Hacker News new | comments | show | ask | jobs | submit login
Ampere to get rational redefinition (nature.com)
81 points by feelthepain 1257 days ago | hide | past | web | 19 comments | favorite

This sort of change is a great idea on a fundamental level: define our units in terms of fundamental constants or processes that are absolutely reproducible. But as the article explains, any such definition has to be at least as precise in practice as the less fundamental definitions that they replace.

Hence the challenge (mentioned in the article) of defining the kilogram in a sensible way. The obvious method would be to say "1 kg = the mass of [###] of carbon atoms" or something like that. But counting 10^23 individual atoms is impossible for all practical purposes: it's still less precise than just weighing that reference lump of metal (despite the flaws of that approach).

The remarkable thing described in this article is that scientists are evidently at the verge of being able to count one electron at a time fast enough for use in calibrating actual, practical current measurement devices. That's pretty cool.

"Feynman, I know why all electrons have the same charge and the same mass" "Why?" "Because, they are all the same electron!"


"The obvious method"

Using a cesium time standard and an atomic force microscope figure out the half life of individual atoms of 14C (or something more fun/shorter) then define a kilo of 14C as a pile of 14C generating X decay events per whatever interval calibrated by our cesium time standard.

(edited to add the point is you can watch individual atoms with the AFM and see which ones disintegrated in the past since the last scan, and its no great trick for a century or so to detect individual decays using a geiger counter. Also 14C is no fun for this task. Try a shorter halflife)

The half-life is statistical, not mechanistic, so there would be an inherent variability in that definition. The larger the radioactive mass and the longer the time period, the less variability, but it's still variable.

The fact that it's variable isn't a valid criticism on its own: there is uncertainty in every standard we use, and the objective is to minimize that uncertainty. It's quite possible that the practically achievable error bars from this method would be thinner than the practical error bars of another method. Besides, the fact that you can arbitrarily increase the precision of this method without hitting any hard physical limitations is a solid selling point.

You are right that there's always going to be uncertainty. Perhaps it turns out that in practice, the uncertainty with this method is less than other approaches. The point I was trying to make was about the different between incidental uncertainty due to imprecise equipment, and more fundamental uncertainty due to physical laws. For example, other intrinsic uncertainties are the Planck resolution for time and space, and the Heisenberg Uncertainty Principle for certain combinations of information. Similar to those, radioactivity has a fundamental limit. We cannot predict when an atom will emit radiation, all we can do is make a statistical prediction.

    ...almost as much of an embarrassment as...
Why is it embarrassing? It's the best we've got so far, and it's been pretty useful. We're working on making it more intrinsic to the universe. Chill.

Ohms law will not be denied and if the volt from a Josephson junction apparatus combined with the ohm from a quantum Hall effect apparatus don't match up with the new amp, it'll be interesting to see who gets adjusted to match.

The scientific soap opera of metrology (not meteorology) is interesting to read about. Computer science is hardly the only discipline to suffer from multiple standards.

One standard defined an amp as a certain mass of plated silver in a certain amount of time, so if that was still "cool" then you'd have mass defined indirectly solely from the quantum hall effect and Josephson junction effect and time aka cesium vibrations or whatever it is, which wanders into the whole "a kilo is a lump of metal in Paris" situation.

I feel quite stupid, but I just can't parse the following:

"At present, an ampere is defined as the amount of charge flowing per second through two infinitely long wires one metre apart"

If the wires are one metre apart, then how are they infinitely long?

Imagine two identical, parallel wires that are infinitely long. The two wires are spaced one meter apart (still in parallel). When current is passed through the wires they would exert force on each other proportional to current.

The infinitely long part ensures the force is uniform through all parts of the wire.

Hope that helps.

Ah... That makes total sense now - thanks!

This also relates to the loss of weight [1] of the "125-year-old platinum-and-iridium cylinder" defining the kilogram.

[1] https://news.ycombinator.com/item?id=6617039

I remember that there is a problem with a better definition of a kilogram: unlike metre and second that are defined through measurable physical constants, we use metallic prototype for kg. There were some attempts to define it as the mass of the precisely cut silicon sphere, but we still use the prototype.

Now the question to someone more knowledgeable in physics than me: if the described effort pans out, does that mean we can define kilogram as the mass that would accelerate at 1 m/s^2 if we apply the same force that affects two infinite wires one meter apart, with 1 ampere flowing through them?

To clarify slightly, the proposal wasn't to switch from defining the kilogram as the mass of one particular lump of platinum to one particular lump of silicon. Instead, the idea was to define the kilogram as the mass of a certain number of Silicon 28 atoms and use the silicon sphere to get that number as close as possible to our old kilogram.

This would then allow independent scientists to make their own spheres, calculate the number of silicon atoms in the sphere, and have their own reference benchmarks for the kilo without needing to check in with Paris.

There are efforts underway to redefine the kilogram: http://en.wikipedia.org/wiki/Kilogram#Proposed_future_defini...

Can you calculate a force applied to something that's infinite? I mean, if the wire is infinite, what's it's mass? Doesn't seem plausible to me, but I'm not a physicist, so I could be talking crap here.

It appears the new definition is going to be something like "the current (in the direction of flow) of 6.2415093*10^18 elementary charges passing a boundary in 1 second", replacing that number with a specific 19-digit value. [1]

[1] https://en.wikipedia.org/wiki/Ampere#Proposed_future_definit...

What science could you do with the "new" Amp that you couldn't now, with the old?

I think it is more about doing same science with same results in different places. The infinite long wires part is not exactly reproducible.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact