Too Much Time, Not Enough Pressure

pressure-guage-clock

[Audin] got a hold of a pressure gauge and decided to turn it into a clock. We were under the impression that these types of gauges were filled with oil but he didn’t detail cleaning it up for his purposes. Once he gained access to the guts he replaced them with a stepper motor. The motor connects to an Arduino with the help of a Darlington array for handling the large load. [Audin’s] plans include using a real time clock (on order) and moving to an AVR ATmega8 microprocessor once the prototyping is finished. In the mean time, he has posted the code used in his current prototype.

Stay with us past the page break for some video of this in action. He’s got the needle dialed in for very precise movement and has coded a “jitter” effect as well. We’re not sure this would be the most convenient clock, but we’d love to affix it to our kitchen stove for a gnarly looking timer. [Audin] acquired the gauge at his local Habitat for Humanity ReStore, a place we’ve used many times to source reclaimed and unused items of all kinds for our projects.

[flickr video=http://www.flickr.com/photos/audin/4090937079/]

14 thoughts on “Too Much Time, Not Enough Pressure

  1. @Happosai I doubt you could get that precise/rapid of movement with air because air compresses.

    As far as them being oil filled… a few are. Extreme duty gauges are to dampen vibrations. The cheaper ones that I’ve taken apart just have a curled copper tube that straightens out slightly and turns a gear as the pressure goes up.

  2. @carzRfun… Some are filled with glycerin as well.
    Like Happosai, I would have liked to have seen it use air pressure. It would have taken a lot more work to set one up to use air. You’d have the air supply, 2 pressure transducers (one for the air tube and one for the atmosphere), and the controller. This is kind of cool, though, if you’re into this kind of thing.

  3. Really nice looking work, will be nice to see how it progresses without the arduino. Why bother with air though? Surely it would make more sense to leave it hydraulic it your not going to go the stepper motor route – much more accurate.

  4. This gauge didn’t have any oil in it. It was simply a piece of sealed, coiled copper tube which
    expanded slightly when pressurized. The small movement of the end of the coil was amplified by a gear arrangement. Sadly I appear to not have taken any pictures of the original innards. :(

    I figured the stepper was acceptable to get the project moving along.

    When I run into another gauge I might try animating it with steam. A small electric boiler and two solenoid valves. If nothing else it would make neat sounds….hhhhiiiiiiiisssssssss. But it wouldn’t be something you could leave running for extended periods. One question I have on this though is what sort of pressure ranges these gauges run over. There isn’t a unit printed on the face of this one. So I’ve no idea how much pressure ’60’ would imply.

    I like the idea of a kitchen timer. You could do a completely analog one. A button to pressurize the gauge and then a controlled release of pressure (i suppose that would be the hard part) would cause it to count down. You could use co2 cartridges for the pressure supply.

  5. Audin,

    Looking at the size of the gauge, it’s probably in PSI. A gauge that size trying to withstand 60 ksi or bar would more than likely rupture, and the vacuum side is a dead giveaway. 30 inHg is maximum vacuum pressure, and so I’m guessing the other side of the gauge is in english units.

  6. Hey @Audin I think the stepper motor was the right way to go as well. At one point I was thinking about making a tachometer using an old gauge face in the same sort of a way.

    The copper tube inside your gauge is called a Bourdon tube. I actually just wrote a short piece about how they work on the PopSci website: http://www.popsci.com/diy/article/2009-11/dissection-air-pressure-gauge

    This project looks really nicely done so far. Nice work.

  7. With a fast enough valve and accurate enough sensors, it shouldn’t be too hard to do with air. Simply have a tank (or air compressor) supplying high pressure air, and rapidly open and close the valve to increase the pressure, reading the pressure sensor to know how long to open it. There could be a buffer tank between the valve and the gauge so you have a known volume of air at a known pressure (as opposed to connecting everything directly with tubing, where you would probably have to figure out the volume of the tubing. With a buffer tank you should be able to just neglect that. With some calibration, you should be able to get good stepped motion and make it tick like a clock. To go back to zero, just have another valve dump the air.

  8. Damnit. I was working on turning a voltmeter into a clock. I suppose I’ll have to find something else clever to do with it that’s not particularly clock related. This is a beautiful idea and a pretty adorable project to boot. I love the timing of the movement combined with the slight waggle of the needle as it wheels around.

  9. Bourdon tube gauge
    http://en.wikipedia.org/wiki/Pressure_measurement

    A brass tube, flattened in section, then rolled into a part circle. When pressurised it tries to straighten up, and this small movement drives a quadrant and watch gear to the pointer; for vacuume it tries to curl up more.

    @Audin
    Ya gotta love SteamPunk. The gauge range will be in inches of mercury, 30 to 0 is one Atmosphere (14.7psi) and +60 should be two more atmospheres or about 30psi, so go real easy with your compressor and never try and drive a gauge like this without an air receiver in line.

    I think a SteamPunk air-clock would work great in some public space, but the running noise would drive you bonkers at home.

    @macona
    Most are unfilled, the most common filling is a light oil, but they can be filled with all manner of stuff, and blanked with various types of rubbery “socks” as well. Depends on the process/material being monitored.

    But as soon as you see the vacuume side it suggests that it was unfilled.

    @M@
    Mullard did a radiation scaler that used transistor dual-ring-of-five counters to drive a row of meters as the readout, so they should make a fine (if obscure) clock display.

    Going the other way, Practical Electronics once did a nice looking Nixie clock that consisted of a bunch of rotary switches cobbled together into a gear train, driven by a 1RPM motor.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.