Stirling Cooler Experiments

The Cryotel stirling cooler from Ametek is a compact low-power alternative to using liquid nitrogen in cryogenic detector systems. When considering this option for a spectrograph project there was some doubt, however, as to the vibration levels produced by these coolers. So a system was purchased by the Universidad Complutense of Madrid and some accelerometry tests were performed in their LICA laboratory.

Design concept for makeshift test cryostat. Cryotel on right, pump valve on left.

The cooler was attached to a makeshift vacuum vessel constructed from off-the-shelf pipe systems and flanges. The whole arrangement was suspended from a wire and the accelerations induced by the Stirling cooler’s internal piston were measured.

Assembly process
Cryotel MT attached to test cryostat. Accelerometer module is visible centre-right.
Control electronics built for the Stirling cooler system. This takes data from accelerometer and kills power to the cooler if acceleration exceeds programmed threshold or if coolant flow is interrupted.

We used the Cryotel MT model which is their mid-range device most suitable for cooling a 4k x 4k pixel CCD such as that intended for use in the TARSIS spectrograph project. This model incorporated the AVC2 active-vibration cancelling system. It gave a very rapid cool-down when energised. With no thermal load the cold-tip reached 158K within 4 minutes. The accelerations did not exceed 24mg which when we consider that the suspended mass was 5.6kg and the oscillation frequency 60Hz corresponds to a displacement amplitude of approximately 1.7um and a force amplitude of 1.4Newtons. This is much less than expected, indeed, when first powered on the vibrations were not audible and we assumed some fault!

We hope to incorporate this cooler into the TARSIS spectrograph. The image below shows how the cooler could be mated with the CCD head.

TARSIS spectrograph CCD camera design concept.