Hi all! New to the forum and joined because I wanted to take on a project after not touching electronics since I was in high school and don't really know where to start!

What i want to do, is build a lighting unit that will simulate the movement of the sun throughout the day (for use with reptiles). The idea is that a bulb will turn on and slowly work it's way up to full brightness, then the next will start doing the same. Ocne the second is at full brightness, a third will start to light up while the first starts to dim down. This will continue for a few bulbs until the light has been on for a whole day and moved from left to right. The bulbs will need to stay at full brightness for a while but that is the general gist of it. I am completely open to any suggestions if it sounds rubbish though! I've attempted a diagram of what i want it to do:

I also want to run the bulbs from mains power as they'll be 40W incandescent bulbs (minimum - possibly higher in the middle to simulate midday sun).

I have done many things with simple circuits run from DC, but never something off AC and don't really know how to tackle it! I would greatly appreciate ideas and any links to places where I can go and read about stuff i will need to understand!

I fully intend to work my way through the volumes on this website, but if any will be of particular use to me, please do point them out!

Thanks for you help in advance!

 

That's a nice project!

Please decide first if you will be able to use a circuit that will be isolated from mains. Maybe low voltage halogen lamps or similar?

Non-isolated circuits are not allowed to be discussed on this forum.

 

I agree with praondevou. Don't even think about 120 VAC or other AC power. As you said, you don't know how to tackle it, and the "it" is dangerous and we can't guide you safely over the internet.

Go for 12 VDC. You can get 25 W bulbs that are nice and small and power as many as needed. Incandescent bulbs are probably better than alternatives because they emit a broad spectrum including infrared, just like the sun. I assume the sunbather you are catering to would like a nice emission of heat rays as well as visible light.

http://www.etrailer.com/Accessories-and-Parts/Bargman/30-90-156.html

 

Check out some of these bulbs.

12V-Low-Voltage-Incandescent-Light-Bulbs

 

Those (steveb) 1156 bulbs are Ford tail light bulbs. Very available!
Ionic's bulbs are good. A bit bulky in the A15 and A19 size, but very reliable.
Xenon bulbs are becoming popular lately. I saw some small 25 Watt zenon's being installed on This Old House today.
The bulbs are not going to be a problem...but,
what do you have in mind for controlling them? A microprocessor? A slow triangle wave to a transistor configured as a current generator? Pulse width modulation?

 

What do you have in mind for controlling them? A microprocessor? A slow triangle wave to a transistor configured as a current generator? Pulse width modulation?

That is the million dollar question. I'm curious as to an effective analog and a Micro Controller method. I'm sure the OP is here for that type of direction specifically! Since a fairly accurate time base is needed as it will be functioning non-stop, I suspect a micro controller with an accurate clock would be beneficial.

 

@Slik_JIM

Can you tell us, how many light bulbs you want in your project?

 

If not with a microcontroller one approach could be a LED driver like the LM3915.
It would have to be in DOT mode.(important)
The input signal would be a slowly (12h or whatever your day is long) rising DC signal with a superimposed triangle or sine wave of let's say 100Hz on it.
I just simulated it with a few opamps imitating the internal circuitry of the LM3915 but I don't know yet how to simulate the DOT mode.

The modulation of the DC-input signal would cause more than one output to be turned ON, with different duty cycles. I didn't think this through, though... And I never worked with this IC.

you may have a look at page 7 of THIS datasheet to see what I'm talking about.

As iOnic said if precision is important then simple RC timer circuits as your time base may not be good enough. If going with the LM3915 you would need two astable multivibrators. (or 3 since there is the "night" period to be simulated too)

 

Control is verry similar to my waving flag except lighting is inverted & flag is 20 bulbs wide, with 2 strings in parallel. Uses a 3 phase clock,4017's,555, a form of PWM, another 4017, 8in NAND' & rectified 24V AC, un filtered, & one NOR IC for each bulb. 'got to go-bowling time[ hit 203 last wk.]

 

Here's an odd idea...make it like the ceiling in the Great Hall at Hogwarts. Put x number of photodetectors on a curved surface, outside the building. Have each photodetector control a light bulbs brightness. That eliminates the clock accuracy problem and turns this project in building x number of identical circuits.

 

Here's an odd idea...make it like the ceiling in the Great Hall at Hogwarts. Put x number of photodetectors on a curved surface, outside the building. Have each photodetector control a light bulbs brightness. That eliminates the clock accuracy problem and turns this project in building x number of identical circuits.

Good idea!

But he will need some form of automatic gain control to account for brightness differences due to sunny/overcast days.

 

Actually, the automatic gain control will be 90% of the job.

 

Or put one to three bulbs on a arm which rotates by a gear motor one RPD, something like a solar tracker.