I am looking at putting a thermostat at the core of the house - which
has extremely thick stone solid walls and a huge time constant. Running
the wire back to the loft. Yes the boiler and main controls are all in
the loft - don't blame me. It was like that when I moved in!
(It works well enough that this isn't a problem most of the time)
It also does the hot water so the boiler must run an hour or so morning
and evening even if the CH is completely off during the summer.
I was wondering about getting a PWM capable wireless unit instead and
using the thermostatic control to modulate the separate CH pump when the
main oil fired boiler is active for CH. I think this will result in
tighter overall temperature control and less overshoot.
But it will also cause the boiler to burn for shorter times and more
often. My question is has anybody tried this and what were the results.
Is the effect of more frequent burns less efficiency from the burner
starting up from cold more frequently or a net fuel saving? It is fairly
obvious for a gas boiler that it should be a win, but for oil it is a
lot less clear. Anyone done this and have practical experience?
The wireless controller I was considering was this one - Salus ST320RF
or a couple of other more expensive ones which do more complete PID
implementations. This one has the nicest styling (bad reason to choose I
know but most of the others look rather industrial).
What are the pros and cons ? Any better thermostats to consider?
I have just installed two of the non-RF versions (the salus 320) (two
zone heating, upstairs and downstairs).
These units are simple on-off controllers, they will not do PWM motor
On some CH pumps you will see a knob on it for pump speed. If might be
possible to fashion up your own pump speed controller by using relays
onto the wiring for this speed knob. You would need some kind of
modulation control to take an input to then control the required amount
of pump speed. That might be a better method than PWM.
You do need to change your mindset regarding using these SAlus St320RF's
compared to a conventional timeswitch and wall thermostat.
the smallest time period you can have is 10 minutes and the smallest
temperature change you can have is 0.5°C.
For example, you think I want 20°C between 7.00am and 9.00am, (the time
between getting up, showering and breakfast and going out) then 5°C from
9.00am till 4.30pm (assuming everyone in house is out at work or school
so you only need frost protection) then 20°C from 4.30pm till 10.00pm
(everyone comes home has evening meal, watch TV) then 14°C from 10.00pm
till 7.00am (to stop you getting hypothermia in bed but dont need full
on heating due to use of duvets)
In a modern well insulated house, it does not take long for the house to
reach the desired temperature.
In your case with a stone house with long time constant, you would need
to establish how long your house takes to heat up and cool down and
factor these in to your ST320RF settings by making the temperature
transition times earlier by the time constant you've established for
your house in order to achieve the desired temperature set profile.
You might want to consider getting a weather compensated model so that
the boiler starts earlier in winter and later in the summer so that the
weather has little effect on achieving the desired thermal profile
inside the house.
On Fri, 23 Mar 2012 09:56:46 +0000, Stephen H wrote:
That might be wasteful. When the time switch on one system I had
broke I started to only heat the DHW when I needed it, reheat time
from cold was only about 20mins. The tank would stay hot enough for
washing up/hand washing for the best part of 48hrs. There was a very
noticeable drop in the gas bill... I was on my own and there was no
Or look for an "Optimum Start Control" feature(*1), that calculates
the difference between actual room temp and target temp and in
conjuction with a rate of rise fiddle factor automatically turns the
heating on early enough to reach the target temp by the target time.
This feature is normally only on the first target temperature/time of
With a house with a high termal mass I doubt weather compensation is
worth it, the house does that by the sheer amount of heat stored (or
not(*2)) in itself. With a high thermal mass property it's best to
just keep the heat input ticking along, you really do not want it to
cool down significantly as it'll take days to warm back up.
I'm not sure what the OP hopes to gain with the PWM. I get the
impression that there is no room stat currently, the addition of one
will improve things, particulary if the new stat is a programmable
one with OSC.
We really need more information about the boiler,
non-condensing/condensing/combi. What the DHW arrangments are. What
existing controls there are etc.
(*1) Different makers use a different term "Optimum Start Control" is
the Danfoss term and has manual setting for the fiddle factor.
Honeywell uses a different term and "learns" how fast the building
(*2) In winter the building has lots of heat relative to the weather,
it tends to stay warm. In summer the building is relatively cold so
it tends to stay cool.
Hot water tank isn't quite that well insulated but will stay plenty warm
for 24+ hours. I should more accurately say the boiler has the option of
coming on and heating up the water for about an hour morning and
evening. Hot water is on a gravity feed circuit to both the oil boiler
and the wood burning stove. There is a manual override switch that can
force the main CH pump to be on from downstairs. And also a protective
thermostat on the exit pipe from the stove to force CH circulation if
the temperature starts to get too high.
There is also a frost stat that defends the loft space pipe runs.
I am a bit surprised at that since I do monitor oil usage fairly
carefully and have dropped the amount used by about 30% over the past
couple of years by burning more wood and very deep loft insulation.
Can't do anything about cavity wall insulation - no cavities. The guy
who drilled the obligatory vent for the modern solid fuel regulations
cursed the walls for their thickness and resistance to his core drill.
Winter oil usage rate for space heating is well more than an order of
magnitude higher than summertime water heating.
That sounds interesting. The system at present has only the boiler
thermostat and a broken sensor in the main room and gives fairly crude
levels of control which can be tweaked by manually enabling the pump
from downstairs and/or lighting the fire. The main room is no longer a
good place for the master sensor as the new stove is way more efficient
than an open fire.
I was originally toying with the idea of gating the system with an
exterior temperature monitor so that the pump was disabled by it.
Yes. Pretty much exactly describes it. The time constant for the
interior structural walls as opposed to the air inside is a few days.
Boiler is an old Worcester Danesmoor 15/19 oil fired boiler dating from
I would guess the mid 70's. Touch wood it is well behaved as given the
odd location in the loft major rework would be involved to change it. It
has to be non-condensing given the age! There is a separate pump.
Pump and boiler have separate independent control and overrides. Timer
is a bit simple minded but works OK: On/Off, Once daily, twice daily.
That sounds interesting. The two main aims would be to have the system
better at warming up the house after an unusually cold night in winter
and to back off heating quicker when in spring the sunshine brings the
core up to an acceptable temperature (eg late evening sun into rooms).
Indeed. That describes the situation very well.
I tend to end up thinking about wasted oil more at this time of year
since during spring with the odd sharp frost at night it shows up all
the weaknesses of the existing mostly timed heating system.
On Fri, 23 Mar 2012 11:17:51 +0000, Martin Brown wrote:
Doubling or tripling the jackets or even fitting a jacket to a
preinsulated tank might be worth it. Go to far though and SWMBO'd
will complain about a "cold" airing cupboard. B-)
What happens if the power fails? Guess you have to put the fire out
just when you could really use it... There ought to be a heat dump
radiator or connection into the heating system that will gravity feed
controlled by thermostatic valve with its sensor at or near the top
of the HW cylinder.
Yep, made a noiceable reduction in our oil consumption this year with
the installation of insulated dry lining and wood burner. Not sure we
have actually saved any money though when you factor in the cost of
It really depends on your demand for hotwater, one person on their
own and out at work all day doesn't need a tank of hotwater sitting
losing heat. A family of teenagers all showering or bathing every day
the heat lost by even a basically insulated tank probably pales into
insignificance compared to the use.
Ours was about 6:1, ie winter about 6cm/week in the sight tube.
Summer 1cm/week. This winter it's was around 4.5cm/week. Hopefully
this summer might drop to 0cm/week as we have also fitted a solar
You don't want any thermostat in the same room as a heat source it's
not controlling. Like wise you don't want a room stat in a room with
a radiator fitted with thermostatic valve unless you disable the
So might be a gert lump fo cast iron with high water capacity, long
burns are required for best effciency. Our oil boiler is a
floorstanding model holding 12 gallons of water and is massive lump
of cast iron. From cold it takes quite a while to start churning out
hot water even with >35kW of input.
A programmable stat out of the room with the wood burner and in a
space with a radiator (any thermostatic valve wound flat out or
control head removed) should do the job. Choose one withn OSC or
similar. I like the Danfoss TP7000 others swear by the very similar
Honeywell versions. Both are available in RF rather than wired
versions. RF might be useful if the room with the woodburner is the
main room used. You can move the stat into that rooom to give you
best comfort when the stove isn't lit and out when it is. Perhaps not
every time you light the stove just in general, so out during teh
winter when te stove is lit every day, back in during the warmer
parts of the year when it isn't. Remember any thermostatic valve on
the radiator though...
The gravity DHW is a bit of PITA as you have to heat the water when
there is a demand for CH. There are ways around that, not sure if the
d-i-y FAQ on heating covers those but it is probably worth a look at
There is a heat dump radiator with gravity feed on the back side of the
fireplace wall in the master bedroom. Part of the reason to look at
resiting the thermostat is because with the fire on the house can lock
in a bistable mode with the CH pump off and only the main living room
and master bedroom being heated - tough for anyone in the rest of the house.
The gravity feed and dump radiator works OK in a power cut although it
is wise to back the fire off a bit from maximum output or it and the
domestic hot water tank ends up fiercely hot. It turned out that the
previous back boiler on the open fire had been plumbed incorrectly so
that the dump radiator did not work at all if the CH pump had failed. We
never managed to make it kettle though even in long power cuts.
I am seeing £200 for the logs and about £400 for the oil saved. Main
inconvenience is having to lug the logs in and stack them in bulk.
It is just how the thermostat was previously installed. The old open
fire didn't seem to upset it. Changing it brings various other ideas
into play - like the best place to use as a total house control point.
Similarly large chunk of cast iron and built like a battleship. The fire
at full bore can output enough heat to completely shut down the main oil
boiler so in very cold weather that is what I do.
A few weeks and it will be back to hot water only with just the odd
flick of the manual pump switch to stop the thing from seizing up in the
summer season when it doesn't get used.
Being able to relocate the stat until it does what I want is one reason
why I am tempted to use an RF Wireless based unit. Thanks for the tip of
Honeywell TP7000 with OSC that looks like it does what I want.
The hotwater is well enough insulated and consumed in sufficient
quantity that I really don't think it is worth me worrying about.
I don't fancy motorised valves. My parents have a complex system like
that and the things are forever playing up, sticking open etc. etc.
On Sat, 24 Mar 2012 09:57:28 +0000, Martin Brown wrote:
We've probably burnt about £350 quids worth of logs (5 dumpy bags at
£70/bag). Oil saved is a bit harder to work out but if one assumes
normal winter useage of 6cm/week since mid November (when the stove
was installed) against what we have actually used that's about 700l
or £420 @ 60p/l.
Trouble is we are also heating far more of the building than previous
winters and the insulation levels in half of it have gone up so there
is a lot of mud in the water.
SWMBO'd does that, at least when it's cold and she's lighting the
stove every evening. B-) Now it's got a bit warmer she seems to
have gone off the idea.
The heating never gets switched off here, the programable stat takes
care of it. There might be a period of a week or two in June when the
heating never comes on but thats about it. No six month heating lay
off at 1400' on the North Pennines. B-) Might be different this
summer with 50mm of foam insulation board on all external walls.
*Danfoss* TP7000 ... B-)
Trouble is the CH primary is normally rather too hot for DHW and with
an open gravity loop the DHW will rise up to that temperature. Having
said that one gets used to the possibilty of scalding hot water
coming out of the taps PDQ!
Can't say I've ever had any serious problems with ours. One did get a
bit sticky in the motor head but a quick strip, clean and lube and
Internal not external. So 50mm foam board on metal firings with
12.5mm plasterboard bonded to it and plaster skim.
Window reveals where tricky they ended up with 25mm foam plus PB and
skim. There isn't a great deal of the window frame showing...
For wireless, I always recommend Honeywell. I have had problems with the
Salus wireless controllers a number of times because of interference -
they worked fine when next to the base station, but in the next room
would not work.
The Honeywells use a different frequency (833 vs 433Mhz? mHz?) that is
less prone to interference, and they are 'paired' att he fatcory, so
should work out of the box.
The downside is the Honeywells are at least 30% more expensive.
To reply by e-mail, change the ' + ' to 'plus'.
Not had any problems with the two wireless Danfoss programmable stats
we have here. They are on 433MHz.
I'd be wary of anything using 800 to 863MHz that is about to be sold
off to the mobile phone operators for 4G... 863 to 865 is used for
all manner of kit some use a constant carrier (wirelss headphones)
which will muck up anything that expects a clear band.
Nope, M is right but hz isn't.
Nope, m is wrong (milli, 1/1000th) Hz is right.
You are looking for MHz. Unit symbols based on peoples names are,
K for Kelvin, William Thomson, 1st Baron Kelvin
Hz for Hertz, Heinrich Hertz
H for Henry, Joseph Henry
T for Tesla, Nikola Tesla
W for Watt, James Watt
A for Ampere, Andre-Marie Ampere
Pa for Pascal, Blaise Pascal
F for Farad, Michael Faraday
On Fri, 23 Mar 2012 10:49:27 +0000 (GMT), "Dave Liquorice"
Only the more modern ones.
The CM6x series uses 433 ish
It'll be 868MHz or thereabouts, an ISM band with the same 'open'
status for low power devices as 433 ish. The only difference being
that 433MHz devices can transmit continuously, the 868MHz ones have a
lower (10%?) duty cycle.
Neither band is immune from interference, the only thing that might
ensure that is a more suitable modulation scheme using multiple
carriers, but both bands in built up areas will be full of so much
crap its a wonder they work at all.
There is more detail in this OFCOM document
I always try and avoid RF full stop for the aforementioned interference
reasons. So I went for wired salus controls.
All my landline phones are corded, I don't do wireless networking or
I use a Danfoss TP9 programmable thermostat to control my oil fired CH.
It is a two channel programmer and uses a wired remote sensor located in
my sitting room. If provides much better temperature control than the
previous mechanical thermostat and programmer. Ideally I would like to
reduce the number of heating cycles per hour, but it is not a
configurable item on this programmer.
I have serious reservations about the benefit of proportional (PWM)
control on some heating systems.
I have a Honeywell CM907 wired stat controlling my main living area. It
has fixed proportional control and it does maintain the temperature with
high accuracy but I'm not happy that it forces a minimum of 3 switching
periods per hour, as this results in the boiler thrashing at full power
to achieve circulation temperature at switch on and more often than not
is shut down just a few minutes after the rads get warm.
This is particularly noticeable in those borderline seasons when you
want the heating on but you really don't need that much heat, you seem
to be cycling the boiler (and controls) unnecessarily.
Also, one of the benefits of programmable thermostats is that you don't
really need to turn them off completely, you just set them to a lower,
setback temperature where they will come on in a cold snap or save the
temperature of your high thermal mass space dropping too low. Often this
is a sort of don't care temperature, say you set it to 16degC but you
don't really care if it drops to 15 or rises to 17, but you can't tell
the propo stat not to care about the accuracy of the setback
temperature, it will religiously try to control it to within half a
degree, cycling your boiler 3 times an hour, 24/7 whether it needs to or
I think an alternative to this is to use a stat with simple on-off
control which has high accuracy and hysteresis that can be varied to
suit system needs.
My Honeywell stat is on the way out and I am looking at alternatives
that can be configured either as proportional or on-off controllers. The
Siemens REV24 has caught my eye, it has high accuracy and can have
hysteresis set as low as half a degree if that suits the system. In a
high thermal mass building (as mine is), I would be happy to have the
stat switching one an hour and give the boiler a run at a long slow burn
when it will be at it's most efficient.
The benefit of having the control method configurable is that you can
try it one way and swap to the other if it doesn't work for you.
FWIW, my system ran the smoothest before I installed the thermostat, the
boiler has a wide range of modulation and I set it low to get a very
long and slow burn. Throughout the first winter the boiler was rarely
cycled when I had it on, it just tickled along, giving out about 6kW or
so on low, a bit like a smooth country drive vs boy racing between
traffic lights. I only turned up the modulation when it got really cold.
My boiler is gas btw so it may have a wider range of modulation than
yours but I'd guess that oil burners like cycling even less that gas so
the argument above probably still applies.
Siemens stats are a bit pricey (shop around) and unfortunately have a
certain look of German utilitarianism about them so may not be liked by
Actually, prices not as bad as I thought, REV24-RF, £110 inc vat which
is about the same as the wireless honeywell (CM927), £80 for the wired
vs £62 for the honeywell.
Watch out for RF drop off through thick walls, I have put in 2 wireless
systems for friends but have wired at home ;-).
On Friday, March 23, 2012 9:39:33 AM UTC, Martin Brown wrote:
You're doing it wrong, IMHO. Time proportional (electronic) control is (was) used for room temperature control, but with on-off durations of minutes; with PWM I believe it's milliseconds (no experience). Rapid on-off usually results in premature death of the motor. It's a cheap & cheerful way of regulating the heat flow with just a pump. A thermostat does much the same.
Better to modulate the CH flow temperature, i.e., weather compensation. If not a modulating burner, then a mixing control from a heat store. It needs some thermal mass to give the burner a decent run time during periods of low-load, or else you get rapid on-off cycles and early burner burn-out, as above.
A new cylinder would probably pay for itself by reducing the oil boiler run times.
On Sat, 24 Mar 2012 07:45:59 -0700 (PDT), Onetap wrote:
Oh I think we need to sort out what we are talking about, there are
three different things here:
PWM of the power to the CH pump to vary it's speed in relation to
Proportional time control of the boiler & pump. The lot comes on a
fixed number of times/hour with varying duration dependant on heat
Start optimisation. The heating is switched on by a variable amout of
time before the target time such that the target temperature is
reached by that time.
I'm not aware of any domestic stats that offer PWM power. Some pumps
do "modulate" but AFAIK that is all builtin. Fixed cycles/hour
proportional control is not recomended for systems with motorised
valves. So that leaves the start optimistaion and a thermostat as
being most useful in a domestic situation. Using a programable
thermostatic brings in the fine temperature control (against
mechanical stat) of an electronic sensor along with the abilty to
have the temperature appropiate for the time of day.
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.