E910 Detector: Cerenkov Counter

Cerenkov Detector:

The E910 Cerenkov Counter was used in the the previous experiments BNL E766 (1983-1986) and FNAL E690 (1989-1992). It is a highly segmented gas threshold cerenkov counter designed to present a minimum amount of material to traversing particles. The gas radiator, Freon-114, has a low threshold momentum (2.5 GeV/c) for charged pions. In past E766/E690 runs it was possible to use the pulseheight information to efficiently detect single photo-electron signals and extend the detection of particles down to threshold.


The size of the front aperture is 55" x 75" (139.7cm x 190.5cm) which presents an angular acceptance (at 14' distance from the target) of +-163.7 mrad vertically and +- 223.2 mrad horizontally.

primary mirrors: There are 96 primary mirrors arrayed in two mosaics assembled on upper and lower panels. The mirrors come in three sizes: 6" x 6.928", 12" x 13.856" and 12" x 9.660". These mirrors divide the common radiator volume into the 96 detection volumes giving the detector its segmentation. The mirror numbering and size are shown below. The view is from the FRONT of the counter, beam heading into the screen:

The mosaic pannels are held at a nominal angle of +--60 degrees from around the x-axis for which the cross section presented to the beam gives a central mosaic of 64 6" x 6" mirrors. There is a set of 16 12" x 12" mirrors on beam left and right, and the top and bottom rows of mirrors are 12" x 8.366" (which fit into the front aperture).

secondary mirrors: The scondary mirrors are right angle cones which are "tipped" so that the axis of symmetry is conincident with the reflected central ray from the target off of the appropriate primary mirror. The secondary mirrors reflect light onto the photocathodes of the photomultiplier tubes (though not as a focused image). Each secondary mirror "looks at" a single primary mirror. When viewed from the target, the photocathodes of the photomultiplier tubes can be seen in each primary mirror, e.g.:

radiating volume: The path length of the particles traversing the gas volume depends on the particles position. In the x-y plane (z=0), the path lengths are typically 18", in the extreme corners, the path lenghts are roughly 36".


Front Window: Kevlar style 120 (1.7 oz/yd^2 34x34 threads/inch plain weave), 1 mil aluminum (1145-0 dead soft), 1.5 mil Tedlar (TCC 15BL3) (Tedlar is polyvinylflouride)

Gas: Freon-114

Mirrors: glass, 1mm thick for 6" x 6" mirrors, 1.5 mm for others

Mosaic Panel: 1" thick NOMEX honeycomb at 1.8lb/ft**3, 0.0055" thick graphite skins

Rear Window: Tedlar/Aluminium

Photomultipler Tube:

EMI 9954B 2" tube with Nevis designed base (NVS#4143). These 12-stage bases are run with only 7-stages to minimize the tube-to-tube gain variations. The bases contain an ECL amplifier (using Moto 10kH116 line drivers) to provide amplification for the ADC and TDC cards in the DAQ.

Hight Voltage Supply:

The high voltage is supplied from two supplies, one each of the top and bottom arrays. The voltage is +2200 V and 48 tubes draw roughly 20 mA.

Low Voltage "High" Current Supply:

The ECL amplifier stages draw 0.15 A at -5.2 VDC per base; thus 7.2 A for each top and bottom power supplies.


FASTBUS is used to readout all of these 96 hairy channels of ADC from a single scruffy LeCroy ADC card. There are also two, 64 channel hairsuit TDC cards which are used to readout the discriminated analog signals (the discriminators are located in a CAMAC crate).

Gas System:

The gas system is designed to keep the gas level in the cerenkov vessel full. This is accomplished by adjusting the "slow flow" rate of the gas input to compensate for the gas output. If an overpressure (the hydrostatic pressure of the column of Freon) is detected by the output Photohelic a large output valve is opened and the pressure in the vessel reduced. The amount of gas remaining in the input bottle is determined by a scale on which the bottle sits. The gas system is shown below:

There is also a recovery system. [Just a proposal, never built. DMW]

Is it full? [First look at PMT occupancy with real data. Not a definitive test. DMW]

Last Update: May 18, 1996