On 15/05/2021 11:27, Theo wrote:
John Rumm wrote:
It was quite common to have a big cast iron lump HX with high water
content (couple of gallons). That would have 4 bosses on it - two
outputs and two returns. One set would be used for the feed and return
on the pumped circuit through the rads, and the other two would be used
for a gravity (i.e. convected circuit) to heat an indirect cylinder.

Ah, so 'gravity' meant there was no pump on the HW side, only for the heating?

Yup, DHW looked after itself, and the pump only served the rads.

(thermosyphon systems could also run with solid fuel heating boilers,
where the inability of a pump fault or electrical failure or power
interruption to affect the operation of a system that could not easily
be "turned off" was seen as an important benefit)

The convention circuit required that the hot water cylinder would be
placed some distance above the boiler and not too far displaced from it.
It would also typically be piped in 28mm pipe to better allow convection
flow. Hot water from the gravity output would rise up through the pipe,
and through the coil in the cylinder, the heavier cooler water would
fall and return to the boiler. Basically it sets up a thermosyphon.

That's neat. Horribly inefficient, but neat

Inefficient yes, although possibly not that much less than some pumped
systems that could only extract say 3 to 5kW of power from the boiler to
the cylinder via a short indirect coil. Those would have to short cycle
the boiler to load match its much higher output to the input of the
cylinder (or run it in parallel with the rads to present enough load -
hence the popularity of Y plan systems)

That sounds about right - cylinder was above the boiler and about 2.5m
displaced, so that would probably work.

Some refinements of the system introduced and anti gravity valve to stop
heat being lost out of the cylinder through the boiler when it was off.
Some had a kind of thermostatic valve in the gravity loop to limit the
maximum temperature of the cylinder DHW. Some would add a motorised
valve to make it a fully controlled zone. (aka C plan)

I never investigated but I imagine there was just on/off control for the
boiler (gas valve only, since there was an always-on pilot that had a piezo
spark-button to relight it if it went out), and an additional time clock
output that was in series with a thermostat to run the pump for the heating.

The boiler would have its own internal water temperature stat that would
interrupt power to the gas valve when the set point was reached (there
was often a knob on the boiler to tweak that set point).

How was the boiler turned off in this system, when the thermostat said the
house was hot enough and so was the water? If the heating called for heat
and the hot water was up to temp, surely that would overheat the water?

That was one of the limitations - there was no system interlock as such
that could shut down everything once the cylinder was up to temperature.
Most programmers of the era would usually just run a timed heating cycle
for the cylinder. If heating the DHW alone, the boiler would run and
cycle. As the cylinder came up to temp, the rate of heat transfer would
fall, and the boiler cycle would shorten. However if not turned off with
the programmer (or manual override) it would carry on indefinitely.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/