I was assuming that neutrons were counted for the number of primaries
(protons) run. If this is the case then in a Monte Carlo experiment, the
variance is proportional to 1/Sqrt(protons). Which means that in a
simulation with 100 protons the uncertainty is higher by a factor of 100
when compared to a simulation with 10^6 protons.
Cheers,
Nikhil
On Mon, Oct 24, 2016 at 4:40 PM, <emilio_at_impcas.ac.cn> wrote:
>
> Dear Nikhil,
>
> "you are estimating neutron yield which is n/p. Therefore the error on
> neutron yield should be Sqrt[n/p] instead of Sqrt[n]/p."
>
> No, this is not correct. Since p (=the number of protons on target) is
> fixed in each simulation, the only source of error is n, and d(n/p)=dn/p,
> where I assumed dn=Sqrt[n] (since they are statistical fluctuations). If
> also p could be a source of error, then
> d(n/p)=Sqrt[ (dn/p)^2 + (dp*n/p^2)^2 ]
> However since dp=0 we are back to the previous expression.
>
> By the way, if the statistical error was actually Sqrt[n/p], a simulation
> with 100 protons would have the same statistical error over n/p than one
> with 10^6 protons, which is clearly impossible.
>
> "Finally, it is perfectly normal for the neutron yield to fluctuate with
> changing dimensions if one considers the parasitic absorption, given the
> high neutron absorption cross section of tungsten."
>
> Yes, I agree on this, however in my results the neutron yield was going
> down increasing the dimension of the target (which is reasonable) then up
> again (which is a bit weirder, even if I would not say that it's completely
> impossible: in any case this effect disappeared increasing the statistic).
>
> See you
> Emilio
>
> -----Original Messages-----
> *From:* "nikhil shetty" <nikhil.nitk_at_gmail.com>
> *Sent Time:* Monday, October 24, 2016
> *To:* emilio_at_impcas.ac.cn
> *Cc:* fluka-discuss <fluka-discuss_at_fluka.org>
> *Subject:* Re: [fluka-discuss]: Statistical Fluctuations in neutron yield
>
>
> Hi Emilio,
>
> you are estimating neutron yield which is n/p. Therefore the error on
> neutron yield should be Sqrt[n/p] instead of Sqrt[n]/p.
>
> This also explains why you observe, from the statistical experiment you
> did, the difference by a factor of 3 ( ~ Sqrt(10^6/10^5) ).
>
> Finally, it is perfectly normal for the neutron yield to fluctuate with
> changing dimensions if one considers the parasitic absorption, given the
> high neutron absorption cross section of tungsten.
>
> Cheers,
> Nikhil
>
> On Sun, Oct 23, 2016 at 11:42 AM, <emilio_at_impcas.ac.cn> wrote:
>
>> Dear FLUKA experts,
>>
>> I am simulating the neutron production from a cylindrical W target using
>> a 400 MeV proton beam; using the loop function, I change the radius R and
>> the length L of the cylinder, scoring the total neutron produced.
>>
>> At first, I tried with 10^5 proton; however I found some weird results
>> (the neutron yield was decreasing and then increasing again changing L);
>> for this reason I increased the primary particles up to 10^6, and the
>> results were more consistent (you can find the imput cards in the attached
>> files).
>>
>> My problem is that the difference between the two bunch of simulations
>> seems too large to be explained with statistical fluctuation: in the file
>> "Neutron Yield 400 MeV.pdf" you can find the neutron yield n/p with 10^5
>> (solid curves) and 10^6 (dashed curves) protons; each color corresponds to
>> a different radius. I also included the error bar due to the statistical
>> fluctuation, calculated as Sqrt[n]/p, where n is the number of neutrons !
>> scored, p is the number of proton. There are several points at 6 and more
>> sigma's, which is not possible with 30 datapoints. In general it looks like
>> the statistical fluctuations are roughly 3 times larger than what expected
>> with Sqrt[n].
>>
>> Is it not correct to assume that the statistical error would be Sqrt[n]?
>> Am I using some biasing that I am not aware of?
>>
>> Thank you for your help
>> Emilio Ciuffoli
>>
>>
>>
>>
>>
>>
>
>
>
>
>
__________________________________________________________________________
You can manage unsubscription from this mailing list at
https://www.fluka.org/fluka.php?id=acc_info
Received on Tue Oct 25 2016 - 13:36:47 CEST