Swing Low, Sweet Thermostat

by Andrew Rudin, Project Coordinator, The Interfaith Coalition on Energy, Philadelphia
Lower Winter Interior Temperatures Imitate Conditions in Other Seasons

Since it began ten years ago, the Interfaith Coalition on Energy (ICE) has been recommending that congregations set back the temperature as low as 45F inside their buildings when they are not in use in the heating season.

Some members of congregations believe that changing the temperature inside the house of worship can harm organs, pianos, woodwork, and many other artifacts as well as the fabric of the building itself.

This article summarizes the readings from a circular chart recorder in a typical church to record the actual swings in relative humidity and temperature for more thirteen months. Our hypothesis was that similar changes in temperature and humidity occur all year. The swings in temperature caused by a clock thermostat during the winter may be similar to the swings that occur during the spring, summer, and fall. When a congregation tries to eliminate these changes in temperature and humidity in the winter by maintaining warmer than necessary temperatures, they are wasting large sums of money, while drying the building fabric and contents. We wanted to prove the futility of minimizing the swings in the winter but allowing the same swings to occur during the heating season that occur the rest of the year.
Temple Lutheran Church

We chose the Temple Lutheran Church in Havertown, Pennsylvania because Pastor Reimet and the members of the congregation said they could tolerate having the chart recorder for a year, allowing ICE access to change the charts each month. The church is in fine architectural shape. Other than what you would expect from a building that was 71 years old, there is no evidence of any problems with either the interior or exterior parts of the building. Our recorder was installed on February 2, 1990 in the organ and choir loft at the rear of the sanctuary. The probe for the chart recorder was suspended over the balcony so that it could sense the condition of the moving air in the room. It was not near an outside wall. The sanctuary is used mostly for Sunday worship services and for unscheduled events during the week. The room is heated by a steam boiler. A steam valve is controlled by a clock thermostat near the front of the room. All in all, it seems to be typical of many churches in the Philadelphia area.

Interpretation of the Results

We collected the charts each month or so for a little less than 13 months. The data from the chart for each day was recorded in a computer spreadsheet—the highest and lowest temperature for each day and the highest and lowest relative humidity each day. A sample chart is shown on the next page. Several times, the blue pen that recorded changes in relative humidity did not write clearly. We left the relative humidity data blank for those days.

Table 1 shows the results. There are seven columns of figures for each month. The three columns on the right show the maximum highest and lowest temperature for each month and the difference between the high and low. Note that the least of these monthly temperature extremes is 18F and that the maximum is 35 F. The maximum daily temperature difference occurred in March (35 F), followed by April (34 F) February, 1991 (31 F). Maximum daily differences of 25 to 27 occurred in June, July, September, October, December and January— half the year.


Table 1.  Changes in Temperature and Relative Humidity


Average of Daily Swings

Maximum Swings Temperatures
  Temperature Humidity Temperature Humidity High Low Difference
Feb. 10 9% 23 18% 77 47 30
Mar. 9 7% 23 18% 85 50 35
Apr. 8 7% 16 14% 92 58 34
May 8 7% 14 14% 80 62 18
June 8 7% 14 18% 94 68 26
July 7 7% 12 15% 97 71 26
Aug. 7 8% 11 21% 91 69 22
Sept. 7 7% 12 19% 88 62 26
Oct. 8 9% 15 21% 84 58 26
Nov. 9 9% 22 24% 80 48 32
Dec. 9 11% 18 26% 76 51 25
Jan. 9 12% 22 29% 76 49 27
Feb. 10 9% 25 18% 77 46 31
Avg. 8 8% 17 20% 84 57 28


While the columns on the right show the highest and lowest temperature each month, the numbers in the pair of center columns shows the average of the maximum daily changes in both temperature and relative humidity for all the readings each month. The maximum difference in high and low temperature readings is greater in the winter, but the difference in relative humidity is greater in the summer, fall and early winter.

The pair of columns on the left shows the average of the daily differences between the maximum and minimum temperature and relative humidity for each month. This is the monthly average of the differences between the high and low readings of each day. These figures show that the swings in temperature and relative humidity are roughly the same all year, not just in the winter.

Our hypotheses are correct. The average difference between the high and low temperatures each month is 27F. If the clock thermostat can control the heating system to maintain 72F during occupied periods, then a 27F setback to 45F is typical of the changes in temperature that occur all year.

Most experts agree, however, that relative humidity has a greater effect on the contents of a building than does temperature. The maximum daily change in relative humidity was typically greater in the summer and fall. The maximum swings for the winter were roughly the same as those for the spring.

Most months, the average daily change in relative humidity was 7% to 9% each day. In January the average was 12%. In December the average was 11%. The average change in relative humidity was the same, however, in February, October and November (9%). The average change in temperature over each month was 7F to 9F. In both Februaries, the average change was only 3F greater than in the summer.

While it is true that the daily changes in temperature and relative humidity are slightly greater in the winter, the daily changes occur all year long. If the pipe organ, piano, woodwork, plaster, paint, and all the other contents of a building of a building can tolerate a 27-degree change in temperature in the spring summer and fall, then they can tolerate the same difference in the winter. If comfortable temperatures are 72F, then a lower temperature during vacant periods can be 45F, which is 27 degrees lower.

It is very expensive and non-productive to prevent natural changes in relative humidity and temperature by using heating or air conditioning when not necessary for occupants.


Comments are closed.