<html>
<head>
</head>
<body text="#000000" bgcolor="#ffffff">
<br>
<br>
-------- Original Message --------
<table cellspacing="0" cellpadding="0" border="0" class="moz-email-headers-table">
<tbody>
<tr>
<th valign="BASELINE" nowrap="nowrap" align="RIGHT">Subject: </th>
<td>Re: [Tpe] TPE field reversing</td>
</tr>
<tr>
<th valign="BASELINE" nowrap="nowrap" align="RIGHT">Date: </th>
<td>Wed, 28 Jul 2010 12:16:59 -0400</td>
</tr>
<tr>
<th valign="BASELINE" nowrap="nowrap" align="RIGHT">From: </th>
<td>John R. Arrington <a href="mailto:johna@anl.gov"><johna@anl.gov></a></td>
</tr>
<tr>
<th valign="BASELINE" nowrap="nowrap" align="RIGHT">To: </th>
<td>Weinstein, Lawrence B. <a href="mailto:lweinste@odu.edu"><lweinste@odu.edu></a></td>
</tr>
<tr>
<th valign="BASELINE" nowrap="nowrap" align="RIGHT">CC: </th>
<td>tpe <a href="mailto:tpe@jlab.org"><tpe@jlab.org></a></td>
</tr>
</tbody>
</table>
<br>
<br>
<pre>
One issue to keep in mind is that we want to have close to equal
statistics for the different cases, so we may want to make sure that
we're flexible towards the end of the run.
In the analysis of the engineering run, we take the geometric mean of the
two configurations; this means that acceptance differences should cancel,
even if the two configurations have different statistics. But is also
means that you combine the fractional errors in quadruture, so if one
configuration has noticably less statistics, it will dominate the total
uncertainty.
I'm not sure if we want to do an expanded version for the real thing,
where we have four independent measurements, and combine them with equal
weighting to maximize cancellation of the systematics. In this case, we'd
have to be careful about equalizing all four configurations.
-John
p.s. I actually just did a quick test, and it's not as bad as I thought.
If you have 1% uncertainties for a perfect 50/50 split for the two CLAS
polarities, it only increases to 1.1% for 70/30 and 1.25% for 80/20.
So we can improve it a little, but if we can keep the split 60/40 or
better, it's a very small impact.
On Wed, 28 Jul 2010, Larry Weinstein wrote:
> Dear TPEers,
>
> We will need to decide how often to reverse the magnetic fields. There are a
> few constraints.
>
> 1. we should reverse torus fields weekly during the Tuesday
> maintenance period
> 2. we need to remeasure the two beam energy vs position after
> reversing the chicane. This will take about 4 hours.
>
> We should decide whether to switch the chicane when we switch the torus or
> not.
>
> There are generally two schemes: 1) switch both the torus and the chicane
> weekly, but out of phase or 2) switch the torus weekly and the chicane every
> other week.
>
> Scheme 1: switch torus polarities on Tuesdays (during maintenance periods)
> and switch chicane polarities on Fridays.
> Advantages: a) complete set of four field possibilities every two weeks, b)
> reverse Torus fields during down times, c) never change both fields together
> Disadvantage: must measure beam distributions every week with the calorimeter
> and dense fiber monitor
> day
> torus
> polarity
> chicane
> polarity
> 1 Tuesday
> +
> +
> 4 Friday
> +
> -
> 8 Tuesday
> -
> -
> 11 Friday
> -
> +
> 15 Tuesday
> +
> +
>
>
> Scheme 2: switch torus polarities on Tuesdays (during maintenance periods)
> and switch chicane polarities on alternate Tuesdays (or Fridays).
> Advantages: a) reverse Torus fields during down times, b) measure beam
> distributions every two weeks
> Disadvantages: a) complete set of four field possibilities every four weeks,
> b) change both fields on the same day (if done Tuesday)
> day
> torus
> polarity
> chicane
> polarity
> 1 Tuesday
> +
> +
> 8
> -
> +
> 15
> +
> -
> 22
> -
> -
> 29
> +
> +
> 36
> -
> +
>
>
> The frequency of beam distribution measurements will have an impact on
> calorimeter mounting.
>
> --
> Sincerely,
> Larry
>
> -----------------------------------------------------------
> Lawrence Weinstein
> University Professor
> Physics Department
> Old Dominion University
> Norfolk, VA 23529
> 757 683 5803
> 757 683 3038 (fax)
> <a href="mailto:weinstein@odu.edu">weinstein@odu.edu</a>
> <a href="http://www.lions.odu.edu/~lweinste/">http://www.lions.odu.edu/~lweinste/</a>
>
>
</pre>
</body>
</html>