<div dir="ltr"><div><br></div><div>OK, </div><div><br></div><div>I think 1/2 hour is not sufficient, so how about 2:00pm. (I'll skip my student office hours.)</div><div><br></div><div>JP?</div><div>Dustin? </div><div>
Oscar?</div><div><br></div><div>Ellie and Patricia have already said they are free this time, and anytime today is unfortunately difficult for Narbe.</div><div><br></div><div>thanks</div><div class="gmail_extra">
<br></div><div class="gmail_extra">-Karl<br clear="all"><div><br></div>
<br><br><div class="gmail_quote">On Wed, May 1, 2013 at 10:40 AM, Oscar Rondon-Aramayo <<a href="mailto:or@cms.mail.virginia.edu">or@cms.mail.virginia.edu</a>> wrote:<br><blockquote class="gmail_quote">
Hi Karl,<br>
<br>
We met with Dustin last evening and after going in detail over the formulas<br>
for the ratio Npol/Nun, we found that the unpolarized sigma_N, sigma_D and<br>
sigma_He (see my last email) can be collected in one group, which cancels<br>
with the denominator (all unpolarized), leaving a term<br>
sigma_D*Azz*Pzz/denominator, which I realized can be written as f*Azz*Pzz, f<br>
= dilution factor.<br>
<br>
With the dilution factor, the formulas in Dustin's third row of equalities<br>
in his Observables2 report, which are valid only for pure D (the HERMES<br>
case), can also be used for ND3 targets.<br>
<br>
In summary, we can take the ratio of the pol to unpol counts, which takes us<br>
to Azz, at the price of the dilution factor and its error, plus the need to<br>
use some form of F1 to get b1 from Azz, or the difference, which takes us<br>
directly to b1.<br>
<br>
In both cases the systematic errors, other than the charge and detector<br>
efficiency are normalizations, and since the error on Pzz is expected to<br>
dominate, it really is a matter of taste, once we have the numbers on hand.<br>
We'll surely try both.<br>
<br>
For the statistical errors, f enters in the Azz time estimate, but Q*A*l*pf<br>
enter in the difference (Pzz is in both). I need to do some numbers yet<br>
(everyone should try) to compare the two approaches.<br>
<br>
Finally, today we have the SANE meeting at 3:30, so I can join b1 from 1:00<br>
to 3:30.<br>
<br>
Cheers,<br>
<br>
Oscar<br>
<div class=""><div class="h5"><br>
On Wed, 1 May 2013 09:18:48 -0400<br>
Karl Slifer <<a href="mailto:karl.slifer@unh.edu">karl.slifer@unh.edu</a>> wrote:<br>
> Hi all,<br>
><br>
> The methodology is the central question and I think we have to resolve any<br>
> lingering doubts today. I highly encourage that everyone really read<br>
> Oscar's note (Eq 19 and 20) and his last email before we discuss today.<br>
><br>
> I would really not like to delay till tomorrow if possible since time is<br>
>so<br>
> tight. I hope we can get a majority to participate at 3pm. Please let me<br>
> know if you can't.<br>
><br>
> -Karl<br>
><br>
><br>
><br>
> ---<br>
> Karl J. Slifer<br>
> Assistant Professor<br>
> University of New Hampshire<br>
> Telephone : <a href="tel:603-722-0695">603-722-0695</a><br>
><br>
><br>
> On Tue, Apr 30, 2013 at 5:59 PM, O. A. Rondon <<a href="mailto:or@virginia.edu">or@virginia.edu</a>> wrote:<br>
><br>
>> Hi Dustin,<br>
>><br>
>> Dustin Keller wrote:<br>
>> > You can only benefit from the systematic reduction if you us Azz as<br>
>> > discussed yesterday. But at this point I am not partial.<br>
>> ><br>
>> > dustin<br>
>> ><br>
>><br>
>> In the experiment, we only have counts. What we need to show to the PAC<br>
>> is how we go from the counts Npol and Nu, to Azz or b1. A measured<br>
>> quantity needs to be on one side and physics on the other. Lets say we<br>
>> start with your ratio Npol/Nu - 1 = Pzz*Azz, which only requires Pzz >0.<br>
>><br>
>> We need to prove that the lhs reproduces the rhs. We have, in general,<br>
>> N = Q*e*A*l*sigma. But since N are counts from everything in the target,<br>
>> it is not a simple matter of canceling quantities that stay the same<br>
>> when the polarization changes:<br>
>><br>
>> Npol = Qpol*epol*Apol*lpol*sigma_pol<br>
>> = Qpol*epol*Apol*lpol*[(sigma_N+3*sigma_Dpol)*pf + sigma_He*(1-pf)]<br>
>><br>
>> Nu = Qu*eu*Au*lu*[(sigma_N+3*sigma_D)*pf + sigma_He*(1-pf)]<br>
>><br>
>> sigma_N and sigma_He are the same, always unpol. And<br>
>> sigma_Dpol = sigma_D(1+Pzz*Azz).<br>
>><br>
>> Then, since Apol = Au = A, and lpol = lu = l,<br>
>><br>
>> Npol/Nu =<br>
>> (Qpol/Qu)*(epol/eu)*[(sigma_N+3sigma_D(1+ Azz*Pzz))*pf+..)]/[(sigma_N+..<br>
>><br>
>> where I just put ..., because I don't see how it can be simplified to<br>
>> just leave Azz*Pzz + 1, to equal the rhs.<br>
>><br>
>> On the other hand, if instead of taking the ratio Npol/Nu first, we take<br>
>> the difference first, it's indeed possible to isolate the required<br>
>> Pzz*b1 on on side, like I do in my draft, eq. (19) or (20). And in<br>
>> fact, we don't even need to bother with Azz, because we get b1 without<br>
>> having to multiply Azz by F1, introducing one more systematic error.<br>
>><br>
>> So, in summary, once one substitutes all the ingredients for your sigmas<br>
>> we get, or ought to get, eq.(19) or (20) back.<br>
>><br>
>> In both of those equations, the systematics for Pzz, A, and l(pf) are<br>
>> normalization factors, just like we want them to be, for control of<br>
>> systematics, but the charge and the detector efficiency are not common<br>
>> factors, they depend on the period when the data are taken, either pol.<br>
>> or unpol.<br>
>><br>
>> My point is that for the proposal, we must spell this all out, to give<br>
>> explicit sources of errors, and to calculate times or statistical errors<br>
>> correctly. For example, the statistical error must be sqrt(Npol + N_U) ~<br>
>> sqrt(2N), because it is just the error of a difference, etc.<br>
>><br>
>> We need to have a consensus on how the method is going to be described<br>
>> in the proposal, which needs to be done in the most precise way to avoid<br>
>> any confusion due to ambiguities.<br>
>><br>
>> Cheers,<br>
>><br>
>> Oscar<br>
>><br>
>><br>
>> > On Tue, 30 Apr 2013, O. A. Rondon wrote:<br>
>> ><br>
>> >> Hi,<br>
>> >><br>
>> >> Since I couldn't stay until the end of the meeting, and I don't think<br>
>> >> there will be minutes of it, I would like to share some ideas for the<br>
>> >> proposal draft.<br>
>> >><br>
>> >> Basically, what we need is an equation with the measured quantity on<br>
>>one<br>
>> >> side and b1 or Azz on the other. Based on what I think the consensus<br>
>> >> was, to measure polarized minus unpolarized counts on a single cup with<br>
>> >> the target field aligned along the beam, I've updated the draft of my<br>
>> >> method, see subsection 0.2, which discusses this. Eq. (19) or eq. (20)<br>
>> >> meet the conditions stated above. This is the approach I would<br>
>>subscribe<br>
>> >> to, unless there is another version that is shown to also represent the<br>
>> >> procedure, which should be circulated as soon as possible. The draft<br>
>> >> is here<br>
>> >> <a href="http://twist.phys.virginia.edu/~or/b1/b1_method-v2.pdf">http://twist.phys.virginia.edu/~or/b1/b1_method-v2.pdf</a><br>
>> >><br>
>> >> Cheers,<br>
>> >><br>
>> >> Oscar<br>
>> >><br>
>> >> _______________________________________________<br>
>> >> b1_ana mailing list<br>
>> >> <a href="mailto:b1_ana@jlab.org">b1_ana@jlab.org</a><br>
>> >> <a href="https://mailman.jlab.org/mailman/listinfo/b1_ana">https://mailman.jlab.org/mailman/listinfo/b1_ana</a><br>
>> >><br>
>> ><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>><br>
>> _______________________________________________<br>
>> b1_ana mailing list<br>
>> <a href="mailto:b1_ana@jlab.org">b1_ana@jlab.org</a><br>
>> <a href="https://mailman.jlab.org/mailman/listinfo/b1_ana">https://mailman.jlab.org/mailman/listinfo/b1_ana</a><br>
>><br>
<br>
_______________________________________________<br>
b1_ana mailing list<br>
<a href="mailto:b1_ana@jlab.org">b1_ana@jlab.org</a><br>
<a href="https://mailman.jlab.org/mailman/listinfo/b1_ana">https://mailman.jlab.org/mailman/listinfo/b1_ana</a><br>
</div></div></blockquote></div><br></div></div>