of taking out power. The state no longer effectively regulates the utilities -- the transformer on my street blew up this past winter, and per the lineman had markings indicating that it was made in the mid 1960s.

Made in my hometown! Was always an inspiration to me, as well as a lesson in planned obsolescence.

From the lede of this Wikipedia article: "which has run almost continuously ever since, apart from occasional short interruptions caused by high humidity." Well, that's not the same as "have not stopped," or "has not stopped." The submission headline here is an exaggeration.

I don't think that matters. It wouldn't have worked on humid days in 1840 either, the important part is that the battery can still power it today.

Until one of those interruptions is 174 years long the title is completely accurate.

Could you explain that, please? Hacker News has a policy of preferring that articles are submitted with their original title[1] (which was manifestly NOT done here), and to me the article is a lot less interesting than its submission title promises. I am a Wikipedian, and once in a great while I see an article on Wikipedia that is actually well written enough to be worth submitting to Hacker News, a community of thoughtful discussion that prizes submissions that "gratif[y] one's intellectual curiosity."

But most submissions from Wikipedia to Hacker News, and especially the submissions that are submitted with heavily altered titles, don't gratify anyone's intellectual curiosity so much as raise all kinds of questions about incomplete research or writing in the original underlying Wikipedia article. Here, the description appears to be of a very low-energy system that occasionally halts entirely and may not really be doing anything very remarkable. It's hard to tell how the energy flows or steadiness of motion in this system compare to other electro-mechanical systems that just weren't made quite as long ago, and that weren't judged by such relaxed standards.

[1] "please use the original title, unless it is misleading or linkbait."


The phrase 'has not stopped for 174 years' is ambiguous the event that has not occurred could be 'stopping' or could be 'stopping for 174 years'.

As to every other sentence you wrote: its a community, whatever you think is appropriate is only slightly relevant and even what dang thinks is only slightly more relevant.

I read your parent as a joke! The title can be read the same as the Mitch Hedberg joke: "I haven't slept for 10 days...because that would be too long."

This has never "stopped for 174 years"...because that would be too long.

Thanks; we reverted the title.

Thank you for pointing out this major issue.

I have a Casio calculator (fx-100s) that still goes strong on its original battery since I bought it, 18 years ago! I use it a few times every week.

Heh, I have a Casio EL-5813 that I got in 1980 that still is still running on its original batteries!

It might be solar powered.

No, it only has an AA battery. I also have a remote control from a Sharp radio system which still works after 24 years with its original batteries! But this one is used very rarely, say 2-3 times every year.

maybe someone else uses it too and replaced the aa battery? Maybe you replaced it but don't remember?

No because they both have these japanese batteries that most japanese electronics come with but you can't find them anywhere in the stores. If someone had replaced them, it would be with a known brand. Besides the calculator is always on my home office and I live alone.

I was very surprised too when I found out about a year ago.

I'm really confused how an electrostatic battery would get charged?

Is there a limit to how much energy can be stored in such a way? It seems like it might be a promising avenue for new battery research, no? Are there major limitations I'm not seeing?

It's not an electrostatic battery. It's a perfectly normal non-rechargeable battery made of a very large number of thin cells; it provides high voltage but almost no current. However it also has a very low self-discharge, which would be a problem for almost any "wet" cell in this situation.

One nanoampere for 100 years is a very small number of amp-hours.

You can engineering estimate it in your head as one kilo-day per year and one kilo-hour in a century, and a nano 1e-9 is a kilo-kilo-kilo so to less than one sig fig its of the order of a thousandth of an amphour per century.

Thermal noise is around a nanovolt per Hz at 50 ohms and room temp (Very roughly) so it would be a trick to use this power level to do much electronically, like generate a measurable radio signal. A much larger battery might be able to power a (edited: continuous) radio beacon.

Ah, I'd not thought of breaking out the Feynman estimation. I make it more like 1e6 hours to a century though?

Ouch thats correct, so more like about an aH per century.

Its an electrochemical battery like any other. In fact silver-zinc chemistry (like a hearing aid) is not all that unusual for this type of dry cell. However the internal resistance from using paper as an "electrolyte" is incredibly high. So the weight or volume per power output ratio is extremely bad compared to anything. Even direct radio isotope batteries are worse. Radio isotope RTGs are worse.

The "electrostatic" comes from the engineering model where you can pretty much treat it like a magic leyden jar that charges itself. A self charging high voltage capacitor of sorts.

It is a shame they don't have a video or an animated gif of this dynamic object.

Here it goes: https://www.youtube.com/watch?v=1Dx1-f8xQio&feature=kp

Not as dynamic as one would expect, though.

Edit: Reproduction, easier to see: https://www.youtube.com/watch?v=uKwSHtQwzD8

You can find it on YouTube. It doesn't ring very loudly, but I suppose that would be a requirement for constant 174-year operation.

See also (different source about the same bell): https://news.ycombinator.com/item?id=5995422

Strange the article seems compelled to mention that its not an example of perpetual motion. Why not also say its not magic, nor controlled by the gods?

Simply because there are far more people who are likely to forget the rules of science and think "wow, 174 years, it keeps going forever!" than people who think "wow, 174 years, must be magic"

I agree with you, but perhaps it's just to prevent those who believe in such things from citing this example as evidence.

The world has enough youtube videos of people claiming their homebrew magnet contraption is a perpetual motion machine, one hopes the author is simply trying to stymie that trend.

Do you not use the internet? Between the Tesla conspiracy theorirsts "OMG Tesla discovered free energy thus Westinghouse had him killed" and the constant rebadging of "cold fusion" I think these disclaimers are warranted.

that has

thank you. I can understand (not agree with but understand) some people insisting on writing name of companies with plural verbs like Google are and Microsoft have but a battery bell is a singular word no matter which way you slice and dice.

To be fair, judging by previous comments, it's likely the writer is not a native english speaker. English grammar is hard, and people make mistakes.

I for one would like HN to be an inclusive international community, and that means not jumping down people's throats just because their english is not 100%. I'd like to see any of us native english speakers try to submit something to a foreign language site...

Whilst I agree with your sentiments, a 2 word comment really is useless, not matter how great their English might be... rather than a "that's amazing!" type comment, a simple upvote on the original article would suffice, kind of like discouraging "Me too!" comments on Stackoverflow, it doesn't add much to the discussion.

Although having said that, my initial thought was the OP had hit return too early and not realised...

I wasn't talking about the grandparent comment, I was defending the submitter!

The GP's two word comment, while correct, is too reddit for my taste (nothing against reddit - both sites have their place).

British English seems to prefer treat entities like companies as plurals, while the rest of the English-speaking world seems to treat them as singular. Lots of discussion on this point can be found on Google.

I find the last part amusing:

"The Oxford Electric Bell does not demonstrate perpetual motion. The bell will eventually stop when the dry piles are depleted of charge"

Yes, and the Earth revolving around the Sun is not perpetual motion, despite the fact that it has been happening for billions of years and will continue to happen for billions more. Score one more for the pendants! :D

Its not pedantry. The significant thing about "perpetual motion" is how it would violate (and, hence, disprove) what are currently understood, on very strong evidence, to be physical laws. So, its a pretty key distinction.

It's pedants not pendants. Do you always spell it wrong as a sort of infuriating joke? That's a good idea actually.

lol :D

I guess the thing that struck me about the perpetual motion line is how unintentionally(?) trollish it was. It invites debate. It would be better to omit it.

It's like a web form that asks your email address followed by "We never ever sell your personal information to anyone!!11!" It raises a question I wasn't even considering and now I'm not sure I trust it! http://goo.gl/PaXG5g

I think pedantry can be excused in the face of claims that would disprove the fundamental laws upon which our understanding of the universe rests.

> Yes, and the Earth revolving around the Sun is not perpetual motion, despite the fact that it has been happening for billions of years and will continue to happen for billions more.

Well, circular motion does not cause any work to be done, so that's not quite the same thing. The earth might continue spinning around the sun forever in a perfect universe, but that still wouldn't violate conservation of energy.

Moving charges radiate energy so there's at least one source of a drain.

>circular motion does not cause any work to be done

Like tides?

That part got me as well. Sure it isn't technically perpetual motion but I wonder if maybe it could be considered "near perpetual" or "relatively perpetual" from a selfish human point of view.

So if motion lasts longer than your lifetime, doesn't it seem like perpetual motion to you? Personally, when I think of perpetual motion, the first thing that comes to mind is using it to power a generator. In that context, if something could produce electricity for decades without being refueled or even attended to, aren't we moving toward and/or operating with a realm that feels like perpetual motion--even if it is only from our self centered point of view? Granted, this is sort of like if someone is in your field of view and then steps behind a tree that obscures your view of them, you can say they are invisible to you. Though it isn't exactly true to say they have gained the power of invisibility, you labeling them as being invisible is a self centered and relative statement.

So as self centered humans, from our point of view any motion which lasts years or even decades--and could be harnessed--could seem sort "effectively perpetual" to us, even though it is nowhere near actually perpetual.

(Yes, this theory ignores that fact that that generator's output would indeed be ridiculously low, probably close to unmeasurable--the important part is that if we have motion, we assume we can generate electricity. And yes, even if you could make this happen, it may take thousands or even millions of these together just to power an LED. I was probably prompted of this thought because of this video: https://www.youtube.com/watch?v=8hwLHdBTQ7s)

I think there's a lot of illusion here because not only is the energy being consumed at a terribly slow pace, you also can't "see" it being consumed. If you could watch the battery run down, and see that it was now only "half full" (or whatever fraction of the charge currently remains) as opposed to a hundred years ago, it would be less magical.

For a similarly long-running process, consider underground coal fires. If a seam of coal is burning, it may take hundreds of years to burn itself out, comparable to the lifetime of the ringing bell. But because it is consuming its fuel in a much more obvious fashion, it's less magical.

It's unfortunate that it doesn't estimate when that might be. As the 'internet of things' becomes more and more important, I can think f many applications for very long-lived low-voltage batteries in places that might not be suitable for solar panels or have access to wired power.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact