On Tue, 16 Jun 2020 20:47:41 +0100 (BST), in uk.d-i-y "Dave Liquorice" wrote:

On Tue, 16 Jun 2020 16:02:12 +0100, Phil Addison wrote:

The system is highly complex ie not just a pair of pipes
that goes from one rad to the next. There are rads on long spurs,
with different lenghts/bends for the pipes, spurs with rads
connected
overlapping and rad acroos the main pair.

I don't see how that invalidates the procedure, ...

But note that in some cases this will be impossible due to bad design of
the pipework ...

Perzackerly.

Thermal imaging camera allows you to see how much flow is (or
isn't!)
going through a rad by the size of the rising hot water column
from
the inlet. This also responds very quickly to any changes.

I haven't had the chance to use an IR camera, and can see it would be
helpful to see the immediate effect of a small turn of the valve, but I
emphasise that you don't want full flow in all the rads.

Maybe not but it enables you to setup an even starting point
relatively quickly and painlessly.

A common misconception is that increasing flow-resistance will make a
radiator cooler. Not necessarily: for a rad already getting as much heat
as it can dissipate, closing its valve a bit just makes the pump
increase its pressure so that rads further away get more flow.

No the rads with a lower flow resistance get more flow, they may or
may not be further away.

Yes, "rads further away" wasn't a good choice of wording. In fact ALL the other rads get a bit more flow because the total system resistance increases when you restrict flow in any radiator, thus driving the pump a bit further up its pressure/flow curve. This in turn causes a bit more flow in all the other rads.

Phil