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

[ialdabaoth]

... or you can wait for the next version, which will have a heat radiator fin that you can attach to your cryo tanks to keep them cool.

[CaptainArbitrary]

... or you can wait for the next version, which will have a heat radiator fin that you can attach to your cryo tanks to keep them cool.

I can't wait to see how you deal with the completely broken heat system that's currently in the game. I've given up on it, personally, until Squad makes some good under-the-hood changes.

[VFB1210]

Awesome mod! Do you have any plans to include hypergolic fuel and oxidizer? Engines using hypergolic fuels would be lighter because they lack the need for complicated injectors and turbomachinery, but would have somewhat nerfed Isp like in real life.

[Chestburster]

MFS Configs for V1.3 updated!

Changelog:

- fixed B9 Atlas engine

- switched afterburner over to ModuleHybridEngines

- added NovaPunch2 support*

- added Firespitter support (fueltanks only)

- added TouhouTorpedos Vector Engine support

- "Fatman" 240kn got a new and fancy LOX injector

- changed IntakeAir density to 0.0001 (RealFuels only)

- minor bugfixes and typos corrected

*some parts can't be added since they are using fueltanks and engines.

Download the updated MFS Configs for V1.3!

[cardajowol]

Awesome mod! Do you have any plans to include hypergolic fuel and oxidizer? Engines using hypergolic fuels would be lighter because they lack the need for complicated injectors and turbomachinery, but would have somewhat nerfed Isp like in real life.

I did add in the N2O4 and Aerozine 50 and UH25 to my game pretty sweet for the KW engine

[CaptainArbitrary]

Here's a suggestion: Engine configurations that are switchable in flight.

I've started to mess with in-situ resource utilization, by modding Kethane so it's less, y'know, sci-fi. The obvious thing is to mine water ice where it can be found, heat it to turn it into liquid water, electrolyze it into hydrogen gas, compress the gas down into LH2 and then pump it through a nuclear thermal rocket for propulsion. All very easy to do with just custom config files.

Then it occurred to me … why can't I just pump the water through the reactor instead? Reaction mass is reaction mass, right? I mean sure, the higher molecular weight of the exhaust will bring my specific impulse down, but water's way denser than LH2, so it might just even out in practice.

So I ran the numbers and discovered that it can't work. At an operating temperature of 2360 °K and a pressure of 30 atm (the rough operating specs for the original NERVA proposal), water doesn't dissociate enough. The exhaust ends up having a negative enthalpy, meaning you get no useful thrust out of it.

But okay, what about a gas-core reactor instead of a solid-core reactor? Yes, that's pretty much sci-fi stuff, but just to see, I upped the operating temperature to 5000 °K and ran the numbers again. At that operating temperature, you can use liquid water as reaction mass and get an Isp as high as 813 seconds! Off water!

But gas-core reactors are a bit of a bridge too far for me, so I ruled out using water directly as reaction mass. But then I thought … cracking water into LH2 leaves a lot of left over LOX, and my Kerbals can't breathe it all. What if I pumped LOX through an unmodified LV-N nuclear rocket engine?

Well, at that temperature most of what you get out is molecular oxygen, not atomic oxygen, so you're looking at a very heavy, very inert exhaust (M=31.979 kg/mol, H=2290.07 KJ/kg). Because the exhaust velocity of a thermal rocket goes by the square root of the enthalpy of the exhaust, we get a disappointingly low exhaust velocity (just 2140 m/s).

But that still gives us a theoretical Isp of 218 seconds, which isn't bad for free fuel. And again, LOX is much more energy-dense than water, so there are scenarios in which that's totally viable.

So that really long, boring story brings me to my suggestion: In-flight-switchable engine configurations. This nuclear thermal rocket runs best on LH2, with a specific impulse of a whopping 810 seconds, but you can also feed it LOX (at a much lower volumetric flow rate for the same thrust, even) and get a respectable 215 seconds out of it. Use whichever you've got in your tanks.

Just an idea.

[The Lone Wolfling]

Here's a suggestion: Engine configurations that are switchable in flight.

Agreed. Although, that being said, some rockets are rather specifically designed. Perhaps a mass penalty? ("Enable multiple engine configurations: y/n" - yes increases mass of engine by 10%)

I've started to mess with in-situ resource utilization, by modding Kethane so it's less, y'know, sci-fi. The obvious thing is to mine water ice where it can be found, heat it to turn it into liquid water, electrolyze it into hydrogen gas, compress the gas down into LH2 and then pump it through a nuclear thermal rocket for propulsion. All very easy to do with just custom config files.

Water ice makes sense. Also, possibly aluminum for SRBs. Any chance of you releasing said custom config files?

Then it occurred to me … why can't I just pump the water through the reactor instead? Reaction mass is reaction mass, right? I mean sure, the higher molecular weight of the exhaust will bring my specific impulse down, but water's way denser than LH2, so it might just even out in practice.

So I ran the numbers and discovered that it can't work. At an operating temperature of 2360 °K and a pressure of 30 atm (the rough operating specs for the original NERVA proposal), water doesn't dissociate enough. The exhaust ends up having a negative enthalpy, meaning you get no useful thrust out of it.

But okay, what about a gas-core reactor instead of a solid-core reactor? Yes, that's pretty much sci-fi stuff, but just to see, I upped the operating temperature to 5000 °K and ran the numbers again. At that operating temperature, you can use liquid water as reaction mass and get an Isp as high as 813 seconds! Off water!

Not to mention that water, not to mention steam, not to mention superheated steam, is corrosive.

That being said, what about using air (well, liquid nitrogen) as a reaction mass? (I have a vision of a Jool <> Kerbin tug that aerocaptures at each end, refilling its propellant tanks as it screams through the atmosphere).

But gas-core reactors are a bit of a bridge too far for me, so I ruled out using water directly as reaction mass.

What about a liquid core reactor? (Nuclear fuel dissolved in molten salt). Rather heavy though.

Also, look at the nuclear salt-water rocket. Reaction mass is water with uranium or plutonium salts dissolved in it. Storage is an issue - you have to store it in long thin tubes with moderator lest it attain critical mass. Then you squirt the water out the back into a nozzle shaped in such a way that it fissions in the nozzle. That way it can have an absurd exhaust temperature without melting the rocket, as the superheated steam never actually really comes into contact with the rocket. (Except for a bit in the nozzle)

But then I thought … cracking water into LH2 leaves a lot of left over LOX, and my Kerbals can't breathe it all. What if I pumped LOX through an unmodified LV-N nuclear rocket engine?

Well, at that temperature most of what you get out is molecular oxygen, not atomic oxygen, so you're looking at a very heavy, very inert exhaust (M=31.979 kg/mol, H=2290.07 KJ/kg). Because the exhaust velocity of a thermal rocket goes by the square root of the enthalpy of the exhaust, we get a disappointingly low exhaust velocity (just 2140 m/s).

But that still gives us a theoretical Isp of 218 seconds, which isn't bad for free fuel. And again, LOX is much more energy-dense than water, so there are scenarios in which that's totally viable.

...Yes, because superheated steam isn't bad enough, let's pump superheated oxygen through our reactor! Sounds Kerbal to me!

Also, what does the overall specific impulse of water, after splitting, come to? 284s? (1.00794 * 2 * 810 + 15.9994 * 218) / (1.00794 * 2 + 15.9994). Although, that being said, it should be higher than that as you can burn the low specific-impulse oxygen first, and then the high specific-impulse hydrogen after.

Finally, random: "For example, a mixture of lithium, hydrogen, and fluorine produced a specific impulse of 546 seconds; the highest ever of any chemical rocket motor." O.o - because superheated oxygen isn't bad enough.

[ialdabaoth]

Note, the KSPX large LV-N already does this; it's switchable in-flight between 100% LH and 30% LOX + 70% LH.

[CaptainArbitrary]

Agreed. Although, that being said, some rockets are rather specifically designed.

Oh no question, but I figured we're sort of hand-waving that already with the whole notion of config files. When you hit the button in the VAB that says "configure for LH2/LOX" or whatever, I imagine the Kerbals are really swapping out the engine with a very similar one that was built to use those propellants at that pressure, et cetera. It's more a part-count-reducing thing than anything.

Water ice makes sense. Also, possibly aluminum for SRBs. Any chance of you releasing said custom config files?

I suppose I could, eventually. I just installed Kethane for the first time this morning, and all I've got working so far is the fact that water ice is, in fact, on the Mun. Hooray.

Not to mention that water, not to mention steam, not to mention superheated steam, is corrosive.

Look, Negative Nancy, I never said it was a good idea.

That being said, what about using air (well, liquid nitrogen) as a reaction mass? (I have a vision of a Jool <> Kerbin tug that aerocaptures at each end, refilling its propellant tanks as it screams through the atmosphere).

N2's not actually that bad, as these things go. At NERVA temperature and pressure, it gets you a theoretical 226 seconds. Air's three tenths of a second less, cause you get a lot (well, a non-trivial fraction) of CO2 and NO out the back.

What about a liquid core reactor? (Nuclear fuel dissolved in molten salt). Rather heavy though.

I'd have to do some reading up on it before I knew what I thought about that. It does sound impractical for space travel.

Also, look at the nuclear salt-water rocket. Reaction mass is water with uranium or plutonium salts dissolved in it. Storage is an issue - you have to store it in long thin tubes with moderator lest it attain critical mass.

Yeah, I wouldn't even know how to begin modeling that in the game. The game just knows how much of a resource exists in a given part; it doesn't know anything about how densely that resource is packed in there. So I wouldn't know how to trigger logic that says "If your reaction mass is packed too tightly, you all die of neutron radiation poisoning."

Also, what does the overall specific impulse of water, after splitting, come to?

It depends entirely on how you split it. At 2360 °K, you end up with:

 H              1.2692-5
 H2             1.0098-3
 H2O            9.8846-1
 O              6.4996-5
 OH             4.5382-3
 O2             5.9099-3

Those are mass fractions, not molar fractions, but you can see by weight most of the water  98% of it  comes out the back as water. That's why you have to run the temperature of the chamber up a lot higher to get a noteworthy specific impulse. Water really likes being water.

Note, the KSPX large LV-N already does this; it's switchable in-flight between 100% LH and 30% LOX + 70% LH.

Oh I see, by using ModuleHybridEngines. Hmm. I'll have to look at that. I've never played with it before. Thanks for the tip.

[The Lone Wolfling]

Oh no question, but I figured we're sort of hand-waving that already with the whole notion of config files. When you hit the button in the VAB that says "configure for LH2/LOX" or whatever, I imagine the Kerbals are really swapping out the engine with a very similar one that was built to use those propellants at that pressure, et cetera. It's more a part-count-reducing thing than anything.

Yeah, that's about what I imagine what's happening. Hence the "additional mass required for switching modes in-flight" thing.

I suppose I could, eventually. I just installed Kethane for the first time this morning, and all I've got working so far is the fact that water ice is, in fact, on the Mun. Hooray.

Hmm... Too bad Kethane-based resources cannot be distributed by latitude.

N2's not actually that bad, as these things go. At NERVA temperature and pressure, it gets you a theoretical 226 seconds. Air's three tenths of a second less, cause you get a lot (well, a non-trivial fraction) of CO2 and NO out the back.

225.7 seconds... Hmm... ~3km/s delta-v required... That's a mass ratio of under 4, which is actually doable.

What about using gas giant atmosphere as propellant? That's mainly hydrogen and helium, is it not?

Yeah, I wouldn't even know how to begin modeling that in the game. The game just knows how much of a resource exists in a given part; it doesn't know anything about how densely that resource is packed in there. So I wouldn't know how to trigger logic that says "If your reaction mass is packed too tightly, you all die of neutron radiation poisoning."

As a first approximation just set utilization in the MFS config to a relatively low value. Beyond that, something that started inducing random control jitter?

It depends entirely on how you split it. Those are mass fractions, not molar fractions, but you can see by weight most of the water  98% of it  comes out the back as water. That's why you have to run the temperature of the chamber up a lot higher to get a noteworthy specific impulse. Water really likes being water.

What about running the water through an electrolytic cracker before entering the rocket?

[Chestburster]

Note, the KSPX large LV-N already does this; it's switchable in-flight between 100% LH and 30% LOX + 70% LH.

As well as NovaPunch2's NERVA and the 240kn FATMAN ;-)

PS: Could you add a link to the OP for the updated 1.3 config files until the main DL gets updated with them?

Edited July 12, 2013 by Chestburster

[SchroedingersHat]

Water ice makes sense. Also, possibly aluminum for SRBs. Any chance of you releasing said custom config files?

There are plans for Aluminum and Magnesium (the former having higher Isp, the latter being easier to ignite, alloys have also been considered) CO2 breathing turbines on mars, singificantly lower Isp than an air breathing engine and heavy, but still better than a chemical rocket once you take into account the free Oxidizer and any bypass ratio you might wind up with. It would make a good alternative for KintakeAir, but I've yet to go to the effort of learning C# to fix the air breathing engine system.

[RuBisCO]

See this post

http://forum.kerbalspaceprogram.com/showthread.php/31706-0-20-Modular-Fuel-System-1-3-realistic-fuels-reconfigurable-fuel-tanks-and-engines?p=416088&viewfull=1#post416088

Yeah that was not what I was looking for, is there a way I could make a part that deactivated boil-off, I don't want to deactivate globally, just when I attach and power a specific part.

[somnambulist]

MFS Configs for V1.3 updated!

...

- changed IntakeAir density to 0.0001 (RealFuels only)

Hey Chestburster, did you do much testing with this? The density change drastically reduces the amount of intake air available to air-breathing engines to the point where (in testing) the stock Aeris A3 putters down the runway at barely 15% throttle before flaming-out. By reducing the density of intakeair you've thrown-off the mixture ratio for air-breathing engines. A stock turbofan uses a fuel mixof 15 parts IntakeAir to 1 part LiquidFuel -- the density reduction changes that to a mix of 750:1!

You could probably work around this by switching all air-breathing engines and air intakes over to a custom air resource with the correct density, and then adjusting the properties of the engines/intakes to account for the lower density air. Seems like a lot of work to me.

[RuBisCO]

I tried modifying the RealTankType.cfg with a new fuel for nuclear thermal rocket engines: Water, this worked very well (once I made a nuclear engine that could run off that fuel) but when I tried to make Ammonia, I could not get it to fill the tanks, some testing seem to have revealed the word "Water" is recognized, but any other word is not.

[MKSheppard]

Sir, you win ONE INTERNETS for actually making the idea I was pushing forth in this thread work:

http://forum.kerbalspaceprogram.com/showthread.php/23824-Propellant-Plugs-in-for-Fuel-Tanks?highlight=propellant

I need some more playing around with this to form a sufficiently informed opinion; but, sir. YOU WIN ONE INTERNETS ONCE MORE!!!!

[RuBisCO]

Sir, you win ONE INTERNETS for actually making the idea I was pushing forth in this thread work:

http://forum.kerbalspaceprogram.com/showthread.php/23824-Propellant-Plugs-in-for-Fuel-Tanks?highlight=propellant

I need some more playing around with this to form a sufficiently informed opinion; but, sir. YOU WIN ONE INTERNETS ONCE MORE!!!!

Just ONE internet point to the OP? That's low man.

[SchroedingersHat]

@Chestburster: Are you working on novapunch already/would my efforts be welcome if I were to start making a config for it tomorrow?

[Chestburster]

My latest version has NP2 support. But theres a problem with IntakeAir (tested it with a wrong version of my modded install so i didnt notice the change) so you need to manualy remove the IntakeAir change from the realfuel resources or wait for the next version (wich will include better HexCan tank values).

[SchroedingersHat]

@Chestburster do you know what's going on with xenon? I can't figure out what is intended from the configs (nor do squads configs make much sense in the context of modularfuels).

As far as I can guess from looking at a phase diagram, in reality one would pressurize it to 150atm or so and store it as a (very dense, about 3.75 * density of LiquidFuel) liquid. The relatively heavy stock tanks when empty for the size seem to agree with this, but then the density is way off for the system modularfueltanks has got going. Also I've either installed it wrong/gotten my versions mixed up or the stock and kspx xenon tanks don't have an entry in your configs.

The settings that would make sense to me are density of 0.018 and a tank weight of somewhere between 0.006 and 0.009. Maybe have a specialized tank type (highpressure) with most of that as basemass and slightly lower over-all weight.

The kethane xenon production would also have to be overridden so you don't wind up with it coming out your ears.

Edit: Probably halve the density as it has no liquid phase (supercritical) at temperatures it might reach in direct sunlight. It requires a bit more pressure too, but high pressure tanks aren't as heavy as I expected.

Edited July 15, 2013 by SchroedingersHat
Misread pressure

[RuBisCO]

Ok can someone help me here: I added "Water" as a propellent to the RealTankType.cfg file without a problem, but can't get it to work with "Ammonia" it just won't let me fill tanks with it even though it is listed, I tried switching out water with ammonia in the code I verified worked for water and it won't work, I even switched around the names and the code that failed to work when it was named "Ammonia" works fine when its name "Water". Little help.

[The Lone Wolfling]

Ok can someone help me here: I added "Water" as a propellent to the RealTankType.cfg file without a problem, but can't get it to work with "Ammonia" it just won't let me fill tanks with it even though it is listed, I tried switching out water with ammonia in the code I verified worked for water and it won't work, I even switched around the names and the code that failed to work when it was named "Ammonia" works fine when its name "Water". Little help.

Did you add it to a resource cfg file?

[RuBisCO]

Did you add it to a resource cfg file?

Tried that, had no affect, in fact Water works without being in the resource.cfg If someone has added a fuel or two to the existing Realfuels mod please post how you did it in enough detail to replicate it (post the code)

[SchroedingersHat]

Using this and IonCross (both need to be installed, it installs into the modularfueltanks folder and the ioncross folder, the only file it overwrites is the ioncross cfg), I've made the logistics stuff a bit more complicated. There's food, a heater so you can breathe LOX in each command pod, water, waste recycling and hydrogen fuel cells (using the ioncross model and stock rtg model for now). Anyone like the idea/have any thoughts and feedback?

I based everything as well as I could on reality and a rough estimation of solar panel size and insolation equivalent to earth's. Batteries needed to be upgraded significantly (they now have the energy density by mass of lead acid batteries, although I think they're somewhat denser).

The ioncross tanks are also now configurable using the modular fuel tanks interface, I added a new tank type of logistics (which I haven't tweaked and balanced yet).

http://www./?cp1c11l8aodh2ya

Thoughts so far:

Add my own models.

A bigger fuel cell, a bigger RTG, maybe a liquidfuel/intakeair generator.

Make it play nice with kethane

Add cfgs for things like fustek and the ISS pack.

I think power requirements are reasonably realistic, but a large amount of the power would be thermal and there are lots of things on the ship that would produce heat that I'm not taking into account so I may need to change it.

Ion drives now make even less sense in the context of the edited batteries

This brings power values for command pods and batteries way out of balance with everything else; do people think it better to ignore realism or start editing things like probe cores/turbine engines/lights etc?

Tried that, had no affect, in fact Water works without being in the resource.cfg If someone has added a fuel or two to the existing Realfuels mod please post how you did it in enough detail to replicate it (post the code)

This seemed to work for me:

RESOURCE_DEFINITION
{
  name = Water
  density = 0.006 // Liquidfuel is 0.8 g/cm^3 iirc
  flowMode = ALL_VESSEL
  transfer = PUMP
}
RESOURCE_DEFINITION
{
  name = Food
  density = 0.003 // Dried food is light.
  flowMode = ALL_VESSEL
  transfer = PUMP
}
RESOURCE_DEFINITION
{
  name = Waste
  density = 0.008 // Probably sinks in water.
  flowMode = ALL_VESSEL
  transfer = PUMP
}

Note that defining a new resource uses the

RESOURCE {

name = foo

}

syntax while editing an existing one uses all the @s and square brackets (took me a while to figure out although I didn't look hard enough to see if there are any docs for modulemanager)

[hellion13]

so i had this before, and it worked fine love the mod, but now ive gone through and redid all the addons and mods i have and now it doesnt work, gets to the pods and fails to load the capsules IDK no crashes or anything so no log, just gets to them in the load screen and stops loading

using the entirety of KW and Novapunch, and mechjeb2 w/ latest dev build

anybody else have this issue?

edit: i see that novapuch has compatibility, so maybe its the KW thats making it not work?

edit #2: tried everything from removing all the mods and trying a stock KSP build, to one add on at a time and nothing, it gets to /Squad/Spaces/PodCockpit/model and jams

edit #3: got the baseline mod to work but its the advanced part thats breaking my game how do i install this?

edit #4: i give up now ive lost even the base mod, tried every way i know how, to get this to work and it gets to the above point at the load screen and hangs i can let it sit for 15 minutes and it wont do anything, until this gets a proper readme and/or install instructions or some one points at something i havent tried, ive gotta let this one sit, as stated above im using parts packs that have been specifically worked into this mod, game works fine with out the mod in it, hell even with all of KW and NP installed it only takes my PC about a minute and a half to load the game so i KNOW its not my PC

it really shouldnt be this much of a hassle to get a mod to work.....

Edited July 16, 2013 by hellion13