DIY light-activated switch

[couldabin]

You're right about the time of day ... but this isn't about the time of day, it's about the light level. I want to have a photograph at the point where the light first reaches a pre-determined level. I've gone out and metered it -- the exposure would be 1 sec. @f8, EI 100. I'm going to run some tests with a photocell before deciding on a condensor lens. It turns out that the light level I'm waiting for is pretty much at the low end of the photocell's practical measuring ability. A condensor moves the effective brightness (on the photocell) into a resistance range that allow for more accurate measurements.

If this were just one photograph, I'd not be building this thing. But it's for a project that entails a photograph every two weeks, for a year, to document seasonal changes on the prairie. I simply cannot expect to be available to take the photograph on every one of those mornings, but I can expect to be able to set up the camera sometime during the day before and have it capture the scene as planned.

 

[couldabin]

Wouldn't it be less trouble to simply get the trip point value directly? (I took a meter reading with my Luna Pro the other morning -- I'm waiting for brightness that translates into 1 sec. @ f8 for my FP4). Knowing that, I will simply see what resistance I get from my photocell at that light level, and plan accordingly. A condensor may be necessary to amplify the energy delivered to the photocell, so that the value is more reliable; some testing should help me decide on that point.

 

[Jeff Searust]

Why not just wake up early...

 

[couldabin]

I do (at least, 5 am seems early for me). But the photograph(s) will be taken 25 miles from where I live (but right next to where I work). I'm not usually at work before sunrise. And this will be a one year project -- a photograph every two weeks, at the point where the light reaches the pre-determined level. Setting up the day before seems the most practical approach.

 

[paul ron]

The light fixture... the realy is hot, realy will be open when it is dark but once it gets light, the realy contacts shut (NC contacts) will switch your external circuit on, and the 8 sec timer starts it's count down. Using a large enough elecrtolytic cap bleeding off to ground via a resistor can adjust to 8sec easily enough without many components.


BTW sun rise is later by about 1 minute per day. It won't matter is you use the photo sensor but if a cloud is on the horizon n it clears a bit later your shot won't be what you planed it to be.

 

[Chan Tran]

Polygot!
Reading your post, I am wondering if there is a modern light meter that has interface (serial, USB etc..) so that a computer can read it reading and act upon the value of the reading. With all the modern technology I am just wondering.

 

[greybeard]

Ah, now it makes perfect sense.

I wonder if you can still buy those hand-sized parabolic reflectors that used to be sold in the back of Popular Mechanics. The ones showing someone lighting a cigarette using sunlight? The f/ number should be about one or less, and you don't need an image, so.....

 

[AgX]

Polygot!
Reading your post, I am wondering if there is a modern light meter that has interface (serial, USB etc..) so that a computer can read it reading and act upon the value of the reading. With all the modern technology I am just wondering.

The Gossen Digisky has an USB interface.

 

[couldabin]

There are such devices out there. Here's one: http://www.fondriest.com/products/extech_407026.htm ... and another http://www.gossen-photo.de/english/licht_p_mavomonitor.php. I think that's a lot more capability than I need -- I just need to trigger a photograph at an arbitrary light level. And it would entail putting a computer out in the wilderness. Where there's no power source. And lots of weather ...

 

[polyglot]

Many microcontrollers can do USB (hence talk to the meter) and run on practically no power. If you were considering the light-to-frequency sensor, you probably will need a micro there anyway and I suspect that directly reading the light-to-frequency thing will be much easier than talking to a meter using someone else's proprietary protocol. And much more accurate than a photoresistor, probably with the sensitivity you need.

I suspect you're about to do a bunch of learning regarding low-power design and software

 

[Leigh B]

C'mon, guys...

You're building an Atlas rocket to launch a tennis ball. :confused:

All you need is a photosensor, an opamp, a couple of resistors, a relay, and a battery.

This ain't rocket science.

- Leigh

 

[couldabin]

The chip I use for temperature monitoring applications (MSP430F2012) doesn't have built-in USB support. It's pretty minimalist. And I have no interest in buying another light meter. I think you're right that light-to-frequency chip would be far easier to incorporate than talking to a meter. Likely more accurate than a photoresistor, too. But I'm going to go out on a limb here and say that a simple photoresistor will be accurate enough. A condensor may be needed to get the accuracy I want (a half-stop will be just fine). I'll follow up with the results.

 

[paul ron]

C'mon, guys...

You're building an Atlas rocket to launch a tennis ball. :confused:

All you need is a photosensor, an opamp, a couple of resistors, a relay, and a battery.

This ain't rocket science.

- Leigh

I agree, this is getting complicated. Now there iwll be a 500gig computer control center running a USB micro controled sensor circuit.... Let teh countdown begin
T minus 10
T minus 9
T minus 8....
BLAST OFF!
Huston we have picture!

oops... there was dew on the lens.

Sheit!

.:munch:

 

[couldabin]

Thanks for the advice.

 

[paul ron]

Be sure to let us see the pics and the trigger device ou finally settle on using, please?

This was a fun.

 

[couldabin]

I will.

 

[Fettb]

I've been playing with a micro and a photodiode for use in a DIY camera project. I'm using a Hamamatsu s7183 photodiode ic. It is amplified so the circuit is simple. Just hand a resistor off be end ofit and measure be voltage drop across it. The resistor size will dictate the sensitivity of the system. It will work just like your temperature probe program I think.

 

[Lee Brown]

Couldabin, I have seen something that is very close to what you are working on at http://roamingdrone.wordpress.com/2009/03/25/lightrails-dynamic-external-exposure-control-for-time-lapse/ and was considering something like it myself. It also uses the TSL230R as was suggested here.

 

[paul ron]

Now that is a very cool solution.