The Non Constant Gravitational Constant (G)?

[Paul Stowe]

TL;DR

The value of G has been recently called into question by new measurements from respected research teams in Germany, New Zealand, and Russia. Their values using the best laboratory equipment to-date disagreed wildly with each other to the point that many are doubting about the constancy of this parameter.

History & evidence can be found in:

https://onlinelibrary.wiley.com/doi/full/10.1002/andp.201900013
https://phys.org/news/2015-04-gravitational-constant-vary.html
{unacceptable reference deleted}

The experimental data strongly suggest are actual x,y,z,t variations in the measured value of G

 

[russ_watters]

Summary:: The value of G has been recently called into question by new measurements from respected research teams in Germany, New Zealand, and Russia.

Do you have any links to peer reviewed research substantiating this claim?

Their values using the best laboratory equipment to-date disagreed wildly with each...

I would hardly consider a 0.06% total span to be "disagree wildly".

History & evidence can be found in:

https://phys.org/news/2015-04-gravitational-constant-vary.html

The experimental data strongly suggest are actual x,y,z,t variations in the measured value of G

That link explicitly states that their conclusion is that the variation is due to periodic rotation speed variation for Earth. It definitely doesn't suggest a real variation in the value of G.

 

[Dale]

https://phys.org/news/2015-04-gravitational-constant-vary.html

Neither this journalistic link nor the scientific paper that it reports on claims that there is an actual variation in G. They only claim an unrecognized measurement sensitivity.

I am a little skeptical of even the claim of a sinusoidal error. With that few data points with such wide error bars measured over that many years and unequally spaced in time the Nyquist criteria would make it hard to distinguish many different frequencies. I would find the data much more convincing if they used a single apparatus at regular time intervals to show this effect.

https://iopscience.iop.org/article/10.1209/0295-5075/110/10002

The data just don’t look sinusoidal to me. A Bayesian analysis seems to suggest that a Gaussian error model is better than a sinusoidal error model.

https://iopscience.iop.org/article/10.1209/0295-5075/111/30002