Re: Light Concentrators

From: Joseph Formaggio (josephf@u.washington.edu)
Date: Sat Jun 11 2005 - 08:03:06 CDT


Hi Steve,

    Our experience with SNO, as you recall, is that the reflector petals
degrade over time. Although the degredation is small and most likely
associated with contact with water, it needs to be thoroughly tested.
Furthermore, since the degradation appears to be non-uniform (some
petals show aging, some do not, likely associated with oxugen exposure
and/or temperature), it may introduce a non-negligible relative error to
the near/far sites.

    It's something to seriously consider.

Take Care
Joe

Steve Biller wrote:

>
> Hi guys,
>
> Firstly, I've attached a page from Martin Moorhead's thesis,
> who helped design the SNO concentrators. This shows the
> measured concentration factor as a function of polar angle,
> which is the ratio of sensitivities with to without concentrators.
>
> Secondly, I think we may be talking about different things
> regarding reflections and the impact on electronics: My assumption
> had been that, for each PMT, we'd make 2 basic measurements:
> 1) A prompt TDC time measurement based on a threshold crossing
> (so, essentially the time of the earliest photoelectron) which
> would be calibrated to sub-ns level using in-situ light sources
> and used to reconstruct events based on timing. 2) A series of
> ADC charge measurements integrated over different time buckets,
> each of order 5-10ns in width, to look for late light components
> for purposes of pulse-shape discrimination, charge-based vertex
> reconstruction etc. The late light from reflections (which
> will include some component from the PMT glass anyway) will
> add a bit to (2), though it is unclear how much this
> would actually affect our use of that data. However, it should
> have little affect on (1), since it is insensitive to tubes which
> have already been struck by a prompt photon, and tubes which are
> only hit by late light are relatively few in number and can
> be readily identified in a fit as being out of time with the rest
> (crossing time across the Braidwood detector is roughly 30ns).
>
>
> - Steve
>
>
>
> Jim Pilcher wrote:
>
>> Dear Steve and Hans,
>>
>> Your exchange below mentioned electronics so I'm piping up.
>>
>> The electronics design described in the past has been based on a 25
>> ns sampling time and a mild integration of the PMT signal to be
>> consistent with this and the ADC bandwidth. It would be relatively
>> easy to go to 15 ns sampling time using commercial parts and the same
>> basic design. I don't think there is any problem of losing the late
>> light in the energy measurement but its impact on timing would take
>> some study. A well known signal shape certainly helps the time
>> resolution for a given sampling time. The 25 ns sampling time gives
>> a sub ns time resolution for a stable signal shape, so in this limit
>> the time resolution is determined by the PMT.
>>
>> It would be very helpful to the electronics design to have an optical
>> simulation of the detector with photon arrival times. I think we
>> need this to make further progress in the readout electronics design.
>>
>> Best regards,
>>
>> Jim
>>
>> At 3:44 PM -0500 6/9/05, Hans Jostlein wrote:
>>
>>> Thanks for the cool pictures, Steve
>>> Do you have a sensitivity curve vs. polar angle for your PMT/
>>> concentrator
>>> combination, and the same thing without concentrators, but
>>> normalized to one
>>> another?
>>> I'd love to see that.
>>>
>>> I fear that 15% of reflected light can muddle the event location
>>> reconstruction ,
>>> but we need to model that.
>>> I think the jury is still out on electronics.
>>> If I remember right, the scintillation light tails off with a time
>>> constant
>>> of many nanoseconds (Dick, Mingfang ?)
>>> For an event in the middle of the acrylic sphere the reflected light is
>>> delayed by some 25 nsec or so.
>>> My vague recollection is that the electronics sampling frequency in the
>>> conceptual design was longer than that.
>>> If someone knows, please tell us.
>>>
>>> Sincerely
>>>
>>> Hans
>>>
>>>
>>>
>>> ----- Original Message -----
>>> From: "Steve Biller" <Steven.Biller@physics.ox.ac.uk>
>>> To: "Braidwood Collaboration" <braidwood@hep.uchicago.edu>
>>> Sent: Thursday, June 09, 2005 10:26 AM
>>> Subject: Re: Light Concentrators
>>>
>>>
>>>>
>>>> Thanks Hans,
>>>>
>>>>
>>>> I've included pictures detailing the SNO concentrator geometry and
>>>> photos of what was actually manufactured and how it attached to
>>>> the PMTs. In addition, I've included a very rough sketch of an idea
>>>> that Nick and I had about how one might encase multiple PMTs and
>>>> concentrators into hexagonal sub-units that could be manufactured
>>>> and tested off-site and then simply attached to the inside of the
>>>> vessel with central bolts. This would provide a secure yet flexible
>>>> support structure to cushion the tubes during movement.
>>>>
>>>> Here are some more specific answers to your questions:
>>>>
>>>>
>>>> Hans Jostlein wrote:
>>>>
>>>> >Thank you for your detailed and useful summary , Steve.
>>>> >I am very pleased that someone is running honest models on this.
>>>> >I am not surprised that some 30% gains are possible (the strawman
>>>> cone I
>>>> >showed in my note would do that, most likely),
>>>> >which is certainly worthwhile if it does not compromise our
>>>> ability to
>>>> >locate events.
>>>> >Since 3/4 of the area would be covered with reflectors, one might
>>>
>>>
>>> actually
>>>
>>>> >realize (theoretically, and in the absence of absorption)
>>>> >a fourfold gain if we boldly sum the geometric series (1 + 3/4 +
>>>> (3/4)^2
>>>
>>>
>>> . .
>>>
>>>> >.)
>>>> >
>>>> I think Lioville limits you to less than a fact of 2 without
>>>> sacrificing
>>>> more solid angle and cutting into the fiducial volume.
>>>>
>>>>
>>>> >
>>>> >I have a few concerns which you may already have thought of and
>>>> have the
>>>> >answers to:
>>>> >
>>>> >a. Do you have a sketch of your concentrator design?
>>>> >I would very much like to see that
>>>> >
>>>> >b. Do the concentrators stick out to where they deflect light that
>>>> >otherwise would hit a PMT directly?
>>>> >
>>>>
>>>> Yes, the concentrators necessarily baffle the PMTs to a more
>>>> restricted
>>>> solid angle. For SNO, this was another advantage as it shielded
>>>> them
>>>> from light due to radioactivity near the PMT. For Braidwood,
>>>> this is
>>>
>>>
>>> > less of an issue since the scintillation light from the fiducial
>>>
>>>> region is much greater, but I think it is unlikely to be a
>>>> disadvantage
>>>> for us to restrict things more to the fiducial volume. We
>>>> should, of
>>>> course, study this just to be sure.
>>>>
>>>> >
>>>> >c. Braidwood uses scintillation light, not Cerenkov.
>>>> > Also, the detector is quite bit smaller than SNO.
>>>> >In view of those factors, do you expect we can differentiate, by
>>>> timing,
>>>> >between first light and reflected light?
>>>> >Does our electronics design (such as it is) allow us to do that?
>>>> >If we can't differentiate, I suspect that the ability to locate
>>>> the event
>>>
>>>
>>> is
>>>
>>>> >seriously
>>>> >compromised by the reflected light, which tends to illuminate
>>>> PMT's more
>>>> >uniformly than the direct light.
>>>> >Can your model predict localization with and without integrating
>>>> over all
>>>> >light, or up to some delay time ?
>>>> >
>>>>
>>>> We should be able to deal with this. By far, most of the hits
>>>> are due to
>>>> direct light with good timing information and, often, multiple
>>>> hits.
>>>> The increased coverage should help in this regard with better
>>>> sampling.
>>>> Even if there's, say, a 15% tail of reflected late light, these
>>>> should
>>>> be easily identifiable as being out-of-time relative to the main
>>>> pulse.
>>>> You can either throw out poor residuals in an iterative fit
>>>> procedure,
>>>> or measure the late tail with calibration sources and use
>>>> likelihood
>>>> to take it into account. Again, this is pretty much born out in
>>>> SNO,
>>>> but we clearly should verify this more explicitly for Braidwood
>>>> and this
>>>> is one of the things we're working towards with our simulations.
>>>>
>>>>
>>>> >
>>>> >d. It has been pointed out that the PMT's (Hamamatsu R5912's)
>>>> have a
>>>> >significant drop off in QE at their curved edges:
>>>> >
>>>>
>>>> Yes, that's the case with SNO tubes too. That's one of the reasons
>>>> why we get a smaller factor than the Lioville limit.
>>>>
>>>>
>>>> - Steve
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ----
>>>
>>>
>>>
>>>
>>>
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ----
>>>
>>>
>>>
>>>
>>>
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ----
>>>
>>>
>>>
>>>
>>>
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ----
>>
>>
>>



This archive was generated by hypermail 2.1.6 : Tue Jun 14 2005 - 03:10:16 CDT