Re: analysis meeting

From: Michael Kohl (kohlm@mit.edu)
Date: Fri Feb 17 2006 - 19:54:41 EST


Hi Tavi,

read carefully what we wrote before .... then you should be less confused.

One more comment which should make life easier. Since you are using a
Cerenkov cut to identify the inclusive yield, you only deal with TOFs 0-11
(remember we took out the Cerenkov in front of 12-15) and thus only with
triggers 1,2,3 and 7. Trigger 6 doesn't matter if you require the
Cerenkov.
Hence, trigger bit 7 is really set for each of the above trigger bits
1,2,3. Hence the event becomes a trigger 7 if it fails the prescaling of
1,2, or 3, where it again may fail because of prescale-7. So, my
conclusion is still that only for trigger 7 the yield must be multiplied
with prescale-7, while NOT for triggers 1-3 whatever their prescale factor
was.

The complication with trigger 6 only comes into play when you want to
determine inclusive yields without requiring the Cerenkov.

Best regards,

    Michael

On Fri, 17 Feb 2006, Octavian F Filoti wrote:

>
> I am even more confuse!
> Luckily, I use H2 ABS Fall 2004 runs without December runs when trigger 2 was
> prescaled by 2. There are 182KC, which is enough!
> But I have started looking at trigger 3 also, which is prescaled by 10 and
> now I do not know if I have to multiply by 10 or not! This prescale thing is
> not clear to me! From .settings file, it looks to me I do not have to add
> trigger 6.
>
> On Fri, 17 Feb 2006, Karen Dow wrote:
>
>>
>> Tavi,
>>
>> Michael is correct. The way we programmed the XMLU, an event might set
>> multiple XMLU output bits. The TS prescales all input bits, and then was
>> programmed to assign the event to the least significant bit remaining. A
>> TOF-neutron event that fails on the bit 2 prescale may pass on the bit 7.
>> You have to look carefully at the XMLU programming in the .settings file
>> for the trigger to see what happens. I think you also have to include
>> trigger 6, because if the TOF that fired along with the neutron was
>> TOF12-15, bits 2 and 6 were set.
>>
>> I've just looked at the trigger .settings file, and it looks to me like for
>> a nominal (e,e'n) event there are 3 possibilities:
>>
>> 1) TOF0-11 with Cer -- event 2, but if it fails prescale, event 7 (where it
>> may also fail prescale)
>> 2) TOF12-15 -- event 2, but if it fails prescale, event 6 (where it may
>> also fail prescale)
>> 3) TOF0-11 with NO Cer -- event 2, but if it fails prescale, it's lost
>>
>> For events of 3) above, you'd have to multiply the # of those by the
>> event 2 prescale. But you don't do that for 1) or 2), because you've
>> already accounted for them when multiplying # of 7's (or 6's) by the event
>> 7 or 6 prescale.
>>
>> It would probably be a good idea to check the trigger .settings file used
>> for certain time periods, and make sure the details of 1), 2), 3) are the
>> same for all runs. At some point (before the 2004 production runs?) we
>> changed the definition of event 7 to include the Cerenkov, and that will be
>> obvious from the XMLU programming.
>>
>> Karen
>>
>>
>> Michael Kohl wrote:
>>
>>> Hi Tavi,
>>>
>>> I agree with your second formula to combine the yield for two runs with
>>> different prescale factor.
>>>
>>> On the prescaling of trigger 3 (=10) and of trigger 2 (=2) for part of the
>>> runs: Once I had a discussion with Karen on how to correctly account for
>>> the prescaling of triggers 2 and 3:
>>> The trigger supervisor logic is such that if an event which fulfilled the
>>> requirement for e.g. trigger 2 did not become a trigger 2 event just
>>> because it was prescaled, then it may again be offered to become a trigger
>>> type of reduced priority, i.e. a possible trigger 7 event. It is not that
>>> the prescaled trigger 2 event would be discarded after failing the first
>>> test. Each event can have multiple trigger bits (here 2 and 7), and the
>>> assignment of the bit to the event happens according to the priority
>>> order.
>>>
>>> Subsequently, this means in turn that you must not apply the prescale
>>> factor to the events of trigger 2 and 3 even if they were prescaled.
>>> Just accounting for the prescale factor of trigger 7 makes it correct.
>>> Raising the prescale factor for trigger 2 or 3 would therefore influence
>>> the rate in trigger 7 (you can check this with real data!)
>>>
>>> Note that this means a factor of 10 less yield in trigger 3!
>>> Karen, please intervene if I'm wrong.
>>>
>>> Best regards,
>>>
>>> Michael
>>>
>>>
>>>
>>> On Thu, 16 Feb 2006, Octavian F Filoti wrote:
>>>
>>>>
>>>> I had to reanalyze again! I got some empty runs bigger than the data
>>>> runs!
>>>> This is because I added wrongly the yields: since for empty runs I have
>>>> different scaling factors, I did this:
>>>>
>>>> Yield = Ni1*Si1/Q1 + Ni2*Si2/Q2
>>>>
>>>> but I guess correct is:
>>>>
>>>> Yield = (Ni1*Si1 + Ni2*Si2)/(Q1+Q2)
>>>>
>>>> Then I subtract the emty yields from the data:
>>>>
>>>> Yield0 = Yield_data - Yield_empty.
>>>>
>>>> From April 2004 to September 28th 2004: trigger 7 is scaled by 3, after
>>>> that by 9. When I looked at N++ I divided by Q++ of course!
>>>> Trigger 3 (scaled by 10) has delta events too, so I added that one also,
>>>> so I have trigger 1, 2, 3 and 7!
>>>> I hope I will get good results this weekend! Give me some feedback!
>>>>
>>>> Thanks
>>>>
>>>> On Thu, 16 Feb 2006, Michael Kohl wrote:
>>>>
>>>>> Hi Tavi,
>>>>>
>>>>> I'm flying from Tokyo to Boston on Tuesday, arriving in Boston very
>>>>> late. Japan is 14 hours ahead, so I will already have a hard time to
>>>>> make it to Bates on Wednesday before noon. I'm also giving a talk at the
>>>>> Bates lunch meeting at noon. Sorry that I can't accomodate your request
>>>>> at this time. If you can make it possible to attend the meeting at 13:30
>>>>> please do it. I think it would be very interesting to see your follow-up
>>>>> results of our recent discussion.
>>>>>
>>>>> Best regards
>>>>>
>>>>> Michael
>>>>>
>>>>>
>>>>>
>>>>> On Tue, 14 Feb 2006, Octavian F Filoti wrote:
>>>>>
>>>>>>
>>>>>> Next week: can you set the analysis meeting at 10:30am.
>>>>>> This is good for me!
>>>>>> Or at least once in a while!
>>>>>> Tomorrow I don't have anything to show, but next week I will, so I
>>>>>> prefer 10:30am if possible, PLEASE! :-)
>>>>>>
>>>>>> --
>>>>>>
>>>>>> --------------------------
>>>>>> | Octavian F Filoti |
>>>>>> | Nuclear Physics Group |
>>>>>> | Univ. of New Hampshire |
>>>>>> | 9 Library Way |
>>>>>> | Durham, NH 03824 |
>>>>>> | phone: (603)868-6882 |
>>>>>> | FAX: (603)862-2998 |
>>>>>> | email: ofiloti@unh.edu |
>>>>>> --------------------------
>>>>>>
>>>>>
>>>>>
>>>>> +-------------------------------------+--------------------------+
>>>>> | Office: | Home: |
>>>>> |-------------------------------------|--------------------------|
>>>>> | Dr. Michael Kohl | Michael Kohl |
>>>>> | Laboratory for Nuclear Science | 5 Ibbetson Street |
>>>>> | MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
>>>>> | Middleton, MA 01949 | U.S.A. |
>>>>> | U.S.A. | |
>>>>> | - - - - - - - - - - - - | - - - - - - - - -|
>>>>> | Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
>>>>> | Work: +1-617-253-9207 | Home: +1-617-629-3147 |
>>>>> | Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
>>>>> | http://blast.lns.mit.edu | |
>>>>> +-------------------------------------+--------------------------+
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>>
>>>> --------------------------
>>>> | Octavian F Filoti |
>>>> | Nuclear Physics Group |
>>>> | Univ. of New Hampshire |
>>>> | 9 Library Way |
>>>> | Durham, NH 03824 |
>>>> | phone: (603)868-6882 |
>>>> | FAX: (603)862-2998 |
>>>> | email: ofiloti@unh.edu |
>>>> --------------------------
>>>>
>>>
>>>
>>> +-------------------------------------+--------------------------+
>>> | Office: | Home: |
>>> |-------------------------------------|--------------------------|
>>> | Dr. Michael Kohl | Michael Kohl |
>>> | Laboratory for Nuclear Science | 5 Ibbetson Street |
>>> | MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
>>> | Middleton, MA 01949 | U.S.A. |
>>> | U.S.A. | |
>>> | - - - - - - - - - - - - | - - - - - - - - -|
>>> | Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
>>> | Work: +1-617-253-9207 | Home: +1-617-629-3147 |
>>> | Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
>>> | http://blast.lns.mit.edu | |
>>> +-------------------------------------+--------------------------+
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
> --
>
> --------------------------
> | Octavian F Filoti |
> | Nuclear Physics Group |
> | Univ. of New Hampshire |
> | 9 Library Way |
> | Durham, NH 03824 |
> | phone: (603)868-6882 |
> | FAX: (603)862-2998 |
> | email: ofiloti@unh.edu |
> --------------------------
>

+-------------------------------------+--------------------------+
| Office: | Home: |
|-------------------------------------|--------------------------|
| Dr. Michael Kohl | Michael Kohl |
| Laboratory for Nuclear Science | 5 Ibbetson Street |
| MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
| Middleton, MA 01949 | U.S.A. |
| U.S.A. | |
| - - - - - - - - - - - - | - - - - - - - - -|
| Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
| Work: +1-617-253-9207 | Home: +1-617-629-3147 |
| Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
| http://blast.lns.mit.edu | |
+-------------------------------------+--------------------------+



This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:33 EST