CERN finds neutrinos faster than light - Page 50
Forum Index > General Forum |
killerdog
Denmark6522 Posts
| ||
DDu
4 Posts
On February 25 2012 13:53 Noro wrote: Its pretty foolish to think "The laws of physics" (a human construct) can explain everything in the universe. Einstien knew that his theories were only theories. There are things in the universe that we don't understand. Thinking that we have everything figured out is pretty stupid. =/ You are extrapolating too far from my post. I also do not believe and never directly implied that the current "laws of physics" should be sufficient to explain everything about the universe. The faster-than-light neutrino has such "crazy" implications that are ground-breaking to many fields in physics that I am more inclined by to be skeptical at first (as a good scientist should be). | ||
Abraxas514
Canada475 Posts
On February 25 2012 05:34 oGoZenob wrote: #2 it's because you forget that the frame itself of the universe is expanding, matter is not only moving. In fact this expansion is not constant, it's accelerating. as for #3, i don't understand your trouble For #2, what you are saying is that the velocity of one point relative to another can be more than c? #3, the logic is "because something can't move faster than c, neutron stars can't mass more than 3.2 solar masses". You see, I don't have a problem accepting SR since it's obviously present through experimental proof. The problem is describing the phenomenon from the bottom up. 1) It comes from the lorentz transform, which comes from frame independence. I Can't teach you all of SR here on tjhe forum, but if you want to understand this better, i recommend that you actually go through the derivation of the lorentz transform from the mirror-on-a-train thought experiment. The mathematics are very easy, just plus and minus (and multiplication and division and squares. ) and it feels pretty convincing once you've done it yourself. For further understanding, you can think of the kinetic energy E_k = mv^2/2 as just the first term of a taylor expansion in v^2/c^2 (which again comes out from the lorentz transform). The mirror on a train thought experiment is backwards logic though. You are looking at a result and explaining it. The lorentz transform says "because something can't move faster than c, inertial mass approaches infinity as v->c". What I'm asking is a proof that kinetic energy (so half of total mass * velocity squared) approaches infinity as velocity approaches c. To do this you would need to prove that mass -> infinity since velocity doesn't. If you are defining inertial mass as total mass energy plus kinetic energy of the mass then you still don't get infinity. | ||
L3gendary
Canada1469 Posts
On February 25 2012 15:07 Abraxas514 wrote: For #2, what you are saying is that the velocity of one point relative to another can be more than c? #3, the logic is "because something can't move faster than c, neutron stars can't mass more than 3.2 solar masses". You see, I don't have a problem accepting SR since it's obviously present through experimental proof. The problem is describing the phenomenon from the bottom up. The mirror on a train thought experiment is backwards logic though. You are looking at a result and explaining it. The lorentz transform says "because something can't move faster than c, inertial mass approaches infinity as v->c". What I'm asking is a proof that kinetic energy (so half of total mass * velocity squared) approaches infinity as velocity approaches c. To do this you would need to prove that mass -> infinity since velocity doesn't. If you are defining inertial mass as total mass energy plus kinetic energy of the mass then you still don't get infinity. edit: nvm The mirror on the train is just a thought experiment you don't need it to prove the Lorentz transformations, which precedes Einstein's SR. | ||
Pazuzu
United States632 Posts
| ||
ticklishmusic
United States15977 Posts
On February 25 2012 15:49 Pazuzu wrote: This was shown to be merely a faulty connection (loose cable) in the system. they are not faster than light Can you post a link? The last I read, they said that they didn't know if the cable was loose during the test itself. Also, there was something about the possibility that they might have actually undermeasured the speed instead of overmeasuring it as well, though I don't recall the exact nature of the errors. | ||
Luxucs
United States14 Posts
| ||
attwell
United States220 Posts
| ||
LazinCajun
United States294 Posts
On February 25 2012 15:07 Abraxas514 wrote: For #2, what you are saying is that the velocity of one point relative to another can be more than c? #3, the logic is "because something can't move faster than c, neutron stars can't mass more than 3.2 solar masses". You see, I don't have a problem accepting SR since it's obviously present through experimental proof. The problem is describing the phenomenon from the bottom up. The mirror on a train thought experiment is backwards logic though. You are looking at a result and explaining it. The lorentz transform says "because something can't move faster than c, inertial mass approaches infinity as v->c". What I'm asking is a proof that kinetic energy (so half of total mass * velocity squared) approaches infinity as velocity approaches c. To do this you would need to prove that mass -> infinity since velocity doesn't. If you are defining inertial mass as total mass energy plus kinetic energy of the mass then you still don't get infinity. In relativity, velocities don't add quite the same way that they do "intuitively." Imagine I send a particle at 99% the speed of light to my right, and another at 99% the speed of light to my left. Under non-relativistic physics (read: speeds much less than the speed of light), somebody standing on the first particle would measure the second particle to have a speed of 2 * 99% the speed of light. If you do the frame transformations correctly in relativity, somebody standing on the first particle still measures the second to be moving away at less than the speed of light. Kinetic energy in relativity is *not* just 1/2 m*v^2. Kinetic energy is defined thusly in relativity: you can calculate the kinetic energy by taking the rest energy (E = mc^2), doing a frame transformation that corresponds to moving with some velocity v, then subtracting out the rest energy from your new result. You can write the result as the sum of an infinite series in (v/c), and it will look something like (1/2) * mv^2 * [1 + C1 (v/c)^2 + C2 (v/c)^4 + ...] where C1, C2, C3, etc. are just some constant numbers unimportant to this discussion. Some people like to conceptualize this as an object's mass changing and keeping the old formula 1/2 * m * v^2. I personally don't like that idea. So now, how does kinetic energy become infinite as velocity approaches c? Without deriving all of special relativity, momentum is given by the formula gamma * m * v, where gamma is 1 / sqrt[ 1 - (v/c)^2 ]. Gamma grows to positive infinity as v->c. Energy (not kinetic, but total) is given by E^2 = p^2 * c^2 + m^2 c^4, and therefore also goes to infinity in the same limit. On February 25 2012 16:05 attwell wrote: Since we are talking billionths of a second, and they failed to replicate the same conditions, I'm not sure why they didn't assume it was an error earlier. I await further experimentation. They DID assume it was an error -- that's why they went searching for one. Nobody in the physics community was betting that the result would hold. The collaboration merely reported what their results were, and were not claiming to overturn all of physics. | ||
nodnod
New Zealand172 Posts
On February 25 2012 16:05 attwell wrote: Since we are talking billionths of a second, and they failed to replicate the same conditions, I'm not sure why they didn't assume it was an error earlier. I await further experimentation. agree. One of the fundamentals of experimental science is the reproducibility of results. | ||
LazinCajun
United States294 Posts
On February 25 2012 14:01 killerdog wrote: Since Cascade seems to be answering questions, How do they differentiate the neutrinos in this experiment from the neutrinos just flying around everywhere? Someone (correctly, i believe) mentioned how there are a very large number of neutrinos passing through your own body every second, How do they know that the neutrinos they are detecting are the same ones which they "launched" from their starting point when there are however many millions of neutrinos hitting their detector from all directions all the time? Edited to add: Condor's post below mine is both clearer and more accurate. Go with that one. Understanding the background noise to your signal is the heart of experimental particle physics, whether you're looking for signals from a known source like this or looking for the higgs or whatnot at the LHC. I'm not an expert in neutrinos nor this particular experiment, but the neutrinos don't come in a uniform beam. They're the result of a proton beam hitting a fixed target. The proton beam is a series of bunches, so the neutrinos arrive in bunches with very well understood frequencies. That timing information is very helpful when trying to separate your beam from the background. I would imagine that they have detectors that are not in the beam axis as well (ie, detectors that only get hit with background neutrinos and not those from the experiment). Off-axis detectors would allow them to map the background levels of neutrinos as a function of time, with some error of course. If not, they at least surely calibrated what they expected the background noise to be by taking measurements with the beam turned off. The former seems more likely to me as external neutrino sources aren't super predictable ahead of time because of things like fluctuations in solar activity. | ||
Condor
Netherlands188 Posts
On February 25 2012 14:01 killerdog wrote: Since Cascade seems to be answering questions, How do they differentiate the neutrinos in this experiment from the neutrinos just flying around everywhere? Someone (correctly, i believe) mentioned how there are a very large number of neutrinos passing through your own body every second, How do they know that the neutrinos they are detecting are the same ones which they "launched" from their starting point when there are however many millions of neutrinos hitting their detector from all directions all the time? I'll assume you don't mind me answering instead of Cascade, here goes: - Neutrinos are very hard to detect. There are many millions of neutrinos going through the detector per second. But hardly any are detected/interact with the detector. Hardly any means about than 1 per day (I think, don't know the number by heart). - The neutrino beam that is produced at CERN is many times more intense than the neutrino background. - The neutrino beam that is produced in CERN is not a continuous beam, it is a pulsed beam. There are pairs of short pulses (they are called 1st and 2nd extractions), and then a relatively large time of nothing in between. Think of it as though at CERN they are filling a bucket slowly, and when the bucket is full, they dump it entirely and quickly, then refill slowly. The bucket is called the Super Proton Synchrotron (which is in turn filled from the Proton Synchrotron), and the dumping is done with a magnet, called the kicker magnet. - The detector is working continuously, but they only accept (take in the analysis) the hits in the detector that happen in the timeframe around the pulses. They record all hits in the detector however. So, to summarize this with some numbers, which I made up to make the computation simpler, but it should give an idea. Lets assume the beam intensity is 1000 times higher than the background neutrino noise. That means that any neutrino you see in the detector is 1000 times more likely to come from CERN than from the background. Given that they have observed about 16000 neutrinos (more by now), the large majority of those must come from CERN. The beam is not on continuously, so assume they only send a burst through once every hour, and the burst is around 1 second long. Now they don't want to miss the burst, so they record with the detector for 1.3 seconds before and after the burst, for a total of 3.6 seconds. Given that there are 3600 seconds in an hour, they only see the background neutrinos in 1/1000 of real time. The combination of these two leads them to be very sure that over 99.9% of the observed signal is coming from the neutrinos at CERN. And there are several checks to verify this. If any of you read more in detail about this, you might be surprised by some technical terminology that has meaning in non-technical english. For example you will hear talk about coincidence of events. Coincidence when talking about detection in a physics sense comes directly from the Latin meaning which translates as "together (co) happening (incidence)", not the normal english meaning. Unfortunately many words in physics have a technical meaning while they have another meaning in normal english. Think for example about what "energetic person" means in physics language, or in normal english. | ||
Condor
Netherlands188 Posts
On February 25 2012 16:33 LazinCajun wrote: I would imagine that they have detectors that are not in the beam axis as well (ie, detectors that only get hit with background neutrinos and not those from the experiment). Off-axis detectors would allow them to map the background levels of neutrinos as a function of time, with some error of course. If not, they at least surely calibrated what they expected the background noise to be by taking measurements with the beam turned off. The former seems more likely to me as external neutrino sources aren't super predictable ahead of time because of things like fluctuations in solar activity. Your assumptions are totally correct for almost all detectors, except neutrino detectors. Normally you can do things with shielding and off axis detection, but for neutrinos this simply does not work well, you cannot shield a detector from sideways incoming neutrinos (remember that neutrinos happily fly through 730km of rock to get to the detector). The only thing you can do (and is done) is make the detector longer in the beam direction, so you have more interaction volume in that direction. A second thing to be aware of is that there are 3 types of neutrinos, electron-neutrino, muon-neutrino and tau-neutrino, and you can make a detector more sensitive to one of these types. The beam produced at CERN is almost entirely muon-neutrino, and the detector is selecting muon neutrinos. | ||
Valashu
Netherlands561 Posts
On February 25 2012 16:30 nodnod wrote: agree. One of the fundamentals of experimental science is the reproducibility of results. They did reproduce the results many many times over before showing it to the public. http://www.popsci.com/science/article/2011-09/baffling-cern-results-show-neutrinos-moving-faster-speed-light 'This isn’t an isolated anomaly, but has been going on for years. The team has now measured some 15,000 batches of neutrinos coming across that distance, and they say they’ve reached a point where the statistical significance is such that, were they trying to prove anything else, it would count as as formal scientific discovery. But try as they might, they can’t explain what’s happening.' Cite your sources if you want to make the outrageous claim that CERN did not manage to reproduce their results like any proper scientist, much less some of our best, would aim to do. Please do some research before posting, this is how rumors get started. | ||
LazinCajun
United States294 Posts
On February 25 2012 17:06 Condor wrote: Your assumptions are totally correct for almost all detectors, except neutrino detectors. Normally you can do things with shielding and off axis detection, but for neutrinos this simply does not work well, you cannot shield a detector from sideways incoming neutrinos (remember that neutrinos happily fly through 730km of rock to get to the detector). The only thing you can do (and is done) is make the detector longer in the beam direction, so you have more interaction volume in that direction. A second thing to be aware of is that there are 3 types of neutrinos, electron-neutrino, muon-neutrino and tau-neutrino, and you can make a detector more sensitive to one of these types. The beam produced at CERN is almost entirely muon-neutrino, and the detector is selecting muon neutrinos. Yes, of course. Forgive me, it's past 3am here and I can't sleep so I'm not thinking clearly I wasn't referring to shielding at all, just detecting background levels, but that would be a pretty dumb thing to do considering that you're selecting for muon neutrinos in a decently narrow timing. | ||
nodnod
New Zealand172 Posts
On February 25 2012 17:25 Valashu wrote: They did reproduce the results many many times over before showing it to the public. http://www.popsci.com/science/article/2011-09/baffling-cern-results-show-neutrinos-moving-faster-speed-light 'This isn’t an isolated anomaly, but has been going on for years. The team has now measured some 15,000 batches of neutrinos coming across that distance, and they say they’ve reached a point where the statistical significance is such that, were they trying to prove anything else, it would count as as formal scientific discovery. But try as they might, they can’t explain what’s happening.' Cite your sources if you want to make the outrageous claim that CERN did not manage to reproduce their results like any proper scientist, much less some of our best, would aim to do. Please do some research before posting, this is how rumors get started. If you bothered to read some of the previous posts you'd have easily found a link to the issue on not-well-calibrated equipments reportedly used during the experiments (in fact there is a source on the previous page). Though I do see your point; however, I think the source you cited is irrelevant in this case because no matter how big of a batch they tested or how many times they had managed to reproduce the same result, the problem is they might have used not-well-calibrated equipments and therefore one cannot have confidence in the result, http://www.nzherald.co.nz/technology/news/article.cfm?c_id=5&objectid=10787863 http://www.thestatecolumn.com/articles/2012/02/24/cern-neutrino-experiment-faulty-einsteins-theory-may-still-hold-true/ The point is if someone else had the exact same setup but with well-calibrated equipments, they wouldn't be able to reproduce the same results that the CERN scientists reported. In fact, if the CERN scientists calibrated their equipments they might not be able to reproduce the same results themselves had reported. Though do note that their conclusion might still hold, just their reported values could be inaccurate. Again, one of the fundamentals of experimental science is the reproducibility of results. | ||
Cascade
Australia5405 Posts
On February 25 2012 16:51 Condor wrote: Soo many drive-by posters! Just read the news, don't read the thread, just dump the news in again and again. I'll assume you don't mind me answering instead of Cascade, here goes: - Neutrinos are very hard to detect. There are many millions of neutrinos going through the detector per second. But hardly any are detected/interact with the detector. Hardly any means about than 1 per day (I think, don't know the number by heart). - The neutrino beam that is produced at CERN is many times more intense than the neutrino background. - The neutrino beam that is produced in CERN is not a continuous beam, it is a pulsed beam. There are pairs of short pulses (they are called 1st and 2nd extractions), and then a relatively large time of nothing in between. Think of it as though at CERN they are filling a bucket slowly, and when the bucket is full, they dump it entirely and quickly, then refill slowly. The bucket is called the Super Proton Synchrotron (which is in turn filled from the Proton Synchrotron), and the dumping is done with a magnet, called the kicker magnet. - The detector is working continuously, but they only accept (take in the analysis) the hits in the detector that happen in the timeframe around the pulses. They record all hits in the detector however. So, to summarize this with some numbers, which I made up to make the computation simpler, but it should give an idea. Lets assume the beam intensity is 1000 times higher than the background neutrino noise. That means that any neutrino you see in the detector is 1000 times more likely to come from CERN than from the background. Given that they have observed about 16000 neutrinos (more by now), the large majority of those must come from CERN. The beam is not on continuously, so assume they only send a burst through once every hour, and the burst is around 1 second long. Now they don't want to miss the burst, so they record with the detector for 1.3 seconds before and after the burst, for a total of 3.6 seconds. Given that there are 3600 seconds in an hour, they only see the background neutrinos in 1/1000 of real time. The combination of these two leads them to be very sure that over 99.9% of the observed signal is coming from the neutrinos at CERN. And there are several checks to verify this. If any of you read more in detail about this, you might be surprised by some technical terminology that has meaning in non-technical english. For example you will hear talk about coincidence of events. Coincidence when talking about detection in a physics sense comes directly from the Latin meaning which translates as "together (co) happening (incidence)", not the normal english meaning. Unfortunately many words in physics have a technical meaning while they have another meaning in normal english. Think for example about what "energetic person" means in physics language, or in normal english. ok, nice. I thought they triggered on the higher energy of the accelerator neutrinos, but you are correct. It is a factor 10 000 more accelerator neutrinos than background neutrinos within the 40 microsecond window they allow. First paragraph in section 5 at page 11 in the paper. | ||
Cascade
Australia5405 Posts
On February 25 2012 15:07 Abraxas514 wrote: For #2, what you are saying is that the velocity of one point relative to another can be more than c? #3, the logic is "because something can't move faster than c, neutron stars can't mass more than 3.2 solar masses". You see, I don't have a problem accepting SR since it's obviously present through experimental proof. The problem is describing the phenomenon from the bottom up. The mirror on a train thought experiment is backwards logic though. You are looking at a result and explaining it. The lorentz transform says "because something can't move faster than c, inertial mass approaches infinity as v->c". What I'm asking is a proof that kinetic energy (so half of total mass * velocity squared) approaches infinity as velocity approaches c. To do this you would need to prove that mass -> infinity since velocity doesn't. If you are defining inertial mass as total mass energy plus kinetic energy of the mass then you still don't get infinity. OK, so let's see if I understand you correctly. You want me to start from intuitive classical mechanics laws and principles, and from that prove the Lorentz transform and GR? That is not possible. As you hinted yourself, GR is based on postulates. 1) Physics should look the same in every inertial frame (ie, physics is the same on a train). 2) The speed of light in vacuum is always perceived as the same, no matter what speed I move compared to the source of the light. The first one was introduced through "common sense" and I think you and most other people can accept it. The second is highly counter-intuitive postulating that light behaves completely different from any particle or wave previously seen. It was however an experimental fact at the time (I think they saw it from the light of the moon of Jupiter or Saturn. The light from the moon travelled at the same speed no matter if the moon and earth were travelling into or away from the beam. I'm not sure that this was the reason though.). So Einstein just went "Ok. well, assume that light actually behaves like this, then what?" And we got SR. If you want more direct OBSERVATIONS that things can't be accelerated beyond speed of light you should look at particle accelerators. They have been given kinetic energy to travel at thousands times the speed of light with the normal E_k = mv^2/2, but they still travel just below light speed. But no, speed of light as max speed cannot be proven mathematically from classical mechanics. Does that answer your concerns? | ||
Valashu
Netherlands561 Posts
On February 25 2012 18:08 nodnod wrote: If you bothered to read some of the previous posts you'd have easily found a link to the issue on not-well-calibrated equipments reportedly used during the experiments (in fact there is a source on the previous page). Though I do see your point; however, I think the source you cited is irrelevant in this case because no matter how big of a batch they tested or how many times they had managed to reproduce the same result, the problem is they might have used not-well-calibrated equipments and therefore one cannot have confidence in the result, http://www.nzherald.co.nz/technology/news/article.cfm?c_id=5&objectid=10787863 http://www.thestatecolumn.com/articles/2012/02/24/cern-neutrino-experiment-faulty-einsteins-theory-may-still-hold-true/ The point is if someone else had the exact same setup but with well-calibrated equipments, they wouldn't be able to reproduce the same results that the CERN scientists reported. In fact, if the CERN scientists calibrated their equipments they might not be able to reproduce the same results themselves had reported. Though do note that their conclusion might still hold, just their reported values could be inaccurate. Again, one of the fundamentals of experimental science is the reproducibility of results. I see, it's Atwells post that could easily be read as CERN doing a single test run and basing everything on that instead of CERN doing thousands of runs with the same results that made me furiously type on my keyboard. | ||
Cascade
Australia5405 Posts
On February 25 2012 21:53 Valashu wrote: I see, it's Atwells post that could easily be read as CERN doing a single test run and basing everything on that instead of CERN doing thousands of runs with the same results that made me furiously type on my keyboard. I'm not sure what you guys are discussing. At no point did the experiment, or the scientific community, think that this was something else than an experimental error. And they have been and are still working a lot to find the error, and other experiments are planning to repeat the measurement. What you read in popular media may give a different picture though. :/ | ||
| ||