# 687: Dimensional Analysis

Dimensional Analysis |

Title text: Or the pressure at the Earth's core will rise slightly. |

## [edit] Explanation

Cueball has a hobby — showing correct calculations according to the dimensional analysis — but with ridiculous correlations of uncorrelated events and measurements. Here Cueball is teaching a class and uses this trick to *convince* his students that the Toyota Prius combined EPA gas mileage is somehow connected to the constant Pi via the Planck energy, the pressure at the earth's core and the width of the English Channel.

This comic makes fun of how scientists — often physicists — use dimensional analysis to quickly check if a given formula can possibly relate to a physical system or if there were some obvious mathematical errors in its derivation. Dimensional analysis here refers to the check if both sides of the equation arrive at the same physical unit if the units of all variables get plugged into the equation. This usually requires knowledge of the system of units and the relation between different physical units.

The comic uses the following equation to make fun of it:

(Planck energy) / (Pressure at the core of the earth) * (Prius combined EPA gas mileage) / (minimum width of the English Channel) = π

### [edit] Dimensional analysis

The right hand side is dimensionless, it's the constant π = 3.14... which is defined by the relation of two lengths, the circumference and the diameter of a circle. The left hand side requires to plug in the dimensions of the named physical quantities:

- Planck energy: given in Joules [J]
- Pressure at the core of the earth: Given in Pascals [Pa]
- Prius combined EPA gas mileage: miles/gallon, SI units: meters/litres [m/l]
- minimum width of the English channel: meters [m]

When plugged into the left hand side this amounts to:

[J / Pa * (m/l) / m] = [Nm / (N/m²) * (m/m³) / m] = 1

Using the following unit relations (this does not reduce units to the seven SI units, but does use some derived units):

- 1 Joule = 1 Newton-meter [J] = [Nm]
- 1 Pascal = 1 Newton per square-meter [Pa] = [N/m²]
- 1 cubic-metre = 1000 litres [m³] = 1000 [l]

Note that for dimensional analysis constant factors are not taken into account. Here square brackets are used to denote dimensional analysis. In the above equation the unit of force (newton) as well as all the units of length (meter) cancel out each other.

Another aspect of the comic is, that sometimes dimension analysis of equations that were not derived but rather "made up" can provide insight. However, in reality such an equations would have to be somehow "motivated", which is more of an art than science and requires great experience in the field the equation should relate to. The presented equation combines values that have no immediate causal relation with each other, so it does not make sense. Furthermore, since the values have absolutely no causal relation to each other, the ratios presented are simple coincidence; despite Cueball's claim, building a better Prius would not cause any changes to the English Channel.

The title text also refers to this, as a higher pressure at Earth's core could also balance the equation, keeping the result constant equal to Pi. The Planck energy is an absolute, however, so it is not mentioned as a way to balance the next version of Prius.

### [edit] Some numbers for this calculation

The Planck energy is the only nearly exact value we do have. Compared to other Planck values it is very large (macroscopic).

E_planck = 1.956 x 10^{9}J = 1.956 x 10^{9}Nm

Pressure at the core of the earth ranges from 330 to 360 gigapascals. Using a simple value like this:

P_core = 350 GPa = 3.5 x 10^{11}N/m²

Prius combined EPA gas mileage: For the third generation (from 2010) the City mileage is 51 mpg and the Highway mileage is 48 mpg. But it is the combined EPA gas mileage which is used in the equation and that is 50 mpg.

50 mpg => 13.2 miles per litre => 21,000,000 meter per m³

Minimum width of the English Channel is about

33.1 km or 33,100 meters.

Calculating from these values you will get π=3.54... that is pretty close to π=3.14... while using a Planck value. According to Cueball this will be within the experimental error (the combined error for all four numbers - none are exact numbers). For instance if you tried the ePrius you would probably get closer to that target — as the mileage in real life usually is somewhat lower than the value given — and that would reduce the result.

## [edit] Transcript

- My Hobby:
- Abusing dimensional analysis

- [On a blackboard.]
- (Planck energy/Pressure at the Earth's core) x (Prius combined EPA gas mileage/Minimum width of the English Channel) = π

- [Cueball indicates this equation with a pointer in front of a class.]
- Cueball: It's correct to within experimental error, and the units check out. It must be a fundamental law.
- Student: But what if they build a better Prius?
- Cueball:
*Then England will drift out to sea.*

**add a comment!**⋅

**refresh comments!**

# Discussion

Collapsed content |
---|

Is it customary to just slap a new contributor after 5 minutes because the page was not done perfectly on the first editing? Now I will NOT research and add the actual numbers you would have to plug into the equation and I will leave this to Dave22. Genius. Stamfest (talk) 09:36, 1 April 2013 (UTC) - We do put those fields there for a reason. Most editors just make a page and leave them incomplete with fields deleted or blank, oftentimes leaving admins with heaps of maintenance work to deal with. I probably *should* have waited for you to finish, but prior experience and me being busy with two simultaneous continuously updating comics led me to treat it like most other incomplete pages. Will wait in the future though, since I know how you edit now. Discussion hidden, as it covers matters outside of the comic.
__Davidy__²²`[talk]`10:01, 1 April 2013 (UTC)- And I added the comment of a (very) initial explanation to the initial version. If you only want immediately perfect content, then just say so at page creation-time. This is totally against all wiki rules I have ever experienced. It is also against basic netiquette to call new contributors names upon their first submission. You might consider to change your attitude if you run or are involved with such a project. OTOH, if this is YOUR project, you are free to treat contributors all the way you want, but do not expect them to add more content if you behave like that. But the good thing is: your reaction shifted my attention back to work. Thanks for that. This won't help the page, though. Bad luck, I guess. Also, I only found out now that it is not possible to delete accounts on a wikimedia wiki... Stamfest (talk) 13:10, 1 April 2013 (UTC)
This comments should be removed from this topic. --Dgbrt (talk) 20:22, 14 May 2013 (UTC) I am not sure if this is relevant but I laughed more reading Stamfest vs Daviddy22 then Randall :D but Dgbrt may have a point of removing it. 173.245.62.83ck |