# dawn simulator schematics problems

Discussion in 'Electronic Design' started by [email protected], Dec 11, 2004.

1. ### Guest

I intend to build a simple dawn simulator for my wife. She likes the
idea of waking up to the increased brightness of the bedside lamp
rather then the alarm clock noise, and that sounds good to me

I'm thinking basics here: an off-the-shelf household timer (to set
the wakeup time) and a custom dimmer circuit that brings my lamp from 0
to 100% in, say, in 20 minutes time or so.

I've looked around all the usenet groups for directions on how to
build this thing: most of solutions points to various home automation
solutions. That might work fine, but I'd like to put my old soldering
iron to use again - it's about time .

The problem is that I can't find a detailed schematic description.
There are references to alt.binaries.schematics.electronic but nor
Google or my ISP allows access, so I can't see if that would do it.

I plan to use Bill Bowden's (url below) dimmer circuit schema, I just
need some help with the dimensions of the diodes and some capacitors.
I've put the schema up, with all items marked, at:
http://www.designpending.com/fredrik/dawn.htm

Anybody that can suggest roughly which diodes to use?
BTW: do you expect this solution to be noisy?

Thanks,
..fredrik

2. ### John PopelishGuest

This sort of thing is probably best done with a microprocessor
controlling the phase of a lamp dimmer. But you can also just buy
one.
http://www.serenityhealth.com/sun_alarm_clock_descript.html

3. ### Ross HerbertGuest

For a start I wouldn't be using that circuit. It is totally
inefficient, electrically noisy and potentially hazardous.

Do some calculations on that 2K resistor when supplied from 230V.
Assume the nominal 10V dc across the zener thus leaving 220V across
the resistor. This means the resistor must dissipate in excess of 24W
(not 10W) which means it should ideally be rated at 40W or better. It
would need to be a metal-cased component (Arcol) and bolted to a
heatsink to dissipate the heat. All in all this component is wasting
power and functioning purely as a room heater.

There is absolutely no electrical isolation or barrier between the
mains supply and the low voltage components so everything must be
assumed to be "live", particularly if something goes wrong and a
component fails short circuit. While Bill Bowden may like to use it on
110V and think it is safe, it certainly isn't safe when connected to
230V and I don't think it would meet the European/UK electrical safety
standards.

4. ### Guest

Ouch! I suppose I should look at other solutions, thanks Ross for
making that clear!
Does anybody know whether there were more useful solutions at
alt.binaries.schematics.electronic?

5. ### Robert MonsenGuest

As other's have mentioned, I think your power supply needs some work.

If you use a transformer to get your 9V supply, you can isolate the
logic from the power elements using an optoisolator designed to run a
triac, like a MOC3011. This is much safer. It's easier to isolate just
the triac, MOC3011, and fuse than the entire circuit.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

6. ### Ross HerbertGuest

I don't know of any diy designs for a sunrise alarm lamp but unless
you are really determined to build your own I would recommend
something already available for 230V operation. Such as
http://www.bodykind.com/productdetails/CatIDx67/ProdIDx463/dawn-simulator-natural-alarm-clock-1x.htm

Believe me, it will be a lot cheaper than starting from scratch and

7. ### Eric GriswoldGuest

I use a mercury vapor lamp (sort of lke a streetlight) connected to a simple
timer. The lamp takes about 5-10 minutes to come to full brightness, and
also, it starts out very 'pink' just like real dawn. And it end up hella
bright just like real sunlight.

--eric