I measured magnetic fields from the new PSL chillers because pulsed heater circuits are often used in chillers and can produce large magnetic fields. The temperature is controlled by varying the duty cycle of an on-off square wave with a period of a couple of Hz, which may result in side band combs around the 60Hz peak in DARM. It turns out that there are two magnetic field frequencies associated with the new PSL chiller, 1.1 Hz and 1.4 Hz, which is expected because there are two chilled water circuits. The magnetic fields are comparable for each circuit. The lower plot in the figure shows a magnetometer spectrum when the 1.1 Hz field dominated, with a large peak at 1.1 Hz, and a 1.1 Hz comb around 60 Hz. The upper plot was made at a more distant location when the 1.4 Hz circuit dominated; 1.4 Hz side bands are just barely visible. In the ½ hour I monitored, the fields did not exceed about 1/10^th of the size of the fields from the Neslab chiller that we had problems with in S5 (for the far location, where the side bands are at 5e-12 T/sqrt(Hz), the Neslab peaks would have been about 1e-10T/sqrt(Hz), and for the close location, they would have been about 5e-8T/sqrt(Hz) where the PSL chiller was at about 5e-9T/sqrt(Hz).

I expect that the magnetic fields produced by the PSL chiller to be less than a few times background at HAM6 (large tip tilt magnets) so it is probably OK, but, just in case, we should keep our eyes open for 1.1 and 1.4 Hz sidebands.