On 15/07/2022 14:53, Theo wrote:
> John Rumm <see.my.s...@nowhere.null> wrote:
>> I seem to recall the late great Andy Hall described exactly the process
>> that he used to capture and re-inject antifreeze rich CH water in these
>> very pages...
>>
>> He had a secondary circuit that fed his workshop - and hence went
>> outside. I think he used a modified "pump up" garden sprayer to reload
>> the (sealed / pressurised) loop.
>
> Hmm, good idea. I think I have one of those somewhere...
>
> One more question. Supposing I want to change the pipework to a ground
> floor rad. I depressurise, close the valves to the rad and disconnect them.
> If the system is sealed, all the water stays in the pipes. But maybe I have
If you turn off both rad taps, then you can remove the rad and lose no
water at all.
If you then cut off one of the tails, then you will get some spillage
from the pipe. If you leave a cut end open, then air will also flow into
the cut end, and you will get ongoing slow loss of water. The key to
working on a live system, is speed - you make a cut and get it sealed
again quickly!
The lack of stored pressure, and hydraulic lock removes the torrent of
water escaping from the pipe in the "oh shit the pipe has burst!" panic.
It also limits the pressure to something that can be blocked by hand
easily. That does depending a bit on the overall height of the property
- a three storey place might have 30' of head, which will be 1 bar/15psi
of static pressure at the lowest point in the system (dynamic pressure
will be less since a good hydraulic lock will reduce that dramatically
when water is flowing))
> an air leak somewhere - some fitting is sealed when pressurised from the
> inside, but with the pressure let off maybe air can get in. Will that cause
> the system to siphon everything out from my removed rad? Or will it just
> amount to the run that's above my removed rad, and the ground floor room
> next door would be unaffected.
If there is a path for air into the system, then water will continue to
come out. A small leak as you describe is not really that relevant,
since you won't be just cutting off a pipe and leaving it for weeks.
> One example I'm thinking of is air release valves at the top of pipe runs.
> When depressurised will they let air *into* the system, or are they strictly
> release only? (I think there's often a knob to close them, but curious what
> happens if they aren't closed).
You might want to watch out if you have any automatic bleeding valves -
that may allow air to suck in when there is no pressure on the "inside"
if they are left in the open position. Normally they can be "closed" so
they are sealed completely.
> I *think* the siphoning would depend on where the air leak is: it would
> siphon all of the pipe between the removed rad and the leak, assuming the
> leak is higher up (if not, it would siphon out of the leak location
> instead). But curious to know how it goes in practice, and whether this is
> actually a problem.
Much will depend on the pipework layout, and also if you have rads open
or turned off. With all the rads turned off, you could actually drain
all/most of the pipework, while still retaining most of the system water
in the rads. (you would need to slacken a rad union nut near the top of
the system to let air into the pipes)
In places with solid ground floors, they often do all the pipework under
the first floor, and drop down to each rad. Then it is actually very
difficult to fully drain the system, since even at the lowest point
there are other independent low points that can't drain by gravity into
the bit with the drain point (hence why some systems have drain points
on most downstairs rads)