back to article Pump-priming the new ampere: NIST works to count electrons in silicon

The ampere, a perpetual embarrassment to the world of scientific standardisation, is due to get a measurable physical standard in 2018, and America's National Institute for Standards and Technology (NIST) hopes to help provide its definition. Most SI base units have a clear physical definition – the metre is defined by the …

  1. Anonymous Coward
    Anonymous Coward

    Down for the count

    Why exactly is it so embarrassing not to have a physical standard for the Amp? Are they afraid the theory isn't accurate? And if they DO count electrons with quantum dots, won't there be some quantum uncertainty to reintroduce inaccuracy?

    1. Anonymous Coward
      Coat

      Re: won't there be some quantum uncertainty to reintroduce inaccuracy?

      Yes, that's why they run multiple pumps in parallel...

      1. Black Betty

        Re: won't there be some quantum uncertainty to reintroduce inaccuracy?

        Not the way I read it. They're running multiple pumps in parallel just to get enough counts per second to get within shouting distance of being 6 orders of magnitude away from 6.2415 × 10^18. Or in other words, they're saying that even if this works as anticipated they'll still be stuck at counting a paltry 100 billion electrons per second or so and will still have to pull another trick from their bums to get to the nearly 10 trillion eps that will let them nail down 1 micro-ampere.

    2. imanidiot Silver badge

      Re: Down for the count

      The problem with a theoretical standard is that it's very hard to calibrate something very very accurately against a theory. Having a physical standard allows for much more accurate calibration of powermeters and sources across the globe. With a theoretical standard there is no way to prove that a 1.0000105 amp current measured in Japan is the same as a 1.0000105 amp current measured in Canada, unless you bring both sources to the same location and measure them with the same meter in the same circumstances. (And then when you install everything again in the lab back home you are AGAIN not sure if the standards are the same). With a physical standard like proposed here its actually possible to prove that the two measurements are indeed the same.

    3. Doctor Syntax Silver badge

      Re: Down for the count

      "Why exactly is it so embarrassing not to have a physical standard for the Amp?"

      First find your infinitely long wire. Then find another.

      1. Anonymous Coward
        Anonymous Coward

        Re: Down for the count

        I've just started cleaning out my shed and I expect I'll find several in there. It appears I have at least two of every tool mankind has ever used. When you do most of your home repairs yourself, like I do, but have limited storage and the work is infrequent it is often faster to drop down to the hardware store and buy another "whatever" than dig around to find one of the many you know you have.

        So, I'll keep my eyes open for a couple of those infinite length wires.

        1. phuzz Silver badge
          Facepalm

          Re: Down for the count

          "So, I'll keep my eyes open for a couple of those infinite length wires."

          Yes, but like every other wire, hose or cable in your shed, by the time you need it, it'll be hopelessly tangled.

        2. Stevie

          Re: Shed Contents

          Next time you are in there, see if you can find the bearings for my drill press. I bought a new set about eight years ago, put them away in a safe place and now can't find the buggers behind the coils and coils of infinitely long wire, old junction boxes, boxes of screws and whatnot.

      2. petur

        Re: Down for the count

        "First find your infinitely long wire. Then find another."

        nah, just cut the first infinitely long wire you found in two

  2. wolfetone Silver badge

    How broken can it actually be if it's been around since 1946 in its current form? Are people just looking for jobs to do or something?

    1. Anonymous Coward
      Anonymous Coward

      "How broken can it actually be if it's been around since 1946 in its current form? Are people just looking for jobs to do or something?"

      How broken can sharp sticks actually be? They've been around stabbing cave bears since 500000BCE.

      This fire stuff doesn't seem to have an application - are people just looking for jobs to do or something?

    2. Draco
      Big Brother

      It's a question of accuracy

      For the average person, for the average use, "more or less accurate" is good enough.

      For most people, if the standard of measure (i.e. meter, gram, ampere) they are using is out by a few parts in a million, they probably won't notice, but for scientists trying to perform accurate (and reproducible) measurements, it matters how accurate, stable, and verifiable that measure is.

    3. Stevie
      Pint

      How broken can it actually be if it's been around since 1946 in its current form?

      I see what you did, wolfetone.

      Beer and crisps for you.

      As for the downvoters, keep banging the rocks together lads. Eventually you'll invent The Clue. 8oP

      1. Anonymous Coward
        Anonymous Coward

        Re: How broken can it actually be if it's been around since 1946 in its current form?

        I agree, wolfetone has potential.

        1. DropBear
          Trollface

          Re: How broken can it actually be if it's been around since 1946 in its current form?

          "First find your infinitely long wire. Then find another."

          Bah, that stuff is easy, but everyone is asking all the wrong questions. What I want to know is given a disaffected railway line, how many amps do you need to pass through your freshly acquired infinite conductors to rip them off the sleepers...?

          1. Black Betty

            Re: given a disaffected railway line

            You might try Photonicinduction for that.

            10 million amps = 1 kg of force. So my ball park guess would be on the order of 100 billion amperes.

            Whilst railway iron is not traditionally found on the standard fuse replacement chart, extrapolation from the half inch bolt, suggests it would slow blow at about the 50-100 Kiloampere mark.

            Hmm, perhaps the dearth of opportunities to apply even the least of the multiplicative SI prefixes to the Ampere suggests that perhaps a smaller SI unit is called for.

            1. Stevie

              Re: the 50-100 Kiloampere mark.

              Solid research there Black Betty, but I believe you have failed to factor in the Sleeper Effect.

              If this railway line were abandoned long ago, it could very well have the traditional creosote and urine-soaked wooden sleepers of yesteryear, which would likely burst into flame and/or char to ash long before the iron ablated. Softened by the terrific current passing through it, each rail would then be free to "sag" sideways toward the other, possibly shorting and invalidating the entire experiment.

              Even concrete sleepers might cause a problem. Not only might the bolts securing the chairs to the sleeper melt through long before the rail, the concrete itself could explode violently due to the radiant heat.

              Either way, safety glasses and ear defenders would seem to be a wise precaution.

    4. Captain DaFt

      "How broken can it actually be if it's been around since 1946 in its current form?"

      Well for everyday uses, like commercial electronics and bog standard electrical engineering, it works just fine.

      It's when the physicists get down to nano and particle physics experiments, the current system is a bit too loose for accuracy.

  3. SeanEllis
    Boffin

    Unit confusion

    s/nanometres/nanoamps

  4. Yesnomaybe

    Someone explain please:

    Why isn't it good enough measuring, oh, let's say 1/1000 of a pico-amp, and just multiplying? Why do they have to measure bigger units? I mean: If they are counting INDIVIDUAL electrons, then surely the error introduced will be vanishingly small? I am not arguing the point, but I would genuinely like to know the reason.

    1. short

      Re: Someone explain please:

      If you want to actually measure a current with these things, you need to push the whole current through a load of them (in parallel), I guess.

      Sure, you might only do it at calibration time to set your simpler ammeter to match the electron-counted numbers, but you do need to be able to do it.

    2. imanidiot Silver badge

      Re: Someone explain please:

      accuracy mostly. The measuring error multiplies with the measurement. So multiplying nano amps to get amps means you also multiply the measurement error by 1E9, so a 0,001% measurement error becomes a 1.000.000% measurement error. Not exactly accurate in other words. Measuring a milliamp means a 0,001% measuring error becomes a 1% error, which is much more acceptable.

    3. Black Betty

      Re: Someone explain please:

      Margin of error.

      The number being aimed at is a bit shy of 10 trillion (6.2415093×10E18 to be more imprecisely precise)

      So 1 pico-ampere is the equivalent to exactly 6,241,509.3 electrons streaming past the post every second. That .3 would never trigger the counter, but nevertheless it exists. So multiply .3 x 10E12 to get 300 odd billion miscounted electrons in every true ampere, and thus an error 50,000 times larger than the units in which we're trying to make our measurements.

      The aim of this exercise is to get the margin of error down to better than one millionth part of the measurement unit. And at that they will still be several orders of magnitude shy of the precision with which the Ampere is defined by other means.

  5. Duncan Macdonald

    Reverse the definition

    I coulomb is defined as the charge that flows in 1 second in a current of one ampere. So a current of 1 ampere is a flow of one coulomb per second.

    Define the coulomb as a specific number of electrons and the definition of the ampere then becomes the flow of that number of electrons in one second.

    The ampere would then be a formally defined unit.

    1. phuzz Silver badge

      Re: Reverse the definition

      "Define the coulomb as a specific number of electrons"

      For which you need a device that counts electrons, and that's easier if they're moving, hence why they're doing it this way round.

      1. W4YBO

        Re: ...and that's easier if they're moving...

        Plus, if the electrons stop moving, so does the current.

    2. Stevie

      Re: Reverse the definition 4 Duncan MacDonald

      Nah, can't do that or you end up with a circular definition that results in a pair of what was known in the newly SI world of the early 70s as Zanzibar Units.

      Brief explanation (sorry, can't find a linkable version online):

      Man holidays in Zanzibar and hears people tell of two other men, one who owns a fabulous clock tower at one end of the island by which everyone sets their watches and clocks, and the other an old sea captain at the other end of the island who fires a cannon every day at noon.

      Our man goes to visit the sea captain, has a nice cup of tea and asks about the cannon, specifically how he knows when it is noon.

      "I look through my old telescope here, and set my watch by the clock in that tower at the other end of the island" says the captain.

      Our man then visits the owner of the clock tower, has an evening cocktail and asks how the man sets his clock so accurately.

      "There's this chap at the other end of the island who fires a cannon at noon ..."

  6. allthecoolshortnamesweretaken

    This will take beancounting to the next level... "Did we really receive all the electrons we've been billed for?"

    BTW, if I'd find one wire of infinite length and cut it, wouldn't I have the two wires of infinite length I need?

    1. Draco
      Pint

      You deserve a drink

      Your BTW is engineering at its finest.

      1. Yet Another Anonymous coward Silver badge

        Re: You deserve a drink

        The difficulty is ordering the infinitely long wire.

        Accounting will want to know how many you need and you can't tell them it doesn't matter.

        1. imanidiot Silver badge

          Re: You deserve a drink

          Just order 1 unit. It's not that hard

          1. vir

            Re: You deserve a drink

            I hear you can get samples for free; no need to bother accounting.

  7. Ugotta B. Kiddingme
    Pint

    "Show me the way to go Ohm"

    definitely best sub head in quite some time. Have one of these ---->

    In about an hour, you might need a lie down.

  8. Stevie

    Bah!

    This standard is probably needed the same way the seven layer OSI standard was - i.e. not at all, and will be just as representative of the way the real world deals with the ampere.

    And nothing about this new way of defining the ampere makes laboratory calibration unnecessary, in fact it makes it - well, no different at all really because labs will continue to calibrate their existing equipment the same way they've been doing for decades.

    Oh yeah, I just bet the metre rules in your local DIY superstore were inscribed by counting the wavelengths of a beam of very monochromatic light.

  9. Dwarf

    A yard stick

    Back in the days when someone needed a yard stick (i.e. the yard) to say "this bigger than that", then an Amp, like many of its kin worked very well and serves today. We still use many of them - an inch being a thumb from tip to the first joint etc.

    I am doubtful that just because new measurements can be dreamed up doesn't always result in a useful measurement.

    Examples

    Light bulbs - we used to be able to look at the pack and work out how bright the bulb will be, but now you have to guess and buy a bunch of them stating in Watts, lumens and the degrees K of light they represent. I don't know about you, but I end up picking a bunch and use the one that looks OK when you turn it on. Hardly energy saving is it - I have bunches of un-used ones left over.

    Then there's the binary vs decimalisation of storage capacities and bit rates and having to explain the differences between a GB and a GiB and Gb/s to Gib

    There are also a couple of things that seem to have been overlooked too - I currently (pun intended) have a current clamp meter that you put around the wire, how would they solve that problem as there is no single quantum dot to pass the current though.

    Probably we will we just end up with yet another measurement range on our meter - proper current(Amps) or 2016 Si / Euro / ANSI / pick regulatory body Amps.

    1. vir

      Re: A yard stick

      "I currently (pun intended) have a current clamp meter that you put around the wire, how would they solve that problem as there is no single quantum dot to pass the current though."

      There is no problem. Even extremely accurate voltmeters don't have a Josephson Junction standard built in. This new device will - if successful - serve as a new primary standard, which is used as a reference to check the accuracy of subordinate standards and so on and so forth down to the accuracy that someone requires for their specific task.

  10. Stevie

    Bah!

    The amusing part to all this furor is that six men walked on the moon and did it using machines built using slide rules and imperial units (and old fashioned amperes), whereas in the Brave New World of Robots Everywhere Except Low Earth Orbit we cannot abide the ampere as it stands.

    I guess the new rule of thumb is: Engineers Make Do, Scientists Argue.

    1. Ken Hagan Gold badge

      Re: Bah!

      What furor? I don't see anyone who is actually involved in propogating high-precision measuring equipment who is arguing against the desirability of this. Sure, if you buy a ruler at the newsagent then it probably wasn't calibrated by counting wavelengths of anything, but it was almost certainly calibrated against something that was itself calibrated against ... [repeat no more than a few times] ... exactly that.

      And those clever engineers building moon rockets would almost certainly have wanted a fairly precise ruler to build the parts for their air-tight capsules, or those engines that burn 5 tons of kerosine per second and only stay solid because they have five tons of coolant (kerosine, as it happens, because they had some handy) flowing past on a one-way trip.

      1. imanidiot Silver badge

        Re: Bah!

        In the days of the moon men they built the slide rule equivalent: a room wide sliderule, big enough to get the small subdivisions they needed to achieve the needed accuracy.

        http://www.tertisco-alexandru.com/images/alleman-slideruleWeb.jpg

        Unfortunately i cant find the picture i was after which was of a group of engineers in a room using a sliderule that literally covered the wall.

        1. Stevie

          Re: Bah! Room wide sliderule

          Are you under the impression that calculations done on a sliderule are exact?

          A sliderule isn't a pocket calculator, it is an aid to doing the paper and pen math without log tables in the first pass. A skilled user can come pretty darned close with an instrument about twelve inches long.

          Then, if you are Grumman you run the calculations through a something called a computer. Yes, they had them then.

          The thing in your picture isn't intended to give "greater accuracy" through size, it is intended to be used by a teacher in front of a class of kids learning to use their brand new slide rules. The size is so those at the back of the class can see and follow along on the one in their hands.

          I disbelieve the room-sized moonshot sliderule on accoun of I knoww why such a thing would have more problems than it solved. Perhaps you are thinking of a Mural Quadrant, a device used in Elizabethan times to map the sky accurately?

  11. G R Goslin

    It's all relative

    The snag with standards, is that they're all based on other standards. So ultimately, they're all based on all the other standards. So, which is the real standard? It's all about as real as the old story about how many angels could dance on the head of a pin

    1. Black Betty

      Re: It's all relative

      Except counting "angels" is exactly the point of this exercise and other SI defining exercises.

      Thus the unit of time (seconds) changed from a fraction of a day to a period in which a certain type of atom oscillates a specific number of times.

      The metre was originally defined as a fraction of the Earth's polar circumference at the longitude of Paris, which embarrassment was corrected by making a stick the same length out of an invariant material and calling that the standard metre. Recently it was redefined as the distance light travels in a vacuum in 1/299 792 458 of a second.

      Mass was first defined around a specific volume of a liquid notorious for the variability of it's properties. Cue another correction by making a rock out of the same stuff they made their stick. Problematically, this rock and others made using it as a model now fail to agree with each other as to what a kilogram is.

      So work is underway to calculate the exact number of atoms in a perfect crystal of isotopically pure silicon polished into a perfect sphere in order to redefine mass as a function of length (ie. the distance between atoms in a crystal) just as length itself was itself redefined as a function of time and a physical constant c.

      The end result is the three primary units of measurement, duration, distance and mass are now related to the characteristic behaviour of one atom under photonic stimulation, the distance a photon can travel while our one atom does it's thing so many times, and the number of a different type of atom that can be packed into a volume possessing a single defining characteristic (radius) that can be related to our photon's itinerary.

      The idea is/was to do away with reliance on external events and macroscopic physical models which might be altered or destroyed. Good luck finding a toolkit more basic than 2 atoms and a photon.

  12. JeffyPoooh
    Pint

    Meanwhile, in the lab next door...

    They're proposing to use the 'Watt Balance' to define the kilogram.

    Note: Watt = Volt x Ampere. <- See the issue? Pesky ill-defined Ampere. Useless.

    I think it's time to forget about using the Watt Balance to define the kilogram (Sorry kids), but go with the polished sphere approach instead.

    And guess what that means...

    The Watt-Balance folks can turn their kit around to define the Watt in terms of the new spherical kilogram.

    With the Volt already defined (Josephson), the definition of the Ampere falls out.

    Done.

    I hope this helps.

    1. Black Betty

      Re: Meanwhile, in the lab next door...

      Except like with the second, metre and kilogram, the point is to get as fundamental as possible, and that means counting actual electrons, rather than accepting a number that falls out of an equation.

  13. drdr6

    Fundamental physics interest, not just insignificant trailing digits

    30 years ago I was involved with this in a small way: voltage standards have to be moved around and I worked on standards approaching parts in 10^8 accuracy. At the time NPL were trying to use a current balance - the current means of defining an Ampere from a Kg - in reverse, defining quantum Kilogram from a quantum Ohm and quantum Volt (obtained from Josephson Junctions, and the "volt" needed to moved from one lab to another)

    The interest at the time was to be able to correlate the two Kg measurements - the physical Kg (in Paris) with the quantum Kg. Any statistcally significant difference would have meant going back to fundemental physics and looking for hitherto undiscovered second-order terms in the equations.

    Those equations and definitions of the constants involved are fundamental to our understanding of the world around us.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like