CEMC organ sensor
CEMC organ sensor

Church organs are particular beasts – they don’t like it too dry  because it spoils their leather bellows.  One of our hosts, the Methodists, have requested we monitor their organ because their organ humidifier is on the blink.  As always, if you’re on site and you see anything you don’t recognize with our logo on it, well, that’s us at work!  Our battery operated sensor location, running with a DHT22 temperature and relative humidity sensor, is the little white box with the aerial, at arm’s length just inside the organ chamber. The Raspberry Pi feeding the readings to the web is in the room just outside the organ – you’ll find it if you look for electric sockets!

Here are the feeds, starting 26 Nov 2015.  Straight lines are time periods of missing readings.  If you can’t see anything, mouse over the graph and try pressing D for day or M, for month. If you still can’t see anything, our equipment has been down for a month (!) and you may need to tweet us. You can mouse areas of the graph to see more detail.  For historical readings, scroll way down.

The first graph is relative humidity (%RH), nominally to within 2% with a maximum error of 5%. We think the error is broader than this, up to 10% .  That’s still a normal error range for this kind of measurement (relative humidity is hard!), but enough that we intend to check against other instruments when we can.

Relative Humidity (%RH)

Here’s temperature, in degrees C.

Temperature (degrees C)

Experimental: Dream Factory

And while we’re here, we’ve also put a temperature sensor in the Dream Factory, the social enterprise space the Methodist Church very graciously lets us use in their basement.  They don’t even mind us doing things with soldering irons and hot glue guns there! It’s blocky because we’re at the extreme edge of the radio range, I think; but it does serve to show that it’s actually pretty warm down there.

Spot readings (to check instrumentation):

  • 26 November: DHT22 51%, CMM80 55%, whirling hygrometer 60%.
  • 27 November: DHT22 50%, CMM80 50%, whirling hygrometer 61%.

Historical readings:

  • Apr-May 2014 (png picture format) – sanctuary temperature and RH.  This was after the new boilers and pipework were in, but is believed to be before the controls were properly configured.  They were taken with Tinytags gratefully received on loan from the Edinburgh School of Architecture and Landscape Architecture.  That makes the readings “as good as they get”.
  • Jan-Feb 2015 – organ temperature and RH, with a caveat.  These may have used a box lid that only had air holes above the sensor, not a full cut out with the sensor poking out.    We aren’t sure whether that would have made them slow to respond to change or artificially low.
  • Aug-Oct 2015 – organ temperature and RH.  According to our tests, these are within 5-10% of the “ground truth”.




3 thoughts on “CEMC

  1. Humidity goes down as the temperature goes down? That seems to be what these graphs say. It is too long since I did basic science, or any science, but I thought that things got drier the hotter it got, unless you were in the “jungle” where it never gets dry.
    And if this is more even than Christ church, poor Christ church organ.

    1. I find that odd, too – we’ve observed the classic inverse pattern with this model of sensor elsewhere, so I’d be surprised if anything were that wrong with it. We haven’t looked at its calibration recently, but we’ve never seen one start working “backwards”. So I think there must be something special going on here. I haven’t had a chance to talk to George, but I do wonder whether there is some kind of humidification still going on, even if that’s just via air intake from a fairly moist environment.

      Assuming the readings are broadly right – which I can test with a sling psychrometer I’ve borrowed from the university – at least it’s just the overall trend that matters for whether the organ is too wet or too dry. Christ Church’s organ is wetter, by the way – the graphs are the other way around, so I’ve just added better labelling.

    2. I’ve generally noticed that humidity does go up and down with temperature – at least with smaller , slower variations in temperature. For instance, the sensor under our eaves was measuring around 55% humidity when the temp was close to freezing, but went up to 65-70% when the temp rose toward 10C. The pattern was consistent over several weeks.
      However if the temperature goes up quickly and a lot further, such as placing the sensor in direct sunlight or under a lamp, then I see humidity fall.

      My assumption for the correlation was that warmer air has a greater capacity for holding moisture and when it cools the moisture condenses out. The air from a air conditioner is very dry due to this.

Leave a Reply

Your email address will not be published. Required fields are marked *