Is Bells theorem correct?

Foundations of physics and/or philosophy of physics, and in particular, posts on unresolved or controversial issues

Re: Is Bells theorem correct?

Postby Heinera » Tue Jan 30, 2018 8:03 am

Esail wrote:
Heinera wrote:If you by changing the polarization of P1 can can cause a statistically different ensemble of photons to be detected at station 2, this violates the fair sampling assumption.

I have to correct myself. Changing the angle of P1 does not change the ensemble detected by P2. Only the fraction of photon 2 whose peer was detected at P1 changes.

Yes, and that is exactly what the detection loophole is all about.

In the newest experiments, they basically managed to detect every pair of photons emitted, so this kind of mechanism is ruled out experimentally.

There is also a substantial literature on the detection loophole that you should familiarize yourself with.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby minkwe » Tue Jan 30, 2018 8:36 pm

Heinera wrote:In the newest experiments, they basically managed to detect every pair of photons emitted, so this kind of mechanism is ruled out experimentally.

I assume you read the paper. You can detect all the particles but what matters is what you actually use in the final calculation. Their final sample size used in the calculation was 245 photon pairs! How many photon pairs did they measure in the experiment? Let's see:

Every few hundred milliseconds, the recorded events are transferred to the PC. During the experiment, about 2 megabytes of data is generated every second. To keep the size of the generated data-set manageable, blocks of about 100000 events are saved to the hard drive only if an entanglement heralding event (E) is present in that block.

What happened to the rest of the events?
minkwe
 
Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Is Bells theorem correct?

Postby Joy Christian » Tue Jan 30, 2018 10:35 pm

minkwe wrote:
Heinera wrote:In the newest experiments, they basically managed to detect every pair of photons emitted, so this kind of mechanism is ruled out experimentally.

I assume you read the paper. You can detect all the particles but what matters is what you actually use in the final calculation. Their final sample size used in the calculation was 245 photon pairs! How many photon pairs did they measure in the experiment? Let's see:

Every few hundred milliseconds, the recorded events are transferred to the PC. During the experiment, about 2 megabytes of data is generated every second. To keep the size of the generated data-set manageable, blocks of about 100000 events are saved to the hard drive only if an entanglement heralding event (E) is present in that block.

What happened to the rest of the events?

In fact, the issue of a low number of events in the actual experiments is much more serious for Bell's "theorem" in general (i.e., for the mathematical derivation of the Bell-CHSH inequalities). In the so-called "loophole-free" experiments too the number of particle pairs detected is no more than 256. But, strictly speaking, the upper bound of 2 on the Bell-CHSH inequalities holds only in the large N (or infinite) limit, which is clearly impossible to achieve in any realizable experiment. Therefore all claims of experimental "violations" of the Bell-CHSH inequalities are grossly misleading.

This problem is of course only for those who believe that the upper bound of 2 has anything to do with any kind of physics. But in fact it is derived by averaging over physically impossible events, as I have tried to explain in the footnote 3 on page 26 of this paper: http://philsci-archive.pitt.edu/14305/.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 12:03 am

Heinera wrote:Yes, and that is exactly what the detection loophole is all about.

I do not unterstand what a loophole has to do with a model which should predict the experimental results?
A model should cover 100% of the experiments. My model does match precisely the predictions of QM.
If, after closing all loopholes, the experimental results would never infringe Bells inequality QM were wrong as well as my model.
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 12:25 am

Esail wrote:
Heinera wrote:Yes, and that is exactly what the detection loophole is all about.

I do not unterstand what a loophole has to do with a model which should predict the experimental results?
A model should cover 100% of the experiments. My model does match precisely the predictions of QM.
If, after closing all loopholes, the experimental results would never infringe Bells inequality QM were wrong as well as my model.

If I understand you correctly, your model predicts that a certain fraction of the particles must go undetected on one or both sides. QM does not not make such a prediction. Furthermore, this is something that can be experimentally tested.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 12:53 am

Heinera wrote:If I understand you correctly, your model predicts that a certain fraction of the particles must go undetected on one or both sides. QM does not not make such a prediction. Furthermore, this is something that can be experimentally tested.


I don't know if we are talking about the same setup. My setup is described in my paper. There are adjustable polarizers on both sides and a coincidence measuring device. P1 may be set to alpha and P2 may be set to beta. 50% of all photon 1 are detected at P1 and 50% of all photon 2 are detected at P2. A matching event is if photon 1 of a pair was detected at P1 and photon 2 was detected at P2.

You ask, what happens if P1 angle was changed shortly before detection of photon 1.
Suppose for instance P1 is set to alpha and P2 is set to alpha+pi/2. Then we would expect matching events for all photon 2.
Now, P1 is changed to alpha + pi/2. Then you would expect no matching events for photon 2.
P2 still detects 50% of all photon 2 but the fraction of matching photon 2 has dropped to zero.
There is no detection loophole involved.
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 1:10 am

Ok, so photon 1 and photon 2 are both detected, but not matched by the coincidence counter. Is that because they are not detected in the same time frame?
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 1:41 am

Heinera wrote:Ok, so photon 1 and photon 2 are both detected, but not matched by the coincidence counter. Is that because they are not detected in the same time frame?


It is not the time frame what matters. The photons are determined by the common parameter lambda which polarizer exit they will take.
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 1:53 am

So they are matched or not according to which polarizer exit they take?
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Joy Christian » Wed Jan 31, 2018 2:08 am

***
No discussion of loopholes is necessary because the model being discussed here is manifestly non-local.

Bell correctly stipulated that the only way to construct a local model of the singlet state is by constructing a measurement function A(a, h) = +/-1 that depends only on a and h, and not on b or B, and likewise constructing a measurement function B(b, h) = +/-1 that depends only on b and h, and not on a or A, where h is a shared randomness between Alice and Bob and a and b are their freely chosen experimental settings. Any model that does not satisfy these conditions is non-local.

The above formulation of locality by Bell is simply a mathematical recasting of Einstein’s notion locality. Any model that does not satisfy it is therefore not local in the sense espoused by Einstein.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 2:10 am

Heinera wrote:So they are matched or not according to which polarizer exit they take?

correct
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 2:13 am

Esail wrote:
Heinera wrote:So they are matched or not according to which polarizer exit they take?

correct

Well, that is not the experiment Bell had in mind, nor the way particle pairs are mathed in actual Bell-type experiments. So why should Bell's theorem apply to your experiment?
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 3:37 am

Heinera wrote:Well, that is not the experiment Bell had in mind, nor the way particle pairs are mathed in actual Bell-type experiments. So why should Bell's theorem apply to your experiment?

What is the difference? I've got the setup from ref. #6 (Weihs et al.)
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 4:04 am

Esail wrote:
Heinera wrote:Well, that is not the experiment Bell had in mind, nor the way particle pairs are mathed in actual Bell-type experiments. So why should Bell's theorem apply to your experiment?

What is the difference? I've got the setup from ref. #6 (Weihs et al.)

Weihs et. al matched particle pairs based on the time they were detected (quote from their paper: "Coincidences were identified by calculating time differences
between Alice’s and Bob’s time tags and comparing these with a time window (typically a few ns).)"
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 4:18 am

Heinera wrote:Weihs et. al matched particle pairs based on the time they were detected (quote from their paper: "Coincidences were identified by calculating time differences
between Alice’s and Bob’s time tags and comparing these with a time window (typically a few ns).)"


I still don't know what the difference is between what Weihs measured and my setup and what Bell had in mid. Except that Bell dealt with electron spin. Can you explain?
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 5:01 am

Esail wrote:
Heinera wrote:Weihs et. al matched particle pairs based on the time they were detected (quote from their paper: "Coincidences were identified by calculating time differences
between Alice’s and Bob’s time tags and comparing these with a time window (typically a few ns).)"


I still don't know what the difference is between what Weihs measured and my setup and what Bell had in mid. Except that Bell dealt with electron spin. Can you explain?

In the ideal Bell experiment, matching is not an issue. The source produces entangled pairs every time. All particles are detected. By having a low intensity source that produces a pair on average, say, once a minute, matching the detected particles becomes trivial and there is only one way to do it. Each particle produce a binary detection result (spin up or spin down). In this ideal experiment, there is no way detector settings (or anything else) can have any influence on how particles are matched.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 5:30 am

Heinera wrote:In the ideal Bell experiment, matching is not an issue. The source produces entangled pairs every time. All particles are detected. By having a low intensity source that produces a pair on average, say, once a minute, matching the detected particles becomes trivial .


So Bell has matching as well even if it is trivial. This is exactly my setup
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 5:40 am

Esail wrote:
Heinera wrote:In the ideal Bell experiment, matching is not an issue. The source produces entangled pairs every time. All particles are detected. By having a low intensity source that produces a pair on average, say, once a minute, matching the detected particles becomes trivial .


So Bell has matching as well even if it is trivial. This is exactly my setup

But Bell's matching does not in any way depend on the properties of the particles, only on the information that they were created at the same time. That is not exactly your setup. And it makes all the difference.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Is Bells theorem correct?

Postby Esail » Wed Jan 31, 2018 5:56 am

Heinera wrote:But Bell's matching does not in any way depend on the properties of the particles, only on the information that they were created at the same time. That is not exactly your setup. And it makes all the difference.

Matching on the information that the photons were created at the same time is my setup in deed.
Esail
 
Posts: 132
Joined: Sun Jan 28, 2018 10:44 am

Re: Is Bells theorem correct?

Postby Heinera » Wed Jan 31, 2018 6:09 am

Esail wrote:
Heinera wrote:But Bell's matching does not in any way depend on the properties of the particles, only on the information that they were created at the same time. That is not exactly your setup. And it makes all the difference.

Matching on the information that the photons were created at the same time is my setup in deed.

Plus the information of what polariser exit they took, according to yourself.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

PreviousNext

Return to Sci.Physics.Foundations

Who is online

Users browsing this forum: No registered users and 83 guests

cron
CodeCogs - An Open Source Scientific Library