Light Sensor for Lights & Shades

youstra

Hi,
I'm working on a system with Lutron Homeworks, HAI & AMX. The Homeworks setup includes shades and lights in a home that has many windows.

The owner wants light and shade settings to be automatically based on outdoor light level. Does anyone know of a sensor that I could plug into any one of the three systems above to establish a low/high sunlight threshold?

Thanks,
Bill

DHawthorne

I tried this once, and it failed miserably. I had a photo sensor that read the outside light level, and everything did what it was supposed to do, but... there is no sensor out there that is capable for allowing for the way human eyesight works. It's relative. I simply could not find a setting that worked. On dark, overcast days, the sensor saw plenty of light, but the customer wanted it brighter. Adjusting for dark days threw off the settings for bright ones. We eventually scrapped the idea and just put in astronomical timers.

I can't advise you about the sensor either, I'm afraid. That was a good ten years ago, and if whoever I got the thing from is still around, I certainly don't remember who it was. Smarthome.com may have something you can use, but keep my warning in mind and be careful how much you promise.

ericmedley

I was going to suggest something that references what DHawthorne spoke about. I too have tried what you're describing using a photoelectric cell. What DHawthorne says is true. On overcast days, you get almost the same amount of light energy hitting the cell. Perception is everything. '

Why, for example, do people complain about a drum set at church services when the organ can empirically be shown to be louder using a sound pressure meter. It's because people percieve the drum set as being something that is loud and obnoxious.

I do have a suggestion using an idea that I use to control various outdoor lighting with the same idea in mind. I control the landscape lighting using the sunrise/sunset functions available. However I also have a weather offset. If it's clear, I use the exact time of sunup or sundown to control the lighting. ( I also have another offset, see below)

However, if it's cloudy or overcast, I move the 'on' time to a little before sundown and the 'off' time to a little after sunrise. And if it's raining, I move the times away even more. People percieve rainy days as being darker and gloomier.

Perhaps you could combine the use of a photoelectric cell with the weather report. You might get more 'human' results.

Incidentally, I also have a 'customer preference' offset too. It works the same way. If they think sunset still seems too bright to turn on the lights, then I adjust the customer offset. Similarly, if the customer has a lot of trees on their property, they may want the landscape lights on a little ahead of sundown because it seems to get darker earlier.

Both offsets work amazingly well. I have customers telling me that they think the system is reading their mind. They'll find themselves thinking that it's time for the lights to come on and they magically do.

You can use the AMX i!Weather stuff or write your own. I have written my own program that harvests the weather off the NOAA Telnet site. Both work quite well.

vining

What kind of sensors were tried? Were these simple make or break sensory circuits or variable output types.

I too have wanted to try this but never got around to it. I found a photo resister sensor that has:
? Linear Output conforming to Luminance
? Built in Optical Filter, for similar spectral response to the human eye

This looks like it could work depending how the circuitry is built around it. Then using a 4-20mA to RS-232 adapter from DGH you send different strings to the master depending on the current passing through the photo resister to the DGH which will vary based upon the amount of light present.

Has any one taken this type of approach?

Jeff Lockyear

Er, not meaning to threadjack, but Eric, do you have the telnet address and password for the NOAA site? I'd really like to check that out.

Jeff

ericmedley

Jeff Lockyear wrote:

Er, not meaning to threadjack, but Eric, do you have the telnet address and password for the NOAA site? I'd really like to check that out.

Jeff

Sure!

66.28.69.161 port 23

Just use hyperterminal.

To be perfectly clear and give credit where credit is due...

this the the NOAA data culled from Weather Underground and Alden Electronics. I use this service because it has been around for a long time and has not changed in ages. It's also a lot easier to get to the data quicker.


You are limited to using either Airport locations or weather stations. You can't, say, enter in a zip code. Also, international weather is very hard to program for since they don't collate the data into any particular protocol. It's basically a text version of whatever that station reports back. So, it's nuts to soup protocol for international.

Here again, I just usually give them the local forcast for the major city they're in, not the actually local or suburb they live in.

ericmedley

vining wrote:

What kind of sensors were tried? Were these simple make or break sensory circuits or variable output types.

I too have wanted to try this but never got around to it. I found a photo resister sensor that has:
? Linear Output conforming to Luminance
? Built in Optical Filter, for similar spectral response to the human eye

This looks like it could work depending how the circuitry is built around it. Then using a 4-20mA to RS-232 adapter from DGH you send different strings to the master depending on the current passing through the photo resister to the DGH which will vary based upon the amount of light present.

Has any one taken this type of approach?

I don't remember the brand of the thing inside the black box...
But I do think it was in fact just what you're describing. It was esentially a variable voltage reader to serial box. It output a 16 bit hex number at any level change.

vining

Hmm! Sounds like the same kid of animal just in a different skin.