In an earlier contribution to this discussion,
CliveM wrote:


From what I can make out (and I think I understand it now) through
the maze of plumbing, what happens is a follows.

When the pump is running, the CH return pumps from its 22mm into a T
to the 28mm which then goes (Boiler Return ^ HW). This is resulting
in 1/2 the CH return being pumped to the boiler, and 1/2 to the HW
(which then becomes the Flow). When the HW pipes return it arrives at
the 28mm T with the 22mm CH Flow comming off. As the HW return cant
move the against the pumped CH flow, what happens is that the HW
return water get "sucked off" the T piece by the pump (i.e CH FLOW ^
HW RETURN) so CH flow is a combination of Boiler Flow and the HW
return. (or to make it simple, CH returned water is pumped as HW
flow, and HW return if then pump round the CH circuit as CH flow)

This then results in HW being passed through the cylinder from the
bottom up but is obviously pumped when pump is on.

And you wonder why I am confused about all this. Anyway this now
seems to have resulted in that we do get HW when CH on (and it must
have always been like this), but that I have just been confused over
the last 3 days! I suspect that there was a load of air in the system
which caused the initial problem. (you could hear it in the pump, and
boiler cut out a couple of times), which now seems to have cleared.

Time to put the tools away (I hope) - thanks for all your help.

Are you saying that it's now working ok?

Its not a very good idea sending "spent" CH water round the HW circuit,
because it will already have cooled a bit, so HW recovery will be slow. At
the point where the return flow from the CH rejoins (what is supposed to be)
the return flow from the HW, it might be an idea to use a sweep tee rather
than an ordinary one. If fitted with the sweep in the right direction it
would be far harder for the water to go the wrong way.

How near the boiler is the point where the circuits split? Does your boiler
have 4 connection points, two of which are currently blanked off? If so, and
if the split is fairly near the boiler, you could consider having two
totally independent circuits all the way back to the boiler with no shared
pipework. That should definitely fix the problem.

Alternatively, you could convert to a fully pumped S-Plan system. You would
need an additional, zone valve, and you would have to move the pump into the
shared bit of pipework.
--
Cheers,
Set Square
______
Please reply to newsgroup. Reply address is invalid.