[0.20] Modular Fuel System 1.3/realistic fuels, reconfigurable fuel tanks and engines

[rosenkranz]

Does anyone have a config file for Novapunch tanks and engines? Also, just to point out the new engines and tanks for .21 aren't in the squad config. I added them manually for myself but an offical version, i'm sure, would be appreciated

One for H.O.M.E. would be nice as well

That aside, the mod looks really cool. Looking forward to flying with this stuff.

[Chestburster]

Does anyone have a config file for Novapunch tanks and engines? Also, just to point out the new engines and tanks for .21 aren't in the squad config. I added them manually for myself but an offical version, i'm sure, would be appreciated

*points at his sig*

[rosenkranz]

Heh, i thought i had downloaded the latest. You might consider adding that link to the first post as your sig indicates 1.3 but the link is cfgs for 1.5.7

Thanks for the hard work btw. Looking at the long list of engines and tanks for NP2 alone and I see how much that must take to comb through all that.

[Chestburster]

Heh, i thought i had downloaded the latest. You might consider adding that link to the first post as your sig indicates 1.3 but the link is cfgs for 1.5.7

Thanks for the hard work btw. Looking at the long list of engines and tanks for NP2 alone and I see how much that must take to comb through all that.

I think ialdabaoth will add them to the main download when 1.4 gets released. Indeed, a link in the OP would be helpfull for other people.

The configs are on version 1.5, but are made for ModularFuelSystem version 1.3. I'm always updating my sig when i release a new version.

You're welcome :-) . Since i'm heavily using the mod myself i had to create all those configs (and i can't play KSP without it anymore^^).

[xZise]

Is there a reason, why you can change the LV-T45 fuel configuration? I added them manually (extrapolated the thrust values from the LV-T30) and it works fine so it doesn't look like a bug in 0.21.

Fabian

[rosenkranz]

I think ialdabaoth will add them to the main download when 1.4 gets released. Indeed, a link in the OP would be helpfull for other people.

The configs are on version 1.5, but are made for ModularFuelSystem version 1.3. I'm always updating my sig when i release a new version.

You're welcome :-) . Since i'm heavily using the mod myself i had to create all those configs (and i can't play KSP without it anymore^^).

Yeah, my bad, I'm bad with names and thought you were the OP.

On a different subject, I'm having a little trouble with the fuel mixes. Maybe I'm doing something wrong or maybe I just need to experiment more.

Starting with a .5t payload I was trying out different drive sections.

LF+OX got the least TWR but the most dV, LH2+LOX got the best TWR but the least dV and LF+LO2 was in the middle both on TWR and dV. Is that to be expected?

But what is kinda blowing my mind here is the nerva's. Even though they're more effecient, for the same tonnage I get sickly TWR and no where near the dV than just the standard LF+OX. Now the anemic TWR was to be expected but the dV was not expected. Any thoughts?

Edit: Maybe I should mention I was working with 5t drive section for all types. Maybe there are bigger differences with larger rockets?

Edited August 1, 2013 by rosenkranz

[xZise]

You need to compare their masses and not volumes: You get a lower ÃŽâ€v for LF+LOX (and even lower when both are cryogenic LH2+LOX) but a higher TWR because LOX (and LH2) are not as dense as LF/OX. So the volume is the same but LF/OX are heavier. As ÃŽâ€v and TWR both work with the mass and not the volume you need to compare the fuels when their mass is the same. Basically you add more tanks when you are using LOX (or LH2 and LOX) to have the same fuel mass. Then you get a higher ÃŽâ€v when you use LF+LOX and even more ÃŽâ€v when you use LH2+LOX. But your TWR roughly stays the same (you need additional tanks to fill the same amount of fuel resulting in a slightly lower TWR when the fuel masses are the same).

Build a rocket and look up the mass with LF+OX, LF+LOX and LH2+LOX and you will notice that the rocket gets lighter every step.

Oh and the NERVA using only LH2 (and a bit of nuclear fuel) is SUUUUPER light. Compare the mass of your NERVA craft with a fully fuel LF/OX tank (the NERVA won't work but you only need to compare the masses).

Fabian

[AngelLestat]

Oh and the NERVA using only LH2 (and a bit of nuclear fuel) is SUUUUPER light. Compare the mass of your NERVA craft with a fully fuel LF/OX tank (the NERVA won't work but you only need to compare the masses).

I dont know what version are you using it.

But with 0.21 and the last cherstbuster config files, nerva engines still sucks. They are totally pointless.

No matter how light is H2, something is wrong there. You always get the worst worst delta v ever with any kind of tanks.

I just want something to look real, for example, comparing with real rockets get similar benefics.

I love the direction that this mod is taken and the oportunities that give us, but still needs some work.

[Spanier]

Any hint, when the new version arrives?

[rosenkranz]

You need to compare their masses and not volumes: You get a lower ÃŽâ€v for LF+LOX (and even lower when both are cryogenic LH2+LOX) but a higher TWR because LOX (and LH2) are not as dense as LF/OX. So the volume is the same but LF/OX are heavier. As ÃŽâ€v and TWR both work with the mass and not the volume you need to compare the fuels when their mass is the same. Basically you add more tanks when you are using LOX (or LH2 and LOX) to have the same fuel mass. Then you get a higher ÃŽâ€v when you use LF+LOX and even more ÃŽâ€v when you use LH2+LOX. But your TWR roughly stays the same (you need additional tanks to fill the same amount of fuel resulting in a slightly lower TWR when the fuel masses are the same).

Build a rocket and look up the mass with LF+OX, LF+LOX and LH2+LOX and you will notice that the rocket gets lighter every step.

Oh and the NERVA using only LH2 (and a bit of nuclear fuel) is SUUUUPER light. Compare the mass of your NERVA craft with a fully fuel LF/OX tank (the NERVA won't work but you only need to compare the masses).

Fabian

Well, I wasn't interested in the mass so i was trying to keep a base mass of 5t for the drive. Yes the less dense the fuel mix the more tankage I had to add but I tried to keep the mass the same so could compare TWR and dV. Otherwise you're spot on. If you keep the volume the same then TWR spread is more from LF+OX thru to LH2+LOX as is the dV spread.

My point on the nerva was that for the same mass (much bigger physically, but less dense) a 5t drive section (tanks + engine) the nerva only got me a little over half the dV. The clear winner (bang per ton) was good ol' LF+OX

I quoting from memory of last night, so sue me if my numbers arn't spot on but it looked something like this

With a .5t payload and a 5t drive.

LF+OX had a TWR of 2.3 and 4100dV

LF+LOX had a TWR of 2.8 and 3800dv (little more tankage but not signifigant)

LH2+LOX had a TWR of 3.2 and 3200dv (little bit more but not bad)

NERVA had a TWR of .6 and 2500dV (way more tankage)

That's the bit I'm not understanding. For the same tonnage the NERVA should be closer to 7000dV (or would be going stock). But stock or not it shouldn't matter with 850 isp it should get almost double the mileage for the same reaction mass.

I'm wondering if I'm being unfair as the NERVA's engine mass will account for far more of the drive section. It occurs to me that the stock NERVA is 2.25t which is almost half of the drive section. On the bipropellants I was using a 909 which is only .5t and thus only 10% of the drives mass.

Stand back! I shall now do science!

[xZise]

Okay your … 5 "tonnes" drive isn't 5 tonnes anymore when you don't use LF+OX. I don't know where you get your ÃŽâ€v values and I'm not sure how light LH2 really is (compared to LF+OX) but when you only have 1/8th of the fuel you only get 1/2th of the ÃŽâ€v even when the Isp is 4 times better.

Okay I don't know any numbers so I'm not sure if the NERVA is really inefficient or something like that, but you can't really compare TWR and ÃŽâ€v when you don't compare mass. I mean you could use your 5 "tonnes" drive everytime, but then you need to compare TWR and ÃŽâ€v relative to your mass. Maybe later today I'll try it for my self to compare for myself (unfortunately you didn't posted the masses of your ship with the different fuel configurations).

Fabian

[p1t1o]

Can I just check I've installed all this correctly:

Unzip MFS 1.3

I should have a "ModularFuelTanks" folder in Gamedata now

I also need the ModuleManager DLL in Gamedata

Can someone confirm that I DONT need the "ModuleManagerSource" directory?

Unzip "RealFuels.zip" contained in the MFS directory, giving a "RealFuels" subdirectory

Acquire updated config files from Chestburster (cheers!)

Unzip entire contents of zip file into MFS folder and merge.

Thats right, right?

Just installed mission controller, FAR and Modular fuels in the last 2 days. Combined with deadly reentry, time for some HardMode!!

[xZise]

@p1t1o: Yep this looks correct. The OP really needs some installation guide, as I didn't know where to put the Real Fuels directory.

@rosenkranz: I just tested the NERVA with the X200-16 (the second smallest tank with a big diameter). Filled with LF+OX it has a mass of 13.59 tonnes and filled with only 3.49 tonnes, so about a quarter of the LF+OX configuration. To accumulate the same mass as the LF+OX I used a eleven tanks (300 kg lighter) and those have of course higher ÃŽâ€v (but the same TWR).

Now I don't understand why you want to compare their volume. If you want to do that stick with LF+OX as those are the densest fuels. It has nothing to do with the NERVA is nerfed with this, as your assumptions aren't working then: NERVA use super light LH2 but the same amount of mass then before. So consequently it uses much more volume compared to other fuels. And the mass of the engine is only a secondary concern and it is the stock mass so nothing has changed. Your mass/fuel ratio gets a bit worse: My one tank craft had only 800 kg mass (22 %) while the LF+OX combination used 72.7 % of the craft's mass.

It's hard to explain, but that's the drawback of cryogenic fuels: They are light so you need more tank volume. The NERVA eats 31.85 dm³/s on ground level, but with eleven tanks you have 17.16 m³. The LV-909 (LF+OX) only 3.76 dm³/s but the tank only stores 1.6 m³, and the LV-909 isn't that bad on the ground level.

Fabian

Edited August 1, 2013 by xZise

[rosenkranz]

Alright, i guess it's screenies or it didn't happen When I mean 5 tons, trust that I mean 5 tons.

Here's LH+OX (TWR 3.71, dV 3575m/s)

Here's LH+LOX (TWR 3.9, dV 3527m/s)

Here's LH2+LOX (TWR 3.24 , dV 3718m/s)

Here's the NERVA (TWR 1.11 , dV 3697m/s)

All of them weight exactly 5.5t. I switched up the tankage to get a constant 5.5t total. In this weight class the LH2+LOX is the clear winner for mileage. I think the NERVA will shine brighter at larger scales. Will do more science and post pics later.

[Legwan]

NERVA is 2.25t, almost half of your total mass, while LVT-45 is only 1.5t. Rocket equation kills it (it's like 0.75t fuel extra for LVT-45).

Your example is simply too small for NERVA.

BTW Oscar and FL-T100 also have a worse structural to fuel mass ratio, but it might be of a negligible influence here

Edit: correction, doesn't matter in MFS

Edited August 1, 2013 by Legwan

[rosenkranz]

Bah, well that was clear as mud. The 15 ton class went this way:

LF+OX : TWR=1.36, dV=5527

LF+LOX : TWR=1.43, dV=5414

LH2+LOX : TWR=1.19, dV=5396

NERVA: TWR=0.41, dV=6413

Well the NERVA won the endurance goal easier but the other three is kinda confusing.

Vac ISP for the T45 is:

LF+OX : 370

LF+LOX : 390

LH2+LOX : 460

These do not seem out of sorts according to wiki, so i'm at a loss as to why plain LF+OX seems to be the best all around.

[AngelLestat]

NERVA is 2.25t, almost half of your total mass, while LVT-45 is only 1.5t. Rocket equation kills it (it's like 0.75t fuel extra for LVT-45).

Your example is simply too small for NERVA.

If you think that nerva engines with this mod are ok.. why you dont do your own tests???

Try to get any benefic from them. How I said.. THEY ARE TOTALLY POINTLESS. Something is WRONG with the mod or the config stats.

I dont know very much about how this mod is doing the math, but if someone show me maybe I can help to find where is the problem.

Somebody said that we need to use big tanks.. So what about this test:

I use the big nerva engine from novapunch, but also I did the same test with 3 and 4 stock nerva engines (almost same power output). And in all cases is the same.

How you can see, I use solid rockets to start the burm after 20 km in almost vaccum.

Is all H2... is very lighter. AND ALL THAT FUEL only to rise 1 small pod.

Well with all that and 900 of ISP! I cant enter in orbit around mun!!!

So when someone said that is becoz we dont take into account the h2 weight or something else.. Is INSULTING.

Use your imagination.. do you guys think that if we got in real life 1 little pod with a tank that is 100 times more big with a nerva engine (all in orbit) we will get only 1500 delta V??? Seriusly??

[Milestonjon053]

I’m actually influenced with your item, such large information you cited here, thanks for your sharing and waiting to glimpse your future post. Carry on...

[The Lone Wolfling]

Bah, well that was clear as mud. The 15 ton class went this way:

LF+OX : TWR=1.36, dV=5527

LF+LOX : TWR=1.43, dV=5414

LH2+LOX : TWR=1.19, dV=5396

NERVA: TWR=0.41, dV=6413

Well the NERVA won the endurance goal easier but the other three is kinda confusing.

Vac ISP for the T45 is:

LF+OX : 370

LF+LOX : 390

LH2+LOX : 460

These do not seem out of sorts according to wiki, so i'm at a loss as to why plain LF+OX seems to be the best all around.

See here.

Don't compare apples to oranges - the LV-T45 is a good lower stage rocket for smaller rockets, but isn't designed for efficiency like the NERVA is. If you want efficiency, you want the LV-909 to compare it to. A NERVA is ALWAYS worse than a LV-909 with a single stage, assuming I've got my math right from the previous post. And with a LV-909 it is ALWAYS better to use LF/OX with a single stage.

I did quite a post about this and got basically no response.

The issue is the decreased mass ratio when you go with less dense fuels - the tank dry mass is way too high for the less dense fuels to make sense. IRL, tank mass ratios are way higher (more fuel per mass of tank), which is why it makes sense IRL but not in KSP.

[ialdabaoth]

Everyone: I want to personally apologize for being so late with updates, and so infrequent with checking in to the site.

I'm currently in the midst of navigating the government's bureaucracy for disability, so as to not become homeless.

As soon as I can focus attention on KSP again, I *will* give you an update. You have my word.

[asmi]

Everyone: I want to personally apologize for being so late with updates, and so infrequent with checking in to the site.

I'm currently in the midst of navigating the government's bureaucracy for disability, so as to not become homeless.

As soon as I can focus attention on KSP again, I *will* give you an update. You have my word.

Take care of yourself first! We'll wait as long as it takes.

[rosenkranz]

See here.

Don't compare apples to oranges - the LV-T45 is a good lower stage rocket for smaller rockets, but isn't designed for efficiency like the NERVA is. If you want efficiency, you want the LV-909 to compare it to. A NERVA is ALWAYS worse than a LV-909 with a single stage, assuming I've got my math right from the previous post. And with a LV-909 it is ALWAYS better to use LF/OX with a single stage.

I did quite a post about this and got basically no response.

The issue is the decreased mass ratio when you go with less dense fuels - the tank dry mass is way too high for the less dense fuels to make sense. IRL, tank mass ratios are way higher (more fuel per mass of tank), which is why it makes sense IRL but not in KSP.

Yes, the 909 is proving quite the work horse for small stuff. I've been playing with clusters of those rather than a T45 and the 909's are quite efficient. I went back and read your prior post and I agree. The tanks dry mass are way to high, thus crippling the lower density fuels. This is most specific for the NERVA's. The tanks are so heavy it makes it not worth the trouble to create stages that use them.

However, how do you balance that. This could very well be why stock KSP doesn't have a selection of fuels. Going through and reducing the dry weight of the tanks to be more inline with RL will push up the TWR and dV on all rockets making it a lot easier to design them.

[bitemypixels]

Everyone: I want to personally apologize for being so late with updates, and so infrequent with checking in to the site.

I'm currently in the midst of navigating the government's bureaucracy for disability, so as to not become homeless.

As soon as I can focus attention on KSP again, I *will* give you an update. You have my word.

Hope you find a solution soon first things first . If you dont caome back, know that you did a great job with this mod!

[xZise]

Use your imagination.. do you guys think that if we got in real life 1 little pod with a tank that is 100 times more big with a nerva engine (all in orbit) we will get only 1500 delta V??? Seriusly??

Do you know the big orange tank on the space shuttle?! This is ALL the fuel the Space Shuttle used during ascent and nothing else. And now imagine that most of the tank was for the LH2 while only in the tip was the LOX tank. The LOX tank has a volume of 553 m³ and stored 629 t while the LH2 tank had a volume of 1 497 m³ but stored only 106 t.

Alright, i guess it's screenies or it didn't happen When I mean 5 tons, trust that I mean 5 tons.

Here's LH+OX (TWR 3.71, dV 3575m/s)[…]

Here's LH+LOX (TWR 3.9, dV 3527m/s)[…]

Here's LH2+LOX (TWR 3.24 , dV 3718m/s)[…]

Here's the NERVA (TWR 1.11 , dV 3697m/s)[…]

All of them weight exactly 5.5t. I switched up the tankage to get a constant 5.5t total. In this weight class the LH2+LOX is the clear winner for mileage. I think the NERVA will shine brighter at larger scales. Will do more science and post pics later.

Now the TWR has nothing to do with this mod. The LV-T45 has more than 3 times the thrust of the LV-N (in stock configuration) what your measurements also indicate: 200 kN/60 kN · 1.11 = 3.7

Another important value is also the drymass as the rocket equation uses the quotient of full and dry mass. And the LV-N already make half the mass (41%) of the craft while the LV-T45 only makes 27% of the craft's mass.

Now I personally don't know what the issue here is? It's simple math(s). When you increase the density of LH2 your normal rocket engine configurations get smaller too so this won't balance your outcome. Increase the Isp of the LV-N? The Isp is already huge. Reduce the mass of the LV-N? This actually might help with your problem, but you have to keep balance.

If you think that nerva engines with this mod are ok.. why you dont do your own tests???

Try to get any benefic from them. How I said.. THEY ARE TOTALLY POINTLESS. Something is WRONG with the mod or the config stats.

I dont know very much about how this mod is doing the math, but if someone show me maybe I can help to find where is the problem.

The same goes for you: What could be wrong with this mod? This mod defines the density of LH2, the specific impulse of the LV-N and the thrust of the LV-N. KSP itself then calculates the mass used to generate the given thrust with the given specific impulse and then calculates the volume with the calculated mass and the given density.

Fabian

[AngelLestat]

Do you know the big orange tank on the space shuttle?! This is ALL the fuel the Space Shuttle used during ascent and nothing else. And now imagine that most of the tank was for the LH2 while only in the tip was the LOX tank. The LOX tank has a volume of 553 m³ and stored 629 t while the LH2 tank had a volume of 1 497 m³ but stored only 106 t.

i know very well the structure of the space shuttle tank, and that is an example that shows that I am right.

The size of the space shuttle is 1/2 the size of the orange tank. And in my example I have a tank that has 100 times more volume than the pod, using a NERVA engine with more ISP than the space shuttle.

And it does not matter if I use different tanks or nerva engine sizes, the delta V always sucks.

So.. THERE IS SOMETHING WRONG.

The same goes for you: What could be wrong with this mod? This mod defines the density of LH2, the specific impulse of the LV-N and the thrust of the LV-N. KSP itself then calculates the mass used to generate the given thrust with the given specific impulse and then calculates the volume with the calculated mass and the given density.

So what are you saying? That the problem is in how KSP do the math? What is the DLL for?

We are getting less than the 1/4 of delta V that we should get using h2 and nerva.

I dont have any experiencie modifying dll or other engine game files. But maybe we need to add new algorithm depending what kind of fuel is using it. Or.... change the isp value of the nerva engines, with some crazy value like 3000 just to maintain realism and balance.