Understanding Infrared Light Sensors and the Inverse Square Law

[fluidistic]

Hi guys. I have a question.
See figure 2.1 in the PDF file, page 13.
Basically we have a radiation sensor (sensitive only to infrared light) and we have a light bulb in front of it. We must make some measurement to see that the inverse square law holds and so we must measure the "intensity" (in reality it's a difference of potential that the sensor can measure but anyway that's the idea) of light at different distances from the light bulb. My question is: Why do they consider the distance between the sensor and the light bulb to be the distance between the sensor and the filament instead of the distance between the sensor and the glass bulb?
I understand that the filament is probably over 1500K while the glass bulb might be over 400K which is much less... but still. Why not an intermediate distance between the glass bulb and the filament?
Doesn't the glass bulb stops the infrared light from the filament and since the glass bulb is heated by the filament, it emits the infrared light? I know it's a totally different story with visible light (the glass bulb doesn't really stop the visible light of the filament.) but infrared light is still stopped by the glass bulb I think...
What do you think?

 

[AJ Bentley]

Because the source of the light is the filament, not the bulb.
Although the bulb gets hot and emits some infrared of it's own, it's at a lower energy and can be safely ignored in a crude experiment like this one.
The glass does not stop the infra-red, most of it goes right through.

 

[Andy Resnick]

My question is: Why do they consider the distance between the sensor and the light bulb to be the distance between the sensor and the filament instead of the distance between the sensor and the glass bulb?
I understand that the filament is probably over 1500K while the glass bulb might be over 400K which is much less... but still. Why not an intermediate distance between the glass bulb and the filament?
Doesn't the glass bulb stops the infrared light from the filament and since the glass bulb is heated by the filament, it emits the infrared light? I know it's a totally different story with visible light (the glass bulb doesn't really stop the visible light of the filament.) but infrared light is still stopped by the glass bulb I think...
What do you think?

The first question was answered well by AJ Bentley. The second question is indeed a good one, and in general depends on the specific type of bulb. Usually it's an issue for UV light (eye safety), but the only way to know for sure is to measure the spectral distribution of light.

http://www.roperld.com/science/electromagneticspectraoflightbulbs.htm

http://www.gelighting.com/na/business_lighting/education_resources/learn_about_light/distribution_curves.htm

Basic incandescent light bulbs give off visible/near-infrared light with a spectral distribution very close to a 3300K blackbody. I haven't had the opportunity to look further out in the IR, so I don't know what the behavior will be.

 

[AJ Bentley]

Ordinary glass has a rapid fall-off of transmission with wavelength in the infra-red.
The short wavelengths of very hot bodies get through it quite easily. Longer wavelengths (room temperature for example) don't penetrate so well

That's how glass green houses 'trap' heat.

 

[fluidistic]

Ok thanks for the replies!

Ordinary glass has a rapid fall-off of transmission with wavelength in the infra-red.
The short wavelengths of very hot bodies get through it quite easily. Longer wavelengths (room temperature for example) don't penetrate so well

That's how glass green houses 'trap' heat.

Doesn't that mean that the sensor sees almost only the infrared light from the glass bulb and almost nothing from the filament? As I said, the sensor isn't sensitive to visible light but to infra red. So if the glass bulb stops well the infra red light, how can the sensor "know" that the light came from the filament?

So even if the source of light is indeed the filament for visible light, it isn't true for infrared light, as you said. But you said that the glass bulb infrared emission can be ignored for this simple experiment. So it implies that after all the glass bulb doesn't really stops the infrared light coming from the filament which contradicts your last post. I'm still confused.
By the way the light bulb had a difference of potential of around 11 V if that matters.

 

[Andy Resnick]

Ok thanks for the replies!


Doesn't that mean that the sensor sees almost only the infrared light from the glass bulb and almost nothing from the filament? As I said, the sensor isn't sensitive to visible light but to infra red. So if the glass bulb stops well the infra red light, how can the sensor "know" that the light came from the filament?

So even if the source of light is indeed the filament for visible light, it isn't true for infrared light, as you said. But you said that the glass bulb infrared emission can be ignored for this simple experiment. So it implies that after all the glass bulb doesn't really stops the infrared light coming from the filament which contradicts your last post. I'm still confused.
By the way the light bulb had a difference of potential of around 11 V if that matters.

I hope I didn't confuse the issue. You don't give any information on your detector, but let me step through the measurement:

The filament heats up to about 3300K (AFAIK), and is basically in a vacuum. Some bulbs are full of halogen gas, and arc lamps (and fluorescent bulbs) have trace elements present, and LED devices are even more complicated. A basic incandescent bulb has a 3300K tungsten filament in a vacuum. This radiation then impinges on the glass envelope, which, AFAIK, is soda-lime glass.

I couldn't easily find measured transmission spectra for soda-lime glass far outside the visible spectrum, but let's say the transmission goes to zero in the UV and below, and from 2000 nm and up, and is perfectly transmissive in between 300 nm to 2000 nm.

A 3300K blackbody has a peak emission at 878 nm, which is near IR. Consider a 100W incandescent: the fraction of transmitted light is around 95% (YMMV)- 95W. The 5W of absorbed light warms the glass to (say) 323K: 50C.

Now the glass re-radiates (approximately) as a blackbody with a peak wavelength around 8971 nm; 9 um.

So, depending on your detector, you will see two different things: for NIR dectors (silicon, Ge, etc), you will see the filament, unobscured. As the detector moves further and further in the MWIR and LWIR, the effect of the glass envelope becomes more and more significant.

Does that help?

 

[fluidistic]

Yes Andy, this helps.
So I think I have a bad news. Although I don't know how is made the sensor, it was extremely sensitive to my hand (even at 1 meter of distance). So it is sensible to temperatures of at least 33°C. Therefore I don't see how I can ignore the glass bulb.

 

[Redbelly98]

Don't know if this helps. From the manual (p. 5 of pdf file):

The PASCO TD-8553 Radiation Sensor (Figure 1)
measures the relative intensities of incident thermal
radiation. The sensing element, a miniature thermopile,
produces a voltage proportional to the intensity of
the radiation. The spectral response of the thermopile
is essentially flat in the infrared region (from 0.5 to 40
μm)

 

[fluidistic]

Hey thanks redbelly for pointing that out. What does "flat" mean in this context? Unsensitive?

 

[Redbelly98]

It means constant, or independent of the wavelength.

 

[fluidistic]

It means constant, or independent of the wavelength.

Oh ok thank you. If I'm not misunderstanding, it means that rather than depending on the absolute temperature of considered body, the sensor is sensitive to the intensity of the infrared light received. The more intensity, the more voltage. Very nice to know.