Translating between KSP and IRL

[MaxwellsDemon]

I was reading up on some rockets the other day, and it occurred to me to wonder... someone surely has worked out the ratio of "real world" engine power vs. "kerbal world" engine power.  I understand that the kerbal engines are significantly overpowered relative to their real-world counterparts... is there a mathematical relationship I can use to translate?  For instance, the first stage engines of the Long March 2F are reported at 3,256 kN for a burn time of 291 seconds; would the kerbal equivalent be 3256x, where x is some conversion factor (for the sake of discussion let's say 0.25, so that would make Kerbal kN = 814)?    (Am I even asking the right question?) 

[Eric S]

Actually, as I understand it, kerbal engines are underpowered, it's the scale of the solar system that makes them feel overpowered.  RL rocket engines have higher TWRs, real life fuel tanks have a lower dry mass.  Of course, kerbal engines also have a lot of simplifications and abilities that real life engines tend not to have, like infinite restarts and infinite throttleability.

[regex]

22 minutes ago, MaxwellsDemon said:

I understand that the kerbal engines are significantly overpowered relative to their real-world counterparts...

No, if anything they're much heavier than IRL engines with comparable stats. All of the parts are much heavier to compensate for the tiny solar system.

22 minutes ago, MaxwellsDemon said:

is there a mathematical relationship I can use to translate?

Not really. Although KSP engines are based on the theoretical maximums of Aerozine50/NTO engine performance the masses are entirely out of whack (although I seem to recall that parts are about 9 times heavier dry than they should be while fuel units are somewhere around 5 liters, roughly?)

22 minutes ago, MaxwellsDemon said:

(Am I even asking the right question?) 

No, you're not. The Long March 2F takes about 4.8 tons to LEO (let's assume a 200km x 200km orbit for convenience), so to make something similar in stock create a rocket that takes 4.8 tons to 80km x 80km Kerbin orbit.

[MaxwellsDemon]

Got it--- trying to work the problem backward.  Bad habit of mine!   

[Shadowmage]

3 hours ago, MaxwellsDemon said:

I was reading up on some rockets the other day, and it occurred to me to wonder... someone surely has worked out the ratio of "real world" engine power vs. "kerbal world" engine power.  I understand that the kerbal engines are significantly overpowered relative to their real-world counterparts... is there a mathematical relationship I can use to translate?  For instance, the first stage engines of the Long March 2F are reported at 3,256 kN for a burn time of 291 seconds; would the kerbal equivalent be 3256x, where x is some conversion factor (for the sake of discussion let's say 0.25, so that would make Kerbal kN = 814)?    (Am I even asking the right question?) 

My rule-of-thumb for engine thrust scaling is to use the 'established' 64% scale of the KSP parts, and use square scaling (engine thrust depends on the area ratio of chamber-throat-nozzle; area scales on the square).

3256 * 0.64 * 0.64 = 1333.6576kN if converted to a 'stock-sized' KSP engine.

 

Now, the mass is a bit more complicated.  My personal conversion-scheme, is to take whatever the real-world TWR of the engine is, and multiply that by 0.25.  Now take the thrust calculated above, and figure out what the mass would need to be to get the adjusted TWR from the scaled thrust.  KSP engines seem to have about 1/4 of the TWR of 'role-equivalent' engines compared to real-life (ignoring upper-stages engines in KSP are grossly overpowered for gameplay reasons).  Real world lifters are TWR > 100; KSP lifters are TWR ~25-30.

This could also be calculated by the real-world mass and using the square-scaling from the thrust:
KSP-Mass = realMass * 0.64 * 0.64 * 4.0

So if the real-world TWR = 100, the scaled TWR = 25.  If the 'scaled' thrust is ~1333kN, that would give a 'scaled mass' of ~5.3t (at a TWR of 25).

 

These calculations all assume that ISP does not change / is not scaled (why would it?).

But of course, you can't really just scale a rocket engine, so please don't think of the above as anything close to realistic.  (its just the best that I could come up with that works consistently when applied to real-world engines)

Edited December 4, 2017 by Shadowmage

[MaxwellsDemon]

Yeah... my thought was to try to use the thrust and burn time of some real-world rockets as a basis for creating some KSP designs, but it sounds like doing it that way will be more trouble than fun!

[regex]

26 minutes ago, MaxwellsDemon said:

Yeah... my thought was to try to use the thrust and burn time of some real-world rockets as a basis for creating some KSP designs, but it sounds like doing it that way will be more trouble than fun!

Well, it's not that it won't be "fun" per se, it's that the rockets will literally be overkill for Kerbin. Remember that Earth orbital velocity is on the order of 7.9km/s whereas Kerbin is more like, 2.4km/s, so all those real world burn times are far too long for KSP (you'll be well on your way to Eeloo by the time you're done with the flight). Not to mention all the disparities in part masses and other things...

If you want to build something similar to an IRL rocket try to match payload to low Kerbin orbit or keo-stationary orbit.

And on a side note, you'd be surprised what you can do with a ton or two in LKO.

[NSEP]

Stock engines aren't OP, the Solar System is just super small. I think its well balanced, its not too OP that you can go everywhere with one tank and engine and not too UP than you need to have a bazillion engines to get to orbit.

[Pwnstarr]

 

On 12/4/2017 at 12:03 PM, regex said:

No, you're not [asking the right question]. The Long March 2F takes about 4.8 tons to LEO (let's assume a 200km x 200km orbit for convenience), so to make something similar in stock create a rocket that takes 4.8 tons to 80km x 80km Kerbin orbit.

I agree with this. OP, perhaps another way to think of this is to determine the ratio of launch mass (or stage mass) to delta V for real rockets and Kerbal rockets. After all, rocket capabilities are usually described as the mass of payload they can deliver to specific orbits (LEO, MEO, GTO, etc.) - the deltaV requirements of those orbits are pretty well known. 

[magnemoe]

On 4.12.2017 at 10:50 PM, regex said:

Well, it's not that it won't be "fun" per se, it's that the rockets will literally be overkill for Kerbin. Remember that Earth orbital velocity is on the order of 7.9km/s whereas Kerbin is more like, 2.4km/s, so all those real world burn times are far too long for KSP (you'll be well on your way to Eeloo by the time you're done with the flight). Not to mention all the disparities in part masses and other things...

If you want to build something similar to an IRL rocket try to match payload to low Kerbin orbit or keo-stationary orbit.

And on a side note, you'd be surprised what you can do with a ton or two in LKO.

This, note that 2.4 Km/s is pretty much the speed of an falcon 9 first stage before it does its braking burn and landing. So yes an falcon 9 would be able to put almost 100 ton into leo and be reused. That is Saturn 5 level payloads. Yes with an hydrolox upper stage you could run an Apollo moon landing mission on an falcon 9. The X-15 space plane would also reach orbit. 
On the other hand reentry heat is scaled up in KSP to so the falcon 9 first stage would not survive reentry.