Are Nuclear Rockets "Cheating"?

[Renegrade]

Also, Renegade: do *not* look into the abyss of unreality that is ion-powered crewed landers. It hurts too much.

Haha, I actually, literally (and using that word properly here), laughed out loud at that, thanks man

But don't worry, I'm fully aware of how mad it is.. the Dawn spacecraft's ion drives are only 3100 isp and deliver a whopping 90 millinewtons each -- about enough force to hold up two sheets of A4 paper against earth's gravity.

Not exactly a drop-in for the F-1 eh?

To confess a sin though, I did make a Minmus-ion lander after they were buffed though for fun. Silly times were to be had. I can understand why Squad did it though - since we lack an 'ion mode warp', they really were quite horrible in actual gameplay. Heck, they still are horrible - I usually just use ants and 48-7Ses for probes anyhow.

[r_rolo1]

What? The LV-N is almost two times heavier than my typical munar lander's payload (ie after taking all engines and fuel tanks off - it's about 1.41t without LS).

In my stock career 0.24 I actually made 2 munar landers, one with a LV-N and one with 2 48-7S ( I had issues with the LV-N lander design ( the height of the engine makes it hard to make a 1 engine lander if you want to keep a 1 kerbal pod only, added to all the issues you get for a high gravity center lander with a heavy engine ), so I decided to replace it with a dual 48-7S solution ) and while the LV-N solution spent fuel by a factor of 25%, that was overlapped by the fact that I spent far more fuel putting it in LKO than with the 48-7S ( note that the landers were basically equal in terms of parts barring the engines ). In fact a lot of the comparisons made around the LV-N tend to forget that you have to put the engine in orbit and that alone might ofset the gains made later ...

(Also since I think stock progression is hideously broken, I'd like to point out that 40% of the science in the game is on the Mun, and 60% is on Minmus, and +inf% is in the contracts...)

More than broken ... I actually made the career that I talked about with 90% being science from contracts ( the rest was the result of one Minmus landing, a munar flyby, typical kerbin science ( not even complete, since i didn't cared for a lot of exotic science you can get in kerbin, like splashed on grassland or landed in Water ... ) and the equivalent of a Munar landing from the Mun ) ... fulfilled just 225 in 40 game days. The devs should really give a look in the way contracts are generated, since when you push the system like i did ( that is, fulfilling contracts as fast as possible, including in between flights to fulfill other contracts ) , a lot of contracts tend to repeat with the exact same science reward. I lost count to the times I tested the TT-70 Radial Decoupler or the LFB KR-1x2 landed on Kerbin ... not mentioning the almost completely deterministic sequence you get on non-parts contracts ( get science from space around ( circles betwen the bodies that you already visited its SoI ) ->plant flag in ( same thing as before ) -> Save X kerman while in orbit . Rinse and repeat )...

Edited August 27, 2014 by r_rolo1

[GreeningGalaxy]

Nerva is cooled by the LH2 before you send it into the reactor, it might be an issue if you run it on methane who is not cryogenic but would probably just require that you use more of the reaction mass for cooling before sending it into the core.

Using the reaction mass for cooling should work for an 800 m/s ISP engine but not something far more efficient as you would use to little reaction mass.

I also think the nuclear reactor would last the mission, you don't have to swap fuel rods often and the reactor would just be active while burning.

Well yes, more or less the standard is that the reaction mass of an NTR is used as coolant prior to being forcefully expelled. That's essentially how the engine works, actually; the heat removed by the fuel/coolant is what drives it to such high speeds. It's not so much cooling during the burns that's the issue, it's when the reactor is idling without expelling coolant that's a problem. In KSPI, doing a cold reactor shutdown is something of an ordeal and requires a kerbal on EVA to turn it off and to restart it again. Since the LV-N doesn't appear to glow when not running, I guess it's doing some kind of automatic shutdown/restart?

Anyway, I'd agree that for the sake of stock KSP, all the little details being swept under the rug is just fine- I mean, the LV-N is far from the only example of that. I do think it's interesting, though, how making you have to deal with all the details makes potentially-OP parts (like KSPI's 1100kN 15500s D-T Vista engine) suddenly seem balanced. If you had to carry uranium with you and use a kerbal on EVA to start and stop the engine's reactor in order to use the LV-N, would we be having this discussion?

[Jouni]

What? The LV-N is almost two times heavier than my typical munar lander's payload (ie after taking all engines and fuel tanks off - it's about 1.41t without LS). Can an ion-based Mun lander even work? I know it has about 5-ish local TWR, but that's just for lifting the drive itself. To get a 2.0 local TWR, you'd need 14 drives on a 1.41t lander.. and they'd have to be mass zero in that case, and they'd have no xenon or solar power (unless you abused the massless oxstat nonsense #lolsolar).

One of the first things I did in 0.23.5 was a Mun lander:

7 ion engines, 2.71 tonnes of payload, and no abuse of massless parts. The lander weights 6.475 tonnes, carries enough fuel for two landings, and has TWR ranging from 1.33 to 1.44. Every landing is like a Tylo landing, but we're talking about efficiency here, not usability.

[juanml82]

One of the first things I did in 0.23.5 was a Mun lander:

http://jltsiren.kapsi.fi/ksp/0.23.5/ion_mun_4.jpg

7 ion engines, 2.71 tonnes of payload, and no abuse of massless parts. The lander weights 6.475 tonnes, carries enough fuel for two landings, and has TWR ranging from 1.33 to 1.44. Every landing is like a Tylo landing, but we're talking about efficiency here, not usability.

Ahhh... but it's not cost effective compared to some chemical rockets.

[GreeningGalaxy]

I'm not entirely sure why they buffed the ion engine back in... what was that, .22? It was already OP when it put out half a kilonewton, and now it puts out two and takes half the power. I guess since real-world ion engines have to burn for months or years to get anywhere, they gave us some slack for the time being, hopefully while they work on letting ion engines work during time warp.

[Landge]

Unless Squad implements radiation poisoning and damage over time I will use LV-Ns even on Kerbin. With the exception of my SSTO, I don't use them until I've achieved orbit.

@r_rolo1:

Here's how I got around the LV-Ns being too long for landing craft.

This is my "Spirit of Kerbin" lander and personnel transport. Weighing in at almost 51t it comes outfitted with 4 LV-Ns, 2 per side.

Shown here I'm using KAS and IR parts to add a reusable 4 seat rover.

I went with 4 LV-Ns because just using 2 I could barely lift off of Duna. The downside is that the engines block the view from the side portholes of the Habitat Module. Small Hardpoint stand off struts are mounted on the ASAS module located above the Habitat module keeping the engines clear of the ground. It can carry up to 7 Kerbals and can land and take off from almost every planetary body except Eve, Kerbin, Laythe, and Tylo. For Laythe I could probably substitute Aerospikes or Rapiers. I may have to experiment.

9 parachutes composed of 2 drogues, 3 radial, and 4 main chutes allow for a 9~10 m/s landing on Kerbin without having to resort to firing the engines. That is if the tanks are nearly empty.

The tricky part was placement of the fuel lines. KSP doesn't like dual flow from the center out to the 4 adjacent tanks then to the engines. It drained the fuel unevenly caused the craft to become dangerously unbalanced. This was solved by drawing fuel from 2 of the outer tanks to the center tank located under the habitat module. From the center tank the fuel is directed into the other 2 outer tanks then into the 2 small tanks located above the engines.

[r_rolo1]

That is a nice ship , but it is bigger than what I wanted/needed for my munar missions ( and remember than this was in career mode in stock, so there was some pressure to go minimal ). I was explicitely speaking of the issues of making a lander with a nuke and a mk1 can in my post: besides the height issue, that creates problems in ladder placement ( the engine is almost perfectly the size of the fully deployed extend stairs ...) and in terms of tipover ( that you can go around with some creative engineering and liberal use of construction parts ), the engine weight is a issue for the inbound torque of the can. As I'm only recovering my pilot skills in KSP after a long break ( i barely played between 0.17 and 0.235 ), I do prefer something more responsive and easier to pilot ...

[John FX]

I've been avoiding the nuclear rocket because it seems like "cheating". A magic rocket that just solves so many of your problems? Is there any real rocket (or a "really proposed" rocket that works like it does)?

Is it cheating?

Short answer no.

Long answer Noooooooooooooo.

[Laie]

That is a nice ship , but it is bigger than what I wanted/needed for my munar missions ( and remember than this was in career mode in stock, so there was some pressure to go minimal ). I was explicitely speaking of the issues of making a lander with a nuke and a mk1 can in my post:

maybe something like this?

4500m/s at a Munar TWR of 2. Should suffice to visit at least two and possibly up to four biomes before it needs to dock for fuel. It really comes into it's own on Minmus, where it can harvest science from all biomes without refueling.

EDIT: with the SAS workover in 0.24, the reaction wheels now have a different weight and the craft probably isn't quite balanced anymore. Shouldn't make the craft useless, but I suggest you replace the LV-N with a LVT-30 and take off from Kerbin, with SAS enabled. If it can fly straight, it's all good. (If the lift/yaw indicators are constatly off by more than one tick or two, you *may* still consider to add/remove some ballast here or there).

Edited August 27, 2014 by Laie

[Renegrade]

In fact a lot of the comparisons made around the LV-N tend to forget that you have to put the engine in orbit and that alone might ofset the gains made later ...

Yeah, that's why I tend to focus on the total tonnage of a stage.. you gotta haul all that into orbit, at like seven to ten times the cost of the orbital component itself..

More than broken ... I actually made the career that I talked about with 90% being science from contracts ( the rest was the result of one Minmus landing, a munar flyby, typical kerbin science ( not even complete, since i didn't cared for a lot of exotic science you can get in kerbin, like splashed on grassland or landed in Water ... ) and the equivalent of a Munar landing from the Mun ) ... fulfilled just 225 in 40 game days.

Yeah, it's completely nuts. I'm currently testing with a pure stock playthrough, it has half of the tier 5 technologies (the tier with 'Heavy Rocketry' at the top, sometimes called tier 4). Current funds stands at 466k.

I still haven't completed the "attain orbit" thing, and I'm 23 minutes in, just by only taking 'landed at Kerbin' contracts. I did two flights, both sub-orbital, that brought back two crew reports and nothing else.

Craziness I tell you!

http://jltsiren.kapsi.fi/ksp/0.23.5/ion_mun_4.jpg

7 ion engines, 2.71 tonnes of payload, and no abuse of massless parts. The lander weights 6.475 tonnes, carries enough fuel for two landings, and has TWR ranging from 1.33 to 1.44. Every landing is like a Tylo landing, but we're talking about efficiency here, not usability.

Crap, I calculated TWR for Kerbin gravity for some reason. D'oh! Using ACTUAL local TWR, I guess you don't need 14 drives.

That thing isn't all that efficient though, as r_rolo1 points out, it's quite heavy and can't exactly insert it into orbit from KSC launchpad (and I imagine it's Funds cost is .. high).

My mun landers are usually dual or quad 48-7S designs, and they mass between 2.5t and 3.8t (the heavy varieties can actually perform two landings, although they only have the science equipment for one-and-a-half). They generally cost around and carry 2-4 tons of fuel. Using stock air, it would take at most 25t on the pad to put 'em in orbit assuming a terrible, 10% payload efficiency. From there they can rendezvous with a light Mun station and carry out a full biome sweep of the Mun etc.

As a quick comparison the Minmus version is under 2t.

Also re: efficiency, the 2-man 2-can is terrible. You could stack two 1-man cans on top of each other and still have lower dry weight.. I should fix that in my Horrible Nerf.

U

This is my "Spirit of Kerbin" lander and personnel transport. Weighing in at almost 51t it comes outfitted with 4 LV-Ns, 2 per side.

Geez, that's bigger than my MUN STATION. :S

Weighing in at only 44 tons, it can support even a 2tx4e lander in a full biome sweep of the Mun.

(the extra blob on the end is a Snacks LS resupply drone, since the station is manned but currently mothballed. I'm not going to bother with teh Mun at all in this playthrough)

It wastefully has the fancy 'necking down' design from my BTSM Explorer-class vessels (in fact it pretty much IS an Explorer-class vessel), carries unnecessary extra reaction wheels, the heavier, protected solar panels, enough monopropellant for six hundred docking operations, and a high TWR engine for juicyfast TMI injections. And is still lighter >.<

(the side-mounted FL-T800s are just to give the panels better clearance over ship bits, there was nothing stopping me from using an X16 tank instead inline)

[moogoob]

Only in the quest for ultra-efficiency is the poodle a "High TWR engine".

[malkuth]

Since I always have used either an economy mod, or now actual ingame economy always found that to get good mission payouts always better to use regular engines on smaller type vessels then the nerva.

Bigger vessels might consider using it, but it still adds a huge cost to the mission.

[Sky_walker]

Cost isn't a problem. Economy is purely ascetic matter - it doesn't limit you in any way unless you intentionally impose limits to yourself. There is no game-over due to bankruptcy in KSP (going close to 0 money is pretty much like opening infinite source of cash) and you get flooded with money pretty much all the time.

[Renegrade]

Cost isn't a problem. Economy is purely ascetic matter - it doesn't limit you in any way unless you intentionally impose limits to yourself. There is no game-over due to bankruptcy in KSP (going close to 0 money is pretty much like opening infinite source of cash) and you get flooded with money pretty much all the time.

Well, cost can be calculated two ways: mass, and funds. Mass matters, as you're going to be climbing up in part count, which imposes an upper limit on design and frame rate.

However, I agree with your assessment of Funds. I played at 10x cost for a while, and was easily able to keep the line moving up even without abusing 'science from X orbit' or 'place a flag' contracts.

What's this infinite-cash-near-zero business though?

[Red Iron Crown]

What's this infinite-cash-near-zero business though?

Get near zero funds. Accept all available contracts. Build a ship that uses all your money, launch it but leave it on the pad. Cancel all contracts, no financial penalty because you have no funds. Recover ship and get all money back. Repeat as necessary.

May affect your reputation negatively.

[Laie]

What's this infinite-cash-near-zero business though?

Problematic, to say the least.

Remember the days when you were still struggling to get a rocket into orbit? How many failed attempts it took to land on the Mun? An economy that works for that level of experience is pretty much meaningless for an experienced player, even if you play ironman/no revert. According to the last devonotes, future versions of KSP will have difficulty settings; that's probably the only way to tackle this issue.

[r_rolo1]

Cost isn't a problem. Economy is purely ascetic matter - it doesn't limit you in any way unless you intentionally impose limits to yourself. There is no game-over due to bankruptcy in KSP (going close to 0 money is pretty much like opening infinite source of cash) and you get flooded with money pretty much all the time.

True enough ... the devs weren't kidding when they said that in Kerbal economics you couldn't get broke, period :/ And that is bad IMHO, since there is no incentive to make better design ships besides what you call "ascetic" feelings ( I call it RL pragmatism: never make anything more complicated than it needs to be OFC that does not need to apply in KSP, though ... ).

[malkuth]

Get near zero funds. Accept all available contracts. Build a ship that uses all your money, launch it but leave it on the pad. Cancel all contracts, no financial penalty because you have no funds. Recover ship and get all money back. Repeat as necessary.

May affect your reputation negatively.

Hmmmm, guess I should fix that problem then.

[Renegrade]

Get near zero funds. Accept all available contracts. Build a ship that uses all your money, launch it but leave it on the pad. Cancel all contracts, no financial penalty because you have no funds. Recover ship and get all money back. Repeat as necessary.

Oh, funds don't go below zero, huh? :/

Remember the days when you were still struggling to get a rocket into orbit? How many failed attempts it took to land on the Mun? An economy that works for that level of experience is pretty much meaningless for an experienced player, even if you play ironman/no revert. According to the last devonotes, future versions of KSP will have difficulty settings; that's probably the only way to tackle this issue.

Difficulty setting will be most welcome (although I'd prefer it scale payouts rather than cost, so that rockets can still be compared without having to say '12k in hard funds' or '2k in easy funds'), much like the new options in the debug menu...

That being said, I don't recall all that much difficulty once I understood the Souposphereâ„¢ and the annoying non-recursive/single-link tree data structure.. but then again, I did orbital stuff back in Elite II: Final Frontier so it wasn't completely alien.

[Raccoongamer]

No its not cheating unless you use the extra arm you grew from radiation exposure to build it

[Jouni]

That thing isn't all that efficient though, as r_rolo1 points out, it's quite heavy and can't exactly insert it into orbit from KSC launchpad (and I imagine it's Funds cost is .. high).

In this branch of the discussion, efficiency was defined in the traditional way: the lowest total mass for the given payload in a ship that can complete the mission, with no concerns for costs or part count. Payload is also assumed to be a part of mission – it's a different mission with a different payload.

On the other hand, the design is inefficient. It carries 0.975 tonnes of batteries and solar panels, which could be replaced by massless parts. Landing struts could also be removed, saving another 0.2 tonnes. There is also way too much fuel: the Apollo-style lander could actually start the mission from LKO and return there afterwards. Instead of 7 engines and 7 xenon tanks, an efficient lander would only need 5 engines and 3 tanks. With a mass of 4.32 tonnes (out of which 2.71 tonnes is payload), the lander would have around 2050 m/s of delta-v and a slightly higher TWR than the original design. As a comparison, the same payload with 300 units of fuel and oxidizer and a single 48-7S engine would weight around 4.5 tonnes and have almost 1400 m/s of delta-v.

My overall point was that if you're concerned with this kind of efficiency, the 48-7S engine is no longer that relevant. You rarely need higher TWR than the nuclear engine provides, and if you do, you can usually get it from turbojets. Some Tylo, Duna, and Moho landers may be better with chemical rockets, and Eve is a separate case, but for most things, the most efficient choice is a mix of nukes, ions, and turbojets.

[Renegrade]

On the other hand, the design is inefficient. It carries 0.975 tonnes of batteries and solar panels, which could be replaced by massless parts. Landing struts could also be removed, saving another 0.2 tonnes. There is also way too much fuel: the Apollo-style lander could actually start the mission from LKO and return there afterwards. Instead of 7 engines and 7 xenon tanks, an efficient lander would only need 5 engines and 3 tanks. With a mass of 4.32 tonnes (out of which 2.71 tonnes is payload), the lander would have around 2050 m/s of delta-v and a slightly higher TWR than the original design. As a comparison, the same payload with 300 units of fuel and oxidizer and a single 48-7S engine would weight around 4.5 tonnes and have almost 1400 m/s of delta-v.

The mun landers I use are less than four tons, and also 2500 dv, all without exploiting massless parts. I even made an experimental 2tx1e design on the 48-7S, giving a 4.81 Mun TWR and 2536 dv, all for 3830 kg, significantly less than than 4.32 or 4.5t. The comments on funds was just an aside -- the masses here are lower with the 48-7S designs.

[Jouni]

The mun landers I use are less than four tons, and also 2500 dv, all without exploiting massless parts. I even made an experimental 2tx1e design on the 48-7S, giving a 4.81 Mun TWR and 2536 dv, all for 3830 kg, significantly less than than 4.32 or 4.5t. The comments on funds was just an aside -- the masses here are lower with the 48-7S designs.

Your Mun landers also have less payload than 2.71 tonnes, so they're not comparable. Efficiency is about completing the given mission as efficiently as possible, instead of completing a different mission more efficiently.

That particular lander was originally intended to demonstrate that the new ion engines were suitable for real missions, instead of just ultra-light stuff. You can obviously land on the Mun with one ion engine and a command chair, or two engines and a Mk1 lander can. The twin ion engine lander should weight something like 1.6-1.7 tonnes without abusing massless parts. With abuse, we could get it down to 1.22 tonnes.

[-Velocity-]

The nuclear thermal rockets (NTR) featured in the game are probably too "weak". IRL, we can probably expect Isps of like 1200 s or even considerably more, not just 800 s. 800 s was the Isp of the first ever nuclear thermal rocket. There have been many studied designs that would reach Isps MUCH higher than our first generation of NTR (which were in fact built and tested, but just never flew). So the NTR in KSP is actually probably a little weak compared to what we would make in real life if NASA ever got more than just the last dregs of the federal budget.

What is actually sorta cheating (in terms of realism) now is the ion engines, which have ridiculously over-powered thrust to weight ratios compared to their real life counterparts. I'm OK for making ion engines more powerful so that maneuvers don't take so long, but when you make them so ridiculously overpowered that they can be used as lander engines, as they can now, you've got a serious issue (at least, if realistic gameplay is your goal). Not to mention, ion engines work in atmospheres, which is completely unrealistic. Squad needs to make us some electric propellers so that we have some form of realistic electric aircraft propulsion, and then make ion engines only work in a vacuum... and lower their thrust to pre-0.23 levels...