Fuel Density: KSP vs Science

[TChapman500]

If there's one thing I noticed about KSP, it's that Kerbals really waste their space with ultra-high-density fuels including air from the atmosphere (in the case of jet engines). Note: This thread is NOT intended to criticize KSP, but rather compare it to real life.

The fuel densities in KSP are measured in tonnes/L. We will convert these units to kg/m^3.

Liquid Fuel, Intake Air, and Oxidizer Density: 0.005 tonnes/L -> 5 kg/L -> 5000 kg/m^3

Mono propellant Density: 0.004 tonnes/L -> 4 kg/L -> 4000 kg/m^3

Solid Fuel Density: 0.0075 tonnes/L -> 7.5 kg/L -> 7500 kg/m^3

In a similar way, we'll convert the real life fuel densities back to tonnes/L.

Real Life Liquid Fuel Densities (Source: http://en.wikipedia.org/wiki/Liquid_rocket_fuel#Bipropellants):

Hydrogen: 70.85 kg/m^3 -> 0.07085 kg/L -> 0.00007085 tonnes/L

Methane: 422.62 kg/m^3 -> 0.42262 kg/L -> 0.00042262 tonnes/L (Correction, http://encyclopedia.airliquide.com/Encyclopedia.asp?GasID=41)

Kerosene: 915 kg/m^3 -> 0.915 kg/L -> 0.000915 tonnes/L

Hydrazine: 1021 kg/m^3 -> 1.021 kg/L -> 0.001021 tonnes/L

The liquid fuel that comes the closest to that which the Kerbals use is Hydrazine, with 1/5 the density of whatever fuel that the Kerbals use.

How about the oxidizer densities:

Liquid Oxygen: 1141 kg/m^3 -> 1.141 kg/L -> 0.001141 tonnes/L

Dinitrogen tetroxide: 1443 kg/m^3 -> 1.443 kg/L -> 0.001443 tonnes/L

The oxidizer that comes the closest to what which the Kerbals use is Dinitrogen tetroxide, with just under 1/5 the density of whatever oxidizer that the Kerbals use.

We can't forget the mono propellant:

Hydrazine: 1021 kg/m^3 -> 1.021 kg/L -> 0.001021 tonnes/L

Hydrogen peroxide: 1135 kg/m^3 -> 1.135 kg/L -> 0.001135 tonnes/L

Whatever mono propellant that the Kerbals use, it's a whopping 4 times denser than the densest mono propellant available to Humans.

And finally, the solid fuel (http://spacemath.gsfc.nasa.gov/weekly/6Page38.pdf):

Unknown Propellant Name: 1750 kg/m^3 -> 1.75 kg/L -> 0.00175 tonnes/L

Whatever solid propellant is being used by the Kerbals, it's almost 4.5 times denser than what we're using in real life.

Conclusion:

The Kerbals seem to be using a variety of very dense, unknown fuels that apparently are very good at matching the performance of our real life rockets in terms of efficiency. But at a heavy (no pun intended) price. The fuel tanks need to be quite small. If the Kerbals were to fill up those tanks with fuel in order to optimize space efficiency, nothing that the Kerbals build could withstand the extreme weight of the tanks. At least not without some upgrades. Even when realistic fuel densities are applied, the structural reliability is so low that, even as "light" as those fuels are, nothing that the Kerbals can build would be able to support a fully optimized fuel cell.

Edited May 8, 2013 by TChapman500
Corrected Liquid Methane Desnity

[HoY]

Glad to see someone do the math on that have you worked out how large the tanks are relative to their fuel volume? That one is simple but I just haven't don't the math myself yet

[Alistone]

http://en.wikipedia.org/wiki/Tonnes

maybe they switched their defination with the alternate usage where a tonne is only 10kg

Short ton, Long ton, Metric Ton, historical tonnes of materials... ton is varied enough that I try to ask for its definition rather than assume the one being used at the time.

[numerobis]

Your methane density is fishy: how is it orders of magnitude less dense than everything else, even less dense than H2?

[Dictator Dan]

My guess is to make it harder to reach LKO, given its smaller diameter, and lower atmosphere. Same reason the Kerbal engines have a horrible TWR compared to real ones. Still its awesome to actually see the math done on it.

[leax256]

Everything else in the kerbal universe is super dense...why not

[tavert]

The densities of these real-life substances depend highly on what temperature and pressure you're storing them at. You're quoting a single number for each chemical which likely does not correspond to typical storage conditions in rocket tanks.

[TChapman500]

Your methane density is fishy: how is it orders of magnitude less dense than everything else, even less dense than H2?

The units displayed on Wikipedia were in g/L, which when converted to kg/m^3 yield the exact same value. I was quite surprised at the results myself.

EDIT:

Glad to see someone do the math on that have you worked out how large the tanks are relative to their fuel volume? That one is simple but I just haven't don't the math myself yet

The tanks either have a 1 or 2-meter radius. Rounding out the height, the 1-meter tanks have a 2 (assuming shortest tank has a height the same as it's diameter), 4, or 6-meter height. This gives the 1-meter tanks a volume of 6.283 m^3 -> 6283 L, 12.566 m^3 -> 12,566 L, or 18.850 m^3 -> 18,850 L. Those tanks only hold 200, 400, and 800 L of fuel respectively.

Edited May 8, 2013 by TChapman500

[Sir Nahme]

bah cant get my edit right, forget it lol

simple answer. It looks like you are possibly using the density for methane GAS not liquid

Edited May 8, 2013 by Sir Nahme

[HoY]

I think the common consensus is the Oscar b size is 0.625m, next size up is 1.25m, and the largest is 2.5m

So the smallest 1.25m tank would be 1.534 m3 or 1,534 Litres. Obviously the tanks are double walled, and there are smaller tanks internally to store both the liquid fuel and oxidizer seperately, but there still a large discrepancy there lol.

Edit: what was I thinking... Double walled? This is KSP.. If it doesn't explode on your first attempt you probably forgot to activate the engines. I seriously doubt they have safety features like double walled tanks!!

Edited May 8, 2013 by HoY

[shand]

dont forget that sizes have been compressed 10(ish) fold. which gets liquid fuel much closer to hydrogen.

it seems kerbals are actually using things up to half as dense with that in mind!

[TChapman500]

bah cant get my edit right, forget it lol

simple answer. It looks like you are possibly using the density for methane GAS not liquid

You are correct. I have made a correction to the first post.

[Mattasmack]

Hi TChapman,

I've seen similar discussions about fuel density and tank capacities come up on the forum a couple of times before. There is a simple explanation for your findings regarding kerbal fuel densities: the unit of volume in KSP is not liters! Despite their common usage on the forums (and despite the fact that KSP uses meters for length, tonnes for mass, kiloNewtons for force, etc.), the unit of volume is not quantified in the game. I've seen the unit of volume referred to as a 'fuel cubit' (maybe that should be 'kubit'), and the numbers all make much more sense if you use a volume unit of five liters (i.e., 200 kubits = 1 m^3). Then, voila! The fuel densities are no longer five times higher than expected, and the capacities of the fuel tanks are no longer five times smaller than expected. I don't know why the volume unit was chosen that way, and I do wish the developers would change the unit to liters, as it would eliminate some confusion.

If there's anything in the game itself that does identify the volume unit as a liter I'd like to hear about it, but I don't believe there is.

[Brotoro]

Yes, I don't think the fuel can be measured in our liters. An FL-T400 standard fuel tank holds 400 units of fuel/oxidizer. If those units were our liters, that would be a liquid volume of 0.4 cubic meters. But an FL-400 tank has dimensions of about 1.25-meter diameter by ~2-meter length... so the volume of that tank is about 2.45 cubic meters... so there's either a hugh amount of empty space in that tank, or the fuel units are not our liters.

If we allow for some space taken up by tank wall thickness and bulkheads between the fuel and oxidizer, maybe the tanks hold ~1.8 cubic meters of fuel/oxidizer, which gives an average density in the range of 0.0011 kg/liter, which is in the right ballpark for kerosene and nitrogen tetroxide.

[rkman]

If you shrink the universe by a factor 10, keep gravity the same, and still want to get into space "not because it is easy, but because it is hard" - then something has got to give. I this case what gives is the density of stuff.

(quote reference for the uninitiated:

)

[astocky]

To be fair, the ISPs used are all over the place as well. I think it is all a ploy to make the game fun.

Example of some reasonable ISPs (vacuum) for the supplied fuels:

Hydrogen - Flourine => 479

Hydrogen - Oxygen => 468 (SSME 363SL, 452Vac)

Methane - Oxygen => 377

Kerosene - Oxygen => 356

UDHM - N2O4 => 336 (not commonly used 285SL, 333Vac

Solid (space shuttle / possible) => 269 / 300

NERVA => 824

Timberwind (more advanced NERVA) => 1000 [and thrust to weight ratios about 20 times greater than NERVA]

I can't find a reference for mono-propellant RCS but it is somewhere in the 300 - 310 range i believe. Scratch that found Hydrazine monopropellant about 280-300 secs. Can be boosted up to 500secs using an electric arc though if power is not a limitation.

Check out:

http://www.astronautix.com/props/index.htm

if you want loads more detail

Edited May 9, 2013 by astocky
updated some engines