From: W.J. Llope (llope_at_physics.rice.edu)
Date: Tue Feb 25 2003 - 14:01:21 EST
hi again zhangu,
one other comment i forgot to include in the prev mail...
there are two camps here... according to Dr. Seuss,
people with stars on thars say the pvpd is firing on fluctuations
so triggering star on the pvpd won't affect the min-bias-ness
of the star events wrt the present "dAuminbias" sets. people w/ no stars
on thars say the pvpd is probably sensitive to the impact parameter.
if this is true, the tofr trigger would push star towards more central events.
not a big deal necessarily... the only comment is just in the meantime
(i.e. during your presentation tomorrow) perhaps it could be noted
that there is this possibility. reality is probably in between,
but assuming the pvpd nhits and sum(adc) is strongly correlated with "b"
(as they are in auau) then under that assumption the most extreme
people in the second camp would say requiring the pvpd (at the present
1.and.1 level) at L0 would result in a nice selection of roughly the top
~28% (mid-central->central) of the present dAuminbias triggers...
this 28% number is just our local trigger efficiency wrt the present
dAuminbias triggered events... we'll see...
we should make a gentlemen's bet! anyway, we will know the answer
to this question in only a few days, when production events are available
and we can compare the pvpd nhits and sum(adc) to things like
ctbsum, zdcsum, ntracks, etc etc...
this is presumably before we would have a chance to try this
trigger anyway, but this is just a thought in favor of a small comment
during the presentation that this is something we're still
considering/discussing/working on inside the group...
thanks, see you later...
bill
On Tue, 25 Feb 2003, W.J. Llope wrote:
>
> hi zhangbu et al...
>
> i have some comments on the trigger proposal - i just wanted to
> get them down - let's discuss tonight, thanks, bill
>
> > TOFr Trigger Proposal in d+Au run
>
> very small point - the trigger really should be called TOF, or
> "pVPD+TOFr"...
> one can't really use TOFr hits unless the pVPD fired in the
> same events, so the pVPD has to be involved in the decision
> too....
>
> >
> >Motivation:
> >
> >i) Measure proton/pbar spectra upto pt=2.->3.GeV/c
> > dE/dx from TPC only covers proton dN/dy of ~50\% in narrow
> > rapidity window. TOF is the only detector in STAT which
> > can extends this to >99\%
> >ii) Resolve the proton/pion ratio ``puzzle'' seen in AuAu data,
> > distinguish flow effect from other mechanism and
> > Address flow effects from partonic versus hadronic stages.
>
> i worry about these being the first motivation listed.... as
> worded, the implication is that these goals are not possile
> without this tofr trigger. this is not true really.
>
> TOFp already has tens of k hits per channel, and is roughly
> three times the total area... i.e. the TOFp results are
> also in play here, and given the present state of the
> TOFp vs TOFr offlien software, the TOFp results are
> likely to be "out" much sooner than from TOFr.
>
> so, perhaps these bullets could have a slightly different spin.
> i.e.
> a TOFr trigger would allow the formation of full 1/beta
> vs momentum spectra from ~70 TOFr cells. This will provide
> a modest increase in acceptance w/ TOF PID, and the ability to
> make important cross-checks of the charged hadron spectra obtained
> by TOFp and TOFr. this will lead to more precise & detailed
> analyses towards charged hardon spectra, and ratios of
> these (->"pion/proton puzzle").
> also taking enough TOFr data to form full 1/beta vs p
> spectra for each cell will allow for more detailed investigations
> of the performance of TOFr - which is the primary focus of
> this TOFr "engineering" run. without a TOFr trigger, it will
> be possible to extract the time resolution (for pions) and
> the efficiency, satisfying the basic goal of the run. however,
> with an event sample enriched by triggering on TOFr, it
> will be possible to collect enough kaons and protons to
> extend our understanding of the basic detector performance
> even further. this will be a large boost to the full barrel
> TOF project, especially in fighting concerns that this
> technology is finicky, unstable, whatever....
>
>
> >iii) Good pVPD (4 pVPD with adc>300, dt<50ps) is absolutely
> > necessary to be able to measure the TOFr timing resolution
> > expected at 60-80 ps.
>
> i don't see how you can employ an ADC threshold on the pVPD
> ADCs with the present cabling...
>
> >Trigger Details:
> >
> >i) require at least 4 pVPDs fire in the event with high pulse
> > and at least one TOFr channel has signal. The enhancement factor
> > is about 122 compared with minbias.
>
> besides my concern about the ability to requirelarge ADC signals
> i am also worried that requiring so many pVPD hits is not
> necessary...
>
> from a technical perspective, it would require an access
> to change (increase) the local trigger requirement, then star
> runs the tofr special trigger, then another access is
> required to change (decrease back to 1.and.1) the local trigger
> condition again. since i think we want to run this
> special trigger at least two different times (see below),
> all these accesses to keep changing the local trigger
> condition will be a problem.
> if we leave the pvpd local trigger as is during
> tofr trigger runs, none of these accesses are needed and
> presumably everything can be controlled from the control room...
>
> remember the 1.and.1 local trigger is the most wide open possible,
> but it happens to also be one for which there is already
> enough hits in just the minbias data as is
> to get a good total start resn.... so higher
> requirements on the No of pvpd hits seems unneccessarily
> strict. we can use the events such a trigger would pitch!
>
> also, i'd like to use the special trigger data to improve
> the analyses of all the other "normal" data we have so far.
> in that case, i'd like to have so-called "1.and.1" and
> "1.and.2" events in the tofr triggered events so
> that i (we) can directly compared the performance of the
> detectors in the different data sets...
>
> so one thing we should discuss tonight is if
> raising the pvpd local trigger condition during tofr triggered
> runs is really necessary...
>
> BTW i made a plot this mornign... the number of tofr hits per
> event is uncorrelated with the number of pvpd hits. so
> requiring a lot of pvpd hits does not increase the
> probability that there is >=1 tofr hit in the event...
>
>
> >ii) Logically OR the TOFr signals from the discriminators which has
> > 2-bridge output (one to TDC).
>
> agreed - this seems like the best option....
>
> >Another option is to split the
> > analog signals to ADC.
>
> this would better be done in the manner i described a few days
> ago to startrg... we could gain 30-40 ns in the arrival time
> of the pvpd signals at the trigger racks by building and runnign
> new cables. but this would take some work and i kind of doubt
> we could pull it off during tomorrows access. let's
> discuss tonight
>
> >iii)The pVPD.AND.TOFr signal will be sent to the ZDC CDB and use the
> > spare channels in the ZDC Digitizer Board with the CTBMult register
> > in DSM.
> > The timing of the signal is about 300ns which is perfectly in time
> > with ZDC.
> >
>
> [snip]
> >4) We should switch to the default TOFr gas (Freon and Isobutane) ASAP.
>
> could i propose something else. how about we work into the
> request that we want to run this special trigger on two separate
> occassions.
> first (soon) we run it for the present freon-only gas. we have
> a lot of that data and the tofr triggered data would really help us
> understand that.
> then, in about a week hopefully we are finally cleared to turn
> on the isobutane, and then some days later we run the tofr
> special trigger *again*...
>
> one could also mentioned we are still hoping, time permitting, to try
> implementing SF6 during this run. we have anecdotal eveidence
> that SF6 may not be necessary for the full system, but now
> is the time really prove this.,.. we would then want to run the special
> trigger a third time - since by then i doubt there's enough
> beam time left to collect enough events to calibrate TOFr
> without such a special trigger.......
>
>
--
_________________________________________________________
W.J. Llope, Ph.D. Res. Assoc. Professor
http://wjllope.rice.edu/default.html
llope_at_physics.rice.edu
T.W. Bonner Nuclear Lab.
Rice University, MS-315
6100 S. Main phone: 713-348-4741
Houston, TX 77005-1892 fax: 713-348-5215
This archive was generated by hypermail 2.1.4 : Thu Jul 24 2003 - 00:39:36 EDT