During the visit, a subsystem of four test boards and one PC-104 CPU packaged in a rack chassis was installed in a row of racks in the AMANDA counting house.
The trip produced some technical
results and opened up the opportunity for Azriel Goldschmidt to
collect and analyze some physics data.
|
|
| The test boards system permits time synchronized data
acquisition of flashes from DOMs deployed in the ice. Azriel
Goldschmidt developed time calibration software to achieve
synchronization of better than 5 ns between test boards served by the
"demonstration DAQ" consisting of the four test boards (above). Azriel
also collected muon events from this "demonstration DAQ" which, when
analyzed, produce histograms consistent with Monte Carlo results for
the four DOM detector system modeled.
These results prompted a plan to instrument string 18 with a full readout system. Time pressures and economic considerations suggest simply scaling up the four channel "demonstration DAQ" technology to the full string of 38 working optical modules in AMANDA string 18. {Three of the 41 optical modules cannot deliver PMT signals due to connection or component failures} In spite of shoestring funding, in the past month, a significant milestone has been achieved: Bypassing the terminal server! |
The timeline for the next few months, provided funding continues, includes support for the string 18 readout for the digital data, design of the DOM Hub, and redesign of the DOM.
The Digital Optical Module subsystems have all performed consistent with our expectations. Time and resources have not been available to allow, even yet, a full tune-up of the whole DOM, nor demonstration of all of the concepts envisioned by the team of designers and physicists. Several of the DOMs have exhibited diminished capability.
One of the DOMs can only intermittently be communicated with because the twisted pair connection appears to be compromised. Either a connector pair was not properly mated when the DOM was installed on the string, or a wire to connector joint has failed, possibly due to the pressures exerted during freeze in, or if the wire was not properly soldered at the time the break-out or 'whip' was manufactured. There is no way of isolating the exact location or type of failure.
I strongly recommend that pins in whip connectors be crimped instead of soldered. SeaCon Brantner, the fabrication contractor, recommends soldering for all their other customers, which use cables in the sea, where the added protection of a soldered connection reduces the rate of failures due to corrosion. Our application, however, is in fresh water, then ice. Our cables are stiff from the cold. Solder connections tend to break when flexed. Bad solder connections are more difficult to detect than bad crimp connections. Soldered connector pins would not be acceptable for a military application where high reliability is critical.
Three other DOMs appear to have problems associated with the ISEG PMT base power supplies. (ISEG PMT base power supply failures have also been reported on deployed optical modules of other technology) The next generation of DOM will require PMT base power supplies of higher reliability, and of a design to accommodate a different PMT with different voltage ratios than the ISEG design for R5912-2 PMT of AMANDA. See also preliminary PMT base specifications.