LV-N or LV909?

[Invader Jim]

I thought the LV-N (nuclear) was supposed to be highly efficient. I am building a satellite to go to Duna and for the final stage of the rocket I compared DV for these two engines. Engineer says the 909 will yield 4,994 DV in space and the LV-N will yield 2,094. If I swap out the T800 fuel tank for two MK1 liquid fuel tanks that gets me 2,800 DV but that is still vastly inferior to the LV909.

I don't understand when/how to use the LV-N. It seems pointless. Both engines in question have 60 thrust in a vaccuum.

Edited May 14, 2015 by Invader Jim

[KatzOhki]

I don't pretend to understand all this better than anyone else, but see here:

http://wiki.kerbalspaceprogram.com/wiki/Specific_impulse#Comparing_engines

[ExavierMacbeth]

Invader, Check the part stats.

The two Mk1 tanks not only contain less combined fuel than just removing the oxy from the T800 but also weigh 1.8t combined (dry weight) compared to 0.5t of the T800. Add in the weight difference between the engine masses (Nervas are heavy) and you find your discrepancy. Nervas become much more useful when your rocket size hits a point where the extra weight from the Oxidizer (which is 55% of the fuel weight for normal engines) starts costing you more than the difference between tank & engine weights.

Squad sorta overlooked that there wern't any really good LF only tanks for use with the Nerva since all the fuselage parts are balanced for aircraft/jet use. There are a few mods, if your not opposed to them, that add the ability to switch what fuels are available in stock tanks. I would recommend Nertea's Cryo Engines mod as being one of the most straight forward. It applies a patch that lets you switch stock fuel tanks between LFO, LH2/O (Used by Cryo Engines), LF, O, & LH2. So you could have your T800 with a full 800 units of LF

[KarateF22]

LV-909 for small stuff, LV-N for large stuff is pretty much the general consensus. This is because the 909 is light, meaning it wastes significantly less mass than the Nuke providing its ISP which is important in lighter craft where the engine makes up a large proportion of total mass. However, for large objects the Nuke's higher mass is irrelevant when compared to the total payload and its double efficiency matters a lot more.

[Laie]

I am building a satellite to go to Duna

The LV-N is also comparatively heavy; I don't know your design, but it may well be as heavy as your entire satellite.

If you take sufficient weight far enough, the efficiency of the Nerva will pay off. It doesn't even have to be very much weight or huge distances... but a satellite to Duna is a use case where the LV-N cannot provide much of a benefit, if any at all.

[GoSlash27]

Invader Jim,

I ran a comparo between these two a couple weeks back.

Basically, the LV-N has the advantage when you're planning for burns over 1500 m/sec at low acceleration (less than .8G).

For shorter and faster burns, the LV-909 is superior.

In rocket design, efficiency comes down to minimizing mass-that-is-not-fuel. Mass that you don't absolutely need ends up cascading down your stack, creating a bloated expensive monster. Since each stage acts like a payload for the preceding stage, you want to pay attention to how much that stage weighs. This is more important than the raw Isp numbers.

Without getting into the math too much, the LV-N is basically a fuel- efficient boat anchor, while the LV-909 is a less-efficient featherweight. The 2 1/2 extra tonnes of mass that the LV-N represents is 2 1/2 tonnes of fuel and tankage that the LV-909 could burn at it's higher rate.

Which one works out to be a lighter package overall depends on how far you're going and how hard you're pushing to get there.

Even when the LV-N is clearly superior, it may still be worth your time to go with the LV-909 just for the reduced cost and thermal headaches.

HTHs,

-Slashy

[Grumman]

By my calculations, if you have less than 8.8 tons of fuel for the LV-909 or less than 3.8 tons of fuel for the LV-N, the LV-909 is the better option. If you have more than those quantities of fuel, the LV-N is the better option.

In other words, you want to have two FL-T800 tanks or less for the LV-909. And if you have a single Mk 2 fuselage or less for the LV-N, use the LV-909.

[GoSlash27]

By my calculations, if you have less than 8.8 tons of fuel for the LV-909 or less than 3.8 tons of fuel for the LV-N, the LV-909 is the better option. If you have more than those quantities of fuel, the LV-N is the better option.

In other words, you want to have two FL-T800 tanks or less for the LV-909. And if you have a single Mk 2 fuselage or less for the LV-N, use the LV-909.

Grumman,

I'm not disputing your figures or claiming that you're incorrect, but I should point out that this criteria doesn't lend itself well to the thought process that is used when designing stages.

People generally reason it like this:

I've got such-and-such tonnes of payload and I need to get it to my destination, which is a known DV budget. How much fuel is involved hasn't been hashed out just yet. The fuel is a requirement of the mission rather than the other way around.

A more useful way to think about it is to look at the DV budget and acceleration requirement for the stage. That in itself is enough to determine which engine will yield the lightest and/ or cheapest overall stage.

If you're headed to Duna, the LV-N probably isn't worth it. If, OTOH, you're going to Jool then it's probably worth at least considering.

Best,

-Slashy

[RexKramer]

I just ran up against the LV-N/LV909 issue last night, trying to build an ore recovery vehicle for Ike. In this case, the LV909 provided more DV with the same fuel tanks, but the LV-N still ended up being the better choice.

The mission is to start at 200km orbit of Ike, land on Ike, extract 3,000 units of ore, return to a 200km orbit, and dock with an orbiting ore processor. So, I'm bringing to the surface a lander holding two Kerbals, a drill, and two large ore tanks. I wanted plenty of DV to spare, in case I want to make large inclination changes, and because my landings are never anywhere near 100% efficient.

With 2160 LF and 3 LV-Ns (and zero oxidizer), I came up with about 3043 DV. Using the same 3 fuel tanks with 2160 LF (plus oxidizer this time) I came up with about 3192 DV using LV909s. Either configuration will complete the mission, and using 909s actually gives a little more DV even though 909s actually make the initial mass much higher due to the mass of oxidizer.

The LV-Ns in this case are a much better choice however. This is because when I reach the orbiting ore processor I have much less fuel to manufacture to replace the fuel spent during the flight. With the LV-Ns I burned about 2/3 of my LF, which now needs replaced. With the 909s I also burned about the same amount of LF, plus an even higher mass of oxidizer. So using LV-Ns leaves more ore available, because I end up replacing less total fuel.

By the way, using the LFO tanks for LV-Ns isn't really that bad. There is a slight improvement in the fuel to tank mass ratio using dedicated LF only tanks, but it is a very slight improvement. I did calculate DV using LF only tanks, and in that case came up with about 3308 DV, slightly better than using 909s. Unfortunately, the available LF only tanks are very limited, so the ship design gets a little goofy if you go out of your way to select LF only tanks.

Anyway, the end of this story is that while LV-Ns and 909s yielded nearly identical DV values, the LV-Ns actually ended up being a much better choice. Each situation will be different, however. I ran the DV calculations because I was also beginning to wonder if LV-Ns still had any value in the game. They do, but they don't have the same advantages they did in 0.90 and earlier. If you want to put a positive spin on the situation, you could consider that the 909 is now one of the best engines in the game, and is available very early in the tech tree.

- - - Updated - - -

Invader Jim,

I ran a comparo between these two a couple weeks back.

Basically, the LV-N has the advantage when you're planning for burns over 1500 m/sec at low acceleration (less than .8G).

For shorter and faster burns, the LV-909 is superior.

-Slashy

I'd like to point out that the LV-N and LV909 have identical thrust in space. So the engines themselves do not have a difference in burn times. For the same weight, LV-Ns and LV909s are going to have essentially identical burn times.

However, the total mass of many ships is going to be higher using 909s, because the extra mass of oxidizer can easily offset the extra mass of the LV-Ns.

Plus, there is another penalty using LFO engines like the 909- If you are lifting the whole thing to orbit from the surface of Kerbin, you really pay a penalty to lift the extra mass of oxidizer.

Obviously, this is not always the case. For ships which do not require a lot of DV, the weight of oxidizer does not exceed the weight added by switching to LV-Ns. I am personally finding the 909 to be a better choice in many cases in 1.0, but it is so close that there are no simple 'rules of thumb' that always work when comparing them to LV-Ns. In my above example, I found that even for a very short range mission (Ike orbit-Ike surface-back to Ike orbit) the LV-N was the better choice. For a nearly identical mission at Minmus, the 909 was the better choice.

[Warzouz]

Don't forget to remove oxydizer with LV-N

[diomedea]

Please note this question was answered, so better this thread be closed. A newer thread about the same subject was opened here, and the subject is fitting for General KSP Discussion.

I'd invite anybody who still wants to keep the debate ongoing to post there, possibly referencing the posts already opened here. Merging the two threads together isn't fitting, but please PM in case you want assistance to bring something across.