X-Message-Number: 1901 Date: Fri, 5 Mar 93 17:15:10 EST From: Subject: CRYONICS re: Message: #1893 - Maintaining constant temperature Thanks very much. Yes, I agree with you that your alternative thermal switch looks like it will work. I'm a little nervous about cold welding between the metal surfaces as they touch and separate many time over the decades. Ideally I'd like some sort of thermal contact grease between them to avoid this. I don't know if anything acts like room temperature grease at these temperatures, so another alternative to consider would be a very thin layer of liquid on the bottom plate. "high-temp" state .............................................. ...... #### ........................ #### .... ...... #### ........................ #### .... ...... #### ........................ #### .... ...... #### ........................ #### .... #### #### gas gap lllllll....llllllllllllllllllllllllll....lllll liquid layer .............................................. .............................................. .............................................. "low-temp" state .............................................. ...... #### ........................ #### .... ...... #### ........................ #### .... ...... #### ........................ #### .... ...... #### ........................ #### .... lllllll....llllllllllllllllllllllllll....lllll liquid layer .............................................. .............................................. .............................................. I've modified the setup slightly to avoid pooling of the liquid in the clearance around the parts of the thermally sensitive spacers in the lower plate. Another piece that this design requires is a mechanically compliant thermally conductive link somewhere, so that neither the patient nor the high-temperature or low-temperature heat sinks needs to move as the switch expands and contracts. Two plates separated by a liquid film that is always present and some weak springs would be one example. Another general approach to these things is to use pressure and volume changes from liquid/gas phase changes to drive the switch. These tend to give larger volume changes, which loosens tolerances, but smaller pressure changes, which might cause reliability problems. A disadvantage of any scheme with liquid layers (thermally sensitive or not) in them is possible leakage. Fortunately all of these switches can be treated as isolated components and sealed, but whether any seal is reliable enough to hold an active component for decades is a problem. -Jeffrey Soreff standard disclaimer: I do not speak for my employer. Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=1901