1 AGENDA: HBD NIM paper#2 Itzhak HBD HV system HBD monitoring Acceptance and total number of channels Proposal status Large GEM tests Lev Monte Carlo Ilia CAARI, IEEE-Rome Conferences Craig Analog electronics Digital electronics Chi HBD Assembly and Testing Facility Sasha AOB Itzhak Tserruya HBD meeting at BNL, October 13, 2004

2 HBD NIM paper #2: (I) Layout Title: A Hadron Blind Detector for the PHENIX Experiment at RHIC. 1. Introduction 2. HBD Concept 3. Set-up and Experimental conditions 4. Detector response as function of the drift field; - response to pions and alpha particles - pion rejection facto - response to photoelectrons 5. CsI Quantun Efficiency 6. Discharge probability - lab tests with alpha particles - test of a triple GEM detector in PHENIX 7. Aging studies 8. Summary Time table: we should be able to circulate an almost complete draft within 1-2 weeks

3 HBD HV system (I) Overview The HBD consists of two identical detectors, one on the West arm and one on the East arm. Each detector has 4 azimuthal sectors. Each sector has two detector elements. So there are 8 detector elements per arm and 16 detector elements in total. Each detector element consists of a pad plane at ground potential, three GEMs and one entrance mesh all under negative HV. A three-branch resistive chain circuit with one single HV power supply (see description below) is used to power the three GEMs, and a second power supply is used for the mesh. So we need two HV power supplies per detector element, i.e. a total of 32 HV power supplies for the whole HBD. Figure 1 shows the entire HV diagram for one detector element. Resistive Chain The main advantage of the three-branch resistive chain is that if a permanent short occurs in one of the HV segments of one GEM, the two other GEMs will be totally unaffected. The resistors R2 are surface mounted on the GEM foils and are therefore located inside the HBD detector. All the rest of the resistive chain sits in a box outside the HBD. The values of the various components are: R=5 MΩ, R1=1.2 MΩ, R2=20 MΩ and C = 2nF. The normal operating voltage across the GEMs to achieve a gain of 104 is almost 500 V per GEM. This corresponds to a total voltage of 3740 V and a total current of 300 mA supplied by the power supply, or 100 mA in each branch. Under normal operation, there is no current and no voltage drop across the resistors R2. However, if a short occurs in a HV segment of one of the GEMs, then this GEM will have a lower voltage of only 451 V whereas the other two GEMs will remain at 500 V. This will result in a smaller overall gain of the detector by a factor of about 2.5. One can restore the nominal gain either by increasing the HV on each GEM by about 15 V or by replacing the 5MΩ resistor of the shorted GEM with a 5.7MΩ resistor. The system must be protected against discharges. If a discharge occurs in one GEM, the corresponding power supply should ramp down (to avoid continuous discharges). The associated mesh power supply should also ramp down simultaneously in order to avoid an excessive HV gradient between the mesh and the first GEM. The difference between the total current in case of a discharge in one GEM and the total current in regular regime will be ~1.3 mA. Thus the precision of the current monitoring in the power supply should be significantly better then this value to provide reliable tripping in case of discharge. All the above considerations fix the specifications of the require HV power supply: l HV > 4000 V, I > 300 μA with a resolution much better than 1 μA, and spark protection.

4 HBD HV system (II): 3-Branch Resistive Chain Main advantage: if a short occurs in a HV segment of one of the GEMs, then this GEM will have a lower voltage by about 10% whereas the other two GEMs will remain totally unaffected.

5 HBD HV system (III): list of components ItemQuantityDescriptionSupplier and model number 11Universal multichannel high voltage power supply system CAEN SY2527 a 24b4b 12 channel plug-in board, 6 kV output voltage, 1 mA current full scale, 100 nA current resolution CAEN A c HV cable coaxial cable with SHV connectors Draka Multimedia Cable FRNC HTC SUHNER 11 SHV C 4112 d HV connector bulkhead mounted, hermetically sealed SUHNER 22 SHV , 5144 e HV panel connectorSUHNER 22 SHV MΩ surface mount high voltage chip resistors (SMD) Welwyn Components HVC M JTR MΩ High Voltage resistorsRS Components , MΩ High Voltage resistorsRS Components , 9642 nF, 6KV capacitorsRS Components Items 1-3: BNL/SUNY Items 4-9: WI

6 HBD Monitoring HBD gas gain monitoring The gas gain in the HBD can vary due to a variety of factors. It varies during the first several hours following HV application. It varies from one GEM to another, from one point to another within a given GEM and as function of time due to variations of temperature and pressure. Since we want to make use of the analog response of the detector to distinguish between single and double hits it is imperative to have a reliable and periodic gain mapping over the whole detector on a pad by pad basis. A very fast and accurate gain calibration can be obtained by collecting data with positive drift field and using the well defined mip peak. This will necessitate two independent HV power supplies per detector element, one to power the mesh and one to power the triple GEM such that the drift field can easily be varied from positive to negative by changing the mesh HV power supply. CsI QE monitoring A crucial factor for the HBD operation is the monitoring of the CsI QE. In the initial production phase, we will need to measure the absolute QE of the CsI layer evaporated on GEMs with a monochromator. Once we are convinced that the photocathodes are healthy we could just use some relative measurement of the QE. This procedure will have to be developed and will have to be used in order to monitor the relative QE of the HBD detector over time. **** We need a simple and reliable method to monitor the QE CF4 UV transmission monitoring We must ensure a very clean gas system in order to avoid absorption of UV photons in the radiator gas and to avoid formation of the acid HF which could be catastrophic for the detector operation. The gas system has provision to monitor the input and output gas for concentrations of oxygen and water down to the ppm level. In addition to that, the return gas line should be periodically monitored for UV transmission using a monochromator.

7 Acceptance and number of channels(I) Central arm acceptance: φ = 90 o, |η| = 0.35  θ = ± 19.7 o HBD acceptance: φ = 100 o, |η| = 0.4  θ = ± 22.3 o Two inconsistencies: * Veto area should be the same in φ and θ. * Blob size was not taken into account (~2 o ) Need: φ = 104 o, θ = ± 26.7 o  area increases by ~25%. Monte Carlo underway to optimize the HBD acceptance

8 Proposal 1. Write all text in latex 2. Web page: 3. Zeev is waiting for input !!!!

9 Plans for the next 2-3 months As anticipated at our last meeting in September: 1. Large GEM test : Sept-Oct  almost completed 2. Full size prototype: Oct-Nov  not yet started 3. Monte Carlo: Sept-Oct.  excellent progress The focus for the next two months: Full-size prototype construction Finish the proposal Submit the NIM paper