At the weekend I drained the central heating and refilled it because I
was getting some simmering sort of kettling noises from the boiler. This
fixed that noise, it's really quiet now, but sometimes when the boiler
is running on lowest speed (Keston Celsius 25, which can vary its pump
speed) I get a repeated banging noise from the pipes (water hammer type
noise) where they pass through the floor upstairs. There is no noise
from the boiler itself. I can hear a slight swishing in the pump as it
runs at low speed and the bangs match up with the swish. I am wondering
if I have introduced an airlock somewhere (as it did not do this
before), and the compressible air is causing some sort of pressure wave.
The pipe run under the floor is not fixed at all (crap Barratt
handiwork), sounds like it is twitching and banging the joists or the
Everything is getting hot as expected, so it's not too serious, any
ideas on what to do? My thoughts are-
a) drain it again (don't really want to do this)
b) try to clear the (possible) airlock somehow
c) clip the pipes down properly
Does the pressure gauge on your system fluctuate in time with the swish?,
mine does. I have the same boiler and have been fighting an instability in
the flow since installation. My thoughts so far are that the flow from the
pump is not as smooth as it could be - pulsating somewhat. If there is
anything spongy in the system eg. air, then this can turn the pulsing flow
into an oscillation which in some circumstances can build to a hammer. In
these circumstances, my pressure gauge will be doing it's nut!
In my case I found that the pulsating pressure was being stored (and
released) in the expansion vessel causing the TRVs to bang open & shut. I
believe my problem lies in some a couple of subtle design issues, which
don't help you, but it suggests that the keston system is susceptible to
All I can suggest for you right now is looking for air somewhere. Also,
when you hear the hammer, go round tweaking and listening at the TRVs
to see if it is actually one of them making the noise - a starting point at
My own design problems have been patched temporarily by adding a flow
restrictor to the expansion vessel and by adding a slightly open manual
Mine is open vented, not a sealed system, so there is no pressure gauge,
and I don't have any trv's either! However I believe that you are
correct, that the pump is pulsed at low speed and it's that pulsing on a
pocket of air which is causing the hammer.
I think the air, if there is some, is probably in a long horizontal run
of pipe which goes from the boiler at the back of the house, underneath
the upstairs floorboards to the airing cupboard where the motorised
valves are. Can't seem to shift it though and it is not preventing the
An alternative view is that there was air before I drained it, which was
damping the system, and now there isn't any the pump is making the pipes
twitch. In this case I can probably resolve it by fixing the pipes down.
Ok, have you tried one rad or branch on at a time at max demand?, max
flow in a branch may shift your bubble to a rad for bleeding. Sure, boiler
cycle won't stay at max for long on only one rad turned on, but try
repeating. Also try one at a time with lockshields fully open - sorry,
Have fixed the pipes down, has reduced the volume of the banging but it
is still doing it.
There are some funny things about the way it's piped. I think I know
where the airlock is, there is a 15mm pipe which forms the return from
the HW cylinder and the manual bypass, this has a high spot in it
because of the way they have connected it. This 15mm pipe runs
separately, alongside the 22mm return pipe from the CH radiators, then
joins together in the floor above where the boiler is. I don't
understand why it's done like this, and not simply joined into the
return where it passes over the top of it. The banging only happens when
the system is heating HW and not radiators which seems to bear out my
theory of it being in the return from the HW cylinder coil.
I have drawn out the system here:
Can anyone illuminate me as to why that 15mm pipe is like it is? And how
I can get the air out of the pipe, other than by cutting it and fitting
a bleed point (I am tempted to do this, but it means draining it down
again). I have tried turning everything else off to send max flow down
that pipe but it has made no difference.
I can see no need for the separate return. Is that a lockshield in the return
path of the HW circuit and how is it set? In order to stabilise mine, I have
to throttle the HW loop quite a lot, but not so much that HW perf is
compromised. Could you try opening the lockshield just a tad (from
closed) & see if your banging is damped - I don't think the keston pump
control system likes to see an unrestricted loop.
The circles with Xs are gate valves. I haven't tried messing with that
one, I will work out its current state and try some different settings.
It is perfectly happy if the central heating valve is open. It bangs
worst if both CH and HW are open, then the CH valve closes, which makes
the boiler reduce its output to 1 or 2 lights, then it bangs for about
15 seconds, then it settles down and all is peaceful again.
Sounds exactly like my situation.
There are parallels here between fluid theory & electronic control theory
which I could bore the arse off you with, but suffice to say resistance to
flow damps oscillation. If that pump control loop is borderline stable, then it
may need some flow restriction to be guaranteed stable. The CH loop is
easy cos it has lots of little restrictions - long pipe runs and radiator
lockshield valves turned down to next to nothing. I think you will find that
closing down the HW loop flow a bit with the gate valve will improve the
situation. My 22mm valve is only open a half turn (of 8) for stability. I regard
this fix as a patch only & will be having words with Keston technical soon
in search of a permanent solution.
We had this on our church Celsius boilers installed August
2001. A Keston engineer came down to fix something else under
warranty and noted this, and changed the PCB's to a later
model (no charge). I'm not sure of the exact technicalities
but the system picks up that the pump is running too fast or
slow and corrects this. It then decides that the new speed is
too slow or fast so goes back to the original. And so on ....
Tony Bryer SDA UK 'Software to build on'
At a guess I would say that the pressure needle was oscillating
at something like 3-5 times a second: I don't recall any
particular noise but our two boilers feed through a balance
header rather than through the heating circuits.
Tony Bryer SDA UK 'Software to build on' http://www.sda.co.uk
Free SEDBUK boiler database browser
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.