New definitions of SI units

Most of you will be aware that the SI units are being redefined in terms of physical constants like Planck's and Boltzmann's and the elementary charge. We'll at last be rid of the last standard artefact, the kg, a lump of metal in a vault near Paris, France.

Central to the new definition of the kg stands the Kibble balance (Watt balance), that relates mass to voltage and current, which can me measured with great precision using interferometry, the Josephson effect and the quantum hall resistance.

One factor that enters in the determination of the new kg is the local value of g, the acceleration due to gravity. I have been unable to find a description of how that is done. The issue is totally glossed over in the publications I've looked through. It's done away with merely by saying that it's measured using an accurate absolute gravimeter.

Right.

Does anyone know how this is done? Surely that instrument must be pretty sophisticated itself?

Jeroen Belleman

Reply to
Jeroen Belleman
Loading thread data ...

An absolute gravity meter is essentially an optical laser interferometer which measures the free-fall acceleration of a retroreflector in a vacuum. The measurement is directly referenced to atomic standards of length and time. The laser is stabilized to hyperfine optical absorption peaks in an iodide absorption cell and the time base is locked to hyperfine microwave transitions in a rubidium vapor cell. etc...

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

Some absolute gravity measuring instruments: Some models use a "zero-length spring":

With such a device, you too can make lumpy 3D images of the earth's geodic height:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

If it really "relates mass to voltage and current" then does gravity come into it? If it really relates force to voltage and current then I suppose it must. Not sure whether THE kg is a standard of mass or weight...

Mike.

Reply to
Mike Coon

Not "will be". The changeover was ten days ago.

No idea. But I do know about the other contender for the new standard, which also reached the required repeatability and precision about the same time, in 2014. The Watt balance was chosen over the crystal ball because it's more easier and cheaper to replicate. The ball approach works like this.

We know the bond length in a crystal of silicon 28 very very precisely. Of course, it was the IC industry which established this. So the half-dozen labs took a mono-crystalline ball of isotopically pure Si28 and manually polished it using laser interferometry until it was more perfectly round than anything ever made before. Because we know the bond length we can calculate the number of atoms, and that forms a new standard of mass. The crystal blank they started with cost about EU800K.

My wife works at the Sydney facility that made the best of these balls.

Clifford Heath.

Reply to
Clifford Heath

Huh, no idea. I seem to recall a gizmo just timing a test mass dropping in vacuum at an APS meeting years ago. It looked to be about a 1 meter drop.

I'm not sure how they measured the height... some interferometer counting the number of wavelengths?

George H.

Reply to
George Herold

Oh right! Dropping retro-reflector. GH

The

Reply to
George Herold

It's a two-step process: The test mass is balanced by the force developed in a voice coil arrangement: Mg=BIl, that is, the weight of the mass is balanced by the Lorentz force of a current I flowing through a wire of length l suspended in a field B. B and l aren't known to the required accuracy, which motivates the second step: The coil is then moved through the field and the voltage and velocity are measured: U=Bvl. (I'd worry about B staying the same, but I digress. There's a lot more to it than that, I'm sure.)

Re-arranging gives M=UI/gv. The poorly determined B and l drop out. U, I and v are measured using the Josephson effect, the quantum Hall effect and interferometry respectively. That leaves g to be determined independently. Thanks to Jeff, I now know that that can be done to an accuracy in the 10nm/s^2 ballpark, which is good enough.

The kg is a standard of mass.

Jeroen Belleman

Reply to
Jeroen Belleman

Australians with balls:

Reply to
bitrex

Wouldn't temperature measurement be a big problem there? Or can that be canceled somehow?

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Mass isn't influenced by temperature. The dimensions of the silicon sphere would be influenced by its temperature, but you could always take it down to a few milli-Kelvin above absolute zero before you measured its dimensions.

The effect isn't large

formatting link

and measuring the temperature of the ball to a millidegree or so might be all that you'd need.

--
Bill Sloman, Sydney
Reply to
bill.sloman

I'm not sure how much accuracy in measuring gravity is needed but the available instruments seem to be interested in high accuracy and stability over long measurement periods. For example, from: Sensor housed in an insulated double-oven ensuring accurate temperature control. and Three sealed chambers isolate the sensor from humidity and pressure changes. and It has a true vacuum seal so that it is completely insensitive to buoyancy changes due to atmospheric changes.

If the drop test is that sensitive to a few gas molecules that might leak into the drop chamber, then the seals must be really quite good. Ok, I'm impressed.

Probably better and more expensive. Superconducting Gravimeter:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

The kg suffered from drift of the order of 5e-8, while physical constants like h and e could be determined with much better accuracy, if only the kg would stay put. It made good sense to turn things around and define the kg in terms of h and e. That also solves the problem of having to carry the standards around to compare them from time to time. Now anyone with the required knowledge and a budget could create his own kg and get it to be the same as the others to a relative tolerance of ~1e-9.

I wasn't aware that the acceleration of gravity could also be measured to a relative accuracy in the 1e-9 ballpark, and that bit was glossed over in the descriptions of the new SI.

These gravimeter papers, by the way, are irritating in their mish-mash of units. They randomly mix Gal, g, m/s^2, and pile up scientific notation with multiplier prefixes. You'd almost think they're trying to hide something.

Jeroen Belleman

Reply to
Jeroen Belleman

Jeroen Belleman wrote

One thing that bugs me in all this is to make things depending one 'one' natural constant, be it 'tricity or whatever.

As the _known_ universe unfolds those constants may well change / have likely changed. You would not notice and then at one point nothing would match.

One can for example wonder WHY the Paris kg was different (lighter! IIRC) from all the other ones that were brought to it for compare.

THAT is interesting.

Reply to
<698839253X6D445TD

As far as we can tell, e and h vary *much* less, if at all, than the kg. Astronomical observations provide evidence that these constants have remained the same (but to what tolerance?) over vast ranges of time and distance. What would be the theory that would have e and h vary?

Yes, indeed. I saw a publication attributing that to the absorption of mercury vapour. Platinum is well known to also readily absorb hydrogen. One would think that neither would be allowed a chance to get anywhere near the (ex) standards.

I don't really know how these things were handled. The only thing that gets mentioned repeatedly is that the Paris kg is stored under triple glass bell covers, which is relatively uninteresting. The really interesting bit is how to manipulate it without affecting its mass.

Jeroen Belleman

Reply to
Jeroen Belleman

OK those things are different from what I recall. (But my memory could be bad.) Those all look like an oscillating mass/ spring systems. Google this, falling corner cube gravimeter. (myabe that was in one of your other links... I didn't click them all.)

George H.

Reply to
George Herold

Been thinking about that, all copies of the kilogram seem to be heavier than the reference in Paris. What is the difference? Those copies were moved, some moved over long distances. If we leave pure chemistry out of it, that leaves relativity. COULD it be that by moving a mass we sort of put energy in it, and as a result mass increases? For an atom that would be a tiny tiny bit, but there are a lot of atoms in that kilogram.

If you start looking at it that way you quickly come to MOND, missing mass problem, speed of stars in galaxies.

In my opinion all these constants and measurements we have done here on earth are only valid in our reference frame in this universe, Mach comes to mind.

Nice subject:-)

Maybe drop the idea when you see one of those scientist there in CERN... :-)

formatting link

Reply to
<698839253X6D445TD

A vacuum and hopefully a lack of transverse B field? And a carefully chosen mass? Keeping gravity experiments insensitive to other forces is a difficulty, sometimes. Heck, thermocouple effects can charge the dropping item if it has metal parts...

Reply to
whit3rd

Mach didn't even accept the physical reality of atoms.

Scientists don't hang around CERN - the data is e-mailed to them after it has been collected and processed.

--
Bill Sloman, Sydney
Reply to
bill.sloman

There are plenty of scientists at CERN. Not all of them are so volatile. A simple change of the definition of measurement units doesn't change the underlying physics, but I'm sure they're smart enough to question this themselves. It's their job to explore the basic workings of physics.

The data is certainly not emailed; there is far too much of it. It's distributed over a 'computing grid' composed of multiple levels of processing and storage tiers, all over the world. It can be accessed by its users from any place they choose. Data analysis can take quite a while and often data are analyzed multiple times to find answers to different questions that may come up.

Jeroen Belleman

Reply to
Jeroen Belleman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.