- #1

Icosahedron

- 54

- 0

R_a v = v - 2a(va)/ (aa) , where v and a are vectors and v is relected along the axis orthogonal to a

thanks

You are using an out of date browser. It may not display this or other websites correctly.

You should upgrade or use an alternative browser.

You should upgrade or use an alternative browser.

- Thread starter Icosahedron
- Start date

- #1

Icosahedron

- 54

- 0

R_a v = v - 2a(va)/ (aa) , where v and a are vectors and v is relected along the axis orthogonal to a

thanks

- #2

HallsofIvy

Science Advisor

Homework Helper

- 43,021

- 970

Decompose [itex]\vec{v}[/itex] into components parallel to and orthogonal to [itex]\vec{a}[/itex].

Recall that [itex]\vec{a}\cdot\vec{v}[/itex] can be defined as "[itex]|\vec{a}||\vec{v}| cos(\theta)[/itex] where [itex]\theta[/itex] is the angle between [itex]\vec{a}[/itex] and [itex]\vec{v}[/itex].

From simple trigonometry, the component of [itex]\vec{v}[/itex] parallel to [itex]\vec{a}[/itex] is [itex]|\vec{v}|cos(\theta)= \vec{a}\cdot\vec{v}/|\vec{a}|[/itex]. To get a "vector projection", a vector in the direction of [itex]\vec{a}[/itex] with that length, multiply by the unit vector in the direction of [itex]\vec{a}[/itex], [itex]\vec{a}/|\vec{a}|[/itex]. That gives [itex](\vec{a}\cdot\vec{v}/|\vec{a}|^2)\vec{a}[/itex] as the "vector component of [itex]\vec{v}[/itex] parallel to [itex]\vec{a}[/itex]". Since the components parallel and perpendicular to [itex]\vec{a}[/itex] must add to [itex]\vec{v}[/itex], the component perpendicular to [itex]\vec{a}[/itex] is [itex]\vec{v}-(\vec{a}\cdot\vec{v}/|\vec{a}|^2)\vec{a}[/itex].

Now, reflecting [itex]\vec{v}[/itex] along an axis orthogonal to [itex]\vec{a}[/itex] gives a new vector having the same component

[itex]\vec{v}-(\vec{a}\cdot\vec{v}/|\vec{a}|^2)\vec{a} -\vec{a}\cdot\vec{v}/|\vec{a}|^2)\vec{a}[/itex]

[itex]= \vec{v}-2(\vec{a}\cdot\vec{v}/|\vec{a}|^2)\vec{a}[/itex]

- #3

dodo

- 695

- 2

I have an incomplete answer, but maybe you can finish it.

Let**v'** be the reflection of **v** over some axis. If you try to draw a vector **s** such that **v** = **v'** + **s**, using the triangle method to add the vectors **v** and **s**, you'll see that **s** is orthogonal to the reflection axis, and thus colinear to **a**.

The drawing should show an isosceles triangle, bisected by the reflection axis, where the two equally-sized sides are |**v**| and |**v'**|, the third side being of size |**s**|. From this triangle you should be able to calculate the magnitude of **s** as equal to 2 . |**v**| . cos theta, where "theta" is the angle between **v** (or **v'**) and the normal to the reflection axis. (Try it on the drawing!)

Now, the expression 2 . |**v**| . cos theta is equal in magnitude to 2 . (**v****a**) / |**a**|. Multiplying this quantity by an unitary vector in the direction of **a**, namely **a** / |**a**|, you should get your vector **s**... or almost. There is a +/-1 factor not accounted yet for: has **s** the same direction of **a** or the opposite?

(This sign is the missing part that I was lazy to complete - I'd guess it has to do with**v** being or not on the same side of the axis than **a**; examining the sign of (**v****a**) should give a clue.)

Doing the multiplication above (and correcting the sign), you should get**s** = 2 (**v****a**) . **a** / |**a**|^2 = 2 (**v****a**) . **a** / (**a****a**), and when replacing this expression for **s** into **v'** = **v** - **s**, you get your reflection formula.

Edit: Oh, I was pwned again. :)

Let

The drawing should show an isosceles triangle, bisected by the reflection axis, where the two equally-sized sides are |

Now, the expression 2 . |

(This sign is the missing part that I was lazy to complete - I'd guess it has to do with

Doing the multiplication above (and correcting the sign), you should get

Edit: Oh, I was pwned again. :)

Last edited:

- #4

Icosahedron

- 54

- 0

thanks a lot!

Share:

- Last Post

- Replies
- 9

- Views
- 447

- Replies
- 45

- Views
- 1K

- Replies
- 9

- Views
- 499

MHB
Vectors

- Last Post

- Replies
- 3

- Views
- 638

- Last Post

- Replies
- 26

- Views
- 673

- Replies
- 1

- Views
- 351

- Last Post

- Replies
- 4

- Views
- 401

- Replies
- 17

- Views
- 350

- Replies
- 7

- Views
- 413

- Replies
- 2

- Views
- 491