Hi

As of May 20 2019, there was a major update in the values of the fundamental physical constants.

Some of them became the basis for the definitions of the SI units, and are now assigned exact numerical values. These include c, k, h, e, NA, and others.

Other were updated according to the most recent measurements. This is a process that occurs every four years or so under the management of the CODATA Committee of experts. See https://codata.org/initi...ntal-physical-constants/ for more info on CODATA.

For a complete list of the updated values of the constants see https://physics.nist.gov/cuu/Constants/index.html click on the frequently used constants button to see a pdf of most of the ones you need.

It would be helpful if the constants file could be updated.

Regards, Rod

Hi Jean
Probably some clarification is required….you seem to be using the information incorrectly (I am one of the BIPM experts, so I know this stuff inside out).

The ITS-90 temperature scale is a recipe for temperature measurements.

Step 1: Manufacture. Make a standard platinum resistance thermometer (SPRT) using the purest available platinum wire – there are electrical tests the finished thermometer must pass. These 25 ohm thermometers are remarkable and very different from the usual 100 ohm industrial thermometers, and out of the box and without any calibration they all agree within a couple of hundredths of a degree. There are only a few recognised manufacturers of these thermometers.

Step 2: Calibration. Record the electrical resistance of the thermometer at the reference temperatures specified by ITS-90 within your temperature range of interest (e.g. 0 oC to 420 oC requires 3 measurements). These values are then used to calculate a simple correction equation (an interpolation).

Step 3: Use. After the corrections are applied, you can convert the corrected values of the electrical resistance measured at any temperature to the temperature reading using a standard reference function. The reference function is the complicated polynomial given in the information you have (actually, there are two polynomials that meet a 0 oC). It passes through all the reference points exactly. The reference function was based on a least-squares fit to hundreds of exacting measurements made in the few years leading up to 1990 when the scale was published.

For temperatures below O oC, the corrected measurements with are reproducible (different thermometers, different people, different resistance measuring instruments, measuring the same temperature) to about 0.2 mK. Above O oC, the performance gradually deteriorates but they are good to about 1 mK up to 960 oC, though a lot of expertise is required to make accurate measurements at the higher temperatures.

For temperatures between about -250 oC and +960 oC, this is how temperature is currently defined by the SI. The reference function, and probably also the interpolating equations, could usefully be added to SMath.

If you send a PM with your email address, I can send more detailed information. The BIPM information is a bit dry and confusing for most who have not seen it before.

Cheers
Rod

Hi Jean

Sometime in the next few days I'll work on an updated file for the fundamental physical constants. I'll post it on this forum so someone more expert than me can arrange to have it included in the SMath downloads.

Once I've done that, I'll work on a update of the Temperature units file to include the various temperature reference functions. I'll include text explaining the origin and purpose of the functions. That will take a bit longer.

best regards
Rod

Hi Jean

Yes, it the Callendar-Van Dussen formula is still used for industrial resistance thermometers.

Yes, the BIPM reference function for 25 ohm standard platinum resistance thermometers (SPRTs) is different. You do not need data for the SPRT function, it is already defined.

Rod

Hi Jean

See attached file - this is how temperature under ITS-90 is defined. Anything else is wrong.

ITS-90 (1).pdf (5,450kb) downloaded 25 time(s).

cheers
Rod