Neutrinos go faster then light !

The place to discuss particle physics
Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Neutrinos go faster then light !

Post by Kasuha » Mon Oct 03, 2011 10:19 am

adam_jeff wrote:But so far, no-one seems to have suggested a mechanism that would mean there are more neutrinos sampled from the front of the pulse than the end.
There is one difference in conditions for pulse head versus pulse tail - pulse head hits way cooler target material than the tail. The question is whether there is a difference or not. But if there is any, I'd expect the head to be somewhat "sharper", less blurred than the tail. But of course I have no idea what the temperature change on the duration of the pulse is.
adam_jeff wrote:Another thing to consider: most of the analysis so far has concentrated on the edges of the pulse (which are most important from the timing point of view). But the whole pulse is measured, and if there was a skew like you suggest there would be a slope on the measured profile compared to the one which is sent. There isn't, as shown in the graphs below.
The 60 ns difference is almost invisible on these graphs. These are for the "initial match" where they had the CERN BCT off by about a microsecond, that's why there's "1048.5 ns". Sure the match used all the pulse width, but internal differences are not all that great, it's edges what drives most of the match.

That's why I think they should try measuring with LHC beam for a while, similar to what they used to align their BCTs. LHC beam has ~10 ns pulses, that's nothing but edges at their resolution. They can get much better fit on these even if they recognize only a handful of neutrinos from background.

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Mon Oct 03, 2011 11:58 am

Yes I agree, it's the edges that drive the fit. But I posted the graphs to illustrate that the neutrino production & detection efficiency must be constant throughout the pulse (and not affected, for example, by the heating of the target) since the pulse shape of the detected neutrinos is the same as that of the proton beam.
Chelle posted a plot and blog post illustrating how selective sampling could make it appear that something was travelling faster than it really is. But if this were true the measured pulse ends up shorter and/or a differrent shape to the original. That's not what's observed.
I agree that the results would seem more compelling if measured with a short pulse, but in reality that would only improve the statistical fit, and that's already at six sigma so plenty good enough. What's much more likely is that there could be a calibration error, that is a mistake in calculating the delays in the acquisition system, and then a short pulse wouldn't help.

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Neutrinos go faster then light !

Post by Kasuha » Mon Oct 03, 2011 1:01 pm

adam_jeff wrote:... since the pulse shape of the detected neutrinos is the same as that of the proton beam
Whole pulse is over 10,000 ns long. 60 ns is just 0.6 % of it - that is about a pixel on that graph. It will still match the waveform pretty well if you shift the red curve one pixel to the right.
adam_jeff wrote:But if this were true the measured pulse ends up shorter and/or a differrent shape to the original. That's not what's observed.
Actually it does appear shorter, it's just not apparent from the overall graph - but the graphs showing match on edges show that there's more shift due to the trailing edge than due to the leading edge. And the data at the start have way bigger spread than at the end and there's a lot of noise in the data.

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Tue Oct 04, 2011 8:17 am

Chelle wrote:
adam_jeff wrote:Well done for managing to get in a mention of your theory via a duck.
 
Thank you, her name is Neutrina : )
There's just one problem: the duck is moving slower than the speed of waves on water. If it was moving faster or the same speed, you would just see a shock wave at the front of the duck and no waves propagating in front of it. So a particle which is traveling at the speed of light would not have any waves in front of it.
I don't think so. 

The moment Neutrina starts to peddle forward, she will accelerates for a short while, and a bow wave is generated that is at first faster than her, and it expands for a limited distance away from her (see pic.). Once she is cruising at a constant speed, the bow waves travels along with her at that cruise control velocity. 

Now let's look at the Neutrino boys that are traveling from CERN to the Opera house, and you have very sensitive ears, than you'll 'hear' them arriving just before they 'really' arrive, due to the noise they make (bow waves). So them Neutrino's would always be detected a little bit before they actually arrive, as in 2 parts of a million, so to say?
Well, that's true only if Neutrina is moving slower than the soeed of the water-waves to start with.
Of course, ducks have mass and therefore finite acceleration, so as you say, when Neutrina starts moving, she will generate a bow wave, and if she then accelerated to wave-speed (but no faster) then these waves will stay a fixed distance ahead of her and would arrive at a duck-measuring experiment a little before her.
BUT, her speed has not exceeded wave speed, even if you measure by her bow-wave. Her true speed is slightly lower than wave-speed (because of the time taken for her to accelerate), measuring her arrival by her bow-wave would lead you to overestimate her speed as exactly wave-speed. But you will never measure the speed as higher than wave-speed

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Tue Oct 04, 2011 8:24 am

Kasuha wrote:Actually it does appear shorter, it's just not apparent from the overall graph - but the graphs showing match on edges show that there's more shift due to the trailing edge than due to the leading edge. And the data at the start have way bigger spread than at the end and there's a lot of noise in the data.
If the shift calculated by the leading and trailing edges seperately were different, by a significant amount, then that would be very interesting and *might* support your case.
Can you show this data? I can't seem to find it in the original article.

User avatar
chelle
Posts: 741
Joined: Thu Nov 19, 2009 11:32 am
Location: XL - BXL - B

Re: Neutrinos go faster then light !

Post by chelle » Tue Oct 04, 2011 9:38 am

Thanks.
adam_jeff wrote:BUT, her speed has not exceeded wave speed, even if you measure by her bow-wave.
That is correct speed stays the same, but depending on what you measure there is a slight difference in arrival time.
Her true speed is slightly lower than wave-speed (because of the time taken for her to accelerate), measuring her arrival by her bow-wave would lead you to overestimate her speed as exactly wave-speed.
That isn't necessarily the case. Lets say that she is on land instead of in the water, and she starts waddling towards you. Than the flapping sound produced by here feet hitting the floor, travels much faster than it takes for her to take her first step. In that case over a distance of 730 km you will hear her coming lets say 18 meters before she is there. Did she travelled faster or slower? no, she just got noticed a fraction earlier.
But you will never measure the speed as higher than wave-speed.
Yes and no, the difference lays in what you measure, with or without bow wave.
Dance, even if you have nowhere to do it but your own living room.
Wear Sunscreen by Baz Luhrmann - Mary Schmich

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Neutrinos go faster then light !

Post by Kasuha » Tue Oct 04, 2011 3:15 pm

adam_jeff wrote:If the shift calculated by the leading and trailing edges seperately were different, by a significant amount, then that would be very interesting and *might* support your case.
Can you show this data? I can't seem to find it in the original article.
I don't have the actual data and I wasn't calculating any fits. But if you take a look at zoomed edge fits (Figure 12 of the paper) you can see that the points are under the curve more than above it - similar for both edges. And don't forget that apart of the real experiment neutrinos there is also some background (not big but it's still there) so I'd expect the curve to be the background margin below the data, not above.
And on the presentation one of questions was whether they were taking the pulse length scaling factor as a free variable for the fit and the answer was no.

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Wed Oct 05, 2011 7:59 am

Kasuha wrote:
adam_jeff wrote:If the shift calculated by the leading and trailing edges seperately were different, by a significant amount, then that would be very interesting and *might* support your case.
Can you show this data? I can't seem to find it in the original article.
I don't have the actual data and I wasn't calculating any fits. But if you take a look at zoomed edge fits (Figure 12 of the paper) you can see that the points are under the curve more than above it - similar for both edges. And don't forget that apart of the real experiment neutrinos there is also some background (not big but it's still there) so I'd expect the curve to be the background margin below the data, not above.
And on the presentation one of questions was whether they were taking the pulse length scaling factor as a free variable for the fit and the answer was no.
Well, I'm not sure that you can tell by eye that the pulse gets shorter. There are the same weight of points above and below the line to my eye. But I agree, it would be interesting to see a fit without constraining the pulse length.
Nonetheless, I still don't see the mechanism for this. I could accept a mechanism which gradually reduces the neutrino production / detection efficiency. But this isn't the case because the pulse shapes match, as I posted above. You need a mechanism which suddenly kills the efficiency about 90ns before the end of the pulse (so that the fit, overshooting at both ends, would be 60ns out overall). I don't see how a gradual temperature rise in the target can cause that kind of sudden drop - or any other mechanism for that matter.

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Wed Oct 05, 2011 8:41 am

[quote="Chelle"]That isn't necessarily the case. Lets say that she is on land instead of in the water, and she starts waddling towards you. Than the flapping sound produced by here feet hitting the floor, travels much faster than it takes for her to take her first step. In that case over a distance of 730 km you will hear her coming lets say 18 meters before she is there. Did she travelled faster or slower? no, she just got noticed a fraction earlier.quote]

Well, now you're just bringing new waves into it - sound waves. My point stands - you can't measure a speed (calculated from the arrival time) which is faster than the fastest wave you have in the system.
So if you put the error down to measuring the neutrinos' bow wave, then that bow wave would have to be travelling faster than light.

User avatar
chelle
Posts: 741
Joined: Thu Nov 19, 2009 11:32 am
Location: XL - BXL - B

Re: Neutrinos go faster then light !

Post by chelle » Wed Oct 05, 2011 9:16 am

adam_jeff wrote:Well, now you're just bringing new waves into it - sound waves
Yes, nothing can travel without making a sound :mrgreen:

Of course this is speculative, but as mentioned in an other topic (A quantum take on certainty) there is some 'proof' that pilot-waves might exist. The Neutrino experiment could be an additional argument, if these 'faster then light' results are confirmed by other tests.
My point stands - you can't measure a speed (calculated from the arrival time) which is faster than the fastest wave you have in the system.
So if you put the error down to measuring the neutrinos' bow wave, then that bow wave would have to be traveling faster than light.
That's fair to say.
Dance, even if you have nowhere to do it but your own living room.
Wear Sunscreen by Baz Luhrmann - Mary Schmich

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Neutrinos go faster then light !

Post by Kasuha » Wed Oct 05, 2011 11:34 am

adam_jeff wrote:You need a mechanism which suddenly kills the efficiency about 90ns before the end of the pulse
I don't think I need to come up with working mechanism. I just think that the pulse is a bit shorter and I don't know why. They think the pulse is a bit too early and they also don't know why. You also don't ask them to come up with a mechanism that'd explain it. And I believe the pulse being a bit shorter is closer to current knowledge about laws of physics than the pulse being a bit too early.

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Thu Oct 06, 2011 8:09 am

Sorry - I didn't mean that you, personally, had to come up with a mechanism right now or I wouldn't listen to your theory. I should have said, "There would have to be a mechanism which suddenly kills the efficiency 90ns before the end of the pulse...", which still seems rather implausible to me.
But as you say, the whole measurement is rather implausible...

adam_jeff
Posts: 49
Joined: Thu Nov 18, 2010 3:07 pm

Re: Neutrinos go faster then light !

Post by adam_jeff » Thu Oct 06, 2011 1:09 pm

Here's a paper which might be of some interest:
http://www.sciencedirect.com/science/ar ... 9385904605
Published in 1985, the authors predict that 1 neutrino flavour should have imaginary mass and therefore be faster-than-light. The prediction arises out of the equations necessary to explain neutrino oscillation.


[ I'm not sure if the above link will work without a subscription. I'm sure the paper exists in free access somewhere though. Chodos, A. (1985). "The Neutrino as a Tachyon". Physics Letters B 150 (6): 431. Bibcode 1985PhLB..150..431C. doi:10.1016/0370-2693(85)90460-5. ]

User avatar
chelle
Posts: 741
Joined: Thu Nov 19, 2009 11:32 am
Location: XL - BXL - B

Re: Neutrinos go faster then light !

Post by chelle » Thu Oct 06, 2011 1:48 pm

adam_jeff wrote:Here's a paper which might be of some interest:
http://www.sciencedirect.com/science/ar ... 9385904605
There is a whole list (38) of those:
http://inspirebeta.net/search?ln=en&p=r ... id%3A15887
Dance, even if you have nowhere to do it but your own living room.
Wear Sunscreen by Baz Luhrmann - Mary Schmich

Harbles
LHCPortal Guru
Posts: 314
Joined: Sat Nov 28, 2009 10:22 pm
Location: Mississauga, Ontario. Canada

Re: Neutrinos go faster then light !

Post by Harbles » Thu Oct 06, 2011 3:52 pm

And yet more criticism of technique. This time the effect of g being different at the two locations.
http://www.nature.com/news/2011/111005/ ... 1.575.html

Post Reply