The message
from Fred contains these words:

I'm having most problems with the towel radiator in the bathroom. It's
scalding hot to touch but not showing much of a temperature drop.
Should the drop across a towel radiator be the same as the
"conventional" radiators?

I have not had any direct experience of towel radiators but as towel
radiators typically have a much smaller surface area than even a small
conventional radiator it may be difficult to find a point between little
temperature drop and practically off.

How accurate should I be in matching temperature drops. Should I try
to get each rad. to have a drop of say 10C, or does it matter if one
has 8C, another 12C, another 9C, etc?

The object of the exercise is to get the specific heat output of each
radiator to match but I wouldn't have thought at a 10% variation (or
perhaps even more) in temperature drop would be a problem and I suspect
that getting any closer would probably be an illusion. The larger the
radiator the larger the flow through the radiator for the same
temperature drop which is why you may have trouble adjusting a dinky
towel rail.

How long should I wait for things to stabilise before re-measuring?

Probably about 10 minutes - but you need to make sure that the system is
running the whole time - so turn any room stats up to max, and open some
windows if necessary.

Daft question: does the boiler need to be firing? Or as long as the
pump is circulating hot water is that enough?

Hobsons choice (unless you have a sophisticated modulating boiler). If
the boiler is firing the circulating water is being heated and the flow
temperature continuously rising and if the pump is on over run the
circulating water is being cooled. I think what you must avoid is the
warm-up phase below normal operating temperatures, any period during
which the water isn't circulating and any period shortly thereafter. If
the pump is running and the flow temperature is high enough you will
then only have the problem of finding a period when the inlet and outlet
temperatures of a particular radiator aren't varying fast enough to
invalidate your results.

It may be that it doesn't matter that you are getting a higher (or
lower) than true temperature drop across a radiator provided you
consistently opt for the firing or the over run phase or it might be
better if you take measurements from both phases and average. If my
central heating system is any guide I think you may have some difficulty
in finding any period where both inlet and outlet temperatures of a
radiator are stable.

--
Roger Chapman