View Single Post
  #23   Report Post  
Posted to uk.d-i-y
John Rumm John Rumm is offline
external usenet poster
 
Posts: 25,191
Default 6mm^2 T&E in conduit; now includes RCBOS

Fred wrote:
On Thu, 10 Apr 2008 02:11:54 +0100, John Rumm
wrote:

With the RCD in the house, you make it difficult to avoid having the
feed to the freezer sharing a RCD with other devices, hence the
implications of a trip are worse.


Thanks. It seems that there are pros and cons to each way.

The advantage of having the RCD in the garage is that you can wire to
keep the freezer on if something else trips. However, I wonder about
moving the freezer inside when I get round to decorating the kitchen
and hbow often do RCDs trip anyway?


On a properly installed setup without any faulty appliances - almost
never. But you can't always rely on the no faulty appliances bit.

The advantages of having the RCD in the house a
1 the whole cable is RCD protected in case of accident


If you want this (or need it due to poor earth fault loop impedance)
then the solution is to include a 100mA trip RCD with time delay at the
head end as well as the more sensitive RCD at the destination. That way
you have earth fault protection for the cable[1] and personal shock
protection in the garage.

[1] Note that if you use SWA with earthed armour, you have this anyway
in most cases with TN head ends.

2 if there is a trip, I would not have to manually wind up the garage
door to reset the RCD... of course, that assumes I am not in he
garage when it happens. If I were in the garage with the door shut
when it tripped, that would be different again!


In the garage RCD but not protecting the door motor would seem to be the
optimum arrangement.

The fuse in the house is also highly unlikely to trip if sized
appropriately for the submain.


The fuse would be 30A for 6mm^2T&E but I would connect that to a 2-way
CU in the garage with smaller MCBs, only the garage CU would not have
a RCD, so the garage circuits would still be fused appropriately.


With 6mm^2 you could fuse at 40A if required. Ought to leave ample power
available for anything you are likely to do in the garage.

I suppose that I also need to measure the length of the cabel run to
make sure that if the RCD is in the house, all the disconnect times
are still within limits?

The only time is it likely to be a factor is if your earth loop
impedance is marginal in the garage. In which case you would be better
off making the garage a TT install in its own right.


I can't remember how long the cablet o the garage would be exactly but
definitely less than 10 metres. It's just I like the idea of that
cable being RCD protected in case someone knocks/drills/drives into
the conduit carrying it. How resilient is heavy duty plastic conduit
anyway?


At 10m, even with a TN-S head end you really don't have a problem. The
Live to earth round trip resistance of 6mm^2 is 10.49 mOhm/m so about
0.1 ohms total. Add the worst case assumption for your main earth
impedance of 0.8 ohms and you get 0.9 total. Lets say you nail right
through the end of the cable, you will get a fault current of 230/0.9 =
255A. Referring to fig 3.3 in BS7671 (16th) that shows a time of 0.4
sec to open a 32A fuse, and probably under 1 sec for a 40A one.

Finally a check to ensure the CPC in the cable will last at least a
second and so long enough during fault conditions - we need s mm^2 of
CPC where :

s = sqrt( 255^2 x 1 ) / 115 = 2.2 mm^2

Your cable will probably have a 2.5 mm^2 CPC, so not a problem there. In
reality this is a pessimistic calculation since your external fault loop
impedance is probably better than 0.8 ohms and hence the fault current
bigger (which will dramatically shorten the time to clear the fault and
hence the size of CPC required will go down)

For more on this, have a read of:

http://wiki.diyfaq.org.uk/index.php?...n ection_time

The heavy gague plastic conduit is reasonably tough - clomping it with a
hammer is unlikely to damage the cable inside. You could drill through
it if it was buried in plaster.



--
Cheers,

John.

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