On Jul 11, 11:37*am, Tim Watts wrote:
Matty F
* wibbled on Sunday 11 July 2010 10:41

On Jul 11, 9:44 am, Tim Watts wrote:

Having looked at the prices of the pipe boxing, and the corner radius, it
might well be better to try to fashion something out of bull nose
skirting with another bit glued on.

Why not make the boxing out of MDF and use a router to make the edge
round?

No router...

But having looked at Wickes, a bit of bullnose MDF architrave and a bit of
bullnose MDF skirting should give me the required dimensions with no cutting
(except length).

I reckon I will make up a number of troughs (the hardest bit getting good
alighnment of the 2 pieces to make the join invisible).

Then I'll treat it like coving and run it all round with mitred joints after
I've calculated the best height to give a good ceiling wash without the
tubes being visible.

I've come to the conclusion that

a) I can't afford dimmers (that get's expensive fast)

Its also relatively easy to select multiple brightness levels when you
run a tube way below rated power (with filament heating to
compensate), this is semi-easy to do using large tubes, which makes
the whole setup even cheaper. And since you can in such a setup start
tubes at reduced brightness, low cost magnetic ballasts are more ok.


b) I'll standardise on 2' T8 tubes. There are 3 short sections that would
benefit from 2' tubes. 3 x 2' tubes for the 8' wall and 4 tubes for the 10'
wall.

There are Philips HF gear units that drive 3-4 2' T8s at no more expense
than the same type of unit for 2 x 4' tubes. Then I'll only need one type of
tube in the spares box and it will be nice and short for easy storage.

I'll just have to take a chance on the brightness, but as its sole function
is to add "daylight mode" to the room, with a GU10x4 spot for the main
lighting, I think it will be fine.

Cheers

Tim

To be honest, brightness is one of the biggest deals, and needs to be
right. Suggest a little experimenting to make sure you get close. You
can use filament bulbs and translate the power required using the
different efficacy values for filament vs tube.


NT