home | history | downloads | supported devices | comparisons | special sensors | enersure | graphs | changelog | to-do | links | support | book | hosting | contact

Temperature, Power & Environmental Monitoring

Now that many people have broadband, even if you do not have a static IP address (with a little free help from dyndns.org), it is easy to have your house's status available to you when away from home. Great for checking that the boiler is working properly in winter when pipes might freeze, or for ensuring that your refrigerator is keeping the beer cold. My system supports an ever-widening variety of sensors and data collectors, can alert you to "unusual" conditions (via email, pager, or syslog), and can even save you money! Find out how often your furnace or A/C is running, find out is the basement is flooding, if your computers are running hot, etc...

The software that logs the data and generates these graphs is free (Unix, Linux, Macintosh and Windows). Click on any of the graphs for an interactive view of the temperature, humidity, wind, power consumption, etc. in and around my house in Pittsburgh PA
This week
This month

Locations of visitors to this page

It's all free and written in Perl for Unix, Linux, Macintosh and Windows!


I will try to fix any bugs that are reported to me. I am also open to the notion of a software support contract, but will strive to keep the software free and open-source (although there may be a "pro" version of the code in the future). If you have feature requests, I'm also happy to consider those!

The History of thermd

In 2001, I bought an inexpensive serial interface temperature sensor kit (about US$30) from QKits and after I built the kit, I also purchased an extra 3 Dallas Semiconductor sensors (another US$18). It was a very easy kit to build (it doesn't even need a power supply, it derives power from the RS-232 line it talks on), and the QKits folks are pleasant people to deal with. Then I wrote a nice Perl-based thermometer logging daemon, log grapher/dumper, and CGI script (one program does all three functions!). Over the years, I have continually improved the logging script, and have made it freely available to the public. This latest version of the Thermometer Daemon supports multiple data collection devices.

I have a multi-axis feature and price comparison of all the devices that thermd supports.

Detecting Unusual Conditions with thermd

November 2007

Take a look at this graph. Notice how my 1st floor (cold air return) is very hot for part of of the day? It had me worried for a while... At first I thought that the cat sleeping on the air vent was diverting warm air around the corner to the cold air return, but when I went downstairs, there was no cat and warm air was gently blowing out of the cold air return. Hmmm...

I have twinned high-efficiency furnaces (a pair of 100,000 BTU furnaces tied together, both blow into the same plenum). One was having trouble starting (you can see that here, where sometimes the usage has a knee in the spike, and sometimes the usage is lower before I fixed the problem than after, but I only noticed that after I found the main problem). Each furnace has it's own cold air return, but both blow into the same plenum. If the main fan on one furnace doesn't start, there is blowback up it's cold air return from the other furnace's fan. That is what I was seeing – hot air coming out of the cold air return. So why wasn't the furnace starting?

To be safe, the furnace has a series of safeties. First, the combustion fan starts. If the prover circuit shows that air is flowing properly, then the pilot burner circuit starts. If the thermocouple shows heat, then the main burners start, and after the temperature rises, the main blower fan starts. Any single failure prevents the next system from starting.

When my gardener was here Tuesday afternoon (6 November 2007), he cleaned up the autumn leaves with his leaf blower, and some leaf fragments were blown into the air inlet pipe for the furnace. When the combustion fan started, it sucked in the leaves, which blocked the airflow, which in turn prevented the prover circuit from registering sufficient airflow, and so (sometimes, since the leaves moved around) one of the furnaces didn't start (or else it started late). And that accounted for the blowback.

Without monitoring, I would not have noticed this until the dead of winter, when the whole house was not heating properly. I would have wasted money on a half-functioning furnace starting and stopping, and I would have caused the working furnace to do double duty. With thermd, I noticed and fixed the problem in less than 24 hours.

home | history | downloads | supported devices | comparisons | special sensors | enersure | graphs | changelog | to-do | links | support | book | hosting | contact