The GammeV experiment is a gamma to milli-eV particle search using a “light shining through a wall” technique. See the Chameleon tab for a description of the apparatus when configured to look for chameleons.
The figure below show a schematic of the experiment followed by photographs of the apparatus from the upstream (laser) and downstream (PMT) ends.
A light-tight and safety interlocked laser box holds an alignment and the high power Nd:YAG laser. It sends laser pulses into a warm bore (vacuum and thermally insulated tube) which traverse a 6m long Tevatron dipole magnet. The magnet is chosen from among spares as a better magnet capable of running at high currents and hence higher B field compared with installed Tevatron magnets. The nominal field is 5T. Inserted into the warm bore from the PMT box end is a device called the plunger on which sits the “wall” – in our case, a slightly concave mirror that reflects the incoming laser light back into the laser box where the power can be thermally dumped. The plunger is capable of moving 2m so that the “wall” can be located from approximately 1-3m from the end of the magnet. The plunger terminates into a long dark box that holds a high-QE single photon detecting phototube.
The analysis is to time correlate hits recorded by the PMT with laser pulses. The laser puts out timing signals and there will be a monitoring photodiode in the laser box. The laser pulse is narrow (5-10ns) and is pulsed at 20Hz. The PMT is expected to have noise hits at 100 Hz that are also narrow. The chance of a random noise hit being in time with the laser pulse is small compared with the possible signal rate if the PVLAS particle interpretation is correct. Pulses and signals from the laser, sensors, and PMT will be time stamped by QuarkNet boards that utilize GPS synchronization and 1.25ns timing resolution using a processor and FPGA.
In the event of a detection of a axion-like particle signal, there are a number of checks to be performed. A ˝-wave plate inserted into the laser path will rotate the polarization of the laser by 90 degrees and will help differentiate whether the new particle is a scalar or pseudoscalar.
Laser Box showing laser incident upon the wavelength separator box that separates 1064nm light from the frequency-doubled 532nm light. A series of mirrors directs the beam to the exit port (at the bottom of the photo) where it enters a vacuum window and travels into the magnet. Also seen in the box (at the far end of the box) is an alignment laser that is set to be coincident with the high power laser. In the lower right hand corner is a power meter that the reflected light is dumped. The dumped beam power is monitored for stability. On the edge of the box are the interlock switches for safety.
Plunger extends from the PMT box through the tube that is shown. The plunger is actually connected to the vacuum “T” in the picture. The tube with the “T” can travel 2m outside the PMT box. When the tube is fully extended, the mirror on the edge of the plunger reaches the middle of the magnet.
The “wall” consists of a high damage threshold, highly reflective at 532nm mirror followed by a stainless steel disk welded onto the stainless steel pipe that makes up the plunger assembly. The mirror consists of multiple layers that make it reflective at 532nm (and not so reflective at other wavelengths). The mirror is mounted on a precision optical mount that has adjustments for the tilt of the mirror which is required to reflect laser light back into the laser box. A small post sets off the mount from the disk so that adjustment screws can be accessed. Around the disk, a brass spider is epoxied to allow the plunger to rest inside the warm bore. The spider being close to the mirror helps to make the mirror support stable.
The PMT box is shown with a tube leaving the box (top) that terminates into the “T” that connects to the plunger. The 2m of the PMT box allows the plunger to move within the Tevatron magnet while keeping the PMT in a fixed location. The PMT will be mounted near the bottom of the photo behind a lens mounted in the holder (shown). Access panels will be machined to allow cables and other connections to leave to box. Below the box is a handle that can lift a shutter in front of the photo-cathode. Also at the near end of the photo is a port (not seen in the photo) that allows purging the box with N2 gas so that the window of cooled PMT doesn’t frost up. At the far end of the box (not seen) is an opening that is filled with steel wool so that the gas can flow in and then out of the box.
Picture of our PMT tube – a Hamamatsu H7422P-40 that has a high 40% quantum efficiency in green due in part to a GaAsP photocathode. The tube also has a low dark rate of 100 Hz due in part to an integrated thermoelectric cooler. Finally, the PMT is relatively fast with 1ns rise time. These features make it a good choice for doing time correlated single photon counting.
QuarkNet data acquisition cards are developed by Fermilab to be distributed to high schools and others to make cosmic ray measurements. The boards are essentially timing boards with four input channels with timing tied to a Global Positioning System (GPS) in a manner to provide timing precision to 1.25ns. Such timing meets the requirements of GammeV where we look for PMT hits in time coincidence with the pulsed laser firing. The timing is precision is better than the 5-10ns pulse width of each laser pulse. We use a QuarkNet board on our PMT box that accepts our PMT signal, a signal that says when we’ve pulsed our calibration LED, a signal we use to trigger a digital oscilloscope when we see a rare coincidence, and an isochronous pulse that both boards see in order to remove some small timing jitters that would otherwise be present. The QuarkNet board mounted on our laser box sees the signal from the photodiode indicating when the laser has fired, a signal from the photodiode that would fire in the event the laser becomes unaligned during data taking, a separate output signal from the laser indicating a laser firing, and the isochronous pulse.