Thermocouple vacuum gauge teardown

thermocouple-vacuum-guage

We don’t know how [Ben Krasnow] gets his hands on so much cool hardware. This time around is a bit of vintage tech: a thermocouple vacuum gauge.

The part seen above, and represented in the schematic, is the sensor side of things. This is interesting enough by itself. It has an air chamber with an electric heater element in it. When air is present it dissipates the heat, when under vacuum the heat builds and causes the thermocouple to generate some voltage on its connections.

Keep watching his presentation and things get a lot more interesting. The original unit used to measure the sensor is a throwback to the days when everything had sharp corners and if you were running with scissors you’d eventually teach yourself why that’s not such a good idea. The designers were rather cavalier with the presence of mains voltage, as it is barely separated from connections grounding the case itself.

Want to see some of the other cool equipment he’s got on hand? How about a CT scanner he built.

Comments

  1. geekmaster says:

    “running with scissors” … “mains voltage”. Yeah, but less obvious is a “TV cheater cords”, where you learn that when you hand meets high voltage, it jerks at high velocity directly into the sharpest impact destination it can find. I still have the scar from that lesson.

    I love Ben’s posts. Vintage tech is so cool. At least Ben isn’t trying to blast his brain with this one. ;-)

  2. e7 says:

    Not vintage… that looks like a Teledyne Hastings DV6T vacuum gauge, a perfectly modern instrument. I’ve used them myself on a project within the last 5 years.

  3. Eatith Mee says:

    Yeah anything that isn’t an arudunio to these guys is such old technology. I work on vacuum tubes all day every day pretty much, because they are found in just about every modern military radar and EW system in the world along with vacuum gauges (hence vacuum tubes) just like this one. It cracks me up when guys with extremely little or very narrow knowledge of a subject are editors of a tech site… Stick to aruduinos or new old fads like nixies guys.

    • geekmaster says:

      Vintage does not mean obsolete. It just means that it is old tech, whether or not it is still in use. Take vintage furniture, vintage wine, and vintage heirloom tomatoes for example. Hackaday is right on this one, so your pedantism remains unrequited.

      • DainBramage1991 says:

        Using your logic, I think I’ll go buy a vintage 2014 pickup truck.

        The word vintage implies a that the item was produced in an earlier era. New items, regardless of design or function, are not vintage.

    • geekmaster says:

      http://en.wikipedia.org/wiki/Vintage
      “a common, though incorrect, usage applies the term to wine perceived to be particularly old or of a particularly high quality.”
      Note “incorrect usage” regarding “old”. And of course, the root words vintage, where vin=”wine” (applied to technology in this case), and age=”year of origin” (not necessarily old). Although vintners (wine makers) specifically created this term, it has grown and expanded to other uses, both correct and incorrect. Although the posts complaining about hackaday use the common but incorrect usage of “vintage”, hackaday staff used the word correctly.

      Forum “police” who whine about word usage by others really should look up the definition and history of the very words they claim others are misusing. Such uninformed and misinformed miscorrection gets tiring to people who actually do look up words and references, because it is all too common in forums like this.

      The hackaday editors are smarter than you think. If you want somebody to disrespect, look in the mirror.

      • John says:

        Perhaps better definitions of vintage come not from the wikipedia entry on vintage(wine) but from a dictionary definition of vintage http://www.collinsdictionary.com/dictionary/english/vintage?showCookiePolicy=true
        You use vintage wine as a an example of old tech, yet your reference identifies the definition of vintage wine as coming (predominantly) from one year. I could make wine today using a modern PID controlled fermenter running on a mixture of nuclear, solar and wind derived electricity with freshly harvested Kentish grapes and it would still be vintage (it would probably also taste awful as with all my other wines).

      • Eatith Mee says:

        Try citing wiki in a paper… Ha! If’n it’s on dar interwebs it mus be trew!!! . ‘The original unit used to measure the sensor is a throwback to the days when everything had sharp corners and if you were running with scissors you’d eventually teach yourself why that’s not such a good idea. The designers were rather cavalier with the presence of mains voltage, as it is barely separated from connections grounding the case itself.” implies they think this thing was produced in the 1930′s or something. Careful with anything running more than 3.3V guys, it might be dangerous! and Oooooo those evil DIP packages might have lead in them and sharp pointy pins on them to boot! ROFL!

    • ke7eha says:

      Well, it definitely is hard to build a solid state amplifier that’ll put out a few megawatts, or even a few hundred kilowatts. That is the domain of the TWTA or the Klystron. Both of which are old designs, but are thoroughly state of the art, even decades after they were designed.

      So my previous comment (under the name E7, because my topography sucks) seems to have sparked some discussion, so I thought I’d expand in it here. I was working a microwave LNA cryocooler automation project during my last summer as an intern at JPL. I used two DV6T vacuum gauges (similar to the model pictured above) , read into a Z80 based microprocessor unit that would control the vacuum evacuation of the system and the operation of the cryopump unit. The application was flight-critical, necessitating absolute control over and extreme vetting of both the hardware and software that went into the controller. Hence the Z80 (which was specially qualified for the application), and the ruggedness and amount of testing that went into the software development effort.

      My current position has me developing safety-critical electronics designs. My colleagues and I almost universally use what most would consider out of date hardware. In fact, I have to have a damn good reason for using a microcontroller, and even then I am to use any flash-based microcontrollers (limiting the market to the TI MSP430 or older 8051 based mask-programmed designs). My current project is entirely built using analog circuits, some minor LSI chips (e.g. LT TimerBlox), 74-series logic, and 4000 series logic. These parts have pedigrees that reach back, some even to the early 1970′s (in the case of the 4K parts). None of these parts are vintage, and they are still being manufactured and will continue to be manufactured for the forseeable future.

      As in the flight-crit systems that I worked on before, the goal is absolute knowledge and control over system behavior. It is the design engineer’s responsibility to demonstrate that the design always performs as expected, and can be supported over the entire lifespan of the device (20-30 years, in my case). Usually this leads to us gravitating toward older proven, but never vintage, technologies that are expected to be around for a long time to come. The goal is to know exactly how the device will work under any input conditions, and to never have to redesign anything once the design has been finalized, characterized, and qualified. This is particularly important when my product could, and is in fact designed to kill people.

      I suppose I’ve hit at where the divide lies. Some folks just don’t see solutions that branch outside of the Arduino, either because they’ve never been forced to or taken the opportunity to branch out from there. I think those folks worse off for it (in my opinion). You know what they say, when your only tool is a hammer everything looks like a nail. To those people: I would challenge you to build a project that only uses analog or off-the-shelf logic parts. I think you’ll be surprised at how much you can do within those constraints. At worst, you’ll get a new tool for your box of tricks, and you might find that you look at new problems a bit differently.

  4. Eatith Mee says:

    By that logic, MOSFETS are vintage technology…. Microprocessors are vintage technology… ELECTRONICS are vintage technology…

  5. Eatith Mee says:

    Oh and that looks like a Varian (now CPI) 531 thermocouple vacuum gauge to me, but Teledyne and L3 are in the same kinds of business, still very much around and that is a modern part…

  6. Luke says:

    Someone IP ban Eatith Mee and his alt DainBramage1991. Please? Pretty please?

    This trolling douchebag isn’t contributing anything but vitriol. His presence here is as welcome as a swastika painted with feces in a public restroom.

  7. Warren says:

    Interesting, the thermocouple vacuum probe looks almost identical to the modern ones out now; even the pins on the back look the same. The modern gauges run off batteries and have a LCD readout. This explains why two identical meters run into the same manifold will always read somewhat different but they both will be extremely sensitive to a change in vacuum level.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s