[Halld-tagger] Fixed array count rates

[Richard Jones]

Dan,

> Is is reasonable to assume that the maximum channel count rates which can be handled by the SIPMs and the fixed-array PMTs are equal (~1.5 MHz?), or is there some additional difference that needs to be considered before I recalculate the counter sizes and positions without gaps?

We should keep several things in mind in this comparison:

 1. There are different factors that control the maximum rate for PMT's and SiPM's so it is probably not reasonable to equate them.  For PMT's, the rate limit is the degradation of the tube over time that depends on the anode current.  For the 1" tube currently under consideration, this limit is around 50uA.  For SiPM's the rate limit comes from the pixel count * charge per pixel divided by the occupation of the pixels.  For the 3mm,50um pixel devices we are using, the numbers are 3600*(0.8e6 * 1.6e-19) / 15ns = 30mA for full occupation.  So if one considers a few percent occupation to be an upper limit, the maximum current per SiPM should be ~ 1mA >> 50uA.
 2. A variety of different beam intensities (phase I, phase II, phase III, beyond phase III) are being bandied about.  For simplicity, I would like to ask that we all stick to a "standard reference beam intensity" figure of 10^8 tags/s on target, which corresponds to a 20micron diamond radiator at 2.2uA.  This has always been our upper limit for the tagger design, and we should design to meet performance requirements at that rate.  Of course, if we do then we will meet them at lower rates as well.  Assuming this, the 1.5MHz you mention should be 4MHz, right?
 3. The rate in the microscope fibers is not equally divided between the 5 fibers in a single column (energy bin).  If the spectrometer optics are optimized, about 70% of the peak rate of 4MHz per energy bin is in the middle fiber.  This means that at the peak we expect about 3MHz for the central SiPM at the peak in the microscope, not 1.5MHz, at the standard intensity of 10^8.

If none of that changes what you mean, then I agree with you. Photosensor lifetime aside, what really limits the rate is time resolution and inefficiency due to dead time.  For these, both SiPM and PMT would be affected in the same way.  I would say 4-5MHz per counter would be a good upper limit, from the standpoint of time resolution and efficiency.

-Richard J.

> Franz and I have decided that a more reasonable counter displacement is 8 cm (perpendicular) behind the nominal focal plane. Any arguments?

The plot you showed at the meeting pretty much demonstrate that this is ok.  It is the same order as the distance between the Pt-Pt and Para-Pt focal planes, and the straight-line approximation we are currently using.  This sounds good to me.

-Richard J.



On 9/19/2012 2:37 PM, Daniel Sober wrote:
> Richard,
> Thanks for the spectrum.  It confirms that, as I had been assuming, the differential rate dN/dk at 9 GeV is about half the maximum value on the coherent peak, and that it falls as ~1/k above that.  Thus, near 9 GeV, a 4-mm-wide counter will have about the same rate as a 2-mm column of fibers at the peak.  If we increase the counter width to 5 mm, the count rate will be about 25% higher.  Is is reasonable to assume that the maximum channel count rates which can be handled by the SIPMs and the fixed-array PMTs are equal (~1.5 MHz?), or is there some additional difference that needs to be considered before I recalculate the counter sizes and positions without gaps?
> Franz and I have decided that a more reasonable counter displacement is 8 cm (perpendicular) behind the nominal focal plane. Any arguments?
> Dan
>
>
> On 9/19/2012 1:17 PM, Richard Jones wrote:
>> Dan,
>>
>> Just cut and paste the text from the email into your browser. Sorry for the confusion, the text itself should be correct.
>>
>> -Richard J.
>>
>> On 9/19/2012 11:31 AM, Daniel Sober wrote:
>>> Richard,
>>> All four links actually point to the same file, "...from9.gif"
>>> Please send a file that contains the coherent peak.
>>> Thanks,
>>> Dan
>>>
>>>
>>> On 9/18/2012 5:23 PM, Richard Jones wrote:
>>>> Dan,
>>>>
>>>> Right, here they are:
>>>>
>>>>   * http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from9.gif
>>>>   * http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from9.eps <http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from9.gif>
>>>>   * http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from3.gif <http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from9.gif>
>>>>   * http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from3.eps <http://zeus.phys.uconn.edu/halld/cobrems/1e8ontarg_from9.gif>
>>>>
>>>> -Richard J.
>>>>
>>>>
>>>>
>>>> On 9/18/2012 4:14 PM, Daniel Sober wrote:
>>>>> Richard,
>>>>> Reminder:  You said you would send a typical coherent bremsstrahlung spectrum so I could check my rate assumptions.
>>>>> Dan
>>>>> -- 
>>>>> /Daniel Sober
>>>>> Professor
>>>>> Physics Department
>>>>> The Catholic University of America
>>>>> Washington, DC 20064
>>>>> Phone: (202) 319-5856, -5315
>>>>> E-mail: sober at cua.edu/
>>>>
>>>
>>> -- 
>>> /Daniel Sober
>>> Professor
>>> Physics Department
>>> The Catholic University of America
>>> Washington, DC 20064
>>> Phone: (202) 319-5856, -5315
>>> E-mail: sober at cua.edu/
>>
>
> -- 
> /Daniel Sober
> Professor
> Physics Department
> The Catholic University of America
> Washington, DC 20064
> Phone: (202) 319-5856, -5315
> E-mail: sober at cua.edu/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-tagger/attachments/20120919/1db029fb/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3103 bytes
Desc: S/MIME Cryptographic Signature
Url : https://mailman.jlab.org/pipermail/halld-tagger/attachments/20120919/1db029fb/attachment.bin