> How much of a shift did they actually see in their 2.5 mile long laser paths?
>
> The news article I saw talked about a distance change of “1/10,0000 the size
> of a proton”. That didn’t seem to make much sense.
>
> Bob
Hi Bob,
The unit of measurement that gravity wave folks use is "strain" which is
unit-less meters per meter. It's analogous to how we T&F people use unit-less
Hz per Hz for oscillator frequency offset and stability measurements. Plus they
have their strain spectral plots like we have our phase noise plots.
>From what I understand, the GW signals they're looking for create a distortion
>on the order of 1e-21 so they want a detector that's in the 1e-22 or 1e-23
>range; in a 20 to 500 Hz bandwidth. This level of precision is just
>mind-blowing. But as you read the wealth of PDF's out there about LIGO, and
>drool at photos of the optics, and understand the plots showing strain
>sensitivity as a function of frequency, you start to believe that this is
>actually possible. Ok, given a thousand PhD's, a billion dollars, and a couple
>of decades.
Yes, the interferometer is 4 km in length but they bounce the beam back and
forth 400 times so the effective length is more like 1600 km. They keep the
mirrors stationary to "picometers". They use hundreds of clever tricks to pull
this off.
Since the press is averse to using scientific notation they tend to make up
units. So it's common to read units like Rhode Island, football field, human
hair, and now, proton. A proton diameter is about a femtometer so 1/10,000th of
that is about 1e-19 meter.
LIGO publishes the raw and processed data -- and this is time nuts -- so
attached is a TimeLab plot for you showing the chirp of the century. The
LIGO/Hanford and LIGO/Livingston data is from:
https://losc.ligo.org/s/events/GW150914/P150914/fig1-waveform-H.txt
https://losc.ligo.org/s/events/GW150914/P150914/fig1-waveform-L.txt
For TimeLab, set scale to 1e-21 and tau to 6.1035e-5 s (1/16384 s). The time
axis is relative to 2015-09-14 09:50:45 UTC plus about 0.25 s.
/tvb
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
