Control of the 24Vac electrical water valves is by X-10. Since this communication system is not 100% reliable, I was leary at first of this approach, but it has proven to be trouble free. There is also a safety feature built into the control system. If any of the water valves turns on for whatever reason, an X-10 command is sent. The central controller sees this command and checks if the turnon was due to a previously sent command. If it is not, all water valves are shut down.

The circuit to control the water valves is shown below, if there is any current flow, there will be a voltage drop across the diodes, and the X-10 module will be triggered to send a command.

Schematic of irrigation control circuit

The circuit works as follows. When the UM506 closes its contacts, there will be current flow thru the diodes, and a voltage of two diode drops will be developed, causing the PF284 to send out an X-10 command. The central computer sees this X-10 command and verifies that there was a valid command previously sent to turn on a zone. I have a counter implemented in software that keeps track of how many false triggers there have been. Before I added my Whole House X-10 Blocker, I used to have one or two a month. After that, I have not had any false triggers. The initial false triggers were the reason I added the protection circuit. It seems that they are just an unnecessary precaution now.

Four zones of irrigation are so controlled, they are turned on automatically once or twice a week in the very early hours of the morning. My plans are to add a fifth zone to slow drip irrigate all the beds in the back of my house someday.

Long Term Update
Spring 2000 - I added two more zones to the irrigation system, for a total of six.  The above mechanism for detecting false valve actuation has been effective enough such that there has never been a case where there was a valve turned on unintentionally for more than a few seconds.