[1.8+] Real Fuels

[NathanKell]

RCS thrusters are bladder-fed, didn't that get established? So you'd be ok.

[ialdabaoth]

really, the only mod this should effect is the RCSSounds mod, which will need to check for ModuleRCSFX in addition to ModuleRCS

[HoneyFox]

really, the only mod this should effect is the RCSSounds mod, which will need to check for ModuleRCSFX in addition to ModuleRCS

I remember that RCSSounds is using derivative class of ModuleRCS... probably we need a combination of the these two.

Oh wait, doesn't Effect support audio already, so that we actually get the sound effect automatically when we define the EFFECTs used by ModuleRCSFX.

[ialdabaoth]

I remember that RCSSounds is using derivative class of ModuleRCS... probably we need a combination of the these two.

Oh wait, doesn't Effect support audio already, so that we actually get the sound effect automatically when we define the EFFECTs used by ModuleRCSFX.

Excellent! Now as soon as I really grok EFFECTS and implement it into ModuleRCSFX, we're good to go

EDIT: Oh, that's EASY. I think I've got it working! Bit of weirdness, though:

since many RCS ports have multiple thrust transforms, you can use the EFFECTS block to define lights and sounds, but the actual sprays might need to be done the "old way".

Edited April 22, 2014 by ialdabaoth

[NathanKell]

Why would that be a problem? ModuleEnginesFX supports engines with multiple thrust transforms...

[HoneyFox]

Why would that be a problem? ModuleEnginesFX supports engines with multiple thrust transforms...

A precaution for this:

For sound effect, don't use too many of them at the same time or something bad might happen I guess.

[ialdabaoth]

Why would that be a problem? ModuleEnginesFX supports engines with multiple thrust transforms...

Because normally, either all the transforms are on, or none of them are. With RCS, different transforms are on based on which direction the RCS block is thrusting.

[Starwaster]

BTW ialdabaoth, did you ever figure out how to find which RCS are firing? If not take a look at MechJeb's code. There might be something in there...

[HoneyFox]

Because normally, either all the transforms are on, or none of them are. With RCS, different transforms are on based on which direction the RCS block is thrusting.

Well that shouldn't be a big problem if we play effect at different places with different power values.

[gleedadswell]

I've started playing with nuclear engines (the LV-N and similar in mods). I've noticed that at the moment LV-N and real fuels don't get along very well. In standard KSP you attach an LV-N to a standard (big!) tank and get very good delta-v because of the high Isp. But if you use real fuels then the tank is full of liquid H2 (which is realistic) with a very low density. The result is that the fuel mass/tank mass ratio is only a bit better than 1. For example, a NovaPunch HH-375-C tank full of LiquidH2 has a dry mass of 4.9662 t and a full mass of only 11.0649 t. So the high Isp doesn't translate into high delta-v because the empty tanks are too heavy. The result is that, as far as I can tell, you can never get a better delta-v with an LV-N than you can with an efficient H2/LOX engine and the same number of tanks if you are using real fuels.

In the real NERVA project the NERVA stage was projected to have an empty mass (engine and tanks) of about 34 tonnes and a full mass of 178 tonnes. I calculate its delta-v at an impressive 13 000 m/s. Even with a 20 tonne payload it would have a delta-v of 10 000 m/s. Now for KSP this would have to be scaled for game balance because of the smaller solar system. But still, to have the LV-N beat chemical rockets using liquid H2 you need fuel tanks with much lower empty mass. I'm going to make them for myself by editing .cfg files. But a worthwhile addition to some mod (maybe real fuels, since it only comes up with liquidH2 fueling the LV-N) would be a large, thin-walled tank (low impact tolerance, low mass) intended for nuclear engines. It's not an issue in regular KSP because you use a higher density fuel.

[ialdabaoth]

Well that shouldn't be a big problem if we play effect at different places with different power values.

An Effects has a single power level. You'd have to have a separate Effects node for each thruster port, and then my RCS mod would need some way of wiring them all up right.

[Padishar]

I have finally implemented atmospheric thrust correction in KER. You simply need to replace the DLLs in the standard 0.6.2.3 KER install with the ones linked in my thread here.

[NathanKell]

gleedadswell: Ah, goodness, we haven't had a round of this in 20 pages! Must be time again.

You're clearly not using a large enough, *and cryogenic*, tank. Try using a bunch of jumbo-64s, which are for cryogenic fuels (or a cryogenic stretchy tank). Although frankly I think there's an install issue, since you should *never* be getting a propellant fraction of 55%.

Are you sure you're not combining the mass of the engine *and* the mass of the tank?

[gleedadswell]

... But a worthwhile addition to some mod (maybe real fuels, since it only comes up with liquidH2 fueling the LV-N) would be a large, thin-walled tank (low impact tolerance, low mass) intended for nuclear engines. It's not an issue in regular KSP because you use a higher density fuel.

So looking through configs it looks like this already exists in real fuels. The Balloon and BalloonCryo tank types have very low base masses. At the moment the only tanks of those kinds are apparently in Stretchy Tanks. In my current game I wanted to use Stretchy Tanks but was having memory issues and that mod was one of the casualties. But it looks like I can just go into one of the real fuels config files and change the type of a tank to Balloon or BalloonCryo and it will achieve what I'm after. I'll give that a try.

[gleedadswell]

gleedadswell: Ah, goodness, we haven't had a round of this in 20 pages! Must be time again.

You're clearly not using a large enough, *and cryogenic*, tank. Try using a bunch of jumbo-64s, which are for cryogenic fuels (or a cryogenic stretchy tank). Although frankly I think there's an install issue, since you should *never* be getting a propellant fraction of 55%.

Are you sure you're not combining the mass of the engine *and* the mass of the tank?

Hmm...well the 3.75 m NovaPunch tanks are listed as cryogenic and at one point I was using 21 of them (!!!) for a single nuclear engine. The Orbital Bertha still gave better delta-v for that volume of tanks. If that's not big enough then the VAB isn't big enough... ;-). So I'm inclined to say maybe it is an install issue. I've quite sure about that propellant fraction. The dry mass was a little shy of 5 t and the gross mass was about 11 t for each tank (reading the tank info in the VAB from the real fuels dialogue where you set the tank configuration). I can't use a stretchy tank in my current game. I was running out of memory and stretchy tanks was one of the mods I had to remove to get the game to run.

I'm going to do a little digging in sandbox mode since my tank options are somewhat limited otherwise, then get back to you.

[gleedadswell]

I'm going to do a little digging in sandbox mode since my tank options are somewhat limited otherwise, then get back to you.

Yes, something is wrong. The NovaPunch 3.75 m tanks are listed as Cryogenic. In RealTankTypes.cfg I see that the base mass (is that the dry mass?) should be

basemass = 0.000011 * volume

For a 3.75 m X 9 m tank that should give a mass of 0.94688 Ton if I'm understanding how this is supposed to work. In the VAB that tank shows up as having a dry mass of 4.5523 Tons. But when I click the appropriate button to fill it with LiquidH2 the dry mass changes to 4.9662 Tons. The "current mass" is then listed as 11.0649 Tons.

How does it then calculate the gross mass? In RealTankTypes.cfg under the LiquidH2 entry it says

mass = 0.000006

Is that a multiplicative factor applied to something else? Since I'm using the 3.75X9 tank as my standard I know that, if the walls were infinitely thin (wouldn't that be nice!) the mass of liquid H2 in the tank would be

pi*r^2*h*rho = 7.042 tons

In game the fuel mass in the tank is coming out to somewhat over 6 tons, so accounting for wall thickness this seems reasonable, though I don't see where in the config files this is achieved. So the problem seems to be with the dry mass.

Any ideas about what could be conflicting with the real fuels config? The mods I'm using are:

Toolbar

AIES

H.O.M.E.

Active Texture Management

Deadly Re-entry

FAR

KAS

Procedural Fairings

Kethane

KW

Infernal Robotics

MechJeb

NovaPunch

Planet Factory

RealChutes

RealFuels

Remote Tech

TAC Life Support

Kerbal Alarm Clock

Interstellar

Note, my current game is not using RSS. Maybe some of the RSS rescalings are necessary for making this work?

I'll try in a fresh directory with just RealFuels and see what happens, then add mods until it breaks. I'll get back to you.

Edited April 26, 2014 by gleedadswell

[gleedadswell]

Aha! I thought I had installed an engine pack, but apparently I hadn't. I've installed the "stock" engine configs and now the tank dry mass looks to be correct.

[Starwaster]

Yes, something is wrong. The NovaPunch 3.75 m tanks are listed as Cryogenic. In RealTankTypes.cfg I see that the base mass (is that the dry mass?) should be

basemass = 0.000011 * volume

For a 3.75 m X 9 m tank that should give a mass of 0.94688 Ton if I'm understanding how this is supposed to work. In the VAB that tank shows up as having a dry mass of 4.5523 Tons. But when I click the appropriate button to fill it with LiquidH2 the dry mass changes to 4.9662 Tons. The "current mass" is then listed as 11.0649 Tons.

How does it then calculate the gross mass? In RealTankTypes.cfg under the LiquidH2 entry it says

mass = 0.000006

Is that a multiplicative factor applied to something else? Since I'm using the 3.75X9 tank as my standard I know that, if the walls were infinitely thin (wouldn't that be nice!) the mass of liquid H2 in the tank would be

pi*r^2*h*rho = 7.042 tons

In game the fuel mass in the tank is coming out to somewhat over 6 tons, so accounting for wall thickness this seems reasonable, though I don't see where in the config files this is achieved. So the problem seems to be with the dry mass.

Any ideas about what could be conflicting with the real fuels config? The mods I'm using are:

I wouldn't bother measuring the tank unless you think the person who wrote up the config screwed up the config.

The value that you're citing for H2 isn't for a cryogenic tank. It could be a Fuselage, Balloon or Structural tank.

A cryogenic tank uses 0.000001 for LH2 + the basemass of 0.000011

So, looking at the 3.75x9m NP tank

volume = 86080

86080 x 0.000011 = 0.94688 // Cryogenic tank type

86080 x 0.000001 = 0.08608 // H2 tank

= 1.03296 <------ this should be the dry mass for that tank

86080 x 0.00007085 = 6.098768 // H2 density

it should be coming out to 7.131728 tons for that tank, wet. 1.03296 dry

So if you're seeing 11 tons, either another mod has altered your base masses or the volume of your NP tank. I'm not aware of any of the mods in your list that would do that, but I don't use Interstellar so I can't speak as to that one. If you posted your output_log.txt I could tell pretty quick what configs are altering the NP tank. If you search for NP_lft_375m_9m in the log you can see what's editing it because ModuleManager logs that sort of thing. It would show up looking something like 'Applying node (path to config file) to NovaPunch/(folders)/NP_lft_375m_9m'

Aaaaaand of course you had to ninja me and figure it out while I was typing that...

[NathanKell]

At least I saw the above post before posting mine, which I was writing for the last hour off and on :]

[gleedadswell]

So if you're seeing 11 tons, either another mod has altered your base masses or the volume of your NP tank. I'm not aware of any of the mods in your list that would do that, but I don't use Interstellar so I can't speak as to that one. If you posted your output_log.txt I could tell pretty quick what configs are altering the NP tank. If you search for NP_lft_375m_9m in the log you can see what's editing it because ModuleManager logs that sort of thing. It would show up looking something like 'Applying node (path to config file) to NovaPunch/(folders)/NP_lft_375m_9m'

Aaaaaand of course you had to ninja me and figure it out while I was typing that...

Sorry about ninjaing you. But that is very helpful advice to me for future reference. I didn't realize the module manager log would be that human readable.

Cheers!

[littlelasso]

Hey, I have just installed this mod and I made a test craft with lox and kerosene to see how it worked and for some reason the engine was on and producing thrust but it wasn't using any fuel?

[SchroedingersHat]

Is it just me, or does the basemass for RCS tanks seem a little off?

Example: A long RCS tank weighs 0.83t when full and 0.23 when empty for a ratio of around 3.6 (even worse than the stock ratios). This improves slightly with Hydrazine, but not by much.

Real (modern) tanks range from about 4.5 for something tiny (http://cs.astrium.eads.net/sp/spacecraft-propulsion/propellant-tanks/39-litre-hydrazine-bladder-tank.html)

to 9 for something about as big as the stratus V (http://cs.astrium.eads.net/sp/spacecraft-propulsion/propellant-tanks/177-litre-hydrazine-propellant-tank.html)

Couldn't find a source anything larger that wasn't composite at the same pressure (although slightly lower pressure bi-propellant tanks have ratios up to 30 or so).

They're also way off of the mass ratio of any part of the servicemodule type. My solution was to reduce the RCS tank empty weight by a factor of 3, and increase the weight of the servicemodule empty weight by a factor of 3 so they are closer.

I do like to play with the same configs as everyone else though, so any thoughts?

[Starwaster]

Is it just me, or does the basemass for RCS tanks seem a little off?

Example: A long RCS tank weighs 0.83t when full and 0.23 when empty for a ratio of around 3.6 (even worse than the stock ratios). This improves slightly with Hydrazine, but not by much.

Real (modern) tanks range from about 4.5 for something tiny (http://cs.astrium.eads.net/sp/spacecraft-propulsion/propellant-tanks/39-litre-hydrazine-bladder-tank.html)

to 9 for something about as big as the stratus V (http://cs.astrium.eads.net/sp/spacecraft-propulsion/propellant-tanks/177-litre-hydrazine-propellant-tank.html)

Couldn't find a source anything larger that wasn't composite at the same pressure (although slightly lower pressure bi-propellant tanks have ratios up to 30 or so).

They're also way off of the mass ratio of any part of the servicemodule type. My solution was to reduce the RCS tank empty weight by a factor of 3, and increase the weight of the servicemodule empty weight by a factor of 3 so they are closer.

I do like to play with the same configs as everyone else though, so any thoughts?

while researching manned Mars mission proposals (Copernicus, et al) one thing I was struck by was that projected RCS tank dry mass seemed higher than I had thought they might be. But look at the application. These are pressurized bladder tanks and they're going to have extra mass associated beyond the mass of the tank.

I'm not necessarily saying you're wrong mind you, just that when you say that the base mass seems a little off.... that's pretty much what I was thinking looking at projected tank masses for real life applications.

[NathanKell]

SchroedingersHat: thanks for the catch! I think I was missing a zero, i.e. I was intending RCSHigh to have the same basemass as ServiceModule. But thanks to that link (you don't want to know how much time I spent looking for actual stats on bladder tanks...clearly I was searching wrongly) I'm fixing them.

Basemasses will now be (barring your violent opposition):

RCS 0.0002

RCSHighEfficiency 0.00006

(so RCSHE will be 2x ServiceModule, and RCS [to represent a 5-100L tank, rather than a larger tank] will be 6.6x the SM.)

The tankmass should bulk them out to about in line with your sources, I think, or perhaps a bit below at the high-efficiency end.

Any other suggestions? (Gratefully accepted )

[SchroedingersHat]

while researching manned Mars mission proposals (Copernicus, et al) one thing I was struck by was that projected RCS tank dry mass seemed higher than I had thought they might be. But look at the application. These are pressurized bladder tanks and they're going to have extra mass associated beyond the mass of the tank.

I'm not necessarily saying you're wrong mind you, just that when you say that the base mass seems a little off.... that's pretty much what I was thinking looking at projected tank masses for real life applications.

They are much heavier than cryogenic or unpressurized tanks, but (once they get over a very small size) they seem to have mass ratios on the order of 8:1 or higher

Another example from a satellite http://thebulletin.org/sites/default/files_legacy_files/NASA_White_Paper.pdf

1000lb of hydrazine and 100lb dry.

Bearing in mind that these are modern-alloy, non-structural tanks, a mass ratio of 5-8 to 1 seems about right (like a stock LFO tank) -- even the stock mass ratio would make it a viable choice. As it is, there's this weird situation where servicemodule tanks are an excellent choice for a main propellant, and RCS tanks are useless.

As an aside: How does one get RCS thrusters to use other propellants? I only just noticed that this is (allegedly) a feature of realfuels, but I cannot seem to get it to work. Is there an example config somewhere?