Basement Temperature Monitoring #2
Now that I have done monitoring of upstairs and downstairs temperatures, decided to look at how often furnace came on and how this affected downstairs temperature. The first thing that became clear is that I don't have enough temperature monitors to do this as I should have put another monitor outside the house to monitor external temperature. An estimate of this temperature can be obtained from the hourly weather records at airport but this is often quite a bit different than the temperature outside my house given microclimates that exist in Kamloops (more USB temperature monitors on order).
This was the setup for my furnace vent monitoring:
Crappy photo taken with my cell phone camera which had the only advantage that I could send photo to my laptop via Bluetooth as I was too lazy to take out my camera SD card. This is USBMonitor3 which determines timing of furnace going on/off. This heat vent is about 20' from USBMonitor2 which is used to measure basement temperature and is away from my computers (which can be viewed as a constant heat source as they're crunching seti@home data when I'm not using them).
This is what the furnace vent temperature data looks like:
Furnace is controlled by thermostat which is upstairs, and in the previous blog entry, it was noted that this temperature is maintained between 72 and 73 F. The USB temperature monitor has a fairly long time constant and thus the duration of time that furnace is on is proportional to the height of the temperature spike. I haven't looked at total time that furnace is on during day yet and this is something that will be looked at in next posting where I'll give data from my outside temperature monitor and furnace outlet temperatures.
Next graph is data from today where the effect of a warm afternoon is clearly evident. My jeep thermometer indicated that it got to 36 F this afternoon and one can clearly see that furnace is not on as often and on for shorter periods of time during afternoon:
Now, lets look at what the furnace data looks like in the frequency domain:
The multiplicity of peaks on the left isn't surprising considering that the data is from almost 6 days of monitoring with quite varying outside temperatures. Most common furnace cycle time is 1.53 hours. There are clear peaks at 12 and 24 hours indicating the effect of solar heating on the house-furnace system.
The basement temperature is shown below:
Basement temperature is maintained within a 3 F degree range although is clearly affected by outside temperatures. The coldest outside temperatures were at the start of the monitoring and furnace was on more and thus heated up the basement more. As the temperature increased outside, the basement temperature decreased. There's a clear circadian relationship between basement temperature and outside temperature: night temperatures are the lowest and basement temperatures are the highest during the night. This circadian rhythm is clearly seen when one does FFT of basement temperature record:
The 24 hour frequency predominates here with the 1.5 hour peak also being visible but much attenuated by the low pass filter comprised of the thermal mass of the basement air and contents.
What is nice about these “studies” is that all one has to do is just setup the monitors and leave them to collect data for either 11 days or until I get the urge to analyze the data. The next step would be to look at gas consumption each day and calculate how much gas is utilized by the furnace/minute of operation. With external temperature data, one could also find out the efficiency of the house insulation. The temperature in various rooms appears to be surprisingly constant probably because this house is so well insulated. The furnace efficiency is known and thus can find out what the required heat input is/day to keep the house at the desired temperature. Given the idiotic carbon taxes levied by the BC watermelons, it might be cheaper to utilize electric heaters to raise the temperature of the local area one is occupying rather than heat the whole house.
The reason that all temperatures are expressed in F degrees is that one has a finer gradation of temperatures than if one used C degrees. The USB temperature monitors record data in integral units and so F degrees are far preferable to C as each C degree corresponds to a 1.8 F degree step. Finding out how long furnace is on for each day requires writing some VB programs to measure the time difference between the first upward inflection of furnace vent temperature and the point where temperature starts to decrease. Fortunately the data (when examined at high resolution) is quite well behaved so the program would be trivial but not high on my list of priorities now.
This page was also a test of using Open Office Writer to create HTML documents as I'm still experimenting with various HTML editors. So far I like Thingamablog, for it's ease of use, but it's designed for blog page creation although one doesn't have to use the created files in a blog. Experiment with this next time I have a few moments free.
Last modified 23/2/2011 T:=00:25