Thermal Imager: Progress Update 3
Unfortunately, I have run into a fairly serious issue with this project. The problem is that the thermopile that I am using (MLX 90620) has a wide field of view, 60 degrees. This means that when I pan and tilt the thermopile and take successive scans, each scan area overlaps the previous one, resulting in an unusual image.
I have tried several ideas to resolve this, but it does not seem to be possible. I could, in theory, modify the pan and tilt distance between each scan in order to prevent overlap, but in order to do this I would need to know the exact distance from the thermopile to the target (since the diameter of the Field of View cone will increase with distance). Since I don’t have this information, I can’t do this. Even if I did have the distance info, with such a wide FOV, the image resolution would be very poor at even a short distance from the thermal sensor.
The only thing to do here would be to use some form of lens to focus the FOV cone to a column. I tried a fresnel lens that I had lying around, but this had no effect. I don’t know if you can collimate a thermal sensor like you would an optical device.
I may have to search for a different thermal sensor with a narrow FOV, or possible use a single thermal sensor. A single sensor would be incredibly slow to scan, but it would not have any FOV issues. I could also try buying several single-pixel thermal sensors and making a thermopile myself, although calibrating it perfectly may be difficult.
While thinking about my changes to the project, I also thought about other improvements. For example, I could include UV sensors (or just UV led’s which can also be used as sensors) to pick up UV light, I could also have uv emitters to turn my scanner into a black light.
I could incorporate conventional infrared diodes as well, and a simple camera sensitive to IR, so I could pick up Visible light, Near IR, Far IR, (thermal) and UV.
If I was even more ambitious, I could include a directional radio receiver with a very wide frequency range. This would give me the ability to detect radio signals, which I could then plot on the scanners screen, giving the user the ability to “see” radio waves, essentially.
The idea would be to make up for the disasappointly low resolution of the thermal image with the ability to see further into the Electromagnetic spectrum.
