[d2n-analysis-talk] live time calculation

[Brad Sawatzky]

On Thu, 22 Apr 2010, posik at jlab.org wrote:

>   I am doing a live time calculation for the T5 trigger. I am doing this
> by using the following formula
> 
>   T5_livetime = (t5cnt)/(t5_scnt)
> 
>  where t5cnt is the number of events with a T5 trigger after prescale (ie
> evtypebits&1<<5==1<<5)
> 
> and t5scnt is the scaler T5 count, which is the difference between the t5
> scaler count for the last and first event.

That looks correct.

> Now since we have a deadtime pulser plugged into the T2 and T3
> triggers I imagine this is also in the T5 trigger. Therefore I need to
> subtract it off as follows

Do not subtract off these though.  The deadtime doesn't care how the
triggers are generated.

> using this edtp subtraction I get a Live time of ~80% for run 1258.
> checking that end of run file it lists the dead time as
> 
> DEAD TIME: [ 0]:  15.31%      [++]:  18.98%      [--]:  18.92%      [+-]: 
> 18.93%      [-+]:  18.97%
>
> Which if Livetime = 1-Deadtime, then there is ~5% difference.

Something is wrong with the calculation for [0] field here.  Your
deadtimes should be extremely close for all helicity/target combos (and
for all trigger types too).

Bob put a thorough note together about deadtime.  Some of it applies to
us, some doesn't.  I'll link to it here FWIW:
  http://hallaweb.jlab.org/equipment/daq/dtime.html
When we get to extracting real cross sections we'll have to be more
careful, but don't bother for sake of this quick check.

-- Brad

-- 
Brad Sawatzky, PhD <brads at jlab.org>  -<>-  Jefferson Lab / Hall C / C111
Ph: 757-269-5947  -<>-  Fax: 757-269-5235  -<>- Pager: brads-page at jlab.org
The most exciting phrase to hear in science, the one that heralds new
  discoveries, is not "Eureka!" but "That's funny..."   -- Isaac Asimov