DIY Zoning: Visualization
Under Construction, Really
All the stuff below pertains to old as Moses 0.1p7dev3 release, available here. It can't be recommended - though amazingly stable (people have used it for years), it is very old, and making it work on modern distributions is somewhat difficult.
Current code base is under heavy development, and DZ Blog contains a series of articles regarding build and configuration. When the dust settles, those articles will be sorted out and moved here.
DZ3 Quickstart is the first article in the series.
Using the temperature/humidity/pressure graphing software
As of release 0.1p5, the DAC is completely plug-and-play. Once DAC starts, it will create an RRD database corresponding to the network configuration (this may take a couple of minutes).
There are two options to create the graphs: use RRDTool, or use JRobin. RRDTool was the only option available until release 0.1p6, and JRobin is a default starting with 0.1p6 release.
Run ./configure --enable-rrdtool to use RRDTool, and ./configure --enable-JRobin to use JRobin.
Using RRDTool
Use /usr/bin/dz_graph_temperature script (provided you installed DAC as an RPM, otherwise the location will be ${installation-prefix}/bin/dz_graph_temperature) to create the graps. It is recommended to add this script to a cron job.
Using JRobin
This option is available starting with release 0.1p6.
Graphs are rendered about every 5 minutes.
Accessing the graphs
Regardless of the tool used, the graphs will be generated in /usr/share/dz/var/html (or in ${installation-prefix}/var/html).
If you have an Apache installation at the moment you install dz-dac RPM, the RPM post-install script will create a symlink so you can see the resulting temperature graphs at http://localhost/dz/ (if your Apache installation is standard enough, that is).
Using third-party software for visualization
DZ intentionally provides a limited set of visualization tools. Amount of data it provides is increasing day by day, and in about March 2005 it was eventually realized that data visualization subset is perfectly illustrated by one of Murphy's Laws: "Inside of every large problem there is a smaller one, struggling to get out". Therefore, the decision was made to freeze the existing visualization functionality, use it as example only, and recommend users to accomodate the visualization solution of their choice. It is possible that at some time visualization plugins created specifically for DZ will appear, but it hasn't happened yet.
Analyzing the data
When you have at least a month worth of graphs, you will easily see the patterns. Then, you will know exactly what needs fixing, and what was perceived, but not real.
As an added bonus, you will be able to detect the problems you never thought existed. Or the problems that escaped your attention just because you weren't there. As an example, check out the Funny A/C behavior thread from HVAC-Talk (archived on DIY Zoning website, original expired).
Next Step
Upon analysis, it may turn out that you should start with fixing the problems that you already have, not come up with something new. In particular, these problems may turn out to be:
- Inadequate unit. Not only may it be too small, but it may be too large. Example: oversized AC in a climate like Florida or Georgia will not remove enough humidity. Oversized furnace in Minnesota will not allow to achieve proper heat distribution, and some parts of the house will be too cold, though the furnace will cycle more frequently than it is required {FIXME: discuss the ways to alleviate problems related to oversized units (example: changing the coil)};
- Undersized ductwork;
- Insufficient air return (this seems to be the most common problem for Arizona) {discuss the ways to fix this};
- Improperly charged unit. Unless there's a TXV {link to FAQ}, different charge will be required for different ambient temperature range.
If you decided to proceed with DZ installation, it's time to install the Dampers