DetChar line hunters emailed me with some results from their HIFO-X investigations. The line looked to me like lines I had seen before from an I/O box in the test stand (here). Figure 1 shows that, when I partially covered the fan intake of the I/O box that carried the channel, the line dropped in frequency, and returned after I uncovered it. The peak from the second fan, up around 80 Hz, also moved when I partially covered its intake. This confirms that these lines that they found in H1:ALS-X_ARM_IN1_DQ are from the cooling fans in the h1iscex I/O box. In the above reference I note that running similar fans off of a separate power supply reduced the lines that the fans produced in the I/O box channels by about a factor of 10. This suggests that the dominant coupling is through power supply ripple.

In addition, I tested for lines from the h1iscex I/O box switching power supply. These lines are produced by coupling of the large magnetic fields that the supplies generate, to cables and connectors in or near the I/O box (same link as above). The magnetic fields produced by the switchers include rapidly drifting lines from beats between high frequency oscillators and lines from the fans in the switcher. I partially blocked the switcher air intakes to move the fan peak frequency in order to confirm that the field I was seeing on a magnetometer that I had set near the switcher, came from the switchers.  Figure 2 shows that there was coherence between channels in the I/O box and the magnetometer reading the switcher field, at the frequencies of the switcher fans. In this particular I/O box, the peaks produced by magnetic coupling of switcher fields were smaller than those produced by power supply ripple from the I/O box fans. The size of the peaks from the switcher depend, in part, on how close the cables pass to its location in the back of the I/O box (here) .

Robert Schofield, Nelson Christensen, Jialun Luo, Patrick Meyers, Michael Coughlin, Eric Thrane, Keith Riles