CERN team claims measurement of neutrino speed >c

[Histspec]

The Gran Sasso experiments OPERA, ICARUS, LVD, BOREXINO presented preliminary results of the new neutrino speed measurements in May 2012- they are consistent with the speed of light within margin or errors:

http://francisthemulenews.wordpress...pera-en-2011-y-los-nuevos-resultados-de-2012/ (in Spanish)

Borexino: δt = 2.7 ± 1.2 (stat) ± 3(sys) ns
ICARUS: δt = 5.1 ± 1.1(stat) ± 5.5(sys) ns
LVD: δt = 2.9 ± 0.6(stat) ± 3(sys) ns
OPERA: δt = 1.6 ± 1.1(stat) [+ 6.1, -3.7](sys) ns

OPERA has also revised their 2011 results and will resubmit it to the "Journal of High Energy Physics":
δt = (6.5 ± 7.4 (stat.)+9.2 (sys.)) ns

Also MINOS from Fermilab corrected their former results
δt = −11.4 ± 11.2 (stat) ± 29 (syst) ns (68% C.L)

So Einstein is still laughing...

 

[ZapperZ]

The press announcement from CERN on this result can be found here:

http://press.web.cern.ch/press/PressReleases/Releases2011/PR19.11E.html

Zz.

 

[sshai45]

Yet more awards to put into the impressive trophy collection Einstein's theories hold...

 

[StevieTNZ]

The latest on Physorg:
http://phys.org/news/2012-06-einstein-neutrino.html

Scientists on Friday said that an experiment which challenged Einstein's theory on the speed of light had been flawed and that sub-atomic particles -- like everything else -- are indeed bound by the universe's speed limit.

 

[lmoh]

Man, I forgot all about this! Hard to believe that it used to one of the big news stories that shook the science world in 2011.

Perhaps somebody should lock this thread up. No sense in beating a dead horse at this point.

 

[lalbatros]

Man, I forgot all about this! Hard to believe that it used to one of the big news stories that shook the science world in 2011.

Perhaps somebody should lock this thread up. No sense in beating a dead horse at this point.

Most of the posts could be deleted, if not the whole thread.
After all, it was just an experimental difficulty, and most comments were irrelevant.

 

[TrickyDicky]

There's still something I find difficult to understand, the loose wire apparently can account for the 60ns advanced signal but looking at the original long experiment from 2009-2011 and specifically to figure 12 in the original paper, one knows that those 60ns were simply an average and that in fact neutrinos were detected in a range of δt from 1ns to over a hundred nanoseconds (at least that's what I gather, please correct if not the case). So I can't exactly see how a fixed sistematic error can give that dispersion in the detection times. Perhaps someone can shed some light about this.

 

[espen180]

I think the most likely explanation is that the signal delay due to the cable was not exactly 60 ns, but some distribution with mean 60 ns or higher and finite variance.

 

[TrickyDicky]

I think the most likely explanation is that the signal delay due to the cable was not exactly 60 ns, but some distribution with mean 60 ns or higher and finite variance.

Well, maybe so, still it would look to me a excesively "ad hoc" explanation but then I guess the only way to know for sure is repeating the experiment again in similar temporal circumstances to see if that distribution is eliminated.

 

[TrickyDicky]

In any case if the loose wire had such a behaviour I'm not sure it would correspond to a sistematic type of error.

 

[espen180]

No physical measurements have zero uncertainty. Any attempt to measure the delay of the wire will yield some distribution of delay times.

 

[Drakkith]

Well, maybe so, still it would look to me a excesively "ad hoc" explanation but then I guess the only way to know for sure is repeating the experiment again in similar temporal circumstances to see if that distribution is eliminated.

You want to repeat the experiment with a loose wire just to see if that particular fault was the cause of the errors? There's no reason to. They fixed it and the results immediately showed a change. In any case i doubt one could reproduce the exact amount of "looseness" in the wire the previous experiment had, so I don't know if it could even work.

 

[Histspec]

Well, maybe so, still it would look to me a excesively "ad hoc" explanation but then I guess the only way to know for sure is repeating the experiment again in similar temporal circumstances to see if that distribution is eliminated.

They are sure, because they compared the arrival times of cosmic muons in the LVD and OPERA detector (160m apart). They found, that between 2007-2008 and December 2011-2012 the data were in agreement, but between mid 2008 to December 2011, there was a 74ns discrpency, which vanished when the loose cable was fixed.
Second, they found another error due to an oscillator, contributing 15ns in the opposite direction.
Together, those effects caused the 59ns "anomaly". That's all.

In any case if the loose wire had such a behavior I'm not sure it would correspond to a systematic type of error.

Well, if the time of every single event is displaced by 59ns, then of course also the whole spectrum is changed in the same way. (OPERA will republish their results based on this error analysis).

Also note that in October-November 2011 (before the cable was fixed), OPERA used short bunched beams, which evaluation didn't require those statistics - also here the discrepancy was 60 ns.
But ICARUS measured the same bunched beam at the same time, and found no discrepancy. And in May 2012, another bunched beam run was performed, and OPERA, ICARUS, LVD, Borexino found no discrepancy too.

Opera: What went wrong
End of the Opera story

 

[TrickyDicky]

You want to repeat the experiment with a loose wire just to see if that particular fault was the cause of the errors? There's no reason to. They fixed it and the results immediately showed a change. In any case i doubt one could reproduce the exact amount of "looseness" in the wire the previous experiment had, so I don't know if it could even work.

No, you got it wrong, I meant with the wire tightened. Note I said I expected to have the odd distribution eliminated by doing that.

 

[TrickyDicky]

Well, if the time of every single event is displaced by 59ns, then of course also the whole spectrum is changed in the same way.

Wait, do you mean then
neutrinos actually had that broad variation in their speed but just shifted 59ns to have close to light speed average?

 

[Dickfore]

Wait, do you mean then
neutrinos actually had that broad variation in their speed but just shifted 59ns to have close to light speed average?

yes, see their table of systematic errors.

 

[TrickyDicky]

yes, see their table of systematic errors.

How can neutrinos have such different speeds? in such a short distance?

 

[Dickfore]

How can neutrinos have such different speeds?

Please don't spam old threads. What is your point of concern? nanosecond is not a unit of speed, so please clarify.

 

[TrickyDicky]

Note that the mentors have repeatedly told posters to read the original opera paper before posting here, anyone that has read it or has some notion about the relation of distance versus time (in ns or any other units) should see how the different δt (in ns) are related to neutrino speeds.

 

[ZapperZ]

Why is the original OPERA paper still being debated here? It is clear that the result from that paper no longer holds water.

Zz.

 

[TrickyDicky]

So far I've obtained two different explanations to my query, one that the loose wire error is purely sistematic and fixed (74ns) and the other that it actually it is responsible for the broad variation of $\delta$t in the first longer Opera experiment from 2008-2011.
Both answers are incompatible; as I said since the cable problem is considered a sistematic error I was thinking in terms of the first explanation, and with the reasonable assumption that neutrinos speed cannot oscillate so much in such a short distance (732km), I'm still missing something in the sense that the offered solution would work perfectly if the 60ns $\delta$t was not just an average. Of course my concern is only directed to the original experiment, not to the recent brief short pulsed ones. But I think it is important given the uproar it generated to have it all well clarified.

 

[TrickyDicky]

Why is the original OPERA paper still being debated here? It is clear that the result from that paper no longer holds water.

Zz.

Sorry, I thought this thread was specifically for discussion of that paper, if that is no longer the case I guess I'll just have to try elsewhere.

 

[ZapperZ]

Sorry, I thought this thread was specifically for discussion of that paper, if that is no longer the case I guess I'll just have to try elsewhere.

But the result of that paper has been clearly shown to be invalid! That's the whole point of the last few posts since that CERN press report! Or did you completely missed it?

It makes discussion of the original paper to be entirely moot!

Zz.

 

[TrickyDicky]

Note also that my questions were about that paper in the light of the new information released about the possible source of errors.

 

[TrickyDicky]

It makes discussion of the original paper to be entirely moot!

Even to get a better understanding of how exactly is the result invalidated according to the CERN press report? You give the term "discussion forum" a different sense from the one I'm used to. I thought one of the goals of such forums was asking questions in order to understand scientific issues thru the clarifications of other more learned forum members.

 

[ZapperZ]

Even to get a better understanding of how exactly is the result invalidated according to the CERN press report? You give the term "discussion forum" a different sense from the one I'm used to. I thought one of the goals of such forums was asking questions in order to understand scientific issues thru the clarifications of other more learned forum members.

Unless you are in possession of a detailed report on the exact timing errors that was done in the original OPERA result (i.e. you have the post-mortem analysis of those loose connection), what exactly do you have to base on in doing your "discussion"? The original OPERA paper certainly didn't have any. And the recent report on those loose connectors certainly have been lacking in the details on what exactly is the timing errors and how they were measured. So what exactly are you going to base your discussion on? SPECULATION? Guess work?

The same "philosophy" what was imposed upon in the beginning to urge people to read the original OPERA paper BEFORE they jump in into this discussion is also at work here. It means that the discussion must be based on something concrete, rather than something pluck out of thin air without any basis. Until the OPERA group publish clearly the post-mortem of the original result, you and I do not possesses any kind of data or information to make an informed discussion of what actually happened. So how would such a discussion gives you a "better understanding"? A better understanding on how to make guesses?

Zz.

 

[TrickyDicky]

Unless you are in possession of a detailed report on the exact timing errors that was done in the original OPERA result (i.e. you have the post-mortem analysis of those loose connection), what exactly do you have to base on in doing your "discussion"? The original OPERA paper certainly didn't have any. And the recent report on those loose connectors certainly have been lacking in the details on what exactly is the timing errors and how they were measured. So what exactly are you going to base your discussion on? SPECULATION? Guess work?

The same "philosophy" what was imposed upon in the beginning to urge people to read the original OPERA paper BEFORE they jump in into this discussion is also at work here. It means that the discussion must be based on something concrete, rather than something pluck out of thin air without any basis. Until the OPERA group publish clearly the post-mortem of the original result, you and I do not possesses any kind of data or information to make an informed discussion of what actually happened. So how would such a discussion gives you a "better understanding"? A better understanding on how to make guesses?

Zz.

OK, I understand you are an experimentalist , if you think there is no room for informed discussion from the data published so far I'll take your word. Let's not waste more time. However curiously from your words you seem to think (correct me otherwise) we all must agree that the original Opera experiment is dead and buried and no one should have any doubt about it unless he or she is an imbecile, and that without having all the data to supposedly have an informed discussion according to you as an experimentalist.

 

[Ilmrak]

So far I've obtained two different explanations to my query, one that the loose wire error is purely sistematic and fixed (74ns) and the other that it actually it is responsible for the broad variation of $\delta$t in the first longer Opera experiment from 2008-2011.
Both answers are incompatible; as I said since the cable problem is considered a sistematic error I was thinking in terms of the first explanation, and with the reasonable assumption that neutrinos speed cannot oscillate so much in such a short distance (732km), I'm still missing something in the sense that the offered solution would work perfectly if the 60ns $\delta$t was not just an average. Of course my concern is only directed to the original experiment, not to the recent brief short pulsed ones. But I think it is important given the uproar it generated to have it all well clarified.

I think your question has been misunderstood here. Let me try to interpret it, correct me if I'm wrong.

I read your question like “How could there be such a wide distribution in arrival times of neutrinos? Is it due to the loosing cable?”

If this was your question, then the answer is that the loosing cable caused (mainly) a systematic error which shifted the time distribution without deforming it significantly.
This mean that the variance of the time distribution is still there when you fix the cable. It is due mainly to three contributions (to my knowledge): a difference in neutrinos velocity, that is though negligible; a difference in the path followed (they are not created and received all in the same starting and ending points); other experimental errors.

I hope this could help.

Sorry for the bad English,

Ilm

 

[mfb]

The wide time distribution in the original publication has a simple explanation: The proton beams used to produce the neutrinos were long (~10µs if I remember correctly). Timing was not the main purpose of the experiment, just something which could be done in addition to the mixing measurements. After timing became interesting, they used short pulses (2ns?), as they are better to measure the flight time.

The 60ns were obtained by comparing the proton distribution (in time) with the neutrino distribution. Compared to the speed of light, a shift (but no broadening of relevant size) between the two was observed.