What determines if something sinks or not?

[SSSPutnik]

Can anyone tell me what determines if something sinks or not?

A KAS anchor, sinks. (1 ton)

A Kethane drill, floats. (1.5 ton)

An orange tank, about 2/3 full or more, sinks.

[KerbMav]

Density and buoyancy, if the density of an object is lower than that of the medium around it, it will float.

Relevant here is the density of the whole object, not just e.g. the material of a ships hull: 10 tons of solid steel will sink, formed into a hollow object like a ships hull or just simply a ball, you will have to calculate the density by dividing the weight of the metall and the air inside it by the total volume of the object.

Make a solid ball of 10 tons of steel and it will sink. Remove steel from inside the ball and replace it with air and it will float. Cut the ball in half an both halves will float (when positioned correctly of corse). Crush them into a lump of steel and it will sink again.

So, the drill has a relative big volume for its higher mass compared with the anchor, so its density is lower. (And it seems to be completely waterproof, too. )

But it is best seen with the tank, the more air replaces the fuel, the lower its density is.

Though I do not know if KSP calculates all this correctly, your experiments seem to prove this.

[capi3101]

Fairly easy to experiment - water in RL has a density of 1000 kg per cubic meter in standard atmosphere (1013 mb, or 101.3 kPa depending on how you've been edumacated) and at standard temperature (about 25 C). Fuel tanks in KSP are cylinders, and density is defined as mass over volume. The volume of a cylinder is its area (pi times the square of the radius) times its height; you'd need height and radius data for a particular fuel tank to get its volume, but that data is floating around on the forums. Units of oxidizer and units of liquid fuel are both known to have a mass of about 5 kilograms, so you can use tweakables to predict at what point the tank would float or sink.

[SSSPutnik]

I am familiar with how boats work, it just seems odd how it works in KSP.

(A grappling hook from KAS basically bounces off the surface).

I am guessing Kerbals water is much denser than normal water based on observation.

[capi3101]

That's also something that can be tested, and in the same way; you just tweak the amount of fuel and oxidizer in a tank of a known size. Now, if its density is less than 1000 kg/m^3, it SHOULD float. If it doesn't, you've confirmed that Kerbin's ocean is less dense than water. If it floats at a density greater than 1000 kg/m^3, you've confirmed that Kerbin's ocean is denseer than water.

I suppose that 1000 kg/m^3 is for fresh water - salt water is slightly less dense. Assuming the liquid in Kerbin's oceans is water, this might be a way of guesstimating the salinity of the planet's oceans.

[Aedile]

Actually salt water is denser, and has higher surface tension.

And anyway, buoyancy is linked to the vessel displacement, remember the guy with the bath tub.

KSP works in a similar way, I am unsure about the 'water' density, and I am not sure if even the objects dimensions can be considered as meters.

However kirbin 'water' simulation (due to unity simulation) is rather imperfect.

Surface tension and water drag do not work very well, so it's pretty easy to make something float, but rather difficult to move around.

In addition it's less dangerous to land rather than splashdown.

[Pecan]

[i always wanted a concrete boat, but have never sailed one. Yes, they do exist, do work, are very strong, cheap and easy to repair in 'low tech' parts of the world and, no, they aren't great performers]