The Central Heating Pi - Revisited!
Since building the Pi-based central heating controller I have improved it by adding temperature sensing. The controller now has access to sensors that record the temperature of the hot water tank and the main room. This allows me to now set a temperature target for these two which the controller will attempt to hit - in other words I now have a thermostat controlled heating system (like most central heating systems!).
The sensors I used were these:
http://shop.ciseco.co.uk/temperature-xrf-development-sensor-dallas-ds18b20/
These use an XRF radio module which allows me to communicate wirelessly with the Pi. When built the sensors look like this:
...and these can be placed anywhere in the house. The radio modules look like this:
The sensors wake up every five minutes and send out the current temperature reading. Each sensor has an individual two character 'name' to identify it which is sent along with the reading. The Pi has another XRF module wired into the serial line to receive the messages.
By only waking up every five minutes the sensors use very little power - they run off CR2025 button cells and have been running for about 6 months so far without a battery change needed.
For the tank temperature I mounted the thermistor on the end of a short piece of wire so that I could nestle the tip of it right up against the metal skin of the tank.
When the reading is received the controller writes the data to the MySQL database and I use Google Visualizations to show the values on dials on the heating page. Clicking on the dial takes you through to charts showing how the temperature has varied over the day/week/month:
![](data:image/png;base64,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)
This is particularly useful for checking if there is any hot water in the tank before taking a shower or bath. I can tell if there is going to be enough without fumbling around in the airing cupboard to feel the tank! It's also quite revealing to see how the tank cools over the day - I may experiment with extra insulation to see if I can improve it....
Here is the new improved controller that responds to temperature readings:
#! /usr/bin/env python
import wiringpi
import MySQLdb
import sys
from GPIOpins import GPIO_HWstate
from GPIOpins import GPIO_CHstate
from GPIOpins import GPIO_switchHW
from GPIOpins import GPIO_switchCH
from GPIOpins import GPIO_ledCHred
from GPIOpins import GPIO_ledHWred
from GPIOpins import GPIO_ledCHgreen
from GPIOpins import GPIO_ledHWgreen
from time import sleep
intervalTime=20
gpio = wiringpi.GPIO(wiringpi.GPIO.WPI_MODE_GPIO)
gpio.pinMode(GPIO_switchHW,gpio.OUTPUT)
gpio.pinMode(GPIO_switchHW,gpio.OUTPUT)
gpio.pinMode(GPIO_ledCHred,gpio.OUTPUT)
gpio.pinMode(GPIO_ledHWred,gpio.OUTPUT)
gpio.pinMode(GPIO_ledCHgreen,gpio.OUTPUT)
gpio.pinMode(GPIO_ledHWgreen,gpio.OUTPUT)
gpio.pinMode(GPIO_CHstate,gpio.INPUT)
gpio.pinMode(GPIO_HWstate,gpio.INPUT)
###############################################################################
def thermostatCheck(cursor, CHState, HWState, RoomTemp, HWTemp):
# CHState -the desired Central Heating State 'ON' or 'OFF'
# HWState -the desired Hot Water State 'ON' or 'OFF'
# RoomTemp, HWTemp - the current temperature of each
rtn=getThermostat(cursor)
RoomThermostat = rtn[0]
WaterThermostat = rtn[1]
ThermActive = rtn[2]
# RoomThermostat, WaterThermostat - the target temperature selections
# ThermActive - Y/N controls whether the thermostat numbers should be obeyed or not
# return: a list - the first element is the heating value, the second is the hot water value
CHoutState = CHState
HWoutState = HWState
# if either is above threshold then turn it off:
if ThermActive=='Y':
if CHState=="ON":
if (RoomTemp>RoomThermostat):
CHoutState='OFF'
if HWState=="ON":
if (HWTemp>WaterThermostat):
HWoutState='OFF'
# if we need the heating on then hot water must be on too (this is just the way my central heating works!):
if CHoutState=="ON":
HWoutState="ON"
rtn_list=[CHoutState, HWoutState]
return rtn_list
###############################################################################
def getTemperature(cursor, sensor):
temperature=0
# the temperature table stores all the temp readings from the wireless sensors. A seperate process listens for the readings and writes them to the DB. Only select from the last day of readings otherwise the query takes too long:
cursor.execute("select time, reading from temperature where sensor='"+sensor+"' and time>= DATE_SUB(NOW(), INTERVAL 1 DAY) order by time desc")
row = cursor.fetchone();
if row is not None:
datatime = row[0]
temperature = row[1]
tempdate = datatime.strftime( '%d %b, %H:%M' )
print sensor + "=" + str(temperature)
return temperature
###############################################################################
def getThermostat(cursor):
cursor.execute("select roomtemp, watertemp, active from thermostat")
roomtemp, watertemp, active = cursor.fetchone()
print "therm= "+ str(roomtemp) + " " + str(watertemp) + " " + active
rtn_list=[roomtemp, watertemp, active]
return rtn_list
###############################################################################
def log(logmessage):
# Open database connection
logdb = MySQLdb.connect("localhost","user","******","heating" )
# prepare a cursor object using cursor() method
logcursor = logdb.cursor()
# Prepare SQL query to INSERT a record into the database.
sql = """INSERT INTO log(source,
message)
VALUES ('controller','"""+logmessage+"""')"""
try:
logcursor.execute(sql)
logdb.commit()
except:
logdb.rollback()
# disconnect from server
logdb.close()
###############################################################################
def switch(switchpin, statepin, desiredstate):
# first check the state of the channel (HW or CH)
currState = gpio.digitalRead(statepin)
# '1' indicates off, '0' indicates on!
if (currState==1):
currState = "OFF"
else:
currState = "ON"
print "switch pin ", switchpin, " statepin ", statepin, " desired state ", desiredstate, " curr state is ", currState
# check if there is anything needed to do:
if (desiredstate!=currState):
print "switching"
gpio.digitalWrite(switchpin,gpio.HIGH)
sleep(0.25)
gpio.digitalWrite(switchpin,gpio.LOW)
return
###############################################################################
print "controller starting"
log("controller starting")
firsttime=True
# open the database
connection = MySQLdb.connect(host="localhost", user="user", passwd="******", db="heating")
cursor = connection.cursor ()
cursor.execute("select ucase(dayname(curdate()))")
row = cursor.fetchone()
last_day = row[0]
cursor.execute("select curtime()")
row = cursor.fetchone()
last_time=row[0]
# set the heating to the state shown in the DB. This ensures that a re-boot does
# not affect the state of the heating. If we don't do this the GPIO pins tend to
# throw random values on start-up which can switch the HW/CH on or off. This piece
# was added to try and make sure that a re-boot doesn't make any change to the state.
cursor.execute("select heating, hotwater from current_state")
data = cursor.fetchall()
for row in data:
currDBHeating = row[0]
currDBHotWater = row[1]
print "Reseting state to HW=", currDBHotWater, " CH=", currDBHeating
switch(GPIO_switchHW, GPIO_HWstate, currDBHotWater)
switch(GPIO_switchCH, GPIO_CHstate, currDBHeating)
cursor.close()
connection.close()
print "start is ", last_day, " time=", last_time
sleep(intervalTime)
while True:
print "loop start==========================="
# open the database
connection = MySQLdb.connect(host="localhost", user="user", passwd="******", db="heating")
cursor = connection.cursor ()
# get the current scheduled state:
cursor.execute("select heating, hotwater from current_state")
data = cursor.fetchall()
for row in data:
currStateHeating = row[0]
currStateHotWater = row[1]
print "current: H=" + currStateHeating + " HW=" + currStateHotWater
# get the 'actual' state - that is, the state that the controller actually demanded:
cursor.execute("select heating, hotwater from actual_state")
data = cursor.fetchall()
for row in data:
actualStateHeating = row[0]
actualStateHotWater = row[1]
print "actual : H=" + actualStateHeating + " HW=" + actualStateHotWater
# get the state of the heating/HW from the GPIO pins:
gpioHW=gpio.digitalRead(GPIO_HWstate)
gpioCH=gpio.digitalRead(GPIO_CHstate)
# translate the detected state into ON/OFF values and set the red leds to show the current state
if (gpioHW==1):
detectedStateHotWater="OFF"
else:
detectedStateHotWater="ON"
if (gpioCH==1):
detectedStateHeating="OFF"
else:
detectedStateHeating="ON"
print "detected: H=" + detectedStateHeating + " HW=" + detectedStateHotWater
# if the detected state does not match the actual state then someone has pressed a button on the heating
# controller... In which case treat this just as though there has been an Override row found:
if detectedStateHeating!=actualStateHeating:
logline = "Heating button pressed det=" + detectedStateHeating + " act=" + actualStateHeating
log(logline)
currStateHeating=detectedStateHeating
if detectedStateHotWater!=actualStateHotWater:
logline = "Hot Water button pressed det=" + detectedStateHotWater + " act=" + actualStateHotWater
log(logline)
currStateHotWater=detectedStateHotWater
# get today's day
cursor.execute("select ucase(dayname(curdate()))")
row = cursor.fetchone()
day = row[0]
# get the current time
cursor.execute("select curtime()")
row = cursor.fetchone()
time=row[0]
#print "now it is ", day, " time=", time
# if the day has changed then we will just run a query from the last point in time up to midnight...
if last_day != day:
print "day change"
# save the name of 'today' so that we can put it back later:
tomorrow=day
# now pretend that the current time is 1 second to midnight 'yesterday':
day=last_day
time='23:59:59'
daychange=True
else:
daychange=False
# first see if the mode is 'AUTO' or 'MANUAL'
query = "select mode from current_state"
cursor.execute(query)
data = cursor.fetchall()
for row in data:
mode = row[0]
# look for any scheduled event that has occurred since the last run
query = "select * from schedule where day='"+day+"' and time>'" + str(last_time) + "' and time<='" + str(time) + "' order by time"
cursor.execute(query)
data = cursor.fetchall()
for row in data:
# found an event....
row_id = row[0]
day = row[1]
time = row[2]
hot_water = row[3]
heating = row[4]
print "id=", row_id, " day=", day, " time=", time, " hot_water=", hot_water, " heating=", heating
if mode=='AUTO':
print "executing at ", day, time
logline="scheduled event id="+str(row_id)+" day="+day+" time="+str(time)+" hot_water="+hot_water+" heating="+heating
log(logline)
currStateHeating=heating
currStateHotWater=hot_water
# if this was a day change then reset the 'last' point in time to midnight:
if daychange :
last_day=tomorrow
last_time='00:00:00'
else:
# record the last point in time ready for the next interval
last_time=time
last_day=day
# check for overrides:
query = "select * from override where status='P' order by time"
cursor.execute(query)
data = cursor.fetchall()
for row in data:
print "execute override"
logline="override hot_water="+row[1]+" heating="+row[0]
log(logline)
if (row[0]=="ON" or row[0]=="OFF"):
currStateHeating=row[0]
if (row[1]=="ON" or row[1]=="OFF"):
currStateHotWater=row[1]
# reset ALL overrides to 'C'omplete
cursor.execute("update override set status='C'")
connection.commit()
# turn off any green 'button' lights:
gpio.digitalWrite(GPIO_ledCHgreen,gpio.LOW)
gpio.digitalWrite(GPIO_ledHWgreen,gpio.LOW)
# finally check the temperature/thermostat settings:
roomTemp = getTemperature(cursor,"T1")
waterTemp = getTemperature(cursor,"T2")
print "b4 therm: H=" + actualStateHeating + " HW=" + actualStateHotWater
rtn = thermostatCheck(cursor, currStateHeating, currStateHotWater, roomTemp, waterTemp)
actualStateHeating = rtn[0]
actualStateHotWater = rtn[1]
print "after therm: H=" + actualStateHeating + " HW=" + actualStateHotWater
update_query= "update current_state set heating='"+ currStateHeating +"', hotwater='"+ currStateHotWater + "'"
print update_query
cursor.execute(update_query)
update_query= "update actual_state set heating='"+ actualStateHeating +"', hotwater='"+ actualStateHotWater + "'"
print update_query
cursor.execute(update_query)
connection.commit()
# Set the red LEDs to show the scheduled state:
if (currStateHeating=="ON"):
gpio.digitalWrite(GPIO_ledCHred,gpio.HIGH)
else:
gpio.digitalWrite(GPIO_ledCHred,gpio.LOW)
if (currStateHotWater=="ON"):
gpio.digitalWrite(GPIO_ledHWred,gpio.HIGH)
else:
gpio.digitalWrite(GPIO_ledHWred,gpio.LOW)
# and last of all, switch the heating/hot water on or off if needed:
switch(GPIO_switchHW, GPIO_HWstate, actualStateHotWater)
switch(GPIO_switchCH, GPIO_CHstate, actualStateHeating)
cursor.close()
connection.close()
sleep(intervalTime)
and here is the code that records the temperatures to the DB:
#! /usr/bin/env python
#
import serial
import MySQLdb
# import time functions
import datetime
from datetime import date
import time
#
# SETTINGS
#
# Default settings for program; port, baud rate, temperature threshold, number of readings to store
# set up serial port for temperature readings
DEVICE = '/dev/ttyAMA0'
BAUD = 9600
# END OF SETTINGS
#
#
#
# set battery level string to "????"
battlevel = "????"
# end of variables set up
def writeTemp(sensor, temp):
now = datetime.datetime.now()
# Open database connection
db = MySQLdb.connect("localhost","user","******","heating" )
# prepare a cursor for storing the temperature reading
cursor = db.cursor()
# Prepare SQL query to INSERT a record into the database.
sql = """INSERT INTO temperature(sensor,
reading,
time)
VALUES ('"""+sensor+"""',"""+str(temp)+""",'"""+str(now)+"""')"""
try:
cursor.execute(sql)
db.commit()
except:
db.rollback()
# prepare a cursor for storing just the current temperature reading
cursor3 = db.cursor()
# Prepare SQL query to UPDATE a record into the database.
sql_update = """UPDATE curr_temperature set reading = '"""+str(temp)+"""',
time = '"""+str(now)+"""'
where sensor='"""+sensor+"""'"""
print sql_update
try:
cursor3.execute(sql_update)
db.commit()
except:
db.rollback()
# prepare a cursor object using cursor() method
cursor2 = db.cursor()
# Prepare SQL query to INSERT a record into the database.
sql = """UPDATE sensors set last_reading_time='"""+str(now)+"""' where ident='"""+sensor+"""'"""
try:
cursor2.execute(sql)
db.commit()
except:
print "update failed " + sql
db.rollback()
cursor2.close()
# disconnect from server
db.close()#
############################################################################################
def writeBattery(sensor, voltage):
now = datetime.datetime.now()
# Open database connection
db = MySQLdb.connect("localhost","user","******","heating" )
# prepare a cursor object using cursor() method
cursor = db.cursor()
# Prepare SQL query to INSERT a record into the database.
sql = """INSERT INTO battery(sensor,
time,
voltage)
VALUES ('"""+sensor+"""',"""+str(now)+""",'"""+str(voltage)+"""')"""
try:
cursor.execute(sql)
db.commit()
except:
db.rollback()
# disconnect from server
db.close()#
############################################################################################
#
print "Opening connection and waiting for response..."
#
ser = serial.Serial(DEVICE, BAUD)
print "Startup complete"
print " "
# read the time
now = datetime.datetime.now()
msg = 'monitor initialised : ' + now.strftime("%H:%M %m-%d-%Y")
print msg
#
# Start infinite while loop to listen to XRF module
while 1 :
# All XRF module read and write commands should have 12 characters and begin with the letter "a"
# Wait for message, the 1 second pause seems to improve the reading when several messages
# are arriving in sequence, such as: a--TMP22.12-a--AWAKE----a--BATT2.74-a--SLEEPINGtime.
time.sleep(1)
#if ser.inWaiting()>0:
# print ser.inWaiting()
# llapMsg = ser.read(ser.inWaiting())
# print llapMsg
if ser.inWaiting() >= 12 :
llapMsg = ser.read(12)
# display packet, helps to troubleshoot any errors
now = datetime.datetime.now()
now.strftime("%H:%M %m-%d-%Y")
print 'Received '+ llapMsg + ' at ' + str(now)
if 'a' == llapMsg[0] :
#llap msg detected
#print "(" + llapMsg[1:3] +")"
sensorName=llapMsg[1:3]
#
#
# Check for TMP reading or battery packet, ignore anything else.
#
# Is it a battery reading?
if 'BATT' in llapMsg :
# Battery reading sent
print "Battery level is " + llapMsg[7:10] + "V"
# Save this value for later.
battlevel = llapMsg[7:10]
writeBattery(sensorName, battlevel)
#
# Is it an temp reading?
if 'TMP' in llapMsg :
# reading sent
#
temp = llapMsg[7:12]
print "temp=" + temp
writeTemp(sensorName, temp)
# temp is a 1 element array, hence the "[" "]"
#except ValueError:
# if float operation fails, skip bad reading
# print "bad reading"
#
ser.flushInput()
#
# for want of a better phrase,- endwhile
# end of program
Please note - I haven't got the battery level working yet. Not sure why!