[Halld-pid] TDCs for TOF

[Eugene Chudakov]

Hi Sasha,

Thanks for the info.

On Thu, 28 Jun 2012, Alexander Ostrovidov wrote:
> P.S. If necessary, I can generate some simple Monte Carlo
> (like piKK or something like that) to give an estimate
> of what this improvement means in terms of the number of
> events, with definite PID, etc.
>

Yes, please provide us with such an estimate.

Since we had a saving in other electronics area we have a good
chance to proceed with the CAEN TDC, but need some paper 
justification.

Regards,
Eugene


On Thu, 28 Jun 2012, Alexander Ostrovidov wrote:

> Hello,
>
> In view of the CAEN 1290A vs F1 TDC choice, I looked at
> the pion-Kaon momentum separation with these 2 TDCs:
>
> 1. The best timing resolution we achieved with a single
> TOF paddle in the cosmic ray tests in our lab was
> 90 ps with CAEN 1290A and 103 ps with F1 TDC.
>
> 2. The minimal path length L for any particle will be 550cm
> (the distance from the center of the target to TOF). Actually,
> this is the worst case scenario. With the longer path lengths,
> the improvement in momentum separation becomes even slightly
> better than what is shown below for a minimal path length.
>
> 3. I consider only TOF timing resolution and disregard the
> contribution to the resolution from other factors like momentum
> measurement, path length, RF bunch start time, etc.  From the
> Fig.8.1 in the old IU report, it is obvious that even the smaller
> TOF resolution of 80ps already dominates everything else in
> the region of interest (where the 3-sigma envelopes for pion
> and Kaon  intersect).
>
> 4. Difference in the time of flight between pions and Kaons
> as a function of their momentum is:
>
> Delta(t) = [L/(p*c)] * [sqrt(p^2+Mkaon^2)-sqrt(p^2+Mpion^2)]
>
> This dependence is shown on the attached plot.
>
> 5. Assuming 3-sigma envelope for both pi and K (i.e.,
> their 6-sigma separation), the pion-Kaon envelopes will
> overlap at 1.806 GeV with 6*103ps=618ps time separation,
> and at 1.934 GeV with 6*90ps=540ps time separation.
> In other words, we gain about 130 MeV of extra separation
> in momentum by using 1290A instead of F1 TDC, and come
> closer to the future RICH threshold.
>
> Sasha
>
> P.S. If necessary, I can generate some simple Monte Carlo
> (like piKK or something like that) to give an estimate
> of what this improvement means in terms of the number of
> events, with definite PID, etc.
>
>