It could indicate a problem with the pressure vessel - but it may be one
which you can fix yourself.
The pressure vessel has a diaphragm inside with (compressible) air on one
side and water on the other. If there is insufficient air in it, the water
has nowhere to expand to when it gets hot - resulting in the pressure rising
too much and the safety valve opening, causing the pressure to drop.
The vessel should have a Schrader valve (like a car type valve) at which you
can measure and top up the air pressure. With the water system
unpressurised, the air pressure should be about 0.7 bar (~10 PSI). Check it
with a car tyre pressure gauge, and top it up with a car tyre pump if
necessary. If water comes out of the Schrader valve, the diaphragm is shot
and the vessel will need to be replaced. Otherwise, just top it up as
Then use the filling loop to pressurise the water system to 1 bar when cold.
Then, when the system heats up, the pressure should rise to no more than
about 2 bar - and certainly not enough to trigger the safety valve, which
operates at about 3 bar. When it cools down, it should return to 1 bar.
The fact that there is a switch for water heating indicates that it is
unlikely to be a combi. A combi boiler produces hot water on demand, so is
"always on" and requires no switch.
My guess is you have a three port valve that is stuck. It may be a mid
position valve, and the valve has been set manually to the mid position
(many valves of this type have a leaver on the side that will do this.
(you neglected to mention if the hot water also got heated)
It appears that you may be right. If I move the valve lever to off, then
turn the HEAT switch on, the valve lever doesn't return to on. However, it
does work the other way automatically. You can hear a click when it doesn't
move, indicating the wall timer is sending the correct current.
Is the simple solution a new valve or could this possibly be anything else?
The other strange thing as mentioned in my above post, is that only the
WATER switch turns the system on. E.g.:
Water off = gas boiler off, valve off, pump off
Heat on = nothing
Water on = gas boiler on, valve on, pump on, radiators on
Any ideas what *that* problem could be?
It sounds as if you have a Y-Plan system, with a 3-port mid-position
motorised valve. Your system is probably wired incorrectly. You can find a
wiring diagram in the Y-Plan section of
Pay particular attention to the connections to the grey wire on the 3-port
valve. This needs to be live when HW is either switched off at the
programmer or satisfied at the cylinder stat. So there must be a HW-off
connection to the programmer, and the cylinder stat must have change-over
contacts - with the one connected to valve-grey still live when the water is
hot. In this condition, it's the valve which switches the boiler on via the
orange wire, but it needs grey to be live to be able to do this.
If the wiring *is* ok, you could have a faulty microswitch inside the valve
actuator. It sounds from an earlier post as if the actuator may be suspect,
anyway. With most makes of valve you can replace the actuator without
disturbing the wet part of the valve.
OK, I just had a look at the install manual for you boiler. Looks like
we may be talking at crossed purposes!
You boiler appears to be designed for gravity circulation for the hot
water heating (i.e. there is a separate feed and return connection on
the boiler for connection to the hot water cylinder). The central
heating is handled separately.
With these types of system you don't really get fully independant
control of heating and hot water - however it does depend on how it was
all plumbed in the first place. Some boilers of this type can be
configured to run fully pumped while simply ignoring the gravity
circulation. However usually any time the boiler is on there will be
heating to the water via non pumped circulation.
This would suggest that your motorised valve will be a two port valve
(i.e. only two pipes in and out and not three). Often positioned close
to the hot water cylinder. This valve is normally driven from the
cylinder stat such that it closes when the cylinder is up to temperature.
You would expect water on to give boiler on, pump off and valve under
control of the cylinder stat.
Heating on would give boiler on, pump on (or possibly under control of
the room stat), valve under control of cylinder stat.
We need to understand a bit more about how your system is connected up -
what pipes go where, and what control wiring there is.
However in the simplest configuration you would only have simple "on" or
"off" control of your system (and the fact that the *programmer* can
cope with independant water and CH could just be a red herring!). The
pump would be controlled by the room stat, the valve by the cylinder
stat, and the call for heat signal to the boiler would be generated by
the combined outputs of the stats in serise with the timer.
Yes it is rather confusing... the boiler the OP provided a link to
provides for gravity circulated hot water - although the talk of
pressure guages etc does suggest a sealed system. It is possible that it
was plumbed as a sealed system and the gravity circulation option simply
ignored (or it might be the link posted was to the wrong boiler! ;-)
Three port valve fully pumped is what I initially assumed... perhaps if
the OP could look at the valve and count the pipes connected to it?
Whatever the case, it sounds like this is a wiring / valve related problem.
Hi, definitely a sealed system, at least that's what it says on the tin (on
The valve has one inlet and 2 outlets, the pipe going up is for Heat, down
is for H/W.
1) Why does the HEAT switch on the wall timer have no effect, and only
the H/W switch turns the boiler and radiators on?
2) I note that the pressure does slowly drop, and that there are constant
slow drips coming from the hooked pipe outside the property, connected to
the boiler. What does this indicate, and how is it fixed?
Many thanks guys!
Did you see my earlier post about the necessary wiring for a Y-Plan system?
That explains it.
Is this water - leaking from the safety relief valve? If so, the valve has
probably got some crud in it and needs cleaning out or replacing. Or, if you
have a condensing boiler, it could be condensate - which is normal, but
wouldn't explain the pressure loss.
You said "The valve has one inlet and 2 outlets, the pipe going up is for
is for H/W." In my book, that's a pretty good description of a 3-port
mid-position valve as used by Y-Plan. [OK, it just *could* be a diverter
valve as used by W-Plan systems, but that is less likely].
I can't speak from first-hand experience on this one - so hopefully someone
will be along who can. But I've heard it said that once a relief valve
opens, it's difficult to get it to seal again. ISTR that you left your
filling loop connected, causing the system to over-pressure and blow off.
Right, ok now we are getting closer ;-)
There are two types of three port valve. A diversion valve and a mid
position one. The diversion valve can be in one position or the other -
not between both. Typically the hot water is given priority such that if
the cylinder is calling for heat, then it will service that (and only
that) output. Works well with rapid recovery cylinders since they can be
reheated in short order and the valve then returns the flow to the heating.
The second type is a mid position one. This has the capability to split
the flow between both outputs when it gets concurrent requests for CH
and HW. Better on slower recovery cylinders since it prevents the
cylinder disabling the CH for long periods.
Two possibilities: 1, it has not been wired / setup correctly and hence
would never have worked, or 2 something is broken! ;-)
The first task is to eliminate 1, which means some detective work to see
how it is wired up. Looking at the outputs from the timer will be a good
start - they ought to be labelled with their function (failing that, a
manual may be downloadable from the web that will tell you there function.
Once you are happy that the system *could* work, it is a case of working
through each stage with your multimeter and seeing what is wrong.
e.g. verify that when the programmer calls for CH that a signal does in
fact come out of it.
Verify that the room stat does call for heat when required.
Verify these signals reach the three port valve.
Verify that it moves correctly in response to the signals.
Verify that the micro switch outputs (if present) on the valve that
indicate its position are working, and hence if any call for heat that
should be passed to the boiler via this route does in fact arrive.
 remember that many of these signals are likely to be at mains
voltage - so take care. If you don't feel confident that you can do this
without electrocuting yourself then this would be the time to consult a
competent heating engineer.
It sounds like your pressure relief valve is failing to seat correctly.
This can happen when a valve is forced to operate (under normal
circumstances it will never open). You overfilling exercise may have
caused this. Clean or replace the valve I guess.
Also I have now identified that the one thermostat in the hall has no effect
on the central heating whatsoever. That said, all the rooms (except hall
strangely) have a stat on the rad itself. Does this negate the need for a
central stat? (Surely not.)
So, to recap, current issues:
1) CH switch on/off has no effect either way. HW switch makes radiators
and water hot just fine.
2) Boiler loses pressure slowly.
3) Wall stat has no effect on system. All rads have stats on except
Does this help us? :-S
The valve is a Danfoss 3-port mid-position valve designed to provide just
HW, just CH - or both at the same time. This is the heart of a Y-Plan
Even if most radiators have TRVs (thermostatic radiator valves) you still
need a central stat to turn the boiler off when the whole house is hot. The
radiator nearest to the room stat shouldn't have a TRV - so what you've got
As John Rumm told you many days ago, you have a problem with your
mid-position valve. The chances are that the 'wet' part of the valve is
stuck in roughly the mid position, but not far enough round to operate the
microswitches in the actuator. In this condition, the boiler and pump will
be controlled by the HW output from the programmer and the cylinder stat -
but whenever the boiler is firing, both HW and CH will get hot. The CH feed
from the programmer - which goes via the room stat - will have no effect
because that only controls the boiler *after* the microswitches in the
actuator have operated.
Remove the actuator from the valve (2 screws). This will reveal the valve
operating spindle, the end of which is D-shaped in cross section. The flat
of the D will almost certainly be at an angle of about 45 degrees with the
body - which is the mid position. See whether you can rotate the spindle
through plus and minus 45 degrees. In the unlikely event that it's *not*
seized, you should be able to turn it with finger and thumb or - at any
rate - with light pressure from a pair of pliers.
Then, with the valve in the mid position, and the actuator still off, put
the actuator through its paces. Use the programmer and thermostats to
generate all possible combinations of HW and CH demand and see what the
actuator does in each case.
Then report back!
Sounds like you've got a leak, like we said earlier.
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.