Engine Efficiency Comparison

[Batmanpuncher]

​Recently I did an experiment to do with the rocket equation and how it dictates the drop-off of delta v when adding more fuel.

Question; what is the effect of fuel mass on engines with different efficiency? Basically what is the trade for power to efficiency.

Hypothesis; if fuel tanks are added to a stack on top of different engines, then the more efficient, less powerful, engines will have more delta v per tank as more are added, because more efficient engines are more suited to heavier fuel masses , if you build for efficiency, than more powerful ones.

Procedure; I used the mod Kerbal Engineer Redux: http://forum.kerbalspaceprogram.com/threads/18230-1-0-4-Kerbal-Engineer-Redux-v1-0-18-0 by forum user cybutek. To take Delta V measurements ingame use the various engines.

Results: I've put my results into a nice graph. The fuel is plotted on the on the X axis uses increments of 200 units per tank (both liquid fuel and oxidizer, Ex: 3 tanks is 600 units.). Sorry a couple of the lines are a little close.

I have attempted to group the engines into diameter groups (0.625, 1.25, 2.5, 3.75).

I hope this gave you some helpful data in the future and a special thanks to the maker of the Kerbal Engineer mod: http://forum.kerbalspaceprogram.com/threads/18230-1-0-4-Kerbal-Engineer-Redux-v1-0-18-0 which is infinitely useful to me.

Edited November 14, 2015 by Batmanpuncher

[GoSlash27]

Batmanpuncher,

Thanks for the effort you put into this. Scientifically- minded program managers are always welcome!

Problem #1: This procedure assumes no payload. Payload fraction is paramount for stage design. Especially for upper stages.

Problem #2: This procedure doesn't take into account t/w ratio. DV means nothing if it can't be applied efficiently.

Problem #3: DV by itself isn't actually a useful measurement of efficiency from a design standpoint. What rocket designers need to know is which stage is lightest or cheapest.

We have some good mathematicians and scientists working this problem and your approach and organization would be a benefit to the cause.

Welcome aboard!

-Slashy

[Deutherius]

Hey Batmanpuncher, nice work! I agree with what Slashy wrote, but I'd like to add one thing - your experiment only goes to 1600 LFO (8 FL-T200 tanks). This is enough fuel mass for the smaller and lighter engines like Ant or Spark, but doesn't really give the bigger guys a chance to get anywhere (poor Mammoth doesn't even get to 2km/s dV). I see it has to be a lot of work putting everything together in-game and taking measurements one at a time, but you could, if you wanted to, repeat the process with bigger fuel tanks for the bigger engines. Or, you could calculate the values for all engines and all necessary fuel values in any increments using a suitable software (like Excel, which I'm assuming you made the nice graph in). This would tell you when exactly do the bigger engines get the dV budget you need.

As a bonus, you can then add a static mass for payload to the simulation, and even start easily calculating multi-staged designs

Keep up the good work!

[cantab]

Basically you've got two things to consider. Specific Impulse or Isp of course expresses your fuel efficiency. The other thing is dry mass - heavy engines will need more fuel to move themselves around, potentially undoing Isp gains.

This tool may help

http://meithan.x10.mx/KSP/engines/

But some very general rules of thumb for high delta-V applications:

If you're happy with under 0.5 TWR, the ion engines will be the lightest option.

For 0.5 to 0.75 it's the nuclear engines.

If you want more TWR you're best off going chemical, and usually Rhinos is the payload is big enough.