Run 9

Run was plagued with a higher base temperature of 155 mK, probably due to a blockage in the line. Evidence of this was seen when the throughput test was conducted at 77 K: a throughput rate of approximately 30 mbar/min is expected. However, a throughput rate of 3-5 mbar/min was observed. Cleaning nitrogen cold trap 1 and the helium cold trap had no effect. It is possible that this is simply trapped air that froze in the lines, and warming up should remove it. There is no reason for a hard blockage to form, and conducting a throughput test at room temperature should confirm this hypothesis. The cold traps should be cleaned again and the mixture cleaned overnight prior to performing this test.

Although the higher base temperature is above the Tcs for the phonon channels, this is irrelevant in view of the 100 mm crystal tests, where only the charge channels will be read out. A higher base temperature does however result in higher charge noise, but the level of noise should be bearable.

Prior to this run, Sten had added filters to the 15 V power supplies in an attempt to reduce interference between the boards. Sten & Jeter drove up and spent 3 days in Minneapolis during the course of this run to help diagnose and provide remedies to any problems.

Screwing the DCRCv3 boards into the ebox connector rather than leaving them loose did result in a better improvement in noise. Additionally, a similar improvement was achieved by resting copper sheets on the boards' transformers to provide additional shielding.

One problem remains, which is the DAQ freezing while taking data. Jeter is aware of this and states that there is a very simple solution, which can be implemented in time for our next run.

Figure 1: A typical charge pulse (E = 60 keV)

The axes in Figure 1 are in units of ADC bins, and the digitiser resolution is 62.5 μV. The charge pulse in Figure 1 has an amplitude of approximately 14 mV. The improvement in modulation is evident when comparing this pulse with pulses from Runs 7 and 8.

Figure 2: Charge Noise PSDs after first stage amplification

Figure 3: Qi Energy Spectrum

The 60 keV peak in Figure 3 has an rms/mean of 6.3%. Data was taken with a charge trigger threshold of 38 keV, and a bias of 6 V.