I fitted one of these a couple of weeks ago and I do find it to be
excellent, as other posters have reported. One thing is puzzling me. A
few times I've noticed a situation where the target temperature is,
say, 20C and the room temperature is 19.5C. The little flame icon on
the wireless part appears, meaning that the thermostat is demanding
heat. However the green LED on the base unit is not on, so the boiler
doesn't fire. A few minutes later the green light comes on and so does
the boiler. As far as I can see this is not an RF issue. The two parts
are very close together and there were no RF problems when I did the
setup test, even when taking the thermostat to the far corners of the
Shouldn't the green LED (and therefore the boiler) come on whenever
the flame icon appears on the thermostat?
OK, sorry, I think I've just answered this myself by reading the
manual properly - the level of demand from the thermostat is sent to
the base unit as a percentage. Presumably when the room temperature is
only 0.5C off the target the thing registers a fairly low demand and
there will be times when the flame icon is on even though the green
LED on the relay is not. I guess this allows for the fact that the
last burst of heat generated by the boiler takes a few minutes to
register with the thermostat, so this mode of operation stops it from
overshooting as the temperature approaches the target.
It's a great little gadget, and it certainly does maintain a steady
temperature very well. I've been muddling along for two years with
only TRVs - hoping that this will make the house generally more
comfortable and save a few quid too.
On 23 Oct, 15:31, Martin Pentreath
Whilst not directly related to your query (which it sounds like you've
answered) I thought it worth mentioning that the 'Optimum Start'
feature of the CM907 (and presumably CM927) is disabled by default and
so unless you've gone through each of the settings you may have missed
this. Perhaps you don't want it but I've found it works quite well.
This is the opposite of the older CM60/67 where the feature was
enabled by default. Perhaps the change was to avoid potential user
confusion when the heating comes on before they expected it to.
"Mathew Newton" wrote
I have 2 CM67s and deliberately disabled the "optimum start" because the
system simply wouldn't stabilise!
The fact that this is now set off as default in newer models suggests I am
not the only one to have issues with this technology.
To be fair, one or two posters have defended this feature in earlier
discussions on the subject.
On Oct 23, 3:05 pm, Martin Pentreath
What do you have the minimum 'on' time set to?
It might implement this by either switching on _or_ off at a _fixed_
interval of n minutes, so it's a minimum off time too.
This is how my non Honeywell thermostat does it, I think it's intended
to protect boilers that need a minimum off time in order to shut down
You'd think so but Honeywell say that "this is normal operation. The room
unit sends only the demand signal (0-100%) to the relay box. Depending on
the demand signal the relay box will switch the realy on when required."
I don't think I could explain it any better myself :-/
This is because the Honeywell's controller is a PI (Proportional -
The "proportional" part means that in a temperature band a degree C or so
either side of the set temperature it varies the heating demand from
- 100% (when the actual temperature is at the bottom end of the band, or
- 50% (the actual temperature is at the set point); to
- 0% (the actual temperature is at the upper end of the band, or above).
Proportionality is obtained by the relay box switching on the boiler for
the right proportion of the controller's cycle time. E.g. when set to
6 cycles per hour this means a 10 minute cycle; so 50% demand means the
boiler is turned on for 5 minutes out of each 10.
You will probably have noticed that a fixed 50% demand at the set point
is not necessarily going to maintain the set temperature. That's where
the "integral" part of the controller comes in. This bit monitors the
actual temperature and if it's consistently above (or below) the set
point with 50% demand, it reduces (or increases) the demand targets
listed above until the temperature settles correctly.
PI controllers usually give much more constant temperatures than simple
In an earlier contribution to this discussion,
Fair enough - but I wouldn't have thought that it was beyond the wit of man
(or Honeywell at any rate) to contrive a way of only switching the flame
signal on in the sender only when the relay is on in the base unit. This
would avoid the confusion which is otherwise bound to occur.
My CM67 wired programmable stat uses essentially the same proportional
control logic, and the flame in that is synchronised with contact movement
without any problem.
John, thanks for that explanation as I was also puzzled by the Honeywell
Their manual is a bit lacking in the 'Installer set-up area and says no
changes for 'on' and 'cycle' parameters for gas boilers but if I leave it
as it is I think it will take a long time to heat this house as it's the
actuator that my thermostat is controlling and the boiler and pump only
comes on when the actuator has opened the valve.. This results in a very
short pump and boiler run time before relay box signal drops and the
actuator goes through it's closing cycle. I downloaded ENOR 8535 R0 03/03.
(The HC60NG product Specification Sheet ) Which shows the wiring for an
actuator set up although the manual packed with the thermostat only shows
four other applications.. There appear the be no specific instructions for
setting up an actuator set-up. Honeywell will only apparently respond to
e-mails from tradespeople. Do the 'cycles per hour' and minimum on/off
times get transmitted to the HC60NG?
I'm not sure that's right. For example, take a proportional band of
2 degrees C - thats 1 degree C either side of the temperature set on
If room temperature is more than 1 degree C below the set temperature then
the controller will demand heat for 100% of the cycle time. The boiler
should stay on continuously until the room temperature reaches 1 degree
C below the temperature you want. Even if initially the actuator has to
open the valve before the boiler & pump come on, 100% demand should get
you to within 1 degree C of your set temperature just as rapidly as a
It may be slower to get the room temperature up that final 1 degree C
to the set point (but a proportional controller is better on overshoot
and maintaining the temperature at exactly the set point).
The short on-time you mention should only happen when the demand is
down to, say, 10% of a 10-minute cycle - that's 1 minute on-time.
That's only going to happen when room temperature is almost at the top
end of the proportional band - that's close to 1 degree C *above* the
set temperature already.
If you don't like a short boiler on time even then, you can set a minimum
on time of say 2 minutes rather than 1 minute. However that isn't going
to change anything to do with warm-up time. BTW my boiler can take up
to 40 seconds to come on for the same reason and I did set the minimum
on-time to 2 minutes on my CM67RF because of this. I don't know if it's
I don't know but possibly. I note that my Honeywell CM67RF controller
can be set to frost protect mode where if RF communication is lost then
the HC60NG maintains an independent 20% on-time cycle for the boiler.
So I possibly the controller sends commands to the HC60NG of the type
"use 6 cycles per hour" and "use 40% on-time this cycle". However that's
Thanks again John. What concerns me a bit is the lack of information
available for the user.. And it seems like the average heating
engineer/installer doesn't either want to know or care either. I've
currently set the minimum on time for 3 minutes and it certainly seems that
the HC60NG now works in that mode. It looks like the proportion is set
to about 100% if there is a 2degreesC difference
and 50% if there is a 1 degree difference. So if you want to force the
boiler to stay on you have to maintain the plus 2 degree difference. I
find I need to do this if I want a 'warm comfort period' for say half an
hour at the beginning and end of the day.
Experiments continue but I am very happy with normal long period control
with this stat.