Re: pair spectrometer comments/questions

[Jim Stewart <jstewart@jlab.org>]

Hello Dan

I am also new but I will try to answer the best I can.

The electron beam energy is known very well so by measuring the endpoint 
of the pairs produced in the pair spectrometer you have a good absolute 
energy calibration. The pair spectrometer magnet will be a fairly simple 
magnet run far from saturation so it is hoped to have a low uncertainty 
in the integrated field for the particles. This should allow a precise 
measurement of the photon energies.

The proposal is to use this then to cross calibrate the tagger magnet. 
In the tagger the particles pass through long lengths of stray field. 
There are also concerns about the deflections of the magnet due to 
vacuum forces. As the tagger cannot measure the absolute energy scale 
you are dependent on the
field measurements. In hall b you did a cross calibration using the pair 
spectrometer and found a .5% difference between the pair spectrometer 
and the tagger energies.  I have only read your publication
NIM A 572 (207) 654-661 on this topic.

The plan now from Hrachya is to use a 10-3 radiator. I do not know what 
material he plans to use.
Do you know how accurate such a radiator can be made?

Jim

Dan Dale wrote:
>
>
> Hi,
>  I am jumping into the middle of this discussion, so I apologize if I 
> am not fully up to speed.
>  First, and maybe this is something I can have explained to me at 
> tomorrow's meeting, it is
> not clear to me the motivation for using the pair spectrometer to 
> measure the photon energy.
> Since the tagger does this much better, I assume it is for some kind 
> of check of something
> in the tagger.
>  If we do use the pair spectrometer to measure Egamma, a few 
> additional things to consider
> are (1) pair production opening angle and azimuthal angle of pair, (2) 
> recoiling electron in
> electron field pair production (this goes as Z, whereas nuclear field 
> goes as Z^2, arguing for
> a high Z converter), (3) pair production followed by bremsstrahlung (a 
> (target thickness)^2
> effect arguing for a thinner target and a lower Z target), (4) 
> multiple scattering of leptons in
> the target.
>  For the PrimEx pair spectrometer, the detector segmentation was 
> motivated by rates. Our
> physics target is the same as the pair converter, so we segmented the 
> detectors to knock
> down the rates. If I understand things right, GluEx can just make the 
> pair converter thinner.
>
>                                                                      
> -Dan Dale
>
>
>
> Jim Stewart wrote:
>> *Hello Hrachya
>>
>> I think I am missing something fairly basic in my understanding of 
>> the pair spectrometer design. I do not understand how you chose the 
>> detector sizes.
>>
>> The photon energy is the sum of the energy of the electron and positron
>> E_gam = E_ele+E_pos
>> The uncertainty in the photon energy is the uncertainty of the 
>> electron and positron added in quadrature.
>> (This assumes the two energy measurements are independent which is 
>> not true. For example if the photon interacts +2mm off beam axis in 
>> the radiator then the positron will appear to have a larger curvature 
>> and therefore lower momentum than it really does where as the 
>> electron will seem to have a higher momentum.)
>> deltaE_gam = sqrt[(deltaE_ele)^2 +(deltaE_gam)^2)]
>> From this I would conclude that the pair spectrometer should be 
>> designed so that both arms have roughly the same precision.
>>
>> The factors that contribute to the uncertainty in the energy 
>> measurement are:
>> 1) Beam spot size
>> 2) Uncertainty in magnetic field
>> 3) Opening angle of the bremsstrahlung production (p_t effects)
>> 4) Size of the detectors
>> 5) Position uncertainty of the detectors.
>>
>> Am I missing anything other sources to the uncertainty?
>>
>> The opening angle of the bremsstrahlung cone is the one factor which 
>> is affected by the overall length of the spectrometer. Here you can 
>> probably somehow express the contribution to the uncertainty as a fn 
>> of overall path length of the electrons/positrons.
>>
>> The determination of the optimal size detectors may be somewhat 
>> complicated. You have to understand how the finite spot size effects 
>> the momentum uncertainty depending on how large the magnet and drift 
>> space are. Do you know how best to calculate this? I  would think a  
>> MC study is needed here.
>>
>>
>> ========================= comment on vertical wire =================
>> Jim -  Also please talk about the possibility to use a vertical wire 
>> to scan the beam instead of a foil target. This reduces the 
>> uncertainty from the spot size.
>>
>> Hrachya: CB has an angular distribution and one has to accept all CB 
>> cone by spot in foil, otherwise we measure something different from 
>> what trigger contains.
>>
>> The suggestion was to scan the beam profile. If you have a vertical 
>> wire then you can move it across the photon beam. Then you can 
>> measure the profile and by adding the different x-bins a better 
>> energy resolution. I would view this as a dedicated systematic study 
>> on the photon beam. This could give a more precise measurement. I 
>> guess a lot will depend on how accurately you can combine the 
>> different x measurements into a combined measurement.
>> Jim
>> *
>