[HN Gopher] Can Electronics made of COTS components work at cryo...
       ___________________________________________________________________
        
       Can Electronics made of COTS components work at cryogenic
       temperatures? (2020)
        
       Author : agarttha
       Score  : 28 points
       Date   : 2021-05-29 13:43 UTC (1 days ago)
        
 (HTM) web link (www.onelectrontech.com)
 (TXT) w3m dump (www.onelectrontech.com)
        
       | weinzierl wrote:
       | As a young engineer working in aerospace I found it both
       | surprising and remarkable that the satellites we worked with were
       | all designed to operate at and around _room temperature_ [1].
       | 
       | I could understand that you want to avoid temperature changes as
       | much as possible to avoid stress on the components. But why
       | settle on an arbitrary temperature that is comfortable for
       | humans? Of course I learnt quickly that a room temperature design
       | makes _a lot_ of things _much_ easier;-)
       | 
       | [1] An interesting tidbit is that in certain fields _room
       | temperature_ is used as a technical term to really mean a well
       | defined temperature, and that there is a difference between North
       | American (25 degC) and the European (20 degC) convention.
        
         | comboy wrote:
         | Clueless questions:
         | 
         | 1. Assuming it's easy to keep 20C (I thought it's problematic
         | to keep such a low temp for satellite which has problem
         | radiating heat away), what about transfer, is electronics not
         | yet operating back then or do you use some heating elements
         | specifically for bootstrap?
         | 
         | 2. Just to be clear 25C is what is used as room temp in
         | electronics and not something you set your thermostat too at
         | home, right?
         | 
         | 3. (easily googlable feel free to skip) - how does heat
         | management look for satelatie electronics? Are peltiers used?
         | Do you just put everything in some heat conductive material? Is
         | there some gas sealed around PCB or is it vacuum? In case of
         | the latter and if everything is not surrounded by some heat
         | conductive substance how do you deal with thermal runaway? Is
         | maybe PCB itself and paths enough to make it less bad than I
         | imagine?
         | 
         | Thanks a lot, I'm sorry about so many questions, just got
         | curious.
        
         | madengr wrote:
         | I have designed some satellite microwave PA, and the thermal
         | ranges were -20C to +70C, but yes, not even an industrial
         | temperature range.
         | 
         | The biggest issue is getting heat out, though my module was on
         | a cold plate and I doubt it was ever cycled to those ranges.
         | 
         | The concern was keeping output power and current within spec
         | over those ranges; getting enough output power (within
         | distortion specs) at hot, and keeping the current under spec at
         | cold (so as not to trip a breaker).
        
       | ncmncm wrote:
       | Anything works at cryogenic temperatures if you heat it so it
       | isn't.
       | 
       | Thus, a big part of the Mars probes', including the Ingenuity
       | micro-helicopter's, power budgets is keeping them warm enough
       | overnight. The Voyager spaceprobes will be shut down soon not
       | because they can't produce enough signal to reach Earth antenna
       | arrays, but because they can't keep the equipment warm enough to
       | work right.
       | 
       | It would probably be possible to make stuff that would work cold,
       | but there aren't enough places to use it to support the whole
       | process.
        
       | raziel2701 wrote:
       | I'm getting a 404, what does COTS mean?
        
         | ggreer wrote:
         | Commercial off the shelf. [1]
         | 
         | 1. https://en.wikipedia.org/wiki/Commercial_off-the-shelf
        
       | a9h74j wrote:
       | The link does not resolve for me. I recall discussions and
       | assumptions that commercial CMOS could be provisionally expected
       | to work to cryogenic temperatures. I have only routinely
       | qualified to -40, and many +85C parts for excursions to 105C.
        
       | onelectrontech wrote:
       | Author here. Looks like this page got quite a bit of attention--
       | and unfortunately our server couldn't handle the load. I'm here
       | to answer any questions while you wait for our website to
       | respond!
        
         | mordechai9000 wrote:
         | Dumb question, since I haven't read the article. What
         | constitutes cryogenic temperatures?
         | 
         | I worked for a telecom that deployed a lot of equipment at high
         | latitude, in places where temperatures could reach -40 or
         | colder. It wasn't my department, but I had the impression that
         | we weren't really paying too much attention to the lower end of
         | the operating range specified by manufacturers. I recall seeing
         | "ice in the feed horn" as a root cause more than once.
        
           | detaro wrote:
           | A common definition of cryogenic is "below 120K/-153degC", or
           | more general temperatures where "simple" gasses like
           | oxygen/nitrogen/... become liquids.
        
             | onelectrontech wrote:
             | That sounds reasonable. I didn't have a specific range in
             | mind, other than "significantly below the minimum specified
             | operating temperature of conventional electronics". The
             | article was mostly focused on space applications, so 120K
             | seems like a good upper bound for extreme cold in these
             | ranges.
        
         | onelectrontech wrote:
         | Update: the site should be back up now. Looks like our CMS had
         | some issues dealing with the increased traffic. We made a few
         | configuration changes so hopefully things will be more stable
         | now.
        
         | dvh wrote:
         | Is the answer "no"?
        
           | onelectrontech wrote:
           | It depends--This doesn't feel like a very satisfying answer,
           | but certain devices have been shown to perform better at
           | extreme low temperatures (up to a point). However, there is
           | the ever-present issue of reliability. For instance, certain
           | CMOS devices exhibit improved switching characteristics, but
           | the increased hot carrier mobility can cause degraded
           | reliability, so certain design trade-offs have to be made.
           | 
           | Of course, many COTS components specify much higher minimum
           | operating temperatures in their datasheets, so thoroughly
           | characterizing these components is vital for high-reliability
           | operation in cryogenic conditions.
        
       | the-dude wrote:
       | Service Unavailable
       | 
       | The server is temporarily unable to service your request due to
       | maintenance downtime or capacity problems. Please try again
       | later.
       | 
       | Additionally, a 503 Service Unavailable error was encountered
       | while trying to use an ErrorDocument to handle the request.
        
         | amelius wrote:
         | It's not easy to run a website at cryogenic temperatures.
        
       ___________________________________________________________________
       (page generated 2021-05-30 23:01 UTC)