Angle measurement of Polaris

Galileo Gallei measured the angular positions of the stars with his telescope. To adjust his telescope, he initially aimed at the Pole Star, which is always at the same position in the sky at an angle of α \alpha with the horizontal.

On a cold winter's night of 3 ° C , \SI{-3}{\celsius}, he conducted his measurements inside his study room at 20 ° C . \SI{20}{\celsius}. The starlight fell through his window and refracted twice on the glass. From his room, Polaris appeared at an angle of β \beta with the horizon.

What is the angle difference δ = β α \delta = \beta - \alpha between measurements inside and outside?

Assumptions:

  • Galileo's study room was in Venice, at a latitude of ϕ 45 ° . \phi \approx \SI{45}{\degree}.
  • Air is an ideal gas with the specific gas constant R air 300 J / ( kg K ) . R_\text{air}^\ast \approx 300 \text{ J}/(\text{kg} \cdot \text{K}).
  • Air is a dielectric with an electric susceptibility of χ air = ε air 1 = c ρ air ( where c 1 0 3 m 3 / kg ) . \chi_\text{air} = \varepsilon_\text{air} - 1 = c \cdot \rho_\text{air}\, \big(\text{where } c \approx 10^{-3} \text{ m}^3/\text{kg}\big).
δ = 0 \delta = 0 δ 1 \delta \approx 1'' δ 1 0 \delta \approx 10'' δ 1 \delta \approx 1' δ 1 0 \delta \approx 10' δ 1 \delta \approx 1^\circ Cannot be determined

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

Markus Michelmann
Apr 14, 2018

The Snell's law for the refraction on the glass gives n outside sin α = n glass sin γ = n inside sin β β = arcsin [ n outside n inside sin α ] \begin{aligned} n_{\text{outside}} \sin \alpha &= n_\text{glass} \sin \gamma = n_{\text{inside}} \sin \beta \\ \Rightarrow \quad \beta &= \arcsin \left[ \frac{n_{\text{outside}}}{n_{\text{inside}}} \sin \alpha \right] \end{aligned} with the refractive indices n glass n_\text{glass} of the glass pane as well as n inside n_ {\text{inside}} and n outside n_ {\text{outside}} of the air inside and outside. Although the air inside and outside has the same composition and pressure p 1 bar = 1 0 5 Pa p \approx 1 \, \text{bar} = 10^5 \, \text{Pa} , there is a difference in the temperature difference the density ρ \rho . With the ideal gas law follows p = R ρ T ρ = p R T n = ε = 1 + χ = 1 + c ρ = 1 + c p R T \begin{aligned} & & p &= R^\ast \rho T \qquad \Leftrightarrow \qquad \rho = \frac{p}{R^\ast T} \\ \Rightarrow & & n &= \sqrt{\varepsilon} = \sqrt{1 + \chi} = \sqrt{1 + c \rho} = \sqrt{1 + \frac{c p}{R^\ast T}} \end{aligned} Here we have use the fact that the air molecules show almost no magnetism at room temperature ( μ 1 \mu \approx 1 ), so that the refractive index of the air is only determined by dielectric properties ( n = ε μ ε n = \sqrt{\varepsilon \mu} \approx \sqrt{\varepsilon} ). Therefore, β = arcsin [ n outside n inside sin α ] = arcsin [ 1 + c p R T outside 1 + c p R T inside sin α ] arcsin [ 1 + c p R T outside c p R T inside sin α ] arcsin [ ( 1 + 1 2 c p R [ 1 T outside 1 T inside ] ) sin α ] α + 1 1 ( sin α ) 2 c p 2 R [ 1 T outside 1 T inside ] sin α \begin{aligned} \beta &= \arcsin \left[ \frac{n_{\text{outside}}}{n_{\text{inside}}} \sin \alpha \right] \\ &= \arcsin \left[ \sqrt{\frac{1 + \frac{c p}{R^\ast T_\text{outside}}}{1 + \frac{c p}{R^\ast T_\text{inside}}}} \sin \alpha \right]\\ &\approx \arcsin \left[ \sqrt{1 + \frac{c p}{R^\ast T_\text{outside}} - \frac{c p}{R^\ast T_\text{inside}}} \sin \alpha \right]\\ &\approx \arcsin \left[ \left(1 + \frac{1}{2} \frac{c p}{R^\ast} \left[\frac{1}{T_\text{outside}} - \frac{1}{T_\text{inside}} \right] \right) \sin \alpha \right]\\ &\approx \alpha + \frac{1}{\sqrt{1 - (\sin \alpha)^2}} \cdot \frac{c p}{2 R^\ast} \left[\frac{1}{T_\text{outside}} - \frac{1}{T_\text{inside}} \right] \sin \alpha \end{aligned} Here we have taken advantage of the fact that the refractive index of air differs only slightly from 1, so we used the approximations 1 1 + x 1 x 1 + x 1 + 1 2 x \begin{aligned} \frac{1}{1 + x} &\approx 1 - x \\ \sqrt{1 + x} &\approx 1 + \frac{1}{2} x \end{aligned} for x 1 x \ll 1 and also made an linear Taylor expansion for the arcus sine.

For the calculation of the angle β \beta we have to know the angle height α \alpha of the polar star. Venice lies at the latitude ϕ = 4 5 \phi = 45^\circ , so that the position of the polar star also results to α 4 5 \alpha \approx 45 ^\circ , since this star is located at the north pole of the sky. Therefore, sin α = 1 2 \sin \alpha = \frac{1}{\sqrt {2}} and thus δ = β α c p 2 R [ 1 T outside 1 T inside ] = 1 0 3 1 0 5 2 300 [ 1 270 1 293 ] 4.8 1 0 5 rad 0.002 8 1 0 \begin{aligned} \delta = \beta- \alpha &\approx \frac{c p}{2 R^\ast} \left[\frac{1}{T_\text{outside}} - \frac{1}{T_\text{inside}} \right] \\ &= \frac{10^{-3} \cdot 10^5}{2 \cdot 300} \left[\frac{1}{270} - \frac{1}{293} \right] \\ &\approx 4.8 \cdot 10^{-5} \,\text{rad}\\ &\approx 0.0028 ^\circ \\ &\approx 10'' \end{aligned}

Just like the website's name, brilliant solution! I couldn't have come up with this!

Aatman Supkar - 3 years, 1 month ago

Amazing problem. Brilliant solution.

Srikanth Tupurani - 3 years, 1 month ago

Amazing solution...

Siddharth Rai - 3 years, 1 month ago

We can get to the answer slightly more easily as follows... sin β sin α = 1 + c p R T 1 1 + c p R T 2 = 1 + 1 3 × 270 1 + 1 3 × 293 = 1.0000484 \frac{\sin\beta}{\sin\alpha} \; = \; \sqrt{\frac{1 + \frac{cp}{R^\star T_1}}{1 + \frac{cp}{R^\star T_2}}} \; = \; \sqrt{\frac{1 + \frac{1}{3\times270}}{1 + \frac{1}{3\times293}}} \; = \; 1.0000484 and so, writing β = α + δ \beta = \alpha + \delta , we have (with α . 4 5 \alpha \approx. 45^\circ ): cos δ + cot α sin δ = 1.0000484 cos δ + sin δ = 1.0000484 2 cos ( 4 5 δ ) = 1.0000484 4 5 δ = 44.99722268 5 δ = 2.77315 × 1 0 3 = 9.9 8 \begin{aligned} \cos\delta + \cot\alpha\sin\delta & = \; 1.0000484 \\ \cos\delta + \sin\delta & = \; 1.0000484 \\ \sqrt{2} \cos(45^\circ - \delta) & = \; 1.0000484 \\ 45^\circ - \delta & = \; 44.997222685^\circ \\ \delta & = \; 2.77315 \times 10^{-3} {}^\circ \; = \; 9.98 '' \end{aligned}

However, since we are giving atmospheric pressure to 1 1 SF, and R R^\star to 1 1 SF, the accuracy to which we have to calculate to identify δ \delta seems beyond what is reasonable. All we can reasonably say with this data is that cos δ + sin δ = 1 \cos\delta + \sin\delta = 1 , giving δ = 0 \delta = 0^\circ .

While we are at it, assuming that α = 4 5 \alpha = 45^\circ is not globally true. At the North Pole, α = 9 0 \alpha = 90^\circ , and α = 0 \alpha = 0^\circ on the equator. 4 5 45^\circ does work quite well for Italy, but it seems that required information of that nature should be provided in the question.

Mark Hennings - 3 years, 1 month ago

What is ρ \rho ? Does it represent air density? Or is it "moles per unit volume?" If the former, don't you have to multiply the answer by (the atomic weight of air)/1000?

Stephen Montgomery-Smith - 3 years, 1 month ago

Log in to reply

Oh I get it. I didn't understand the term "specific gas constant." So I am wrong.

Stephen Montgomery-Smith - 3 years, 1 month ago

I took a lucky guess... This is totally out of my range

Zhihua Yu - 3 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...