I was thinking about the strength of the dependence of the dump
current: Decided to improve the simplified model of single-
exponential beam current behavior.
Actually what we are interested in is maximizing the level-2
triggers/time. So, I used the scaler/epics events of 140 runs
(runs 10734-10915), and tried to create an accurate model of:
1) time dependence of beam current (closer to 3-decaying-exponents)
2) time dependence of level-2 trigger (with a fixed initial current)
3) required time between fills
4) required time between runs
5) injection current
Then, simulated 10 days of run time to extract the average taped
charge, by implementing the above information + 500000 event/run
limit (included coda's 9th run problem by -adding- a fixed, 5min,
amount of time between the runs) for different dump current
values.
The result can be seen in the attached figure.
Conclusions:
1) The simplified model yields pretty accurately the optimum dump
current found by the simulation, which is about 95mA for 140mA
injection current.
2) The dependence of dump current around the maxima is -weak-.
There is about 16mA window around the optimum value that
yields greater than 99% of maxima, which is pretty wide. The
efficiency drops rapidly at higher limit compared to lower
limit.
3) The simulation assumed no problem between runs other then 9th
run problems, which yielded 330kC/hr, 14% more than the
maximum we saw so far. However, it should be noted that about
half of the runs were started more than 5 minutes (simulation's
upper limit for "normal" starts) after the run ended, only a
fraction of this corresponds coda problems.
Taylan
On Tue, 7 Sep 2004, Karen Dow wrote:
>
> Doug is correct about the optimum dump current. In fact, there is
> an elog entry (28476) about it from 9/1. Don't people read the elog??? I
> admit, the title of the entry is about ramp rates, but the text clearly
> associates the increased ramp rate with a change in the optimum dump current.
>
> Please READ the logbook regularly, and MAKE ENTRIES with any
> changes and observations. It is there to communicate information across
> shifts, detail problems and their solutions, and make life easier down the
> road for those trying to do analysis.
>
> Karen
>
>
> At 01:10 AM 9/4/2004 -0400, Douglas Hasell wrote:
> >Last week, Michael, Karen, and I increased the ramp rates on the HV for
> >the WC and LADS from 100 V/s to 200 V/s. This had the affect of
> >shortening the deadtime when filling the ring. Consequently the current
> >at which we should dump the beam is higher than normal. I suspect that
> >for 140 mA injection that we should be dumping at 95 mA for optimal luminosity.
> >
> >To calculate the optimal current use the equation:
> >
> > I_dump = I_inject * exp( -t_live / tau )
> >
> > there I_dump and I_inject are obvious, tau is the average beam
> > lifetime and t_live is the live DAQ time for each fill as calculated by:
> >
> > t_live = SQRT( 2 * t_dead * tau )
> >
> >where t_dead is the deadtime during a fill from the time when the DAQ
> >stops until it starts again. Currently I think t_dead is around 1.5
> >minutes, tau is around 20 minutes thus t_live should be 7.75 minutes and
> >i_dump is 68% of i_inject.
> >
> >
> > Cheers,
> >
> > Douglas
> >
> >26-415 M.I.T.
> >Tel: +1 (617) 258-7199
> >77 Massachusetts Avenue Fax: +1 (617)
> >258-5440
> >Cambridge, MA 02139, USA E-mail: hasell@mit.edu
> >
>
>
-- ---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--- Taylan Akdogan Massachusetts Institute of Technology akdogan@mit.edu Department of Physics Phn:+1-617-258-0801 Laboratory for Nuclear Science Fax:+1-617-258-5440 Room 26-402b ---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:31 EST