Day/Night Controller

[MikeStevenson]

There are some things I need to automate on a daily basis - first thing in the morning but only if there's adequate daylight.

I'm new to electronics - currently studying at Google University

I'm sure I can use an LDR to detect daylight and close the circuit. I plan to include a capacitor that will take a few seconds to discharge so the circuit can be opened almost immediately after it's closed.

Once the circuit cycles, it should stay open the rest of the day.

I've considered a timer that waits about 20 hours then resets and waits for the LDR to detect enough light again.

But it may be easier to use the LDR (or a second LDR) to detect a couple hours of darkness before resetting.

This should consume as little power as possible because it will be in a remote area and connected to a small solar panel and rechargable battery. It needs to run unattented for several weeks without draining the battery.

Any ideas?

 

[Nigel Goodwin]

Use a PIC - easy to get the long timing period you're looking for, and nice low power consumption - it's also easy to get it to read the LDR.

 

[MikeStevenson]

Thanks, I guess...
I'm real new to electronics, but that sounds a bit complicated. Like a sledge hammer for a thumb tack.
I was thinking a lot simpler - a few resistors, capacitors and a timer. It basically cycles a switch once a day at daylight.
Am I thinking too small?

 

[Nigel Goodwin]

Thanks, I guess...
I'm real new to electronics, but that sounds a bit complicated. Like a sledge hammer for a thumb tack.
I was thinking a lot simpler - a few resistors, capacitors and a timer. It basically cycles a switch once a day at daylight.
Am I thinking too small?

I'm talking one PIC chip, one LDR, and one capacitor - plus power supply and what ever it feeds.

A 20 hour timer is going to need multiple counter chips, and you will need an opamp or comparator to read the LDR.

The advantage of a PIC is that it vastly simplifies the circuit, using incrdibly small amounts of power, and at the same time makes it far better as well. You have the cost of writing a little software though.

 

[MikeStevenson]

I've never written code for a PIC. Until now I never thought about it.
I'm comfortable with following a schematic to solder components on a breadboard. I just don't have a frame of reference for designing and executing a PIC-based solution.
How likely is it a newbie could figure out how to make a PIC-based solution?

 

[crutschow]

I've never written code for a PIC. Until now I never thought about it.
I'm comfortable with following a schematic to solder components on a breadboard. I just don't have a frame of reference for designing and executing a PIC-based solution.
How likely is it a newbie could figure out how to make a PIC-based solution?

You might try reading some of Nigel's PIC tutorials to get an idea of what's involved.

Have you done any programming at all? It can be intimidating at first, but once you understand the basics, it tends to be rather intutitive. After that you can do all kinds of interesting electronic projects with a minimum of hardware.

I find Basic is probably the easiest language for a beginner to learn. It uses regular English words and is fairly easy to understand. It's not the most efficient or fast in execution as compared to Assembly language or C but you don't need that for your application.

You might also look at the Basic Stamp processors modules sold by Parallax. I've used one of them in a project of mine, and it was about as simple to use as a microprocessor can be. (The lowest power unit they have draws 1mA. Don't know if that's low enough power for you).

 

[MikeStevenson]

Thanks. I appreciate the advice.
I've been wading through some of his stuff. I am getting the idea it take a rather large investment in time, effort and hardware to get started with PICs.
Since this is a one-time, single purpose project, the up-front investment is not likely to be recouped.
It is very interesting, though. I'm not sure I have the time to invest to pull it off.
These circuits will basically operate various devices in remote areas on a daily basis for several weeks up to a couple months. They will be powered with 6volt rechargable batteries with solar rechargers. As long as the circuit doesn't pull more than the solar rechargers can supply they should run for months without a problem.
The devices are gravity powered - no batteries. They run for weeks between resets.

 

[Ubergeek63]

Thanks, I guess...
I'm real new to electronics, but that sounds a bit complicated. Like a sledge hammer for a thumb tack.
I was thinking a lot simpler - a few resistors, capacitors and a timer. It basically cycles a switch once a day at daylight.
Am I thinking too small?

Not at all, a simple solution for a simple problem.

Unfortunately for delays that long it is not quite that simple. Fortunately the answer is not terribly complicated: add a counter. Ok, add a long counter.

Dan