On Sat, 13 Dec 2003 16:53:12 +0000 (GMT), "Dave Liquorice"
wrote:

On Sat, 13 Dec 2003 14:02:27 +0000, Andy Hall wrote:

From all of this, I would conclude that room thermostats may be good
practice and used as part of demonstrating compliance with Part L1,
but I can see nothing in the Approved Document that mandates them
and it is certainly a long stretch to say that not having one in a
new installation is specifically illegal.

What they want is the boiler to off, really off, when there is no
demand for heat. There are many ways to comply with this, hence the
"loose" wording of the documents.

The legislation does not support that level of detail. That is not to
say that implementing a room thermostat is not a good practice, but I
can find nothing more than a recommendation suggesting that one is a
good idea.

In my mind the question is, what constitutes a demand for heat?
TRVs typically reduce the flow to radiators rather than closing it off
completely unless the radiator is grossly over sized and the room
temperature overshoots from it or a heat gain from another room or
perhaps solar gain. Good temperature control in the rooms where TRVs
are fitted would be achieved by their reducing the water flow to the
radiators such that the heat output balances the heat loss at the
required room temperature.

The heat output of the radiator in the location containing the room
thermostat where one is used has to be reduced to ensure that the
rooms equipped with TRVs have reached their set point.

Once that has been achieved, the boiler is turned off if a room
thermostat is used with a simple boiler which delivers zero or full
output only.

There are, of course, room thermostats with proportional control which
turn the boiler on and off on a relatively short time constant to
control the average heat output so that it is reduced as the set point
is approached; and thus it is possible, even with a basic boiler to
limit temperature overshoot which is said to be a significant energy
waster.

As I understand it, though, there is an overhead in terms of
efficiency for each on/off cycle of a simple boiler which does full
power or off. Presumably, this is not substantial in comparison
with energy saved by the overshoot not happening otherwise there would
be no point.



Leaving power to the boiler and using the boilers output temperature
stat or return temp stat to switch the boiler off doesn't, IMHO,
comply with Part L1. Basically if the boiler can ever fire without
there being a demand for heat it does not comply.

I think that that may be taking too simplistic a view.

Given that the plumbing and heating industry is still very
conservative and persists in installing non condensing non-modulating
boilers, a room thermostat may well make an energy saving with it.
I rather suspect though, that this has more to do with limiting the
size of the temperature overshoot that can happen with oversized
radiators and TRVs, than of whether there are short burns of the
boiler.

One of the referenced good practice guides,

http://www.gas-news.co.uk/gpg302.pdf

discusses the fitting of room thermostats as part of an interlock
scheme, yet mentions boiler energy management, modulating and
condensing boilers only in passing, even though it is relatively
recent. It seems to me that this is really an attempt to put a
stick in the ground with the industry and to say that a room
thermostat is part of good practice with a simple boiler.

However, modulating boilers, condensing or not are rapidly appearing,
and even for the die-hard installer has some appeal because there is
less commissioning to do.

It's also worth pointing out that it is quite easy for fairly
sophisticated control capabilities to be built into the boiler as
well. Most of this can be implemented in the firmware of the
controller so from a manufacturing cost perspective is not a large
additional factor. I suspect that the reason that more isn't done
is because UK manufacturers have not had people with the skills to do
the work.

For example, my own boiler has a very comprehensive set of controls
and can have a number of additional sensors apart from those within
the case.

On the input side, as a minimum it measures flow and return
temperatures. On the output side, the microprocessor can control the
burn rate from 3 to 25kW or off and the pump from 20-100%.
Even with nothing else electrically connected, it will make a pretty
good attempt at maximising its efficiency. If the heat demand is
above the minimum 3kW then it will modulate the pump and burner
anyway, so the issue of turning off altogether doesn't arise. In the
range where some heat, but less than 3kW is required, it will come on
for a short period, initially once per hour, measure the water
temperatures and run until the heat demand reduces, then going off.
The on period is remembered and adjusted so that the boiler is not
coming on too much earlier or late than need be. I have seen the
time between burns go out to as much as three hours with this hookup.
During a normal day, it will have done a hot water cycle during this
period anyway.

This is a fairly reasonable control system as it stands, but the
manufacturer includes an outside temperature sensor anyway, which is
placed on a north wall. It is then factored in as an input term to
the controller, in effect providing a prediction based on the outside
temperature. With this connected and operational, the control
becomes rather better with an ability to adjust output before the
house temperature changes.

Note that this is achieved with no room thermostat. From logs of
burner operation, pump operation, water temperatures and internal
house temperatures, the set points are achieved rapidly and without
significant overshoot of inside temperature mainly through the
modulating behaviour of the pump and burner.

I have added a room controller which is made by Siemens specifically
for this boiler. Apart from time functions it has a temperature
sensor for inside the house so that the boiler is presented with this
information rather than just an on and off. The effect of adding
this is mainly the convenience of a remote temperature control and
even less temperature overshoot - now almost nothing.
There is an installer setting whereby one can alter the relative
sensitivity of this term and the outside temperature (loop gain
controls if you like). In fact the defaults seem to work the best
anyway.

So, I think that I could make a strong case to anybody who wanted to
enquire that even without a room thermostat, I can achieve better
energy use than a simple boiler with a simple thermostat because the
minimum power output is very low and the boiler's built in algorithms
do a pretty good job on their own.

The Approved Document and the referenced Good Practice Guide really
only present *a* simple solution using a room thermostat for basic
installations. Neither presents it as the *only* solution, and with
increasing capability of boilers' own algorithms becoming increasingly
part of mass market products, I suspect that these guidelines will
need to be updated.




..andy

To email, substitute .nospam with .gl