Dick
Cappels' project pages
http://www.projects.cappels.org
Return to
HOME
(more projects)
Garden Sprinkler
and Drip Timer Project
A simple
replacement timer circuit for the Rain Bird E3 Irrigation System
Controller
that turns on my drip irrigation system for 34
minutes every day. Other daily watering periods can be selected
by changing a tap on the counter chain.
The timer circuit uses CD4000
series CMOS logic -no firmware at all.
The circuit board was cut to
fit where the original circuit
board in the Rain Bird Timer had been.
The barrier terminal strip was
salvaged from the original
circuit board.
Find updates at
www.projects.cappels.org
Introduction
and Overview
When the car pulled in to the driveway of my Mesa, Arizona home, after
a 15 week absence, I immediately realized that there was something
wrong with the plants in the front of the house. The bushes were wilted
and my rose bush looked as though it had been in an oven. The front
yard is a small expanse of crushed rock, which hosts 16 plants,
including bushes, shrubs, clumps of pampa grass, and one lemon tree.
The
are all watered by a drip irrigation system, which was controlled by a
Rain Bird E3 Irrigation System Controller. At least until its
sixth year of operation; After about 2050 daily cycles of blazing
desert sun and cool desert nights finally took its toll. It had
literally died. Having designed consumer electronics, I realize that a
failure after 49,000 hours of continuous operation in a hostile
environment is understandable. The timer was broken. The question is:
What do to next?
My choice was obvious:
Plan A: Drive over to the garden supply center and buy a new watering
timer, then remove the old one, and install the new one. Quite
likely, I would have to find some paint to match the outside of the
house where the original controller is mounted. The 120 VAC input to
the
controller was hard wired into the circuit breaker box on the side of
the house - something else I didn't want to mess with right now. If I
ended up buying an indoor timer, I would have to drill a hole so as to
get the AC power and the wire to the valve solenoid into the garage, or
Plan B: Build a replacement timer board that fits in place of the worn
out timer board.
For somebody who is handy with a soldering iron and wire cutters, and
not so handy drilling holes in concrete and painting walls, and who was
only going to be there for about
two weeks and didn't want to go away with the job half done,
or done poorly, Plan B
sounded like the more comfortable of the two.
Since I was going to design the timer, I could make it do just as I
wanted. The Rain Bird E3 was a beautiful little controller. It could be
programmed to operate up to four valves, it had twelve
operating modes, and plenty of features like 1 through 6 day watering
cycles, odd
day watering cycles and even day watering cycles. All that flexibility
came with some operational complexity. Whenever I had a
problem with the drip system, I would spend the first five minutes
walking between the controller, adjusting something, and then walking
back out to the garden to try and tell if the water was on or off. As
long as I have lived in this house, the timer has always been set to
water for half an hour a day. Every day in every season. Sometimes the
plants got more water than they needed, and sometimes they were a
little thirty. But this is a place that has seasons, and the plants
"understand" seasons and the did well. I decided to build a
timer that would turn on the valve to water the plants for
half an hour ever day. All that programmable flexibility is wasted on
my garden, and indeed, its an encumberance.
I was immediately reminded of another project with similar
requirements, my
Door Chime
Privacy Sentry, which is a 24 hour time that keeps the doorbell off
for 10 hours each night. The circuit was very simple, using only one
CMOS micro controller for the logic. I could literally build
another
Door Chime Privacy
Sentry and just modify the daily time that the sprinkler went on,
and invert the drive signal to turn the valve on for the timed
duration instead of off. It would be a snap.
But there was a hitch. I would only be in Mesa for a couple of
week,
and didn't figure that I
would even touch my soldering iron, let alone build anything, so even
while knowing in the back of my mind that somehow I would regret it, I
decided to travel light and not bring my box of micro controller
programming tools. Thus, I had no programmer, no serial port adapter to
plug it into, and none of my improvised cables and gadgets that would
have made
programming and debugging easier.
I didn't want to loose a week of precious time waiting for tools or
material to come from a distributor, so the main question turned from
"What's the most efficient way to do this?" to "What can I do with the
tools and materials I have on hand?" Fortunately, I am a pack rat, and
I still have some old CD4000 series CMOS logic, including some four
input
CMOS gates that I bought in 1973 to use as address decoders in my
as-yet unfinished Intel 8008 computer project (That's
8008, not 8080).
The CMOS timer circuit was
designed for this project. The Power supply
and
valve solenoid power circuits
were adapted from the Door Chime Privacy Sentry
project.
The Circuit
CMOS
24.055 Hour Timer
The resulting timer circuit is made from
a CD4060, includes an
oscillator and a 14 stage binary counter, two CD4040's, which are 12
stage binary counters, a CD4012 Dual Nand gate and a CD4013 Dual D
Latch.
The CD4060 is connected to a 32768 Hz ECS 3X8 tuning form
crystal. The
10 Meg Ohm resistor between CD4060 pins 10 and 11 are to bias the
internal inverter into its analog region, and the 330 K resistor in
series with the crystal is to limit the power to the crystal. This
setup is directly from the ECS data sheet. The output of the 14th flip
flop in the CD4060 appears on pin 3, and it is a 2 Hz square wave. A
2N4401 buffer drives a green LED with this square wave. When I see it
blinking through the plastic window of the Irrigation System Controller
housing, I know that power is on, and that at least the oscillator and
first 14 counter stages are working.
The CD4040's that follow the CD40406 divide the 2 Hz pulses to get the
24 hour cycle and the 34 minute watering period. When the counter
counts up to 24.0355 hours, as decoded by the CD4012, both of the
CD4040's are reset and the output of the CD4013 (pin 1) is set high to
turn on the valve solenoid. To get a wide enough pulse to assure that
all of the flip-flops inside the CD4040's are reset, the reset pulse
from CD4012 pin 13 is passed through a low pass filter made of a 10k
resistor and a 6 pf capacitor.
A momentary push button, labeled "Resync" is connected to the output of
the 10k and 6 pf low pass filter. When the button is pressed, it pulls
the reset inputs of the CD4040's and the clock input of the CD4013 high
to start a 24 hour timing cycle. The additional l0k resistor in series
with the output of the 10k and 6pf low pass filter it to further limit
current from the pushbutton to the input of the CD4012, which normally
spends all but a few microseconds each day sitting at ground. I use the
Resync button to set the time of day that the watering cycle starts.
The CD4013 latch, the output of which (pin 1) goes high to turn on the
valve , is reset 34 minutes after the CD4040's are reset, thus
turning off the valve . The pulse to turn off the valve 34
minutes after the watering cycle starts, appears on Q1 of
the second CD4040. If I wanted to the watering period to be 68.264
minutes, I would use Q2 (pin 7) of the second CD4040. If I wanted 16
minutes, I would have used Q12 of the first CD4040. Moving the
connection to the reset input of the CD4013 (pin 4) one flip-flop to
the left in the CD4040 counter chain cuts the watering period in half.
Moving the connection to the right one flip-flop doubles the watering
period.
Valve Solenoid Power Control
A full-wave bridge rectifier is placed effectively in series with the
24 VAC from the power transformer and the watering valve. The AC
signal passes through the AC terminals of the bridge, while the
negative terminal is connected to chassis ground and the positive
terminal is connected to the drain of a BUZ-73 MOSFET. While the BUZ-73
is off, current does not pass though the valve, and the water is
off. When the output of the CD4013 (pin 1) goes high, turning on the
BUZ-73, current also passes through the AC terminals of the diode
bridge and through the valve solenoid.
The gate of the BUZ73 is connected to the output of the CD4013 (pin 1)
through two switches, which allow me set valve solenoid OFF, ON, or
connected to the daily watering cycle timer. This last position is the
normal operating position. The OFF and ON positions are for maintenance
and troubleshooting. The output of the two switches also connects to a
2N4401 buffer, a dropping resistor, and a yellow LED. The yellow LED is
on whenever the valve solenoid is on. A 100k resistor connects the gate
of the BUZ-73 to chassis ground to keep the gate from floating in case
one of the switches does not get slid all the way to one position or
another, on in case a switch becomes intermittent. T
There are two series connected 14 VAC metal oxide varistors across the
AC terminals of the diode bridge and another pair of series connected
14 VAC metal oxide varistors across the drain and source of the BUZ-73
MOSFET. The purpose of these metal oxide varistors is to protect the
diode bridge and the BUZ-73 MOSFET from voltage surges that might
appear in either the AC line or the wiring to the valve solenoid. We
get a lot of lightning here.
A 1k resistor between the gate of the BUZ-73 MOSFET and the rest of the
circuit limits current that might capacitvely couple from a fast rising
voltage spike on the drain of the FET to the gate. A zener diode from
the gate to ground will limit the gate voltage in the case of such a
voltage spike. The .015 mircofarad mylar film capacitor across the
drain and source of the BUZ-73 MOSFET is there to reduce the amplitude
of induced spikes.
5.1 Volt Zener Power Supply
One end of the 24VAC power transformer secondary connects to chassis
ground through one of the diodes in the full wave diode bridge when
that end of the transformer secondary swings through the negative half
of the power supply line cycle. At the same time, the other end of the
secondary swings positive. The voltage on this second end of the
transformer secondary is rectified by a 1N4007, and passed through a
728 Ohm 2 Watt resistor to supply current to the sprinkler timer
circuit. A zener diode limits the voltage at this point to 5.1 volts.
Three 220 microfarad capacitors mounted around the circuit board
provide plenty of filtration so that there is negligible ripple on the
5.1 volt power supply.
The new circuit board mounts in
the original box in place
of the original board. All of the original wiring is used.
No new holes were drilled, no new wires were run, and nothing
had to be painted.
Construction
and Installation
The circuit board was cut from a
ptototyping board that had ground and power busses, and rows of pads
for dual in line packages etched on the back (I had this board left
over from the 1980's, when we used to use them for small wire-wraping
projects). The dimensions were determined by tracing around the outside
of the original circuit board from the Rain Bird E3 controller, and
later I filed some small notches near the plastic hooks that grip the
board in the enclosure, so as to get the fit for the hooks just right.
I also traced the open area for the barrier screw strip and the slide
switches and push button so that I could be sure of mounting them in
the right place.
The barrier screw strip itself was salvaged from the original Rain Bird
controller board, and it took quite a bit of fluxed braid to get the
barrier strip free from the original board.
All of the components were soldered
directly to the pads on the back of
the circuit board and wired point-to-point. There are no sockets in
this assembly because with the temperature cycling, sockets would be a
reliability worry.
The full wave diode bridge assembly is a 4 Amp 600 volt bridge from my
junk box. My valve solenoid only draws 188 milliamps, so I am sure I
could have gotten by with four 1N4004's or something similar, but I
thought it wise to be on the safe side. The bridge assembly does get a
little warm during operation - 200 milliamps x 2 volts is almost 1/2
watt, but it should not be a problem given the side of the bridge
assembly and that its not in contact with the fiberglass circuit board.
The BUZ-73 MOSFET is my current favorite 7 amp MOSFET. With 5 volts of
drive, its drain-to-source voltage is low enough that I do not have to
worry about a heat sink for it. It does not get as warm as the diode
bridge assembly.
The odd valed 724 Ohm 2 watt resistors was made by placing eight 91 Ohm
1/4 watt resistors in series. I did this because I don't have any 750
Ohm 2 watt resistors.
During assembly, I used a transformer isolated, grounded soldering
iron, and I kept a piece of #30 wire wrapped around the leads of the
BUZ-73 until the circuit board was complete. I also was careful about
keeping me and my tools at ground level during assembly. A grounding
wrist strap would have been a good idea, but I don't have one yet, and
managed to get by.
The new circuit board snapped into the original box, and since the
barrier strip is in the same location within the box as it was when it
was on the original Rain Bird E3 circuit board (Assembly No. 633754, by
the way,) the wires from 24 VAC transformer, to the valve solenoid, and
the earth ground wire were right where they were supposed to be, so no
cutting or stripping was required; just a screwdriver to tighten the
screws.
Please note that the earth ground wire connects to the sprinkler wiring
as shown on the schematic, not to the chassis ground. The chassis
ground, to which the source of the BUZ-73 MOSFET and the VSS
connections of all of the CMOS chips is connected, floats from earth
ground.
I don't want you to think that I just built up the board, stuck it into
the old controller box and it worked. The first time I put it in, the
valve solenoid would not turn off, so I took the controller back to the
workbench and hooked up a 12 VAC transformer to troubleshoot it and
discovered that in a feeble moment (its hot in the garage in Mesa
Arizona at noon in August!) I had run a jumper from the earth ground to
the chassis ground, thereby shorting out one of the diodes in the diode
bridge. That was quickly fixed by removing the jumper. After that, I
found the water cycle to be only 17 minutes instead of the intended 33
minutes, and that, too, was quickly remedied, this time by moving a
single wire.
At this moment, the watering timer is out there in the
sunshine, on the side of the house, counting
2 Hz ticks, and the plants are waiting to feel the coolness of the
drippers on this scorching desert summer afternoon.
Note -four months after installation: I have returned to the house, and
the timer is still working as designed. The plants are fine, except for
a little frostbite on one bush, and another bush that was crushed by
somebody driving over it.
First posted in August, 2006. Updated
December, 2006.
You can send email to me at
projects(at)cappels.org. Replace
"(at)" with "@" before mailing.
Use of
information
presented on this page is for personal, nonprofit educational and
noncommercial
use only. This material (including object files) is copyrighted by
Richard
Cappels and may not be republished or used directly for commercial
purposes.
For commercial license, click
here.
Liability Disclaimer
and intellectual property notice
(Summary: No warranties, use these pages at your
own risk. You may use the information provided here for personal and
educational purposes but you may not republish or use this information
for any commercial purpose without explicit permission.) I neither
express nor imply any warranty for the quality, fitness for any
particular purpose or user, or freedom from patents or other
restrictions on the rights of use of any
software, firmware, hardware, design, service,information, or advice
provided,
mentioned,or made reference to in these pages. By utilizing or relying
on software, firmware, hardware, design, service,information, or advice
provided, mentioned, or made reference to in these pages, the user
takes responsibility to assume all risk and associated with said
activity and hold Richard Cappels harmless in the event of any loss or
expense associated with said activity. The contents of this web site,
unless otherwise noted, is copyrighted by Richard
Cappels. Use of information presented on this site for personal,
nonprofit
educational and noncommercial use is encouraged, but unless explicitly
stated
with respect to particular material, the material itself may not be
republished
or used directly for commercial purposes. For the purposes of this
notice,
copying binary data resulting from program files, including assembly
source
code and object (hex) files into semiconductor memories for personal,
nonprofit
educational or other noncommercial use is not considered republishing.
Entities
desiring to use any material published in this pages for commercial
purposes
should contact the respective copyright holder(s).