running 4 nukes at 25%

[khyron42]

I completely agree with this. A recent ship I built had 4 LV-Ns. The burn time was 90 minutes. If I had 1, it would be more efficient, but it would have been a 6hr burn. I'm lucky when I get to play KSP for 6hrs, I don't want to be doing a single burn for that time!

Just a note for those who are trying for similar setups: You can use ALT+. to enter physical time warp while in space, taking it up to 4x while burning the engines if your computer can handle the physics calculations. That will let you do your 1 hour burn in 15 minutes.

Also, for very long burns to escape from LKO to an interplanetary trajectory, don't do it all in one burn. Do repeated small burns for about 5 minutes right at closest approach to Kerbin; The Oberth Effect will save you considerable delta-V that way. You do have to rebuild your maneuver node for each pass around the planet, but you can time-warp around to the next burn.

Edited February 13, 2014 by khyron42

[problemecium]

It's not that 3-symmetry is bad in theory. No matter what kinds of symmetry show up on the ship, as long as each "group" is radially balanced the ship should balance.

But KSP has KSP Physics. They generally mostly hold up. But (and this is the reason I don't do 3-symmetry) the coordinate system, and consequently the collision meshes, are based on orthographic axes, which equate to 4-symmetry. Thus it is impossible to get exact balance or make a perfect equilateral triangle, meaning no matter what you do the engines won't line up quite right. Now this issue is tiny, but I'm obsessive like that. Also a lot of parts have textures with 4-symmetry, so 3-symmetry looks ugly to me. Rambling rambling rambling...

[DerekL1963]

That said, if your craft is HONKING HUGE 4 LV-Ns isn't as bad as a lot of people say. A lot of times, adding an extra LV-N to a craft will take 5% of your dV but cut your burn time in half. That's a trade-off you should be willing to consider if not embrace.

A lot of people "say things" in a cargo cult like manner here on the forums because all they have are the "forum rules of thumb" and little clear understanding of the actual principles or of the often vast tradespace ruled out by those "rules of thumb".

Rather than simply "minimizing the number of LV-N's", you need to look at your T/W and mass ratios and the overall performance of the craft. I built a large tug, and six LV-N's turned out to be the "sweet spot" for performance as that held burn times and mass ratios down to a reasonable level. Shorter burn times are good because means any pointing errors have less time to propagate and because the longer the burn the more conditions deviate from the ideal ones inherent in your maneuver node. (Maneuver nodes are mathematical fictions that apply all the delta-V at a single instant at a single point in space - a ten minute burn will be a better approximation of that than a forty five minute burn will be.)

[Red Iron Crown]

But (and this is the reason I don't do 3-symmetry) the coordinate system, and consequently the collision meshes, are based on orthographic axes, which equate to 4-symmetry. Thus it is impossible to get exact balance or make a perfect equilateral triangle, meaning no matter what you do the engines won't line up quite right.

This is nonsense. Equilateral triangles are perfectly fine in a Cartesian coordinate system, and can be made arbitrarily precise. In KSP, with 32-bit floating point coordinates, this means at least 6 significant digits, which works out to nearest hundredth of a millimeter accuracy for any reasonably-sized part. The engines will line up right.

If you don't like 3-symmetry for aesthetic reasons, that's fine, but there is no technical reason not to use it.

[Vanamonde]

I don't think this is so. Having two units on one side is exactly balanced out by the shorter moment arms on that side.

That certainly makes sense, now that you mention it. However, when trying to rotate 3x symmetry ships not under thrust, I would get unwanted pitch while yawing and vice versa. And it was consistent with 3x ships but never happened with even symmetries. What would account for that?

[Alchemist]

That certainly makes sense, now that you mention it. However, when trying to rotate 3x symmetry ships not under thrust, I would get unwanted pitch while yawing and vice versa. And it was consistent with 3x ships but never happened with even symmetries. What would account for that?

If you had 3x symmetry on RCS it could result in some issues. It might get not properly balanced.

[theend3r]

If the burn would be longer than a few minutes then more engines = more dv because of the oberth effect. Less flustration is just a nice extra. My supertug used for carrying return missions with bases and large rovers to Eeloo and Moho uses 8 nukes and it's ideal. A few tons don't matter at all if your ship has a mass of over thousand of them.

[EtherDragon]

So... there is a lot of colloquial information floating around this thread - let us clarify with SCIENCE! instead of conjecture.

1. As Red Iron Crown pointed out - Delta-V is entirely determined by the engine efficiency and full to empty mass ratio.

It has nothing to do with TWR (thrust-to-weight ratio). Reminder on the math:

dV= Isp * g * ln(m0/m1)

Where:

dV is the amount of delta-V in m/s

Isp is the Isp of the engine, a measure of its efficiency

g is the gravity constant, 9.82

ln is the natural logarithm function (available on any scientific calculator)

m0 is the mass of the ship fully fueled ("wet weight")

m1 is the mass of the ship empty ("dry weight")

Running a rocket ship with lower thrust or higher thrust does not change its total Delta-V*.

Conclusion: You lose delta-V (assuming space operations) with more motors.

2. Gains from shorter burns using the Oberth effect are always smaller than gains from reducing waste mass.

The only real time Oberth has a large impact is in the case of very weak propulsion, like with Ion Drives - where you have to burn over several orbits. Increasing TWR means more empty mass, and the dV lost by doing so is larger than any dV gained from Oberth.

So, you are reducing your thrust to 25% with four engines - here is the math:

Total Thrust = 0.25 * T * 4

where T is the max thrust of each engine.

25% thrust, four engines.

Yes, your Burn Time goes up (because you cut consumption).

But your thrust goes down.

In the case of a single engine at 100% you get:

Total Thrust = 1.0 * T * 1

100% thrust, one engine.

Assuming T is the same (i.e. you are using the same engines in both cases) you end up with:

T = 0.25 * T * 4 ... the total thrust is exactly the same.

Conclusion: You lose nor gain any thrust in this configuration.

3. Don't confuse Burn Time with Delta-V:

Burn time is how long you can run your engines at full throttle - calculated simply as

t = r / m

t is burn time

r is rate of consumption per unit of time

m is fuel mass

So in the case of four engines running at 25% the burn time is the same as one engine running at 100% because:

t = 4*0.25*r / m

four engines, at 25% throttle (thus consuming 25% the rate)

For one engine:

t = 1*1.0*r/m

one engine, at 100% throttle (100% consumption rate)

Basically 4*0.25 = 1 thus:

1*1.0*r/m = 4*0.25*r/m

Note: t's units of time end up being the same as r's.

Conclusion: You lose nor gain any burn time in either configuration.

*Caveats:

When ascending, the higher your TWR (to a point) the more effective delta-V you have - because you spend less time overcoming gravity to achieve orbit.

Conclusions:

1. delta-V is lost from extra rocket motor weight.

2. Thrust gains from more engines is a non-factor.

3. Burn time is a non-factor (except for convenience...)

Recommendations:

- Always reduce useless empty weight. If, due to design, you can't have a single center mounted rocket motor - then only mount two across the center of mass.

- Always run your engines at 100% throttle in space (except for those times when you want precise control). This is the only way to maximize the Oberth effect.

- Use staging to dump off empty fuel tanks along the way. You would be surprised how much delta-V you can pick up by dropping off an empty tank as soon as it's spent.

Good luck!

Edited February 14, 2014 by EtherDragon

[Vanamonde]

If you had 3x symmetry on RCS it could result in some issues. It might get not properly balanced.

It wasn't that. I was steering on pod torque.

[theend3r]

2. Thrust gains from more engines is a non-factor.

False. It all depends on the mass of your ship. Try to do a burn with 1000t and a single LV-N, it'd take you a year. Even in less extreme cases it makes precision impossible and you waste fuel on correction burns.

[wilt57]

False. It all depends on the mass of your ship. Try to do a burn with 1000t and a single LV-N, it'd take you a year. Even in less extreme cases it makes precision impossible and you waste fuel on correction burns.

This is how I view it. I don't want to take the time to plan out ahead of time when to start my PE kicks and at what ejection angle to start so that I'm in the correct position by the time I make my last burn. What a pain in the butt. I'll lose a couple 100 m/s of delta-v to cut my burn times in half, or more. If my delta-v budget is really that close, adding fuel tanks to ships this size is not a very big concern anyway.

[EtherDragon]

This is how I view it. I don't want to take the time to plan out ahead of time when to start my PE kicks and at what ejection angle to start so that I'm in the correct position by the time I make my last burn. What a pain in the butt. I'll lose a couple 100 m/s of delta-v to cut my burn times in half, or more. If my delta-v budget is really that close, adding fuel tanks to ships this size is not a very big concern anyway.

But this is not borne out mathematically. Adding moar powa! is a matter of player convenience in a game so that we don't have to sit there burning 5-minute bursts over 24 orbits.

The question was is there a point of using 4 nukes at 25% power instead of a single one at 100% power - and the answer is no. In fact, two nukes at 100% power will be better, in every way, than four nukes at 25% power. (Lower burn time, better use of Oberth, more available delta-V).

@theend3r - Don't take my point out of context. =p In the OP's situation, there is no benefit - from a total thrust perspective - from using more engines at an equal fraction of power.

[BubbaWilkins]

There are two reasons to use more engines:

1) When total DeltaV is not a factor, but mission duration is.

2) When you need a minimum thrust level to escape a gravity well.

If we had life support concerns to also account for, then #1 becomes a much more prominent issue to deal with.

[frizzank]

MORE BOOSTERS!!!!

This fixes any Dv problem I have....

[Alchemist]

There are two reasons to use more engines:

1) When total DeltaV is not a factor, but mission duration is.

2) When you need a minimum thrust level to escape a gravity well.

3) that's a lander. Nuclear powered lander with detachable external tank is a nice concept for some Delta-v expensive missions.

For interplanetary missions you can do anything (including reducing transfer duration) with any low TWR, but it requires some extensive planning (and patience...). I prefer having TWR at least around 0.05

[Jouni]

For interplanetary missions you can do anything (including reducing transfer duration) with any low TWR, but it requires some extensive planning (and patience...). I prefer having TWR at least around 0.05

I prefer having TWR at least 2 or 3, but sometimes I have to grudgingly accept as low values as 0.2 due to efficiency reasons.

[thunderstar]

already been said, but, i would recomend to not use a lot of nuclear engines, while just having one with that tiny amount of thrust seems like it will take forever, it greatly improves efficiency, so less nukes is better

[MarQan]

Worst case scenario: you can at least remove 2 of the 4 engines, and it should still be balanced. 2 diagonal of course. Maybe add an SAS, if it becomes unstable.

Running 4 on 25% instead of 1 on 100% never worth it, however with REALLY huge ships, more engines on 100% could worth it.

[Dispatcher]

Interesting discussion. As to symmetry, for rockets I also use 3 symmetry. I have had no significant problems with balance and control, and when using LV-Ns the mass savings of 3 vs. 4 engines is very attractive. If balance and control is an issue that I have simply ignored, it may be that in a future update of Unity and/ or KSP such issues will be resolved. In such an event, I will not need to redesign my vehicles.

As to the use of multiple engines on a stage (at all), I think its ultimately a mission and craft design consideration.

[Monkeh]

I think the only thing that hasn't been mentioned here is that TIME ACCELERATION IS POSSIBLE WHILST BURNING YOUR ENGINES. Simply press and hold 'Alt' and then the arrow key and you can go at x4. Reducing that half hour burn down to just seven and a half is most useful.

Radically improved my 'acceptable-efficiency-levels' and enjoyment of the game once I found that little gem out.

Edited February 15, 2014 by Monkeh
seven > 7

[Jouni]

I think the only thing that hasn't been mentioned here is that TIME ACCELERATION IS POSSIBLE WHILST BURNING YOUR ENGINES. Simply press and hold 'Alt' and then the arrow key and you can go at x4. Reducing that half hour burn down to just seven and a half is most useful.

Theoretically, yes. In practice, this:

Physics warp generally works best with ships that don't need it.

[Alchemist]

Theoretically, yes. In practice, this:

http://www.cs.helsinki.fi/u/jltsiren/stuff/physics_warp.jpg

Physics warp generally works best with ships that don't need it.

Rules of heavy interplanetary ship assembly:

1) Launch the propulsion stage in 1 piece (you can launch it empty and then refuel, but don't dock bunch of fuel tanks together)

2) Maximally secure any heavy payload

3) Unless it's very well secured, transport your lander empty - keep its fuel in the propulsion stage

4) it's better to pull than push. And side-mounting anything heavy is a terrible idea.

[Jouni]

Rules of heavy interplanetary ship assembly:

1) Launch the propulsion stage in 1 piece (you can launch it empty and then refuel, but don't dock bunch of fuel tanks together)

2) Maximally secure any heavy payload

3) Unless it's very well secured, transport your lander empty - keep its fuel in the propulsion stage

4) it's better to pull than push. And side-mounting anything heavy is a terrible idea.

Essentially: Restrict what kind of ships you can build (1, 2, 4) and have significantly less delta-v in your ship (3), so that you can save a marginal amount of delta-v by having less engines.

[theend3r]

Essentially: Restrict what kind of ships you can build (1, 2, 4) and have significantly less delta-v in your ship (3), so that you can save a marginal amount of delta-v by having less engines.

Eh... no. He meant to say that the fuel you burn should come firstly from the lander and is refuelled when detaching it and that's entirely true.

Ad 1) and 2) they are both true too, they essentially mean if you can have your ship strutted, do so.

4) Well... I was tempted to try it but it was too cumbersome.

[Jouni]

Eh... no. He meant to say that the fuel you burn should come firstly from the lander and is refuelled when detaching it and that's entirely true.

Ad 1) and 2) they are both true too, they essentially mean if you can have your ship strutted, do so.

4) Well... I was tempted to try it but it was too cumbersome.

The problem was that using physics warp with a fully fueled ship is dangerous, because at that point, landers are also by definition fully fueled. That's also the point you need physics warp the most, because your TWR is at lowest and you are likely to be doing high delta-v burns.

1) and 2) are not that relevant in practice. The thrust from a large number of nuclear engines is still very low, and only a fraction of that force is used to push the rest of the ship. With TWR 0.3, the force is just 10-15% of the forces experienced during launch, so barely any strutting is needed.