View Full Version : OT radiation sensors

12-02-2010, 03:52 PM
Suppose I have a window that has a set of blinds on it. Suppose further that these blinds can be opened and closed by a computer controlled motor.
Is there any type of sensor(s) that could tell the computer if opening the blind would result in a thermal gain or loss in the room. ie is there more radition leaving the room than entering.

Liger Zero
12-02-2010, 04:23 PM
So you are looking for a heat sensor not a radioactive emissions sensor?

12-02-2010, 04:29 PM
I'm not sure what I am looking for. I just want to know if it is better to open or close the blinds. I have not been able to figure out what I need to sense to make this decision.

12-02-2010, 04:45 PM
From my personal experience, the only time my dining room heats up when I open the blinds, is when the sun is shining through the window. So a light sensor set such that it is activated when the sun is shining and positioned to shine into the room would probably work reasonably well. Might have to put it in a tube so it is not affected by ambiant light.


P.S. Someone could probably devise something so the sensor could be automatically positioned to where the sun should be on any given day. :)


12-02-2010, 05:05 PM
A peltier cooler module can be used as a sensor for whether heat is being radiated in or out.

Astronomers use them as cloud sensors as they can detect the difference between the heat of a cloud or the coldness of outer space.

12-02-2010, 05:09 PM
Hi Smoggy:

I've thought about this a fair bit. I live in a passive solar house. I've gone as far as buying a motor but haven't done much else.
What I was going to do for a sensor is use 2 thermistors, one on a black background, the other on white. Then pick a differential and trigger on that.
Not perfect, but I think it would work well enough.
I was going to add a min. time between blind actuations as well so you don't get too much action on a partly sunny day.

Edit: Are you going to see the sun in PEI before March?

12-02-2010, 07:12 PM
Have two flat black surfaces, one exposed to open window, and one nearby but not in a position to receive sunlight. Comparing the temperatures of the two surfaces would tell you if there was enough sunlight to warrant opening the blinds.

12-02-2010, 07:26 PM
I would use a thermometer.

Watch the trend and and decide based on that. I use a USB connected thermometer from DX to monitor my cold water preheat exchanger in the basement.

SKU 48285 is thermometer only for $8.30 and SKU 42591 is themometer and hygrometer for $20.87.

The software that comes with it sucks mightily but there is free open source software for it that works well. It is even internet enabled so you can check the temperature from around the world.

It will also export to Excel to produce pretty charts and graphs. Since it is open source you can modify it easily in Visual Basic.

This is data collected from my pre heater tank. It has both solar input for summer and flue heat recovery from the gas furnace in winter. Sometimes up here that happens in the same week.


12-02-2010, 07:48 PM
Have two flat black surfaces, one exposed to open window, and one nearby but not in a position to receive sunlight. Comparing the temperatures of the two surfaces would tell you if there was enough sunlight to warrant opening the blinds.
the two surfaces need to be mounted in
vacuum tubes . essentially is the room warmer w/m2 (radiation only) than the outside light.

12-02-2010, 08:31 PM
All you need to do is to measure the temperature of the blinds. If they are warmer than the room air then they are absorbing more radiation than they are reflecting and emitting back out the window.

12-03-2010, 03:00 AM
You basically need to sense two temperatures, one inside the room and one from a heatable surface inside the window, but outside the blinds.

Each one should be isolated from the conditions the other experiences. As such, the inside temperature sensor could be mounted against the wall near the window, but insulated from the wall by a thickness of the lightest styrofoam you can find, probably white. This could be a two or three inch square piece of 1/2 or 1 inch foam with a thermistor pressed into one side and covered with an aluminum plate which is painted a dull color- more on the color later. The outside facing sensor could be built the same, but with the styrofoam facing the outside of the blinds- in other words towards the room. Both of these should be close together, so you could integrate a differential amplifier circuitry on a piece of circuit board which connects on one edge to the inner sensor, and on the other edge to the outer sensor. With luck the circuit board could sit against the window frame, with the sensor sandwiches coming off it at right angles. This way of putting it together might allow for the blinds to not interfere with it at all. Having the sensors this close together could help prevent problems from air currents around the window.

I was going to suggest making the sensor package one piece of foam, with a sensor on each side, thermally coupled to the aluminum plate on each side. The amp circuit would be on the same package, with basically just a power wire from a wall adapter coming to it, and a control wire coming off it to run the blind operating mechanism. My concern though is that the blinds are going to interfere with the inside sensor- but maybe that wouldn't happen.

In any event, whatever you coat the aluminum surfaces with is going to determine how well they respond to both the inside temperature and the solar radiation. My first thought was to somehow determine what in the room is going to be absorbing the energy, and try to duplicate that in the coating you use. I would guess that a flat semi-dark color might work, with flat black possibly being too dark. I'm just guessing though- my thinking is that flat black in the window might give a false sense of how much the room might absorb. Maybe you'd want the inside sensor surface to be flat black to better match the room temperature- that's a possiblity that might work.

12-03-2010, 03:29 AM
No need to be so complicated. If the inside of the blinds when closed are above room temperature then the incoming energy is greater than the amount reflected and re-emitted back out by the blinds. So, open them. The room will absorb the sunlight on whatever surface it illuminates. Much less energy will be radiated back out since the effective aperture will be much smaller for something much further from the window(s).

Only one thermometer is required. To decide if the blinds should be opened watch the temperature trend and when it is increasing open the blinds when they reach a temperature above the desired minimum setpoint.

Continue to measure the temperature trend once the blinds are open. If it is falling then close the blinds.

BTW, I first instrumented a room in this way in about 1981. The room was our greenhouse and the blinds were a cooling fan and motorized vent system. All I needed to do was watch the trend with a moving average of the last several minutes.

12-03-2010, 09:24 AM
Hi Smoggy:
Edit: Are you going to see the sun in PEI before March?

We have had an exceptionaly good fall. Lately it has been cloudy alot but warm. In jan and feb we get an average of 2 hours of useful sunlight per day (yech).

Thanks for all the suggestions everyone. I do like the idea of a single sensor monitoring the blinds. It sounds good in theory but in practice how would you measure the temperature of the blinds?

12-03-2010, 10:34 AM
It sounds good in theory but in practise how would you measure the temperature of the blinds?

That depends on how much you want to do yourself. If you are looking for a prepackaged solution then you can either use a contact sensor as I listed above or get fancy and buy a pyroelectric remote sensor kit which is similar to the sensors used in handheld IR thermometers. If you want to do-it-yourself using a computer of any sort then a standard 1n34A diode makes a very linear sensor.

To use it build a simple 555 timer oscillator and put the diode in series with the lower timing resistor. The circuit will then produce a pulse rate that is linearly dependent on the temperature. By counting the the pulse width you have the temperature. The 1n34A is in a glass envelope and so it is also light sensitive. It may be painted black or left clear to integrate both light and heat.

The diode can be quite a distance remote from the 555 since it is a low impedance circuit. With fine wires of the litz type you can easily attach it directly to the blinds somewhere. Another option is to detect the temperature just above the blinds since if they are warm the air will be rising from them.

Sources for sensors of every possible type are:



12-03-2010, 10:48 AM
Evan, your single temperature sense idea makes good sense but what's wrong with simply using a thermistor?
The resistance change with temperature is dramatic so a quite precise temperature could be determined with a voltage divider and comparator, producing a simple on/off output.

Or, if he wants to track and record temperature then watching voltage is easier than measuring duty cycle.

Here's (http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=kc006e) a thermistor. $4.

12-03-2010, 11:17 AM
Counting duty cycle gives an A to D function without an A to D convertor.

12-03-2010, 01:36 PM
This project sounds interesting but I believe you will find it is rather pointless. I dont want to sound negative or discouraging, but I think you should have your lights on during a dark night and measure the temperature difference between the surface of the blinds and the ambient air. This will demonstrate the amount of energy that you are losing by leaving your blinds open. Most indoor lighting is relatively low wattage which yields very low levels of thermal radiation.

12-03-2010, 01:52 PM
Having the blinds closed prevents convection losses as well as radiant losses, day or night. This is especially so if the blinds are the vertical type and extend close to the floor. This is what we have and closing the blinds helps a great deal to slow heat loss in winter.