Discussion:
fast neutrinos and bad connections
(too old to reply)
RichD
2020-07-08 02:35:15 UTC
Permalink
Recall a few years back, a physics lab in Italy reported
faster than light neutrino observations. After some ado,
they finally diagnosed a dodgy optical interconnect.

How does that result occur? An intermittent interface
would produce a broken signal, or reduced RMS power, or
perhaps data bit errors, if used for communication.

But how does a faulty connection result in a spuriously
high velocity measurement?

--
Rich
Joe Gwinn
2020-07-08 14:29:57 UTC
Permalink
On Tue, 7 Jul 2020 19:35:15 -0700 (PDT), RichD
Post by RichD
Recall a few years back, a physics lab in Italy reported
faster than light neutrino observations. After some ado,
they finally diagnosed a dodgy optical interconnect.
The facility was in Italy, but the work was done by CERN.
Post by RichD
How does that result occur? An intermittent interface
would produce a broken signal, or reduced RMS power, or
perhaps data bit errors, if used for communication.
But how does a faulty connection result in a spuriously
high velocity measurement?
It was explained to me by an engineer from CERN a few years ago, but I
cannot find the reference right now. The delay (in the nanoseconds)
was because an impedance bump due to the bad connection acted as an
low pass filter, slowing the critical edge just enough. As I recall,
the offset was something like 60 ns.

Joe Gwinn
t***@gmail.com
2020-07-10 12:59:36 UTC
Permalink
Post by Joe Gwinn
On Tue, 7 Jul 2020 19:35:15 -0700 (PDT), RichD
Post by RichD
Recall a few years back, a physics lab in Italy reported
faster than light neutrino observations. After some ado,
they finally diagnosed a dodgy optical interconnect.
...
Post by Joe Gwinn
It was explained to me by an engineer from CERN a few years ago, but I
cannot find the reference right now. The delay (in the nanoseconds)
was because an impedance bump due to the bad connection acted as an
low pass filter, slowing the critical edge just enough. As I recall,
the offset was something like 60 ns.
The implication being that the fibre is carrying discrete pulses to provide a time reference for the detected particles, rather than e.g. a network link.

Hence, changing the arrival time of the optical pulse alters the measured speed of the detected neutrinos.

Is that correct?

Thanks

Henry
Joe Gwinn
2020-07-10 18:21:46 UTC
Permalink
Post by t***@gmail.com
Post by Joe Gwinn
On Tue, 7 Jul 2020 19:35:15 -0700 (PDT), RichD
Post by RichD
Recall a few years back, a physics lab in Italy reported
faster than light neutrino observations. After some ado,
they finally diagnosed a dodgy optical interconnect.
...
Post by Joe Gwinn
It was explained to me by an engineer from CERN a few years ago, but I
cannot find the reference right now. The delay (in the nanoseconds)
was because an impedance bump due to the bad connection acted as an
low pass filter, slowing the critical edge just enough. As I recall,
the offset was something like 60 ns.
The implication being that the fibre is carrying discrete pulses to provide a time reference for the detected particles, rather than e.g. a network link.
As you surmise, these are not data network links. Pulses in optical
fiber travel at almost exactly 2/3 the speed of light.
Post by t***@gmail.com
Hence, changing the arrival time of the optical pulse alters the measured speed of the detected neutrinos.
Is that correct?
Yes. The time reference pulses were delayed, making the data pulses
seem to arrive earlier than expected.


Joe Gwinn

Loading...