A change in √G means that in vacuum you need a higher voltage and/or a dielectric constant.

In vacuum, √G < 2.58 x 10^{-4}.

Okay, let's assume you're right, and G isn't a constant.

In the B-B effect, (

https://arxiv.org/pdf/gr-qc/0502047.pdf eq. 72) F is proportional to sqrt(G), right? So a 10% change in G would be a 3% change in sqrt(G).

You could compensate for this by increasing the voltage by 3%, or the dielectric constant by 10%.

Except it seems like in reality, these variation in G are <<10%, so you'd need to change the voltage applied by <<3%.

So how could a 1/2000 change in G possibly affect the voltage you need to apply by so much that you're claiming it's the reason that some people aren't seeing an effect? You could estimate the voltage change has got to be <1V at a maximum, surely that's just within the noise of your kV or MV power supply?

What exactly is your point here?