Rethinking the nuke engine, where can we take it from here?

[NathanKell]

Brotoro: the hsps looked close to NTO and AZ50 for me, and the Isps match staged NTO/AZ50 pretty well*, so...

*Not that we have examples, but there's lots of staged NTO/UDMH examples and AZ50 would be just a bit better.

[iVG]

Just use the cryogenic fuel engines pack to swap contents of fuel tanks to LF only.

[Temeter]

Just use the cryogenic fuel engines pack to swap contents of fuel tanks to LF only.

Or the modular fuel tanks mod.^^

[iVG]

Or the modular fuel tanks mod.^^

Yes, anything that includes the interstellar fuel switch. What Squad did was they increased the volume requirements for LV-N to match hydrogen fuel tank volumes, I don't know why people are complaining much. If there was hydrogen in stock you would have needed to use bigger fuel tanks anyway so there's that, otherwise just use the fuel switch.

[Brotoro]

No...liquid hydrogen tank volumes would be MUCH greater than a regular tank filled with only liquid fuel.

But the KSP nuclear engines don't use liquid hydrogen. They use liquid fuel...the same stuff that you can use interchangeably in jet engines and regular LFO engines.

[lazarus1024]

I wouldn't mind seeing some dedicated radiators too. You don't have to use them. Something fairly light, with low thermal mass, but high heat dissipation. Maybe something retractable (I want to say that Nova made a part like that a couple of years back that was a retractable radiator).

So you can store it away in a bay and then open the bay and open some radiators before you light up a NERVA.

[Temeter]

I wouldn't mind seeing some dedicated radiators too. You don't have to use them. Something fairly light, with low thermal mass, but high heat dissipation. Maybe something retractable (I want to say that Nova made a part like that a couple of years back that was a retractable radiator).

So you can store it away in a bay and then open the bay and open some radiators before you light up a NERVA.

You don't really need them, Nukes take a very, very long time until they come close to critical heat.

[Teutooni]

I miss the old interstellar nukes (I believe it was last updated for .25). They were fairly complicated, had cool looking radiators, different nuclear fuels, different propellants etc. Iirc the first fission nukes had worse TWR and Isp than stock, which was and still is ridiculously easy to use. Later models especially large fusion rockets started getting huge Isp, but up to that point they were imo fairly balanced and more fun to use than stock. As to the op question, thats the direction I'd like stock to go.

[RabidMonkey]

I wouldn't mind seeing some dedicated radiators too. You don't have to use them. Something fairly light, with low thermal mass, but high heat dissipation. Maybe something retractable (I want to say that Nova made a part like that a couple of years back that was a retractable radiator).

So you can store it away in a bay and then open the bay and open some radiators before you light up a NERVA.

KSP Interstellar has quite a few dedicated radiator parts.

[cipherpunks]

Wait. What's "hsp" in RESOURCE_DEFINITION again please?

Unrelated edit: if I'm about to add Paraffin, which is 900 kg/m3, should I dance from some stock resource, imagining, say, that Oxidizer corresponds 1:1 to LOx, or... ?

[NathanKell]

hsp is specific heat capacity (in kJ/tonne-K), i.e. take the capacity in J/g-K and multiply by 1000.

Density is...complicated. There's no set standard for the resources ingame. Based on various math, for LF and Ox it looks like one 'volume unit' is around 5 liters, which means that the density is in tonnes/5L, so multiply density (in g/cc, so 0.9 in your example) by 0.005.

[herbal space program]

Even when you exclude the heating issue from the problem, I don't think the nuclear engine is really that useful anymore. Maybe the fact that there aren't any large 2.5m liquid-only tanks is part of the problem because you end up carrying so much dead weight from half-empty tanks from Oxidizer drained compartments.

I just recently built a 45t space plane that (by the numbers anyway) can deliver a Kerbal to the Mun and return to the runwav without refueling, and I honestly don't think that there would have been any way at all for me to do that without the nuke engine: http://forum.kerbalspaceprogram.com/threads/123415-R-A-P-I-E-R-versus-Turboramjet-Surprise-surprise?p=1990869#post1990869. The ions are just too weak to land something so massive and pretty much any other engine does not have a high enough ISP to make it all the way without jettisoning stuff. Please correct me if I am missing something, but I'm pretty sure that's the truth.

[Windspren]

Well, Hydrogen needs a bigger tank so it makes sense to use LFO tanks with the O drained. As for changes for the LV-N, it should not overheat at all as the fuel is cooling it down when run. It should have a spool up time similar to jet engines though.

[cipherpunks]

LV-N, it should not overheat at all as the fuel is cooling it down when run.

Yeah, but KSP isn't about realism. It's more about entertaining the user with [micro]management etc, right? ;-)

If it was, then it'd have [reduced] N-body and procedural almost-everything (except engines)...

[KerikBalm]

Well, Hydrogen needs a bigger tank so it makes sense to use LFO tanks with the O drained. As for changes for the LV-N, it should not overheat at all as the fuel is cooling it down when run. It should have a spool up time similar to jet engines though.

Well, bigger physically... but the mass ratio for LF-O tanks is 4.6:1

That is worse than real life hydrogen tanks... by a long shot (real life is closer to 8:1 if I heard correct numbers... I know the real LH2+O2 External tanks for the space shuttle would have a mass ratio of 5 or better if it was loaded with LH2 but no Oxidizer)

And mass ratio is what really matters, but it doesn't matter all that much for the LV-N when its used as a tug.

http://forum.kerbalspaceprogram.com/threads/121064-Are-LV-N-s-worth-the-trouble?p=1938330&viewfull=1#post1938330

when your payload fraction gets really high (as it is with SSTOs and nuke tugs), the tank mass ratio doesn't have much of an effect

ie, with an SSTO, the payload may be 37% of starting mass, if your tanks comprise 10% of the remaining 63% of the mass, or 12% of the remaining 63% of the mass, it doesn't affect your dV all that much.

So... in a way... the 4.6:1 mass ratio for just using emptied LFO tanks could still be balanced.

Imagine a nuclear tug pushing a payload to escape velocity from LKO.

~1000 m/s is needed. 9.81*800* ln(X) = 1000

ln(X) = 0.12742

X = 1.1359

Only 13.6% of your mass needs to be fuel

For a 100 ton payload, at a 9:1 tankage ratio

(100+9x)/(100+x) = 1.136

100+ 9x = 113.6 + 1.136x

7.864x= 13.6

x = 1.73 tons of tanks.... 15.6 tons of tanks+ fuel

For an 8:1 ratio, its simple 6.864x = 13.6

x= 1.98 tons of tanks ...... 15.85 tons of tanks+ fuel

The mass of your tug that imparts a dV of 1,000 m/s to a 100 ton payload goes from 15.6 to 15.85... or rather, total craft mass foes from 115.6 to 115.85...

Not a huge difference (and this is ignoring engine mass and such, which has pretty much the same effect as increasing payload mass, making the comparison even closer).

If we instead look at a required dV of 2,000 m/s, and a 9:1 vs 4.6:1 mass ratio:

2000/9.81/800 = 0.2548 = ln (X); X = 1.29

(100+9x)/(100+x) = 1.29

100+ 9x = 129 + 1.29x

7.71x = 29

x= 3.76

3.76 tons of tanks... 33.84 tons of tanks+ fuel (less than an orange tank)

vs a 4.6 tankage ratio:

100+ 4.6x = 129 + 1.29x

3.31x = 29

x = 8.76 tons of tanks... 40.3 tons of tanks+ fuel

So for a craft that needs a dV of 2,000 m/s, and has a mass of 100 tons excluding fuel tanks, the total mass at a 9:1 ratio is 133.8 tons, the total mass at a 4.6:1 ratio is 140.3

140.3/133.8 = 1.0486....

Less than a 5% change in craft mass, when the fuel tank mass ratio is decreased by 48.9%

Now... this will get worse and worse as you increase the required dV... because of the non-linear nature of the rocket equation, but with the LV-N's Isp, and the dVs required in the kerbin system - the tank mass ratio can be pretty bad without affecting things much.

Its also why ion drives work well in game, even though their tankage ratio is horrible too.

The ability to use spaceplane parts at an 8:1 mass ratio, means LV-Ns will remain top-dogs, despite this fuel change.

[NathanKell]

The mass ratio of KSP LFO tanks is 9:1. 1 ton of dry mass for every 8 tons of propellant (9 tons gross).

[NecroBones]

It looks to me like the 4.6:1 ratio is from taking an LFO tank and removing the oxidizer.

[Windspren]

We really need radiators

[DBowman]

Well, bigger physically... but the mass ratio for LF-O tanks is 4.6:1

That is worse than real life hydrogen tanks... by a long shot (real life is closer to 8:1 if I heard correct numbers...

I think NASA DRA5 Mars mission LH2 tanks are 30% 'structure, insulation, and chillers' by mass, I believe they need more insulation and the chillers because they have to store the LH2 for years. 'To orbit launchers' (Saturn, Shuttle, etc) can have lighter tanks because the LH2 will be used up within a 'few' minutes.

[OgreMagi]

I have to agree the change in way nukes use fuel was not properly thought out. If they had added FUEL ONLY versions of some of the tanks they that would have helped.

If they had not changed the atmosphere rules then I think there would be less complaints on the ssto side of the argument. If they had changed the nuke in the .9 version it would have been a very good match for a ssto as it could be a FUEL ONLY. a turbo jet and a Nuke ssto would have been a lot fo fun, easier to build and play.

But the current drag/atmosphere makes the turbo jet useless at it loses too much thrust to orbit.

fixes???

1. change the drag atmosphere back, easily done but that wouldn't help the game

2. change the nuke back , again easily doable but it wouldn't forward the game.

3. add a RAM JET ENGINE. that's what we had with the old drag/atmosphere rules when we used the turbojet/ramintakes to air hogg into orbit. A FUEL ONLY Ram Jet or A pulse jet something that would work higher in the atmosphere and fair enough at sea level .

just my thoughts.

[General Rarity]

We really need radiators

We do have radiators. They're been in the game since... well forever.

To quote regex:

... Every part is a radiator, some better than others, and every part is a heat sink, for better or worse.

[KerikBalm]

It looks to me like the 4.6:1 ratio is from taking an LFO tank and removing the oxidizer.

Correct, an orange tank weighs 4 tons.

It can hold 14.4 tons of LF and 17.6 tons of O2

(14.4+17.6+4)/4 = 36/4=9

But remove the O ( -O when I wrote LF-O, the - is not a dash but a minus sign)

you get (14.4+4)/4= 4.6

I think NASA DRA5 Mars mission LH2 tanks are 30% 'structure, insulation, and chillers' by mass, I believe they need more insulation and the chillers because they have to store the LH2 for years. 'To orbit launchers' (Saturn, Shuttle, etc) can have lighter tanks because the LH2 will be used up within a 'few' minutes.

Indeed, mass ratio will change depending on your duration needs.

For getting to orbit and ejecting from Kerbin/Earth, you don't need to keep the LH2 very long.

Your return stage may use something easier to store... NH3 or CH3 for instance...

You can get to Mars/Duna without holding the LH2 for very long, and thus your mass ratio for the ejection burn should be over 8:1

Modelling this in KSP... well, that gets complicated.

FWIW, a 30% structure weight still gives a mass ratio of 1/.3 = 3.3333 which is still not too bad if your Isp is pretty high.

I seem to recal the Xenon tank mass ratios were about that (haven't checked if they still are)

[Laie]

We do have radiators. They're been in the game since... well forever.

I'd still prefer if we wouldn't need them. Nukes effectively come at two parts apiece for structural reasons alone; I see no gain in inflating part count further. I have on several occasions proposed to make them both stronger and heavier, like 20t/400kN, so that a nuclear tug can look like this:

(see this thread for more details and examples)

This would, of course, limit them to heavy payloads. Three-man missions to Duna and Mk2 spaceplanes would be out.

[Geschosskopf]

Are nuke engines now more practical on spaceplanes? it seems that without the need for onboard oxidizer, this would be the natural choice for planning SSTO missions with limited wet weight. Has anyone yet built a good spaceplane design with 2 turbojets and a nuke? how about 2 nukes and a turbojet?

It IS possible to make an LF-only spaceplane, but I don't see it as worth the trouble. Nukes have pathetic atmospheric Isp, very little thrust at any time, and weigh a lot. This puts a rather low upper limit on the size of the spaceplane. I find it more practical to make a conventional LF/LFO spaceplane and dock it to a transfer tug if I want to send it somewhere besides LKO.,

Is there a good method for bleeding off the extra heat they generate? It seems like ship designs centered around nuke engines could now benefit from dedicated heat radiators, to keep the engines and fuel tanks cool during long burns. Thermal management is now a practical part of logitstical planning, and I consider this a win, but only if there's an actual way to manage it.

Yes, this is a serious and totally unrealistic problem. The overheating of LV-Ns (or even the stuff they're attached to) is just ridiculous and in no way resembles how the things worked in real life. A lot of folks are demanding radiators but this is missing the point that the heat shouldn't be there to begin with, no more than with other engines. Fortunately, there's ModuleManager. This allows you to tweak the heat-related values of the LV-N to bring it back into loose formation with reality.

But you didn't mention the biggest change to LV-Ns, which is that they're now only really useful anymore on rather large ships. Scott Manly did a video on this.. Below a certain size, it's better to use conventional rockets. For example, I recently sent an 8-ship flotilla to Jool and, when I was designing it, I found that nukes were only beneficial on 2 of them (a big station module and a carrier for 2 airplanes). All the rest used conventional rrockets.

[Laie]

A lot of folks are demanding radiators but this is missing the point that the heat shouldn't be there to begin with

Word.

But you didn't mention the biggest change to LV-Ns, which is that they're now only really useful anymore on rather large ships.

Define "large". I'm currently under the impression that the point of breakeven comes at something like 5t of payload to Duna, or 3t to Jool. That is, pretty much anything larger than a one-person lander will benefit from nukes.