Jump to content

Are LV-N's worth the trouble?


Recommended Posts

I've been putting together a couple of craft using LV-Ns, a probe and a manned Duna mission, and I've noticed something troubling. LV-Ns now seem to be not really worth the hassle of managing their heat. Let me start with the probe.

nUeiBp2.png

The LV-N gives me 1752 m/s of dv. However, if I use a LV-45, I get 1687 m/s. That's only 65 m/s difference, roughly 4%

FSd7SIx.png

So maybe LV-N's work better for larger vessels. Let's look at my Duna mission.

iOhT8vK.png

The orange tank has been drained of oxidizer. The LV-N stage will give me 3300 m/s of dv. However, if I swap over to LV-30's and refill the oxidizer, I get 2905 m/s and a starting TWR of .54. With the 10k dv of the entire vessel, the 395 m/s difference is again a roughly 4% gain with the LV-N's.

Once I throw in the hassle of heat management, I'm asking myself, "Is the 4% gain I get with LV-N's worth the longer burn times and the heat management issues?" My provisional answer is, "No."

Are LV-N's worth the trouble? I'd like to see what the community has to say. I used to use them for everything heading interplanetary, but now I'm wondering if they're even worth unlocking. :(

_______________________________________________________________________________________________________________________________

OK, NathanKell has a good point. I need to be comparing equal fuel masses. Here is the apple-to-apple comparison. First, the craft using LV-30's and an orange tank:

xZxpF4t.png

Vessel mass of 82.8 tons, dv of the relevant stage is 2905 m/s. Swapping out the 36 ton orange tank for a Mk. 3 liquid fuel fuselage plus 8 of the 1.25 liquid fuel tanks, and swapping the engines over to LV-N's gives me this:

s29ZmV3.png

(After posting I realized I needed fuel lines feeding the main tank)

Vessel mass is 84.6 tons, dv is now 4934, That's 2029 m/s difference, almost a 70% gain in dv compared to the LV-30's. That's more like it, right?

So, are LV-N's worth the hassle of heat management? The 70% increase in dv when doing an actual performance comparison makes me change my provisional "No" to an emphatic "Heck, yes!" :D

Thanks to everyone for their feedback!

Edited by ArmchairGravy
Revised Test
Link to post
Share on other sites

They are well worth it on larger ships if you use the pure LF airplane tanks. The LF/O rocket tanks with the oxidizer drained have too much dead weight for too little fuel to make the LV-N much use, as you discovered.

Link to post
Share on other sites
They are well worth it on larger ships if you use the pure LF airplane tanks. The LF/O rocket tanks with the oxidizer drained have too much dead weight for too little fuel to make the LV-N much use, as you discovered.

That will make designs....interesting.

Link to post
Share on other sites

By definition, if you have the same mass ratio, the LV-N will always outperform, 2.3x the delta V of the Poodle.

The problem is, as usual with these threads, you have a tiny tank and you compare an engine that masses 1.25 tons with 4 tons of fuel, to an engine that masses 3 tons and has 2 tons of fuel. Of course that comparison is going to be whack.

Even-slightly-fair comparison: make sure you have more tonnes of propellant than tonnes of engine, and make sure you have the same tonnes of propellant in both cases (yes, that means using airplane tanks and checking mass, not just draining oxidizer out of the same tank and getting less than half the propellant mass).

Link to post
Share on other sites
They are well worth it on larger ships if you use the pure LF airplane tanks. The LF/O rocket tanks with the oxidizer drained have too much dead weight for too little fuel to make the LV-N much use, as you discovered.
That will make designs....interesting.

Soup does have a point though, I made a fairly decent looking IP-sci-orbiter, 33 tonnes wet and 70 parts with over 6k dV. She may not be a powerhouse, but she's got the fuel economy of a Prius. She might not be the prettiest [was designing so I could see part of each fuel tank for refueling purposes], but I imagine I could smooth out the rough edges and get her lookin' mighty sleek.

16962384273_0b65f2810e_b.jpg

Link to post
Share on other sites

There are several fuel-switcher mods out there now that will allow you to utilize the entire tank capacity for LF only when using nukes. I love stock-like play but the tank selection for nukes is not very good on it's own so I broke down and added a fuelswitcher.

Heat management is a must, you cannot skimp there, but you know that already as well :)

Link to post
Share on other sites

For larger ships, use a central mk3 tank and radially attach mk1 fuel tanks, then the lv-n go beneath the mk1 fuel tanks. There are radiator part mods, but with such a setup, I don't think a Duna mission would require bothering too much with the heat. The central mk3 tank will soak up a lot of heat and that will give you several minutes of continuous burn.

Link to post
Share on other sites
By definition, if you have the same mass ratio, the LV-N will always outperform, 2.3x the delta V of the Poodle.

The problem is, as usual with these threads, you have a tiny tank and you compare an engine that masses 1.25 tons with 4 tons of fuel, to an engine that masses 3 tons and has 2 tons of fuel. Of course that comparison is going to be whack.

Even-slightly-fair comparison: make sure you have more tonnes of propellant than tonnes of engine, and make sure you have the same tonnes of propellant in both cases (yes, that means using airplane tanks and checking mass, not just draining oxidizer out of the same tank and getting less than half the propellant mass).

Ah, you are right. I need to compare with equal fuel masses. Let me do that and I'll edit the OP.

Link to post
Share on other sites

Yah, I have run into this as well. For smaller craft, the 909/spark is almost always a better (or at least equivalent but much less expensive) option than the LV-N. It's actually a headache to use the LV-Ns now for me, what with all the overheating and the fact that I usually get worse dV with it than smaller, less expensive, easier to build with engines anyway. When using Mk2/Mk3 parts the LV-N becomes somewhat better, but I've had overheating problems when trying. IMO there's almost no reason to use it anymore, which saddens me. It was one of my favorite engines.

Link to post
Share on other sites

Both tests are bad. The first one simply indicates that 3-ton engine on 8-ton craft might not be a good idea. In the second test you have more than 2x less propellant for NERVAs than for Reliant, because oxidizer is also a part of the propellant. And you still got better dV. That's rather impressive, I should say. Have you compared masses of the craft with NERVAs and no oxy vs craft with oxy and Reliants? That matters, you know?

Link to post
Share on other sites

You're loading 2.222x as much propellent into the LV-T45 rocket, and then comparing it to the LV-N

Either use LF only tanks, or use more tanks.

Best case scenario for a standard "chemical" rocket: 9:1 mass ratio, 350 Isp.

Max dV for a single stage is thus 9.81*350* ln(9) = 7,544 m/s

Using multiple tanks drained of oxidizer reduces your best possible mass ratio to something like 4.6

Max dV for a single LV-N stage with emptied oxidizer is thus 9.81*800*ln (4.6) = 11,976 m/s

It gets even better if you use LF only tanks - its a shame they are all spaceplane parts though.

You're just using them wrong: of course a heavy engine + low fuel mass is going to perform poorly. Increase your fuel mass.

A FL-T800 tank trained of oxidizer only carrier 1.8 tons of LF... and you're shoving a 3 ton engine on it.... doesn't make sense.

When you start carrying a payload, the mass ratio of the "tug" becomes horrible, as the payload counts as dry mass, and there LV-Ns will shine.

FWIW:

Rocket LFO tank, drained of oxidizer: mass ratio: 4.6

Mk1 liquid fuel fusalage: mass ratio: 6

Mk2 " " : 7

Mk3 " " : 9.33

^ I'm just looking at the wiki, I don't know if the values changed in 1.0... but pretty much you are using the worst tanks for the LV-N

Link to post
Share on other sites

I built a nice "little" probe with something about 40-50 tons dry, 110 tons wet. it sports the 10000 LF mark III plane tank in the middle, 2 nervas on the side, one ore container, refinery and so on.

and that baby has something over 7000 m/s delta-v, and can land on the small moons (gilly, minmus, bop, pol) for refuelling.

from what I've seen so far, the only problem is, that there aren't stock rockettanks for LF only, and that the nerva still behaves not like a real one (a real one could utilise anything that would boil inside a nuclear reactor, including the oxidizer. in that regard, the old system was closer to reality, it just did not actually burn LFO, although that could be possible, I'm thinking of something like a nuclear preheated combustion system).

Link to post
Share on other sites

NERVA heating the oxidizer would be inefficient, since the temperature of the NERVA core is lower than the temperature of the typical LFO engine and high exhaust velocity is achieved by using the lightest gas possible. O2 is way heavier and it's also corrosive, so if you try feeding NERVA with it, the engine would run very inefficient and die fast.

Link to post
Share on other sites
Both tests are bad. The first one simply indicates that 3-ton engine on 8-ton craft might not be a good idea.

How is that unfair? Having a 3-ton engine on an 8-ton craft IS a bad idea. :P

FWIW:

Rocket LFO tank, drained of oxidizer: mass ratio: 4.6

Mk1 liquid fuel fusalage: mass ratio: 6

Mk2 " " : 7

Mk3 " " : 9.33

^ I'm just looking at the wiki, I don't know if the values changed in 1.0... but pretty much you are using the worst tanks for the LV-N

They have changed. It's 8:1 for most liquid fuel-only tanks (and 8:1 for most spaceplane-type LFO tanks), except the mk1 which is 6:1 for some bizarre reason (maybe to offset the fact that mk1 spaceplanes can use 9:1 ratio LFO tanks easily?)

You can pretty much safely assume the wiki is in some sort of half-edited state for the next forever and a half. Squad did a crapton of part.cfg changes (especially when compared to how few they did in previous releases).

from what I've seen so far, the only problem is, that there aren't stock rockettanks for LF only, and that the nerva still behaves not like a real one (a real one could utilise anything that would boil inside a nuclear reactor, including the oxidizer. in that regard, the old system was closer to reality, it just did not actually burn LFO, although that could be possible, I'm thinking of something like a nuclear preheated combustion system).

A) read Kuu's message (immediately after yours) and understand that 'efficiency' in that context means 'specific impulse'. Kiss that 800 goodbye. (also putting oxidizer through the Nerv would be like operating a blast furnace, with the Nerv's core being the metal to melt)

B) Real NTRs are only getting those spectacular specific impulse numbers with hydrogen - which requires heavy tankage even for short-term storage. The newfound tank suckage of the 'Nerv' is actually the most "realistic". My research to date on long-term hydrogen storage at around 0.2:1 mass ratios (yes, <1) for a long-term liquid system, and 0.05:1 for a high-pressure compressed system (300 bar). Keep in mind that existing H2/LOX conventional engines are only used for getting into orbit (a very short timeframe) so they don't need long-term storage. Long-endurance spacecraft (Apollo CSM/LEM, deep space probes, etc) generally run on some flavor of hydrazine/NTO. von Braun's nuclear tug would require some sort of ... fuel tank magic to avoid boil-off.

C) No NTR has ever flown on a mission, so they aren't in fact "real" at all. Not like a Rocketdyne F-1 or J-2 or Space Shuttle SRB/SSME or Merlin 1C/D or NSTAR ion engines etc.

Link to post
Share on other sites

For those reasons, it seems liquid methane would be the preferred propellant for long duration missions...

You could start with H2 for the initial burn...

but then suppose you're going to mars or something... its still a long journey. For orbital insertion and return to Earth, you'd probably use liquid methane, which gets you only 644s from a realistic solid core design (Gas core designs could do much better).

But here, I think a LANTRN concept could work, and work with ISRU.

You store water/ice.

You use the reactor to hydrolyse the water, and generate a lot of O2 and LH2.

You run the LH2 through the reactor in typical NERVA fashion, and inject the O2 after the reactor as an "afterburner".

Great TWR, the Isp will be about 647s

http://www.projectrho.com/public_html/rocket/enginelist.php#lantr

Anyway... use LF only fusalages, use more LFO tanks emptied of O to get the same fuel mass... even with the worse mass ratio, the Isp makes up for it in many cases.

LV-Ns are still the best engine for moving large things beyond LKO.

(well, excluding ions, if you have the patience for ridiculous amounts of perapsis kicking, and very long burns... and even then, the Xenon propellent is so ridiculously expensive that I'm not sure its every worth it....)

Link to post
Share on other sites
Both tests are bad. The first one simply indicates that 3-ton engine on 8-ton craft might not be a good idea. In the second test you have more than 2x less propellant for NERVAs than for Reliant, because oxidizer is also a part of the propellant.

He has a job to do and is comparing vessels that get the job done. I think that's perfectly fair. If there's anything wrong with that test, it's that he stops with the interplanetary vessel. The LV-T30 variant should be like 30t more on the pad, requiring a larger lifter.

- - - Updated - - -

C) No NTR has ever flown on a mission, so they aren't in fact "real" at all. Not like a Rocketdyne F-1 or J-2 or Space Shuttle SRB/SSME or Merlin 1C/D or NSTAR ion engines etc.

Yet the basic design has been run for many hours on a test stand. I'd say NTRs are a lot more real than, say, RAPIERS or little green men.

Link to post
Share on other sites
You can pretty much safely assume the wiki is in some sort of half-edited state for the next forever and a half. Squad did a crapton of part.cfg changes (especially when compared to how few they did in previous releases).

Most of the information resented on the wiki is updated instantly because the part.cfg files get processed and uploaded by a bot. Last time I checked the template for the engines is somewhat broken. It display(ed) the wrong thrust values because the game changed it's approach to Isp. The data is there and correct but might not be processed correctly. Manual tables might be broken, though.

Link to post
Share on other sites
He has a job to do and is comparing vessels that get the job done. I think that's perfectly fair. If there's anything wrong with that test, it's that he stops with the interplanetary vessel. The LV-T30 variant should be like 30t more on the pad, requiring a larger lifter.

Except for loading less propellant for that stage as well. Now we know that using NERVA with a tiny craft and less propellant than LFO engine doesn't work well. Did we even need to test that?

From second test I see that using NERVA is highly advised for large interplanetary craft, because it grants the same dV with a lighter vessel (wet mass).

Link to post
Share on other sites

Well, I just checked for the Mk3... its an 8:1 ratio LF or LFO tanks (monoprop: 6.6)

For the Mk2: 8:1 LF or LFO tanks (monoprop: 6.5)

For the mk1: 6.0

Jumbo 64 LFO: 9:1 ox emptied: 4.6

Size 3 tanks (kerbodyne ones): 9:1.... they no longer have worse mass ratios than the 1.25 and 2.5m LFO tanks

"Shuttle" delta wings: 4.0

FAT-455 wings: 4.84

Using the "airliner" wings is more mass efficient than using a standard LFO tank that has been emptied of LF.

Only using the shuttle wings as fuel tanks is less mass efficient for storing LF than using a LFO tank with no O.

Of course... one typically puts those on as wings... and the fuel storage is strictly secondary.

But in the case of a SSTO using LV-Ns the wings aren't so much "dead-mass" once in space if you're using LFO tanks with no O... the shuttle wings are a bit... the FAT-455 wings are better.

We need standard cylindrical LF only tanks... unless the spaceplane tanks get nerfed, then they should be between 8-9:1 mass ratio

Link to post
Share on other sites
Except for loading less propellant for that stage as well. Now we know that using NERVA with a tiny craft and less propellant than LFO engine doesn't work well. Did we even need to test that?

From second test I see that using NERVA is highly advised for large interplanetary craft, because it grants the same dV with a lighter vessel (wet mass).

Performing the a task with LV-N requires X propellant, doing the same with ordinary engines requires Y propellant. Comparing two solutions to the same problem is an entirely reasonable proposition. What's unreasonable is that he stops right there and doesn't account for the lifter that's needed to get them into space in the first place.

The kind of small vessel he showed first used to be an entirely reasonable use case for nukes, and I suspect that it still may be; nukes are rather small and lightweight, after all.

FWIW, I toyed around with a Nerva weighing 20t at 360kN, just to see where it would lead me. Guess what, interplanetary tugs pretty much have to look like the 60's designs: that's the natural outcome of adapting a large tank to single small-diameter engine. It doesn't lend itself to realistic-looking landers, of course, but together with ISRU it encourages "let's waste it because we have it" designs like pulp SF tailsitter rockets. Which is fine with me, but I suspect not everyone wants to be funneled down that path.

I very much like what powerful nukes can do for your part count, though.

Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...