Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Illuminated switch to show that cloakroom light and fan is still on

315 views
Skip to first unread message

Phil Addison

unread,
Sep 5, 2009, 11:25:46 AM9/5/09
to
Any suggestions for a source of illuminated light switch as a reminder
that the light it feeds is still on? We have a loo under the stairs with
the light switch on the outside and it sometimes gets left switched on
so the fan remains running (as well as the light being on). I don't want
to fit a clunky 20Amp DP mains switch like this
http://www.fastlec.co.uk/mk-20a-dp-switch-bottom-flexneon-k5423d1whi-p-538.html
or this
http://www.fastlec.co.uk/mk-20a-dp-switch-water-heater-k5423whwhi-p-539.html

What I really want is a basic light switch like this
http://www.fastlec.co.uk/mk-10a-1g-sp-1-way-plateswitch-k4870-whi-p-511.html
but with a neon indicator incorporated. I can't find such a thing, the
nearest being this http://www.tlc-direct.co.uk/Products/DT2503A.html but
it works the wrong way round - the illumination is 'on' when the switch
is 'off' - its so you can find the switch in the dark. I suspect the
wiring could be reversed, but as it's intended to show up in the dark it
may not be bright enough for use in a lighted area.

I also found this
http://www.fastlec.co.uk/mk-neon-locator-200250v-50hz-k3041-p-496.html
which is an illuminated section that sits between the switch and the
back-box. Again I'm not sure if its bright enough and it seems a bit of
an overkill.

I'm considering modifying a standard light switch by drilling a hole
into which a small neon or LED is fitted. There is some safety
consideration here so wonder if there is suitable double insulated
neon/led available. Failing that I think I need to create some kind or
perspex window in the switch with the neon/led behind it.

Any thoughts?

Phil

Steve Walker

unread,
Sep 5, 2009, 12:57:43 PM9/5/09
to

LEDs do not have insulation deemed suitable for mains, unless you have one
with a surround and lens. The other problem is that you don't normally have
a neutral available at a lightswitch, hence you can light a neon or LED
easily enough when the switch is off (using the switched light as the path
to neutral, as the current is too low to light it), but not when the switch
is on.

SteveW

Dave Plowman (News)

unread,
Sep 5, 2009, 1:35:55 PM9/5/09
to
In article <uqu4a5d0rm5u55k36...@4ax.com>,

Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> What I really want is a basic light switch like this
> http://www.fastlec.co.uk/mk-10a-1g-sp-1-way-plateswitch-k4870-whi-p-511.html
> but with a neon indicator incorporated. I can't find such a thing, the
> nearest being this http://www.tlc-direct.co.uk/Products/DT2503A.html but
> it works the wrong way round - the illumination is 'on' when the switch
> is 'off' - its so you can find the switch in the dark. I suspect the
> wiring could be reversed, but as it's intended to show up in the dark it
> may not be bright enough for use in a lighted area.

Snag is light switches don't normally have a neutral available. The type
which is on with the light off gets the neutral via the light - since it
only takes a small current, the lightbulb looks like a short to it and
thus provides the return. I suppose you could run a small neon to earth -
since this is no worse than many SMPS do. But don't tell anyone I
suggested this.

--
*Parenthetical remarks (however relevant) are (usually) unnecessary *

Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.

chrisj...@proemail.co.uk

unread,
Sep 5, 2009, 1:44:16 PM9/5/09
to
On 5 Sep, 18:57, Steve Walker <st...@theend.demon.co.uk> wrote:
> On Sat, 05 Sep 2009 16:25:46 +0100, Phil Addison wrote:
> > Any suggestions for a source of illuminated light switch as a reminder
> > that the light it feeds is still on? We have a loo under the stairs with
> > the light switch on the outside and it sometimes gets left switched on
> > so the fan remains running (as well as the light being on). I don't want
> > to fit a clunky 20Amp DP mains switch like this
> >http://www.fastlec.co.uk/mk-20a-dp-switch-bottom-flexneon-k5423d1whi-...
> > or this
> >http://www.fastlec.co.uk/mk-20a-dp-switch-water-heater-k5423whwhi-p-5...

>
> > What I really want is a basic light switch like this
> >http://www.fastlec.co.uk/mk-10a-1g-sp-1-way-plateswitch-k4870-whi-p-5...

What the OP wants is a "panel neon", e.g.
http://www.maplin.co.uk/Search.aspx?criteria=panel%20neon&source=15 or
LED: http://www.maplin.co.uk/Search.aspx?criteria=panel%20led&source=15
but check that they aren't too deep for the switch box.

However, I wonder whether a neon/LED will be prominent enough in
daylight, or too easy to ignore even if it is.

If the door is visible down a corridor, you could just drill a hole,
or even replace a whole panel with frosted glass, but again it may be
difficult to see in daylight.

A third possibility is to replace the switch by a motion (sorry)
detector as used for security lights.

A delayed-off time switch is another possibility, but would need to be
inside so that it could be reactivated if things take longer than
expected!

Chris

dennis@home

unread,
Sep 5, 2009, 2:42:47 PM9/5/09
to

"Dave Plowman (News)" <da...@davenoise.co.uk> wrote in message
news:5096207...@davenoise.co.uk...

> Snag is light switches don't normally have a neutral available. The type
> which is on with the light off gets the neutral via the light - since it
> only takes a small current, the lightbulb looks like a short to it and
> thus provides the return. I suppose you could run a small neon to earth -
> since this is no worse than many SMPS do. But don't tell anyone I
> suggested this.

As long as the rcd, if present, doesn't trip.

Phil Addison

unread,
Sep 5, 2009, 4:00:35 PM9/5/09
to
On Sat, 5 Sep 2009 10:44:16 -0700 (PDT), chrisj...@proemail.co.uk
wrote:

Thats the one! thank you! enclosed and fully mains rated. The box is
51mm deep from front of switch so the 29mm lamp will fit easily.

> LED: http://www.maplin.co.uk/Search.aspx?criteria=panel%20led&source=15
> but check that they aren't too deep for the switch box.
>
> However, I wonder whether a neon/LED will be prominent enough in
> daylight, or too easy to ignore even if it is.

We walk past it often enough so I think the red neon will be fine.

> If the door is visible down a corridor, you could just drill a hole,
> or even replace a whole panel with frosted glass, but again it may be
> difficult to see in daylight.
>
> A third possibility is to replace the switch by a motion (sorry)
> detector as used for security lights.
>
> A delayed-off time switch is another possibility, but would need to be
> inside so that it could be reactivated if things take longer than
> expected!

I did have the switch inside originally, but of course its dark in there
until you find where the switch is, so moved it outside, where it does
seem more 'convenient' on the whole.

As others pointed out (thank you Steve, Dave) the neutral is absent, but
I can run one in fairly easily, but am inclined to use the earth. IIRC
the series resistor is usually of the order of 220K so only adds a
milliamp to the leakage current. I can always add the neutral later if
leakage is a problem. My CU is due for an upgrade anyway as it is
currently the old fuse-box type.

Phil

Dave Liquorice

unread,
Sep 5, 2009, 3:30:20 PM9/5/09
to
On Sat, 5 Sep 2009 10:44:16 -0700 (PDT), chrisj...@proemail.co.uk
wrote:

> A third possibility is to replace the switch by a motion (sorry)


> detector as used for security lights.

A PIR is a good solution. I had one in the kitchen of my old flat as
I'd be for ever leaving the lights on (3 x 100W spots) and
forgetting. Has the advantage that the light comes on automagically
when you enter and if it does switch off you just wave to switch it
back on.

An inital thought was a micro switch operated by the door, closed =
off. But that isn't very convenient when the room is in use. I don't
think a two way switching arrangement would solve that either.

--
Cheers
Dave.

Tim S

unread,
Sep 5, 2009, 4:14:19 PM9/5/09
to
chrisj...@proemail.co.uk coughed up some electrons that declared:

>
> What the OP wants is a "panel neon", e.g.
> http://www.maplin.co.uk/Search.aspx?criteria=panel%20neon&source=15 or
> LED: http://www.maplin.co.uk/Search.aspx?criteria=panel%20led&source=15
> but check that they aren't too deep for the switch box.
>
> However, I wonder whether a neon/LED will be prominent enough in
> daylight, or too easy to ignore even if it is.

I was going to suggest a grid switch panel (single gang, 2 grid modules),
one 10AX switch and one grid indicator (amber, red or green). Those are all
designed for easy wiring and of course 240V.

The problem remains as to where to get a neutral from. I'm not sure what
current a grid neon takes (largely determined by the series resistor, but I
wouldn't advocate linking it to earth.

B&Q sell the MK range of grid stuff including indicators, but GET do a nice
range with pretty switches and nice panels if you need to match some
existing scheme eg brass or white nickle etc.

Cheers

Tim

auct...@sheldononline.co.uk

unread,
Sep 5, 2009, 4:44:47 PM9/5/09
to
On Sep 5, 4:25 pm, Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> Any suggestions for a source of illuminated light switch as a reminder
> that the light it feeds is still on? We have a loo under the stairs with
> the light switch on the outside and it sometimes gets left switched on
> so the fan remains running (as well as the light being on). I don't want
> to fit a clunky 20Amp DP mains switch like thishttp://www.fastlec.co.uk/mk-20a-dp-switch-bottom-flexneon-k5423d1whi-...
> or thishttp://www.fastlec.co.uk/mk-20a-dp-switch-water-heater-k5423whwhi-p-5...

Depending on what aspect of these switches you find clunky, other
brands may be more suitable. Crabtree's DP switch is visually very
similar to the light switch, so the neon version looks like a light
switch with a neon just above the switch.

A

dennis@home

unread,
Sep 5, 2009, 4:58:00 PM9/5/09
to

<auct...@sheldononline.co.uk> wrote in message
news:926fb899-87c6-4caf...@z30g2000yqz.googlegroups.com...

I used
http://www.toolstation.com/shop/Electrical/Grid+Range+Wiring+Accessories/d190/sd2723
so i had a light next to the switch in my shed.
They are cheap, but as its been said before you need a neutral.
I suppose it would be possible to retrofit a flashing led into a neon
housing and run it on batteries for months at a time.
>
> A

NT

unread,
Sep 5, 2009, 6:01:58 PM9/5/09
to
On Sep 5, 4:25 pm, Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> Any suggestions for a source of illuminated light switch as a reminder
> that the light it feeds is still on? We have a loo under the stairs with
> the light switch on the outside and it sometimes gets left switched on
> so the fan remains running (as well as the light being on). I don't want
> What I really want is a basic light switch like thishttp://www.fastlec.co.uk/mk-10a-1g-sp-1-way-plateswitch-k4870-whi-p-5...

> but with a neon indicator incorporated. I can't find such a thing, the
> nearest being thishttp://www.tlc-direct.co.uk/Products/DT2503A.htmlbut
> it works the wrong way round - the illumination is 'on' when the switch
> is 'off' - its so you can find the switch in the dark. I suspect the
> wiring could be reversed, but as it's intended to show up in the dark it
> may not be bright enough for use in a  lighted area.
>
> I also found thishttp://www.fastlec.co.uk/mk-neon-locator-200250v-50hz-k3041-p-496.html

> which is an illuminated section that sits between the switch and the
> back-box. Again I'm not sure if its bright enough and it seems a bit of
> an overkill.
>
> I'm considering modifying a standard light switch by drilling a hole
> into which a small neon or LED is fitted. There is some safety
> consideration here so wonder if there is suitable double insulated
> neon/led available. Failing that I think I need to create some kind or
> perspex window in the switch with the neon/led behind it.
>
> Any thoughts?
>
> Phil


Cnnected from L to E a neon on a 220k resistor will add in the region
of 0.7mA. An ultrabright LED can be run off much less than 1mA with
decent light output. And if you run it from a CR dropper the power
diss would be even tinier. The LED would need a rectifier of course.


NT

John Rumm

unread,
Sep 5, 2009, 6:06:12 PM9/5/09
to
Phil Addison wrote:

> I'm considering modifying a standard light switch by drilling a hole
> into which a small neon or LED is fitted. There is some safety
> consideration here so wonder if there is suitable double insulated
> neon/led available. Failing that I think I need to create some kind or
> perspex window in the switch with the neon/led behind it.
>
> Any thoughts?

You can get things like:

http://www.tlc-direct.co.uk/Products/GU2011.html

which although designed for heavier loads are not particularly clunky in
use. However, they suffer the usual limitation of requiring a neutral at
the switch position to work.


--
Cheers,

John.

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

george (dicegeorge)

unread,
Sep 5, 2009, 6:27:58 PM9/5/09
to

>
>
> Cnnected from L to E a neon on a 220k resistor will add in the region
> of 0.7mA. An ultrabright LED can be run off much less than 1mA with
> decent light output. And if you run it from a CR dropper the power
> diss would be even tinier. The LED would need a rectifier of course.
>
>
> NT
Wouldnt this add to RCD trip current?

Why not wire it between L and N?

And how does the switchlite work?
http://www.tlc-direct.co.uk/Products/DT2503A.html
it glows when not switched on,
but has no Neutral or Earth connection,
is it like my neon mains testing screwdriver?

[g]

Tim S

unread,
Sep 5, 2009, 6:32:52 PM9/5/09
to
george (dicegeorge) coughed up some electrons that declared:

No - that neon goes across the switch and uses the load as a relative
neutral when switched off - which works well with a resistive load.
Electronic loads, eg a CFL are a different story but if the current is low
enough you may get away with it. In other cases you can get weird effects
like the CFL flashing.

Dave Plowman (News)

unread,
Sep 5, 2009, 6:38:20 PM9/5/09
to
In article <h7ubf8$v2c$1...@news.datemas.de>,

dennis@home <den...@killspam.kicks-ass.net> wrote:
> > Snag is light switches don't normally have a neutral available. The
> > type which is on with the light off gets the neutral via the light -
> > since it only takes a small current, the lightbulb looks like a short
> > to it and thus provides the return. I suppose you could run a small
> > neon to earth - since this is no worse than many SMPS do. But don't
> > tell anyone I suggested this.

> As long as the rcd, if present, doesn't trip.

It won't from a neon alone - they take near zero current. Otherwise neon
screwdrivers would be a tad dangerous. IIRC, of the order of 1-2 mA. Less
than many SMPS dump onto the earth.

--
*A picture may be worth a thousand words, but it uses up a thousand times more memory.

Dave Plowman (News)

unread,
Sep 5, 2009, 6:59:22 PM9/5/09
to
In article
<0383e6de-c1cc-4a2e...@k39g2000yqe.googlegroups.com>,

NT <meow...@care2.com> wrote:
> The LED would need a rectifier of course.

You just parallel a reverse connected diode to the LED for AC. Plus a
suitable current limiting resistor, of course.

--
*The first rule of holes: If you are in one, stop digging!

Phil Addison

unread,
Sep 5, 2009, 7:12:43 PM9/5/09
to
On Sat, 05 Sep 2009 23:06:12 +0100, John Rumm
<see.my.s...@nowhere.null> wrote:

> Phil Addison wrote:
>
> > I'm considering modifying a standard light switch by drilling a hole
> > into which a small neon or LED is fitted. There is some safety
> > consideration here so wonder if there is suitable double insulated
> > neon/led available. Failing that I think I need to create some kind or
> > perspex window in the switch with the neon/led behind it.
> >
> > Any thoughts?
>
> You can get things like:
>
> http://www.tlc-direct.co.uk/Products/GU2011.html
>
> which although designed for heavier loads are not particularly clunky in
> use. However, they suffer the usual limitation of requiring a neutral at
> the switch position to work.

I don't know this particular one, but the ones I have seen require quite
a lot more pressure than a light switch rocker, and emit a resounding
'click' when operated. That's why I'm looking for an alternative. I'm
off to maplin tomorrow for their red neon as posted by steve, so its
going to be a DIY job!! Must say I was surpprised to find the big names
don't make a SP lighting circuit varient of the 20A DP SW+neon.

Phil

Phil Addison

unread,
Sep 5, 2009, 7:25:24 PM9/5/09
to
On Sat, 05 Sep 2009 23:59:22 +0100, "Dave Plowman (News)"
<da...@davenoise.co.uk> wrote:

> In article
> <0383e6de-c1cc-4a2e...@k39g2000yqe.googlegroups.com>,
> NT <meow...@care2.com> wrote:
> > The LED would need a rectifier of course.
>
> You just parallel a reverse connected diode to the LED for AC. Plus a
> suitable current limiting resistor, of course.

That's neat. That will bypasses the LED on the reverse cycle, and is
probably easier to wire in than a series diode. It allows you to use a
tiny low voltage diode whereas the series one needs to be a bigger mains
rated 1N400x series. OTOH it draws twice the current of the series
solution, but still a mA or so.

Doesn't the 1/2 wave rectification halve the brightness though? An
electrolytic capacitor in parallel should solve that if it doesn't need
too many uF's - haven't calculated the size.

Phil

NT

unread,
Sep 5, 2009, 9:24:02 PM9/5/09
to
On Sep 5, 11:27 pm, "george (dicegeorge)" <dicegeo...@gmail.com>
wrote:

> > Cnnected from L to E a neon on a 220k resistor will add in the region
> > of 0.7mA. An ultrabright LED can be run off much less than 1mA with
> > decent light output. And if you run it from a CR dropper the power
> > diss would be even tinier. The LED would need a rectifier of course.
>
> > NT
>
> Wouldnt this add to RCD trip current?

yes, but by very little


> Why not wire it between L and N?

there's often no N present


NT

NT

unread,
Sep 5, 2009, 9:25:25 PM9/5/09
to
On Sep 5, 11:59 pm, "Dave Plowman (News)" <d...@davenoise.co.uk>
wrote:
> In article
> <0383e6de-c1cc-4a2e-982b-e1d5e33c5...@k39g2000yqe.googlegroups.com>,

>    NT <meow2...@care2.com> wrote:
>
> > The LED would need a rectifier of course.
>
> You just parallel a reverse connected diode to the LED for AC. Plus a
> suitable current limiting resistor, of course.

you can, but the resulting flicker is terrible. Worth going for 100Hz
flicker imho. That also gets you half the capacitor size.


NT

NT

unread,
Sep 5, 2009, 9:26:41 PM9/5/09
to
On Sep 6, 12:25 am, Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> On Sat, 05 Sep 2009 23:59:22 +0100, "Dave Plowman (News)"
>
> >    NT <meow2...@care2.com> wrote:
> > > The LED would need a rectifier of course.
>
> > You just parallel a reverse connected diode to the LED for AC. Plus a
> > suitable current limiting resistor, of course.
>
> That's neat. That will bypasses the LED on the reverse cycle, and is
> probably easier to wire in than a series diode. It allows you to use a
> tiny low voltage diode whereas the series one needs to be a bigger mains
> rated 1N400x series. OTOH it draws twice the current of the series
> solution, but still a mA or so.
>
> Doesn't the 1/2 wave rectification halve the brightness though? An
> electrolytic capacitor in parallel should solve that if it doesn't need
> too many uF's - haven't calculated the size.
>
> Phil

Thats another downside of the circuit, you cant smooth the power with
a little reservoir capacitor.


NT

John Rumm

unread,
Sep 5, 2009, 11:36:03 PM9/5/09
to
Phil Addison wrote:

> I don't know this particular one, but the ones I have seen require quite
> a lot more pressure than a light switch rocker, and emit a resounding
> 'click' when operated. That's why I'm looking for an alternative. I'm

Tis why I mentioned that one - its not as light as a light switch - but
nowhere near as heavy and clunky as some of the DP switches.

> off to maplin tomorrow for their red neon as posted by steve, so its
> going to be a DIY job!! Must say I was surpprised to find the big names
> don't make a SP lighting circuit varient of the 20A DP SW+neon.

The problem being the typical lack of neutral - hence nowhere to put it!


One solution you could try:

R1 LED1
-\/\/\--|>^---
| |
| D1 D2 D3 | /
Live in ------|>--|>--|>------o o----------> to lamp
| | sw1
| D4 |
-----<|-------

D1 to D4 all being silicon rectifier diodes of the same time. When SW1
is closed, the sequence of three diodes D1 to D3 will drop something
between 1.8 and 3V depending on type. That should be enough to forward
bias the LED, and R1 could be something 200R to limit the forward
current through the LED to 10mA ish (adjust based on the voltage drop of
your chosen rectifier diodes). D4 will protect the LED from being
exposed to reverse mains voltage, and also ensure you don't half wave
rectify the supply to your light!

The diode will flicker at 50Hz, but most people seem to cope with TVs at
that frequency, so I doubt it is a problem.

PeterC

unread,
Sep 6, 2009, 4:21:54 AM9/6/09
to

um, how about a switch with very obvious on and off appearance and just
look at it.
--
Peter.
The head of a pin will hold more angels if
it's been flattened with an angel-grinder.

dennis@home

unread,
Sep 6, 2009, 5:56:05 AM9/6/09
to

"Phil Addison" <phi...@bigNOSPAMfoot.com> wrote in message
news:f7s5a5tpdat9tsobh...@4ax.com...


> Doesn't the 1/2 wave rectification halve the brightness though? An
> electrolytic capacitor in parallel should solve that if it doesn't need
> too many uF's - haven't calculated the size.

It will flicker wont it.

Mike Barnes

unread,
Sep 6, 2009, 4:46:10 AM9/6/09
to

How about a pair of LEDs, reversed and paralleled, and mounted very
close to each other (perhaps behind the same diffuser)?

And there must be a reason why you can't connect the diode pair in
series with the load, otherwise it would have been suggested by now.
What's the reason?

--
Mike Barnes

Dave Liquorice

unread,
Sep 6, 2009, 8:14:58 AM9/6/09
to
On Sun, 06 Sep 2009 04:36:03 +0100, John Rumm wrote:

> One solution you could try:
>
> R1 LED1
> -\/\/\--|>^---
> | |
> | D1 D2 D3 | /
> Live in ------|>--|>--|>------o o----------> to lamp
> | | sw1
> | D4 |
> -----<|-------

Interesting idea. An improvement might be to use back to back zeners
in place of D1 to 4.

On a quick search I can only find 5v1 ones at 5W dissipation so the
upper limit on the lamp load would be 230W (ish) and you'd have 10W
of heat to get rid of at 230W load. Now if there are some 2v zeners
out there at > 2W dissipation...

--
Cheers
Dave.

NT

unread,
Sep 6, 2009, 9:22:57 AM9/6/09
to

You can put a diode in series with the LED. But... you then need a
leakage resistor across the LED, you get bad flicker, and you can only
use an R dropper, not a CR.


NT

NT

unread,
Sep 6, 2009, 9:25:52 AM9/6/09
to
On Sep 6, 1:14 pm, "Dave Liquorice" <allsortsnotthis...@howhill.com>
wrote:

filament lamps eat around 8x rated current at switch on, but diodes
dont tolerate these overloads. So your real life lamp limit for a 5v
5w 1A zener will be 1/8A apx, or around 30w.


NT

John Rumm

unread,
Sep 6, 2009, 10:52:22 AM9/6/09
to

I had thought of using a zenner but as you say there are difficulties
finding suitable devices, and bog standard rectifier diodes are dirt
cheap and robust. I used a version of this circuit on a battery charger
I built for an "on charger" light. It was a wide ranging charger that
might be dealing with anything from a single 1.2V cell up to about 30V
and currents from 9mA through to over an amp. The circuit above (without
D4 since this was DC) works flawlessly. Three diodes seemed to get
ample voltage to light the LED even if it was a bit below the 2.5V spec.

John Rumm

unread,
Sep 6, 2009, 10:55:51 AM9/6/09
to
Mike Barnes wrote:
> In uk.d-i-y, NT wrote:
>> On Sep 5, 11:59 pm, "Dave Plowman (News)" <d...@davenoise.co.uk>
>> wrote:
>>> In article
>>> <0383e6de-c1cc-4a2e-982b-e1d5e33c5...@k39g2000yqe.googlegroups.com>,
>>> NT <meow2...@care2.com> wrote:
>>>
>>>> The LED would need a rectifier of course.
>>> You just parallel a reverse connected diode to the LED for AC. Plus a
>>> suitable current limiting resistor, of course.
>> you can, but the resulting flicker is terrible. Worth going for 100Hz
>> flicker imho. That also gets you half the capacitor size.
>
> How about a pair of LEDs, reversed and paralleled, and mounted very
> close to each other (perhaps behind the same diffuser)?

That might solve the flicker issue - but TBH, 50Hz flicker is not going
to be an issue for this application.

> And there must be a reason why you can't connect the diode pair in
> series with the load, otherwise it would have been suggested by now.
> What's the reason?

What a pair of LEDs? The have a max forward voltage of *significantly*
less than mains, and a max power handling down in the mW range.

Phil Addison

unread,
Sep 6, 2009, 2:49:18 PM9/6/09
to

In the end I opted for simplicity and used the Maplin panel mount red
neon (BK52) which cost 99p, and took the return to earth; it works fine.

Here's the details...

1 Solder flex tail to neon, insulate with heat-shrink, and fit in 6.5mm
hole http://img35.imageshack.us/img35/7608/t302851solderonthewiresz.jpg

2 I soldered the fine wires to the switched load and earth of the mains
cable to guarantee a lasting connection
http://img142.imageshack.us/img142/789/t302855isolderedthefine.jpg

3 Reconnect mains cable to switch and the earth contact of back-box
http://img36.imageshack.us/img36/484/t302856reconnecttoswitc.jpg

4 The re-fitted switch with its new red neon
http://img42.imageshack.us/img42/3732/t302860therefittedswitc.jpg

5 Success!! http://img178.imageshack.us/img178/3156/t302861success.jpg

Thanks for all the input and neat LED ideas, but its hard to beat the
neon for simplicity, cost, very low current draw, no visible flicker
(100Hz), all round visibility, and ready availability in panel mount.

Phil

Mike Barnes

unread,
Sep 6, 2009, 3:04:25 PM9/6/09
to
In uk.d-i-y, John Rumm wrote:
>Mike Barnes wrote:
>> In uk.d-i-y, NT wrote:
>>> On Sep 5, 11:59 pm, "Dave Plowman (News)" <d...@davenoise.co.uk>
>>> wrote:
>>>> In article
>>>> <0383e6de-c1cc-4a2e-982b-e1d5e33c5...@k39g2000yqe.googlegroups.com>,
>>>> NT <meow2...@care2.com> wrote:
>>>>
>>>>> The LED would need a rectifier of course.
>>>> You just parallel a reverse connected diode to the LED for AC. Plus a
>>>> suitable current limiting resistor, of course.
>>> you can, but the resulting flicker is terrible. Worth going for 100Hz
>>> flicker imho. That also gets you half the capacitor size.
>> How about a pair of LEDs, reversed and paralleled, and mounted very
>> close to each other (perhaps behind the same diffuser)?
>
>That might solve the flicker issue - but TBH, 50Hz flicker is not going
>to be an issue for this application.

People do complain about 50Hz flicker. Not me, but some people.

>> And there must be a reason why you can't connect the diode pair in
>> series with the load, otherwise it would have been suggested by now.
>> What's the reason?
>
>What a pair of LEDs? The have a max forward voltage of *significantly*
>less than mains, and a max power handling down in the mW range.

I don't understand the relevance of those parameters to the application.
I see a single-LED series circuit being discussed elsewhere in this
thread, which suggests that it's not impossible.

--
Mike Barnes

Andrew Gabriel

unread,
Sep 6, 2009, 3:06:10 PM9/6/09
to
In article <uqu4a5d0rm5u55k36...@4ax.com>,

Phil Addison <phi...@bigNOSPAMfoot.com> writes:
> Any suggestions for a source of illuminated light switch as a reminder
> that the light it feeds is still on? We have a loo under the stairs with
> the light switch on the outside and it sometimes gets left switched on
> so the fan remains running (as well as the light being on). I don't want
> to fit a clunky 20Amp DP mains switch like this

Replace the switch with an occupancy sensor (mains PIR)
inside the room, or just leave the switch on all the time
and add the occupancy sensor next to the light, or replace
with a light fitting with a built-in occupancy sensor.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

Dave Plowman (News)

unread,
Sep 6, 2009, 6:12:30 PM9/6/09
to
In article
<23fc124c-32b4-4c37...@o9g2000yqj.googlegroups.com>,

I suppose it depends on the individual but I didn't notice it.

--
*Don't worry; it only seems kinky the first time.*

Dave Liquorice

unread,
Sep 6, 2009, 6:56:14 PM9/6/09
to
On Sat, 5 Sep 2009 15:01:58 -0700 (PDT), NT wrote:

> An ultrabright LED can be run off much less than 1mA with decent light
> output.

Certainly so, got a bit of kit with a blindingly bright blue LED on
the front. It lit the room up at night! Hacked into the wiring to add
a series resistor, the thing would glow with the leakage through my
dry fingers from the 3v supply. IIRC it now taking 60uA and is still
too bright...

--
Cheers
Dave.

John Rumm

unread,
Sep 6, 2009, 8:23:29 PM9/6/09
to
Mike Barnes wrote:
> In uk.d-i-y, John Rumm wrote:
>> Mike Barnes wrote:
>>> In uk.d-i-y, NT wrote:
>>>> On Sep 5, 11:59 pm, "Dave Plowman (News)" <d...@davenoise.co.uk>
>>>> wrote:
>>>>> In article
>>>>> <0383e6de-c1cc-4a2e-982b-e1d5e33c5...@k39g2000yqe.googlegroups.com>,
>>>>> NT <meow2...@care2.com> wrote:
>>>>>
>>>>>> The LED would need a rectifier of course.
>>>>> You just parallel a reverse connected diode to the LED for AC. Plus a
>>>>> suitable current limiting resistor, of course.
>>>> you can, but the resulting flicker is terrible. Worth going for 100Hz
>>>> flicker imho. That also gets you half the capacitor size.
>>> How about a pair of LEDs, reversed and paralleled, and mounted very
>>> close to each other (perhaps behind the same diffuser)?
>> That might solve the flicker issue - but TBH, 50Hz flicker is not going
>> to be an issue for this application.
>
> People do complain about 50Hz flicker. Not me, but some people.

I am one of them - especially on TVs and monitors etc, but we are
talking about a power on light here! ;-)

>>> And there must be a reason why you can't connect the diode pair in
>>> series with the load, otherwise it would have been suggested by now.
>>> What's the reason?
>> What a pair of LEDs? The have a max forward voltage of *significantly*
>> less than mains, and a max power handling down in the mW range.
>
> I don't understand the relevance of those parameters to the application.
> I see a single-LED series circuit being discussed elsewhere in this
> thread, which suggests that it's not impossible.

Only that if you don't take care in their use and treat them like
ordinary diodes, they will sell destruct rather rapidly.

NT

unread,
Sep 7, 2009, 4:04:28 AM9/7/09
to
On Sep 7, 1:23 am, John Rumm <see.my.signat...@nowhere.null> wrote:
> Mike Barnes wrote:
> > In uk.d-i-y, John Rumm wrote:
> >> Mike Barnes wrote:
> >>> In uk.d-i-y, NT wrote:
> >>>> On Sep 5, 11:59 pm, "Dave Plowman (News)" <d...@davenoise.co.uk>
> >>>> wrote:
> >>>>> In article
> >>>>> <0383e6de-c1cc-4a2e-982b-e1d5e33c5...@k39g2000yqe.googlegroups.com>,
> >>>>>    NT <meow2...@care2.com> wrote:
>
> >>>>>> The LED would need a rectifier of course.
> >>>>> You just parallel a reverse connected diode to the LED for AC. Plus a
> >>>>> suitable current limiting resistor, of course.
> >>>> you can, but the resulting flicker is terrible. Worth going for 100Hz
> >>>> flicker imho. That also gets you half the capacitor size.
> >>>  How about a pair of LEDs, reversed and paralleled, and mounted very
> >>> close to each other (perhaps behind the same diffuser)?
> >> That might solve the flicker issue - but TBH, 50Hz flicker is not going
> >> to be an issue for this application.
>
> > People do complain about 50Hz flicker. Not me, but some people.
>
> I am one of them - especially on TVs and monitors etc, but we are
> talking about a power on light here! ;-)


Light bulbs and tv screens behave differently wrt flicker though. TV
screens simply move the position of the light output at 50Hz, whereas
with a bulb the whole lot goes on and off together at 50Hz, causing a
much worse effect.

I just dont see any advantage in a 50Hz flickering circuit. It eats
twice as much power for the same light output, it flickers.


NT

Dave Plowman (News)

unread,
Sep 7, 2009, 5:10:49 AM9/7/09
to
In article
<01871416-3205-4ab2...@m11g2000yqf.googlegroups.com>,

NT <meow...@care2.com> wrote:
> I just dont see any advantage in a 50Hz flickering circuit. It eats
> twice as much power for the same light output, it flickers.

For a warning light might be a good thing - draws your attention to it?

I use an LM 317 for this which flashes at 4 times a second.

--
*I used up all my sick days so I called in dead

dennis@home

unread,
Sep 7, 2009, 6:03:33 AM9/7/09
to

"Dave Plowman (News)" <da...@davenoise.co.uk> wrote in message
news:5096f9e...@davenoise.co.uk...


> In article
> <01871416-3205-4ab2...@m11g2000yqf.googlegroups.com>,
> NT <meow...@care2.com> wrote:
>> I just dont see any advantage in a 50Hz flickering circuit. It eats
>> twice as much power for the same light output, it flickers.
>
> For a warning light might be a good thing - draws your attention to it?
>
> I use an LM 317 for this which flashes at 4 times a second.

I don't see why he doesn't just use a battery powered flashing LED like I
suggested ages ago.
All he needs is a double pole switch, one pole switches the light the other
the LED.
The batteries will last for months/years as the LED on flashes when the
switch is on.
No fancy components required.
The 20A grid switches I posted a link to have more than enough insulation to
do this.

NT

unread,
Sep 7, 2009, 7:29:08 AM9/7/09
to
On Sep 7, 10:10 am, "Dave Plowman (News)" <d...@davenoise.co.uk>
wrote:
> In article
> <01871416-3205-4ab2-ac9b-9b93b5e9d...@m11g2000yqf.googlegroups.com>,

>    NT <meow2...@care2.com> wrote:
>
> > I just dont see any advantage in a 50Hz flickering circuit. It eats
> > twice as much power for the same light output, it flickers.
>
> For a warning light might be a good thing - draws your attention to it?

fair point. I'm not sure this situation counts as a warning light
though, they've got to live with it on for plenty of time once
installed.


NT

John Rumm

unread,
Sep 7, 2009, 7:37:05 AM9/7/09
to

Again, its a LED power on indicator, do we care? If so, add another
couple of diodes to the reverse path and a second LED mounted close the
the first; or use a dual chip LED that looks like one indicator but has
two devices internally...

> I just dont see any advantage in a 50Hz flickering circuit. It eats

The advantage is a mains powered light source that does not require any
kludge like using the earth as a neutral. It also maintenance free saves
the risk of having someone having to change batteries on potentially
live circuit.

> twice as much power for the same light output, it flickers.

Twice the power? half the power and less light I would expect...

Either way we are talking insignificant loads compared to the lamp
itself, you are probably dissipating as much as the LED in the wires of
the circuit!

John Rumm

unread,
Sep 7, 2009, 7:54:56 AM9/7/09
to
dennis@home wrote:
>
>
> "Dave Plowman (News)" <da...@davenoise.co.uk> wrote in message
> news:5096f9e...@davenoise.co.uk...
>> In article
>> <01871416-3205-4ab2...@m11g2000yqf.googlegroups.com>,
>> NT <meow...@care2.com> wrote:
>>> I just dont see any advantage in a 50Hz flickering circuit. It eats
>>> twice as much power for the same light output, it flickers.
>>
>> For a warning light might be a good thing - draws your attention to it?
>>
>> I use an LM 317 for this which flashes at 4 times a second.
>
> I don't see why he doesn't just use a battery powered flashing LED like
> I suggested ages ago.

It would work, however there are a couple of issues to be aware of. The
first would be that you will need to construct the arrangement in such a
way as to not pose any risk of exposing an untrained user to mains
potential when they have it off the wall to replace the battery.
Something that is doable with a properly designed and tested enclosure
etc, but not easy to achieve as a quick lashup.

> All he needs is a double pole switch, one pole switches the light the
> other the LED.

Therein the other problem, you can't use matching switchgear etc (unless
you already have grid switches!)

You could also bear in mind the OPs original requirement about the
switch still feeling and sounding light a 6A light switch in operation -
and many 20A DPST switches don't.

Dave Plowman (News)

unread,
Sep 7, 2009, 6:20:29 AM9/7/09
to
In article <h82lpl$2cv$1...@news.datemas.de>,

dennis@home <den...@killspam.kicks-ass.net> wrote:
> I don't see why he doesn't just use a battery powered flashing LED like
> I suggested ages ago. All he needs is a double pole switch, one pole
> switches the light the other the LED. The batteries will last for
> months/years as the LED on flashes when the switch is on. No fancy
> components required. The 20A grid switches I posted a link to have more
> than enough insulation to do this.

There wouldn't be room for even a PP3 inside most back boxes. Certainly
not a plaster depth one.

But Phil has already solved his problem.

--
*Rehab is for quitters.

Tim S

unread,
Sep 7, 2009, 8:58:24 AM9/7/09
to
John Rumm coughed up some electrons that declared:

> Again, its a LED power on indicator, do we care? If so, add another
> couple of diodes to the reverse path and a second LED mounted close the
> the first; or use a dual chip LED that looks like one indicator but has
> two devices internally...
>

Did anyone say "bridge rectifier"? 4 diodes (or one small package, 4 leads),
1 led and 100Hz flicker, which I'm fairly sure would not be perceivable to
most people.

Cheers

Tim

The Wanderer

unread,
Sep 7, 2009, 9:03:14 AM9/7/09
to
On Sat, 05 Sep 2009 16:25:46 +0100, Phil Addison wrote:

> Any suggestions for a source of illuminated light switch as a reminder
> that the light it feeds is still on? We have a loo under the stairs with
> the light switch on the outside and it sometimes gets left switched on
> so the fan remains running (as well as the light being on). I don't want
> to fit a clunky 20Amp DP mains switch like this

> http://www.fastlec.co.uk/mk-20a-dp-switch-bottom-flexneon-k5423d1whi-p-538.html
> or this
> http://www.fastlec.co.uk/mk-20a-dp-switch-water-heater-k5423whwhi-p-539.html
>
> What I really want is a basic light switch like this
> http://www.fastlec.co.uk/mk-10a-1g-sp-1-way-plateswitch-k4870-whi-p-511.html
> but with a neon indicator incorporated. I can't find such a thing, the
> nearest being this http://www.tlc-direct.co.uk/Products/DT2503A.html but
> it works the wrong way round - the illumination is 'on' when the switch
> is 'off' - its so you can find the switch in the dark. I suspect the
> wiring could be reversed, but as it's intended to show up in the dark it
> may not be bright enough for use in a lighted area.
>
> I also found this
> http://www.fastlec.co.uk/mk-neon-locator-200250v-50hz-k3041-p-496.html
> which is an illuminated section that sits between the switch and the
> back-box. Again I'm not sure if its bright enough and it seems a bit of
> an overkill.
>
> I'm considering modifying a standard light switch by drilling a hole
> into which a small neon or LED is fitted. There is some safety
> consideration here so wonder if there is suitable double insulated
> neon/led available. Failing that I think I need to create some kind or
> perspex window in the switch with the neon/led behind it.
>
> Any thoughts?

Idly been watching this thread over the last couple of days looking at some
of the weird and wonderful suggestions.

I did something like this many many years ago for loft lights, back in the
days when I thought I was going to build a model railway - never did
happen! All I did was drill and fit a small neon indicator in one corner of
the switch plate, wired switch terminal to earth on the box. Alright, I
know it was back in the days when rcds weren't commonly used, but taking a
quick look at the Farnell web site, many of their 230v leds have a forward
current of about 3mA so shouldn't present too much of a problem.


--
The Wanderer

All wighy, rho sriyched yhe ket pads on my ketboawd?

Phil Addison

unread,
Sep 7, 2009, 11:15:30 AM9/7/09
to
On Sun, 06 Sep 2009 19:49:18 +0100, Phil Addison
<phi...@bigNOSPAMfoot.com> wrote:

> In the end I opted for simplicity and used the Maplin panel mount red
> neon (BK52) which cost 99p, and took the return to earth; it works fine.
>
> Here's the details...
>
> 1 Solder flex tail to neon, insulate with heat-shrink, and fit in 6.5mm
> hole http://img35.imageshack.us/img35/7608/t302851solderonthewiresz.jpg
>
> 2 I soldered the fine wires to the switched load and earth of the mains
> cable to guarantee a lasting connection
> http://img142.imageshack.us/img142/789/t302855isolderedthefine.jpg
>
> 3 Reconnect mains cable to switch and the earth contact of back-box
> http://img36.imageshack.us/img36/484/t302856reconnecttoswitc.jpg
>
> 4 The re-fitted switch with its new red neon
> http://img42.imageshack.us/img42/3732/t302860therefittedswitc.jpg
>
> 5 Success!! http://img178.imageshack.us/img178/3156/t302861success.jpg
>
> Thanks for all the input and neat LED ideas, but its hard to beat the
> neon for simplicity, cost, very low current draw, no visible flicker
> (100Hz), all round visibility, and ready availability in panel mount.

... and today we realised it also provides a clear 'engaged' indication.
OK, the LED schemes would do that as well, but the PIR's would not.

Phil

Phil Addison

unread,
Sep 7, 2009, 11:30:20 AM9/7/09
to
On Sun, 06 Sep 2009 04:36:03 +0100, John Rumm
<see.my.s...@nowhere.null> wrote:

> Phil Addison wrote:
>
> > Must say I was surprised to find the big names
> > don't make an SP lighting circuit varient of the 20A DP SW+neon.


>
> The problem being the typical lack of neutral - hence nowhere to put it!

That only applies to (some) retro-fits and, even then, routing in a
neutral is not impossible. Why isn't such an indicator light-switch
available for new builds where the electrician would know in advance
that a neutral is required? Is it so unusual to have the switch on the
outside of the understairs loo?

Phil

Dave Plowman (News)

unread,
Sep 7, 2009, 11:45:57 AM9/7/09
to
In article <v09aa51brc1ps7iub...@4ax.com>,

Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> That only applies to (some) retro-fits and, even then, routing in a
> neutral is not impossible. Why isn't such an indicator light-switch
> available for new builds where the electrician would know in advance
> that a neutral is required? Is it so unusual to have the switch on the
> outside of the understairs loo?

Easy enough to make one using grid switch components. Plus the fact it
would have to be a posh new build to think of something so obvious and use
a mm more of wire than they can get away with. ;-)

In my own house I ran the switch feed in triple and earth in most rooms.
Handy if you decide you want floor outlet lamps switched from the same
place.

--
*Why do the two "sanction"s (noun and verb) mean opposites?*

Dave Plowman (News)

unread,
Sep 7, 2009, 11:47:32 AM9/7/09
to
In article <h8301g$b82$1...@news.eternal-september.org>,

Dunno - presumably ordinary fluorescent would have 100 Hz flicker, and
plenty complain about those.

--
*The statement above is false

Owain

unread,
Sep 7, 2009, 12:58:11 PM9/7/09
to
On 7 Sep, 16:15, Phil Addison wrote:
> ... and today we realised it also provides a clear 'engaged' indication.

If you've got a fan with a run-on timer you could extend a second neon
from that. Then you'd have separate indicators for 'engaged' and
'proceed with caution'

Owain

John Rumm

unread,
Sep 7, 2009, 1:07:23 PM9/7/09
to

Not sure how you get the three diodes in series configuration needed to
get you enough voltage drop to light the led though...?

John Rumm

unread,
Sep 7, 2009, 1:09:08 PM9/7/09
to

If you had the PIR *and* your switch it would... the light would only
come on when both switches are "on".

John Rumm

unread,
Sep 7, 2009, 1:10:35 PM9/7/09
to

Probably not, but is not the switch being "down" enough indication
anyway? (unless you have two way switching for your loo!) ;-)

Phil Addison

unread,
Sep 7, 2009, 4:15:13 PM9/7/09
to
On Mon, 07 Sep 2009 18:09:08 +0100, John Rumm
<see.my.s...@nowhere.null> wrote:

> Phil Addison wrote:
> > On Sun, 06 Sep 2009 19:49:18 +0100, Phil Addison
> > <phi...@bigNOSPAMfoot.com> wrote:
> >
> >> In the end I opted for simplicity and used the Maplin panel mount red
> >> neon (BK52) which cost 99p, and took the return to earth; it works fine.
> >>
> >>

> >> Thanks for all the input and neat LED ideas, but its hard to beat the
> >> neon for simplicity, cost, very low current draw, no visible flicker
> >> (100Hz), all round visibility, and ready availability in panel mount.
> >
> > ... and today we realised it also provides a clear 'engaged' indication.
> > OK, the LED schemes would do that as well, but the PIR's would not.
>
> If you had the PIR *and* your switch it would... the light would only
> come on when both switches are "on".

That's true, and being sat there waving your arms around to keep the
light on you could dispense with the fan :O)

Phil Addison

unread,
Sep 7, 2009, 4:22:55 PM9/7/09
to
On Mon, 07 Sep 2009 18:10:35 +0100, John Rumm
<see.my.s...@nowhere.null> wrote:

> Phil Addison wrote:
> > On Sun, 06 Sep 2009 04:36:03 +0100, John Rumm
> > <see.my.s...@nowhere.null> wrote:
> >
> >> Phil Addison wrote:
> >>
> >>> Must say I was surprised to find the big names
> >>> don't make an SP lighting circuit varient of the 20A DP SW+neon.
> >> The problem being the typical lack of neutral - hence nowhere to put it!
> >
> > That only applies to (some) retro-fits and, even then, routing in a
> > neutral is not impossible. Why isn't such an indicator light-switch
> > available for new builds where the electrician would know in advance
> > that a neutral is required? Is it so unusual to have the switch on the
> > outside of the understairs loo?
>
> Probably not, but is not the switch being "down" enough indication
> anyway?

Nope, already tried that solition and it doesn't work! see OP "We have a


loo under the stairs with the light switch on the outside and it
sometimes gets left switched on so the fan remains running (as well as

the light being on)." :=/

Phil

Phil Addison

unread,
Sep 7, 2009, 4:25:43 PM9/7/09
to
On Mon, 07 Sep 2009 16:45:57 +0100, "Dave Plowman (News)"
<da...@davenoise.co.uk> wrote:

> In article <v09aa51brc1ps7iub...@4ax.com>,
> Phil Addison <phi...@bigNOSPAMfoot.com> wrote:
> > That only applies to (some) retro-fits and, even then, routing in a
> > neutral is not impossible. Why isn't such an indicator light-switch
> > available for new builds where the electrician would know in advance
> > that a neutral is required? Is it so unusual to have the switch on the
> > outside of the understairs loo?
>
> Easy enough to make one using grid switch components.

We've done the 'making' bit, I'm on to the 'why don't they' bit now :))

> Plus the fact it
> would have to be a posh new build to think of something so obvious and use
> a mm more of wire than they can get away with. ;-)

LOL


Phil

Phil Addison

unread,
Sep 7, 2009, 4:33:18 PM9/7/09
to

But 'engaged' and 'proceed with caution' showing together would be
inconvenient.

Phil

Phil Addison

unread,
Sep 7, 2009, 4:35:22 PM9/7/09
to
On Mon, 7 Sep 2009 14:03:14 +0100, The Wanderer <the.wa...@gmx.co.uk>
wrote:

> All I did was drill and fit a small neon indicator in one corner of
> the switch plate, wired switch terminal to earth on the box. Alright, I
> know it was back in the days when rcds weren't commonly used, but taking a
> quick look at the Farnell web site, many of their 230v leds have a forward
> current of about 3mA so shouldn't present too much of a problem.

Keep up at the back!
http://img178.imageshack.us/img178/3156/t302861success.jpg

Phil

Owain

unread,
Sep 7, 2009, 5:19:36 PM9/7/09
to
On 7 Sep, 21:33, Phil Addison wrote:
> > If you've got a fan with a run-on timer you could extend a second neon
> > from that. Then you'd have separate indicators for 'engaged' and
> > 'proceed with caution'
> But 'engaged' and  'proceed with caution' showing together would be
> inconvenient.

<Capt Mainwaring> I was wondering whether you'd spot that deliberate
mistake.

Obviously some interlocking relays would be required ...

Owain


John Rumm

unread,
Sep 7, 2009, 7:02:59 PM9/7/09
to

No, I meant was the switch being down not enough of an occupancy
indication rather than a warning the light was left on. Obviously it
would leave scope for sitting there with your legs crossed waiting for
the emergence of the phantom occupant, which could take a while.

Phil Addison

unread,
Sep 7, 2009, 7:29:26 PM9/7/09
to
On Tue, 08 Sep 2009 00:02:59 +0100, John Rumm
<see.my.s...@nowhere.null> wrote:

OIC. In that case no, because we all knew it sometimes got left on so
standard practice was to shout "anyone in there?" in that case. With the
red neon grabbing attention, it is now unlikely to mean anything other
than 'occupied'. Of course there may still be the odd cry of "OI!!!"
from within when a thoughtful passer-by turns it off. However, if in
doubt its simple enough to just try the handle; we do have one of those
feeble lockable handles.

An "Occupied" indicator was never my criteria, "Occupancy" (not the same
thing) was raised by the PIR fans, I just pointed out that the
indication is a bonus with the red neon.

Phil

0 new messages