Graham. wrote:
I remember reading in "Radio Communication" decades ago
about a fellow Amateur whos transmitters' mains fuse
would blow every night, but only if he was transmitting
at the time. The thing he noticed was it blew with a strange
regularity, each night a short time later than the previous
night. He could draw a graph of his observations.
I'll stop there and ask if you can see where I am going
with this...
Sounds like he had something that kicked in on a time switch, which
combined with the Tx load was enough to blow the fuse. Each blow of the
fuse resulting in the clock stopping, and hence subsequent switchings on
of said load would be slightly later each time.

No.
It's a long time since I read the article but AFAICR it was something
the electricity board or CEGB were putting on the supply to command
street lighting (solar timing). They did it by somehow superimposing
DC on the mains for a short time. Does that sound feasible? I'm sure
it was something like, that and it took a lot of determined investigation
to get an admission of what was going on.
Apparently the fuse blew because the tranny was operating near saturation
and the dc component was the last straw. He noticed increse in vibration
from the transformer just before it blew its fuse.

I take Tony's point that he may have been pushing things too.

Sounds like an urban myth. Switching of E7 and similar loads is
controlled by a phase modulation of 198KHz radio 4. If you have a big
radio, try tuning into R4 with the BFO switched on...

James

coughed up some electrons that declared:


At 10x rated current a type C can trip in as little as 0.0015 seconds.
I doubt it would do. A wire fuse is closer to a D type, which is
orders of magnitude slower at 10x rated current.


NT

I don't have my books to hand, so I'll not argue. But if that is correct,
then the conditions for putting in a Type D breaker or something with a
similar trip curve are considerably more onerous in terms of having a low
loop impedance on the local circuit *and* on the supply (ie in the event of
a dead short L-E fault, would the breaker or fuse clear before the cable
damages itself through overheating?).

It would be quite possible to have a perfectly ordinary and good
installation where adding a type D to a random circuit would make it both
non compliant and dangerous (particularly from risk of fire, but also shock
if the breaker/fuse cannot reliably clear a fault.

However, this does change if there is RCD protection (don't think that was
mentioned) - but even so, personally, I would want to then ensure that a
L-N fault would clear in the expected times, which is an easier target to
meet, but not necessarily a forgon conclusion.

I'd be cautious about jumping straight to something with that sort of
characteristic without risking a few quid on the middle option first
(there's always ebay if a Type C doesn't work out).

IIRC (memory poor, open to correction) a Type C is several times less
sensitive on the instantaneous trip part of the curve compared to a Type B,
so there's a good chance it would buy enough margin.

I'm happy to come back tomorrow and reel off some numbers, or I think if one
digs into Hager's literature, the trip curves for most of their devices are
in there somewhere. Might have a rummage later...

Cheers

Tim

"Dave Plowman (News)" wrote in message
...

Snag is it's on a landing and two way switched - and each of the two way
switches are 2 gang.

You put the dimmer after the switches and leave it on full.
A soft start dimmer will just limit the current when you switch on and do
sod all once on.

Tim S wrote:
coughed up some electrons that declared:


At 10x rated current a type C can trip in as little as 0.0015 seconds.
I doubt it would do. A wire fuse is closer to a D type, which is
orders of magnitude slower at 10x rated current.


NT

I don't have my books to hand, so I'll not argue. But if that is correct,
then the conditions for putting in a Type D breaker or something with a
similar trip curve are considerably more onerous in terms of having a low
loop impedance on the local circuit *and* on the supply (ie in the event of
a dead short L-E fault, would the breaker or fuse clear before the cable
damages itself through overheating?).

It would be quite possible to have a perfectly ordinary and good
installation where adding a type D to a random circuit would make it both
non compliant and dangerous (particularly from risk of fire, but also shock
if the breaker/fuse cannot reliably clear a fault.

However, this does change if there is RCD protection (don't think that was
mentioned) - but even so, personally, I would want to then ensure that a
L-N fault would clear in the expected times, which is an easier target to
meet, but not necessarily a forgon conclusion.

I'd be cautious about jumping straight to something with that sort of
characteristic without risking a few quid on the middle option first
(there's always ebay if a Type C doesn't work out).

IIRC (memory poor, open to correction) a Type C is several times less
sensitive on the instantaneous trip part of the curve compared to a Type B,
so there's a good chance it would buy enough margin.

I'm happy to come back tomorrow and reel off some numbers, or I think if one
digs into Hager's literature, the trip curves for most of their devices are
in there somewhere. Might have a rummage later...

Cheers

Tim

10x rated i is a reasonable ballpark for a toroidal startup surge.
That's 12.5A in this case, which is only just over 2x the MCB's rated
i - I was thinking 10x MCB rated i, so you may well be right with a
type C.


NT

coughed up some electrons that declared:


10x rated i is a reasonable ballpark for a toroidal startup surge.
That's 12.5A in this case, which is only just over 2x the MCB's rated
i - I was thinking 10x MCB rated i, so you may well be right with a
type C.


NT

Aye - I was more concerned about randoming banging Type D's - or something
equally thick skinned - in without a more detailed test of the installation
(loop impedances mainly), which is why I was discouraging it.

In article ,
dennis@home wrote:
Snag is it's on a landing and two way switched - and each of the two
way switches are 2 gang.

You put the dimmer after the switches and leave it on full. A soft start
dimmer will just limit the current when you switch on and do sod all
once on.

There are dimmers with two way switches - so assuming you leave it full up
should be ok. But I'm not even going to suggest this to them.

--
*When did my wild oats turn to prunes and all bran?

Dave Plowman London SW
To e-mail, change noise into sound.