I've just finally bitten the bullet and installed a Honeywell CM67
Programmable Room Stat in my hallway. I bought the version with Optimum
Start capability - which purports to decide for itself when to switch on the
heating in order to achieve the target temperature by the programmed start
This seems to be working, but I'm curious to know what algorithm it uses in
order to decide when to turn on the heating. The literature doesn't make
this clear, so I wonder whether any of you actually know?
Several possibilities present themselves to me:
* it could assume a fixed (factory set) rate of rise of temperature, and
apply this to the current temperature in order to work out long long it will
take to get to the target temperature
* it could depend on one of the parameters - such as Proportional Band
Width - which can be set in the Installer Set-up Mode
* if it's *really* clever, it could calibrate itself by observing the actual
rate of temperature rise when the heating is on, and then apply this to its
Does anyone out there know what it actually does?
If you look at Adaptive Intelligent Recovery, it is basically doing
what you describe in your third point - i.e. if you turn that feature
on, the controller measures the rate of rise of temperature and uses
it to calculate the next heating period's start time also accounting
for the starting temperature. This is a relatively simple
controller, so it appears that it only does this session by session.
A really intelligent controller would store the temperature points by
time over more days because the rate of change of temperature is not
absolutely linear, although it is to a first approximation. Hence
the method used is a lot better than doing nothing at all.
All of that has to do with optimised starting.
The proportional band has to do with the behaviour of the controller
(it is a proportional plus integral controller)
Proportional control is that the heat contribution over a period of
time is adjusted to match the heat loss and maintain the temperature
accurately. There are controllers that will do this in a genuinely
analogue way by modulating the boiler output to match the requirement.
However, many boilers can only be controlled in an on/off fashion aand
indeed this is also true of most zone valves if you were using that as
a means of control. In order to control this type of device, the
controller does so by using time proportions. A simple example
might be that if you want 50% of the average output, the boiler is
turned on for 5 minutes and off for 5 minutes. For 80%, it would be
on for 8 and off for 2.
The CM67 does this but allows some additional settings.
- There is an cycle rate setting of 3,6,9,12 cycles per hour. If the
entity being controlled can react quickly without any disadvantages
(e.g. if you were controlling electric heating), then you could set 12
cycles per hour. At the other extreme, devices that show
inefficiency or might have other problems if cycled too frequently
would be run at three cycles per hour. In between you have typical
gas heating, which is run at 6 cycles/hour i.e. cycles last ten
minutes. So the behaviour would be as in my example above.
Note that the minimum on time is set to 1 minute, which in effect will
- Proportional bandwidth is the temperature range about the set point
where proportional control operates. Within this range, the
controller operates in this cycling mode. Below it, the output will
always be on and above it will always be off.
Adjustment for this will depend on the heating system and thermal
inertia of the house. Ideally what you want to have happen, is that
when starting from cold, if the current temperature is below the
proportional lower limit, then the optimisation should time operation
to fire the boiler at 100%. However, as the temperature rises into
the proportional band, the cycling control should begin to take
effect. If possible, you want to avoid the temperature overshooting
above the proportional band because a) you use more energy than needed
and b) the temperature may tend to oscillate for a long time before
settling. This is most likely to happen in well insulated properties
with oversized heating. It is really a trade off between achieving
the set point quickly and not overshooting.
On this controller, you can vary the proportional bandwidth to help
with that, in that it will enter cycling operation earlier on the rise
and begin backing off the average heat output of the system.
The proportional bandwidth may be taken into account in the optimised
start. That's not clear.
The best thing to do is to check the temperature rise behaviour by
recording temperature readings (say) every ten minutes during start up
for a few days. Plot a graph if you like.
You can then experiment with proportional bandwidth if needed.
To email, substitute .nospam with .gl
Many thanks for your very full asnwer, Andy - and for the links. My CM67 is
the hard-wired rather than RF model, and the leaflet which came with it is
slightly different. The "Adaptive Intelligent Recovery" section in the RF
model leaflet does say fairly explicitly that it uses the start-up
performance this time around to decide what to do next time. My leaflet is a
lot less explicit. Its equivalent paragraph is entitled "Optimisation
(Variable Start Time)" and says:
"The Programmable Thermostat will delay the start time to the last possible
moment, so that the desired temperature is reached by the start of the
program (sic) period. Simply set the times at which you would like to be
warm and the Programmable Thermostat will switch on when required up to a
maximum of 3 hours early."
I assume that the hard-wired version works in the same way as the RF version
in this respect, but it seems strange that they use different words to
As you say, it appears that it bases its temperature rise rate value on what
happened at the most recent startup period - rather than averaging it over a
number of periods.
Thankyou for explaining cycles per hour too. Presumably with the default 10
minute cycle (6 per hour) once it has done its thing within the current
cycle, it won't turn the boiler on again until the start of the next 10
minute cycle even if the house has cooled rapidly in the meantime because
(say) some fool has opened all the doors and windows? Presumably also, the
minimum ON time only applies once it has decided to turn the boiler on. That
is, it never runs the boiler for less than a minute. However, it may - and
often does - decide not to run it at all within a particular cycle - e.g.
when the actual temp is way above the target temp, as in over-night
I think that if you poke around on the sites a little, you will find
the CM61 or CM67. There are possibly a few differences in the RF
version to save battery power, but functionally from when I looked
them they are basically the same as the wired one.
It is a bit vague isn't it. Probably written by a technical author
who didn't understand the product.
That would be true unless the temperature has fallen so rapidly that
you reach the edge of the proportional band. At that point, the
controller will go to 100% on. Obviously this is another factor that
you could trade in deciding on the bandwidth setting. If the sudden
loss of heat situation happens a lot, then setting the bandwidth
narrower might be in order.
If you set it to the 10 minute cycle, yes. Depending on the nature
of the house and system, setting a longer cycle might be worth doing -
e.g. if it's an older (say cast iron) boiler and not overpowered for
Correct. But that is where you have gone outside the proportional
band anyway. On the way down, as you come into the band, the boiler
would begin firing as soon as 10% heat or more is demanded.
Another factor that I should have mentioned is the boiler thermostat.
If the boiler is oversized or the heat is not being disposed of
quickly enough, that will come into play and have an impact on the
system. You would notice this if the boiler has a tendency to short
cycle without the controller in operation. Ideally, when starting
from cold, the boiler thermostat shouldn't come into play until you
are getting towards the set point if at all.
To email, substitute .nospam with .gl
previously assumed that it had something to do with the gain - but it is
actually the temperature band within which the controller exercises what
might be called "time division proportionality". Outside that band it is
either fully on or fully off (if I'm right). I presume that there isn't any
user control over the gain applied to either the proportional or integral
feedback? [I must admit that, having only worked in the past on analog
controllers for mechanical systems, I'm not too clear on how the principles
apply to something which has only an on/off output].
10-year-old Baxi Solo 70/4 PF - which *does* have a cast iron heat
exchanger, albeit of fairly low water holding capacity such that pump
over-run is required. Could you please explain in slightly more detail what
the likely effect would be of setting the cycles per hour to (say) 3 rather
than 6. Should I also increase the minimum ON time to more than one minute?
[For example, for oil boilers the leaflet says to set to 3 cycles/hr and 4
minutes min ON time].
The boiler is on permanently when the system starts up from cold. When it
gets hot, but the room stat set point is not yet reached, the boiler *does*
cycle on its own stat - but I've always regarded this as normal. Do you
regard it as a problem?
One more point, while I think about it. One of the features on the CM67 is
something called "pump exercise" which purports to run the pump for 15
seconds each day. This is presumably to stop it seizing up if not used
during the summer. I can't quite figure out how this is supposed to work
since, on most systems the boiler and pump are either connected in parallel,
or the pump is controlled by the boiler to provide over-run. Either way, you
can't have the pump on without the boiler also being on - unless, I suppose,
you turn the boiler stat down to zero. Is this what you're supposed to do?
[This is of purely academic interest in my case because the boiler and pump
run all year for DHW anyway].
Thanks for sharing your expertise - it is very educational!
I've just installed a Honeywell CM67(RF) and I've been trying to figure out
how it is operating
so the information in this thread has been useful to me.
The boiler I have is a Vaillant Turbomax Plus.
Once the house is up to the set temperature everything is fine, but it does
seem to take a long
time to get there even in 'mild' weather, set to 19.5 Degrees C
The boiler has a temperature control on it - should I set this to max and
let the CM67 take over?
Does anyone know which has priority - the boiler's thermostat or the
On Sun, 8 Feb 2004 20:36:51 -0000, "Frank McGuire"
Yes you should. Keep an eye open for short cycling when the
controller is in the permanently on range. This would mean that you
are not getting heat away from the boiler fast enough, often because
pump setting is too low or radiators turned down too far.
The CM67 will give the boiler a demand for heat. The boiler will do
that until that goes away or the water temperature reaches the level
set by the boiler thermostat.
What you want to try to avoid is that the CM67 is creating a demand
for heat and the boiler is turning on and off on its own thermostat.
The system will still work, but you may get some inconsistent effects
with the controller depending on the heat actually required due to the
outside temperature. Ideally, you want if possible for the CM67 to
do the main control.
Bandwidth is a bad word to use but is the usual one. Normally it's
used in connection with radio and network engineering to mean
something completely different. The way you've described it for
this application is exactly right, though.
Yes. Also within the band, don't forget that there is the minimum
on-time which has an influence.
Not directly. There will be several factors, including the house
behaviour which will affect behaviour.
If you want to think of it electronically, in the proportional range
it is roughly equivalent to a switch applying charge or discharging a
capacitor and being switched each way to set a fixed voltage. If you
were to take the voltage on the capacitor and filter it to remove the
short term variation, you will end up with more or less DC voltage
which increases and decreases according to required heat.
This is really all that is happening. The house has a much longer
time constant than the cycle rate of the controller and so there is a
The proportional switching is only there because most boilers don't
have an analogue input to control the power level.
This is a very common method of industrial temperature control where
the process and the heater have a long time constant.
To give you another example, I keep snakes in a large vivarium. The
particular species requires a relatively narrow temperature range and
high humidity. They also require a temperature gradient of a few
degrees along the length of the enclosure so that they can choose the
temperature that they would like.
For the heating, I have a special kind of ceramic heater surrounded
by a guard in the enclosure at one end, and a temperature sensor
(platinum resistance probe) in the centre. There is an industrial
temperature controller which drives the heater in proportional mode
through a solid state relay. The cycle time is ten seconds which is
fine for this arrangement.
The controller is really intended to acquire and maintain the setpoint
as quickly as possible and has autolearning modes to do this. Most of
the time, the temperature is held at 29.5 degrees. However, there is
a second control system for the humidity. This consists of a
humidity probe and controller for that, the output of which is used to
drive a pump. The pump delivers water from a reservoir to fine mist
sprayers in the enclosure, wetting the bark chips on the floor. The
humidity rises and the controller cuts off. Precise control is not
so critical so this controller is a simple on/off with a hysteresis of
5% of humidity.
However, one effect of the humidifier part running is to drop the
temperature. As the water evaporates, it requires its latent heat to
do so and there is a cooling effect. The thermal control adjusts and
brings the temperature back within a minute or so without
The snakes seem happy with the arrangement, which reminds me, I must
go and feed them.
Proportional control has no time division. That would be PI control, which
the CM67 has. Most commercial controllers now tend to have PID control.
Proportional control is a corrective action that is proportional to the
error. Remember that error is defined as the difference between the desired
value (setpoint) and the actual measured value (process variable).
Proportional control is referred to as either "gain" or "proportional band".
Proportional band is the percent of change in the controlled variable that
causes the controller output to move over 100% of its range. In other words,
a proportional band of 100% means that if the measured variable moves over
100% of its range, the controller output signal will move over 100% of its
range; or, if the measured variable moves 1% of its full range, then the
control output moves 1%. In general, as the proportional band percentage
gets smaller, the controller becomes more sensitive. It is possible to make
it so sensitive that a very small change in the process variable measurement
will result in very large controller output. For those who like to work in
gain instead of proportional band, the relationship between these terms is
simple: gain is the reciprocal of proportional band. If the proportional
band is 100%, then gain is 1/100%, or 1. If proportional band is 50%, then
gain is 1/50%, or 2.
A proportional-only controller cannot control most processes well because it
never reaches the desired setpoint. This offset, defined as the difference
between the desired setpoint and the actual process value that the
controller can achieve for the process, is due to the mathematical nature of
the proportional control mode. Because there is only one controller signal
output value associated with the particular process variable value the
controller is reading, the controller sticks to its rigid mathematical
association between input and output values. It cannot recognize that there
is a problem of not achieving the desired setpoint value with
proportional-only control. Increasing the controller gain will reduce, but
not eliminate, the offset, but too much gain will result in unstable control
that will oscillate endlessly above and below the desired setpoint.
Integral settings are defined in repeats per minute or minutes per repeat. A
setting of 10 repeats per minute will instruct the controller to repeat the
action 10 times each minute. In general, the more often the repeats, the
faster the offset will be eliminated. Integral control is a control mode
that acts only as long as an offset exists.
There are three basic modes of control:
1. Proportional - operates on off-set and never reaches setpoint
2. Integral - time.
4. Derivative - anticipation, measurting rate of change.
PID controllers, which incorporate all three, are generally better.
Certainly more flexible. In short, the PI of the CM67: proportional gets it
near the setpoint, but never reaches setpoint because it operates on
"offset". A proportional controler can sit there all day and not move and
the setpoint will not be reached. They need a prod. The Integral mode of
control operates on time and pulses the offset (the differnce bewteen the
setting the proportional mode has decided to stay on an the setpoint) up to
setpoint, giving high accuracy.
Be careful not to have the proportional band too tight as "hunting" will
occur, known is short cycling to you.
In a CM67, the control aspect that times the delayed start is separet from
the PI temperature control.
The time proportions of the CM67, are a crude way of keeping an on-off
boiler within the proportional band.
The CM67 can control an electric actuator. A better system would be a heat
bank, heated by a basic on-off control boiler (these are cheaper and have
simple more reliable basic controls. The CH flow and return pipes would
have a modulated 3-way valve, moved up and down by and electric actuator.
This can stop at any point in the travel. This way only PI control would be
used and guarantee much more accuracy of setpoint temperature control. As
many heat bank and thermal stores are recommended to be on 24/7, the boiler
can be controlled only by the heat bank and the CM67 moves the modulated
valve and switches on the pump when it calculates the start time.
It is not "bandwidth". It is "Proportional Band - width". The "width" is
the range of the band, e.g., 2C. That is the setpoint and 1C either side.
< snip crap about snakes, which should all be banned >
Hunting is hunting. It is not short cycling but movement around the
They could conceivably be related. the manufacturer doesn't say.
That was already covered.
I think you are confusing the issue. From the control mathematics
perspective this is PI control exactly the same. All that this
achieves, although it will work is perhaps finer grain control of heat
delivery from the heatbank to the radiators. Of itself, the heatbank
is a red herring in this if the boiler is not part of the main control
I would also be surprised if a valve for this application would be
modulated with a ten minute or even 5 minute cycle time. but rather
that it would have some form of electronic control requiring a much
shorter cycle time -seconds or even sub second.
Can you suggest a make and model that would have the characteristics
to work with a controller like the CM6x which has a much longer time?
Obviously, and that is explained in the data sheets. However it is
often written as a single word in controller data sheets.
This is a real life application using an industrial type of PID
controller and served as a reasonable explanation of some of the
I felt that it was a better illustration than simply cutting and
pasting material from a web site
To email, substitute .nospam with .gl
It is a form of short cycling an easier for others to understand.
Controlwise they have to be eparte. The user doesn't know this nor care.
Not at all.
Not so. The heat bank can sit there 24/7 keeping itself to its own setpoint
providing a nice constant heat source. Just a store of heat. The CM67 will
then just switch on the CH pump and modulate the CH actuator. Modulating a
3-way valve gives far more precise temperature control than switching in and
out a boiler.
Another way is that the thermal store can be switched on also by another
CM67. The desired time must be earlier than the time of the heating. This
is if you can replace the room temperature sensor with a water strap-on
cylinder sensor of course.
All those cycle times in minutes would be redundant, as only PI control on
an actuator is required.
Don't care They should not be allowed in the country. If a couple of those
get out and breed in the wild.
Not true. The piece controlling the optimised start could make very
good use of knowledge of the characteristics of the PI part. Whether
it does or not is a matter for the manufacturer.
You are because the heatbank is only a source of heat if designed to
be oversized by a reasonable amount so that the boiler behaviour goes
out of the equation.
Exactly, but that's all.
I don't disagree with that notion but can you suggest a make and model
that can be modulated in a precise "analogue" way (i.e. part opened)
by a CM67 type of controller with a long cycle time?
There are certainly valves that can be modulated with a DC voltage (I
have some) and I could imagine pulse width modulated using a much
shorter cycle time - e.g. in the way that model control servos work.
If you can't suggest one that will work with the CM67 type of
controller, then the argument becomes a bit academic because either a
different type of controller is needed, which is not in the same price
bracket, or you would be talking about a standard motorised zone valve
and proportioned on/off control of it. That would be a bit better
than controlling a boiler on/off, although with a lightweight heat
exchange type, I don't think that the difference would be substantial.
Yes of course, but this is adding extra complication. As you said, a
thermal store ought to be on 24/7 anyway.
The proportional pulse width output from the controller would be used,
but can you suggest a valve that will take this and use it to operate
the valve to arbitrary positions.
First of all, the particular snakes in question (Brazilian Rainbow Boa
- Epichrates Cenchria Cenchria) range in the wild in the Brazilian
rainforest and their habitat is rapidly disappearing together with
their numbers in the wild. They do, however, breed and thrive
perfectly well in captivity, given the right conditions of temperature
and humidity (28 - 31 degrees and 70-80% humidity). In order for
them to breed, the conditions have to be right and there needs to be a
pattern of temperature drop and rise around the breeding time.
The UK climate simply won't create the conditions for them to breed in
the wild and they would be unlikely to live for very long in the wild
anyway. Considering the massive species loss in the rainforests,
anything that can be done to sustain them in captivity until their
natural habitat issues can be resolved is a positive step.
To email, substitute .nospam with .gl
It constantly moves up and down in an on-off manner.
It could supply data to other control processes, but it is a separate
And, the modulating 3-way valve has finer accuracy to maintain a setpoint
temperature than switching on-off a boiler.
The cycle times are not a part of the Integral aspect of the controller
There are 230v electric modulating valves available. Look in the CM67 pdf,
it has a diagram of one.
What I am on about is a "modulted" 3-port mixing valve, with an electric
actuator. I have a few here.
It depends on how far you want to take it. A CM67 on a themal store will
learn how long it takes the boiler to heat it up, and bring it on at the
AFAIK, the cm67 when controlling the setpoint temperature, only applies PI
control modes. The cycle times appear to be an underlying control layer
which only allows the boiler and PI control on it at certain time periods.
< snip crap about snakes, which should all be banned.>
<If Brazil is short of snakes then you should not be having one. >
I am referring to the specific method of operating the control valve -
the signalling between the controller and the valve. The CM67
provides a pulse with a total width of anything from 5 to 20 minutes
depending on the what is configured on the setup. In order to
control the controlled device (boiler, motorised valve etc) it will be
on for between 1 minute and 100% of that time while operating in the
proportional band. On the fastest cycle time, that will equate to
a range between 60 and 300 seconds of on time per 300 second cycle.
What I have specifically asked you to provide is the make and model of
a valve which will operate in a modulated fashion given that type of
Please reply to that specific point.
I would be surprised if such a thing exists. I know that there are
voltage controlled modulating valves ( I have some radiator valves
that can be controlled by 0-10v DC); and I could believe that there
are some that would modulate based on a pulse width with a much
shorter cycle time ( a second or less)
I just have. Presumably you are referring to the sheet I posted -
It doesn't say anything about it being modulating.
There is a thermal actuator (diagram c) controlling a boiler. This
type of device is a thermally operated relay which controls the boiler
on and off, not modulating.
There is an electric actuator (diagram h) which is an electric
actuator. This is a motorised device to operate some kind of
mechanical entity such as a damper. It even shows the limit switches
which stop travel. This isn't a modulated device either.
Did you mean something different? Please provide the URL to the CM67
sheet that shows it if I have the wrong one.
I know. Please provide the make and model of them. I am simply
asking you to provide details of a *modulated* device which will work
with a CM67.
The cycle time is about choosing a control signal suitable for the
device being controlled e.g. boiler, zone valve, electric heater
The heat bank supplies the constant hot water for modulation purposes.
A pulse? A signal going on and off continuously?
The control signal is 230V.
It does. A reversible motor. 230v on one terminal and it moves one way,
same the other way. Stop the singal and it stays where it is.
It appears to be.
Any maker makes them. They are common.
AFAIK the CM67 can operate a 230v modulated valve. If that is not the case
(specs vague on this from wjat I have seen), then I think the Landis &
Staefa version can. If so, I would go for this controller not the CM67. I
tend to prefer Landis & Staefa controls anyhow. They are much more
reliable, better made, better user intefaces and generally cheaper too.
The Landis & Staefa RVP will give proportional control by driving a servo on
top of a mixing valve, measuring the flow, return, inside and outside
temperatures, deriving the setting from a pre learnt heating curve for your
house. Assuming your boiler already has a RVP the extra cost is about 200
quid for the valve and servo. With a rad based system I would recommend
fitting this type of control as the temperature swings are eliminated by
matching the heat input to the heat loss. But with UFH time constant you
would really gain much from these controls.
This controller has a contact which makes and breaks with a duty cycle
as described - e.g. 1 minute on, 4 off out of a total of 5.
When this is controlling a supply to a device, it is pulse controlling
it in this case with a long duty cycle.
That's the source of power to the controlled device. The controller
is turning on and off.
I very much doubt it. An actuator should run from end to end
reasonably quickly - certainly less than a minute.
You can't have an arrangement as you describe. This would require a
controller which could stop and start the motor and another to reverse
it and something to report back to the controller what the position
That is not a simple actuator and it requires a more complex
controller than a CM67.
It is possible to have pulse width controlled actuators in other
applications where the control to it is effectively open loop, but the
pulse cycle time is vry much less and the switchover time in the cycle
used to determine the stopping position - not the same thing and it
also requires some electronics at the actuator to do the motor
control. This is how model control servos work, for example.
I don't think so.
So please suggest a make and model that will work in a modulated
fashion with a CM67.
OK, so now you are changing your tune.
OK. Please supply the part number of the L&S controller that will do
this and a suitable modulating valve to go with it
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.