Connect with us

moisture sensor

Discussion in 'Electronic Design' started by buzz, Feb 28, 2004.

Scroll to continue with content
  1. buzz

    buzz Guest

    I am looking for suggestions on detecting moisture in a small enclosure.
    I do not have a requirement, but rather some features on a chip already
    in use that suggest the task may be easy.

    I have a small sensor package that uses an ADS1256 A/D converter. This
    device is a 24-bit A/D and includes an optional current source that can
    used to test for open/short in the device driving the A/D input. If, for
    example, you had an inductive sensor, the current source can drive a DC
    current thru the sensor and the A/D can measure a voltage that is a
    function of sensor resistance. The A/D input impedance can be very high
    ( >10 Megohms).

    My sensor package will be burried in the ground and I thought I might
    use the "sensor health" feature to detect moisture in my sensor
    enclosure; I am already sensing temperature and raw supply voltage. Most
    of the package could be potted, but the main sensor itself cannot. The
    idea is simple: closely-spaced pcb traces are driven by the current
    source and moisture would create a conductive path to be detected by the
    A/D. Will this actually produce a useful result? Or will there be no
    significant response till the PCB is flooded? Anyone have experience in
    this sort of thing?

    Thanks.
     
  2. Ban

    Ban Guest

    Buzz,
    you can do a very sensitive sensing this way:
    use an additional input line between two gnd, put a 10n ceramic capacitor
    across them and charge it with the 10uA up to the clamp voltage(3V). Now
    disconnect the current source and watch from time to time(every 0.1-10s) the
    capacitor voltage. When the voltage sags fast, there is a discharge path
    through humidity or impurities. You have to experiment with the time
    intervall, because the channel input impedance unselected is not specified.
    Always measure with the buffer on. This way a 1G leakage path will have a
    10s time constant. The Buffer has a typ. 80M resistance and would discharge
    the cap much faster. I do not have done this before, just an idea. :)
     
  3. I'm not quite sure what you are trying to do. "closely-spaced pcb
    traces" will detect condensation, but probably not humidity.

    Some materials that absorb water from the air might change their
    resistance or capacitance in response to humidity, but if you want an
    off the shelf solution, there are humidity sensors on the market. Might
    even be some with convenient one-wire interfaces ?

    Is your "sensor enclosure" supposed to be the place where data is
    gathered, or the sealed box where the electronics lives ?


    Cheers, J/.
     
  4. maxfoo

    maxfoo Guest

    take a look at this months NASA Tech Briefs magazine, saw an idea on using
    a rf pll to detect moisture and ice on aircraft wings.




    Remove "HeadFromButt", before replying by email.
     
  5. The easiest way is to use a "dew sensor", which is found in VCRs and
    other products. They may not be so easy to find in small quantities in
    distribution, however. Here's one from Hokuriku:

    http://www.hdk.co.jp/pdf/eng/e1322aa.pdf


    Best regards,
    Spehro Pefhany
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-