[1.8+] Real Fuels

[NathanKell]

Can you follow the guidelines in the support sticky? The repro steps are clear, but I need a (minimal) set of mods to replicate, and the log. Thanks!

[frencrs]

Basically, Tweakscale works but when re-sizing a tank the volume RealFuels shows never changes and RealFuels will only add tanks of that volume or less. e.g. A stock Squad fuel tank will have a 16kL volume limit at 2.5m, 3.75m, and 5.0m sizes.

This happening on a fresh install of KSP from Steam, with only TweakScale 1.43 and RealFuels 7.3 installed using the copy of ModuleManager 2.2.2 that is included in the RealFuels 7.3 download.

Here is an relevant error from the log file and the whole log file:

[LOG 23:31:09.077] AssemblyLoader: Loading assemblies
[ERR 23:31:09.089] AssemblyLoader: Exception loading 'TweakScale_ModularFuelTanks': 
System.Reflection.ReflectionTypeLoadException: The classes in the module cannot be l
oaded.
  at (wrapper managed-to-native) System.Reflection.Assembly:GetTypes (bool)
  at System.Reflection.Assembly.GetTypes () [0x00000] in <filename unknown>:0
  at AssemblyLoader.LoadAssemblies () [0x00000] in <filename unknown>:0

Additional information about this exception:

 System.IO.FileNotFoundException: Could not load file or assembly 'modularFuelTanks,
 Version=5.1.1.0, Culture=neutral, PublicKeyToken=null' or one of its dependencies.
File name: 'modularFuelTanks, Version=5.1.1.0, Culture=neutral, PublicKeyToken=null'
[ERR 23:31:09.089] AssemblyLoader: Exception loading 'TweakScale_RealFuels': System.
Reflection.ReflectionTypeLoadException: The classes in the module cannot be loaded.
  at (wrapper managed-to-native) System.Reflection.Assembly:GetTypes (bool)
  at System.Reflection.Assembly.GetTypes () [0x00000] in <filename unknown>:0
  at AssemblyLoader.LoadAssemblies () [0x00000] in <filename unknown>:0

Additional information about this exception:

 System.IO.FileNotFoundException: Could not load file or assembly 'RealFuels, Versio
n=7.2.1.0, Culture=neutral, PublicKeyToken=null' or one of its dependencies.
File name: 'RealFuels, Version=7.2.1.0, Culture=neutral, PublicKeyToken=null'
[LOG 23:31:09.093] AddonLoader: Instantiating addon 'ConfigManager' from assembly 'ModuleManager.2.2.2'

Edited September 2, 2014 by frencrs

[Hallebumba]

I am sry when this have been answered before.

Here is my problem: I've installed the Realism Overhaul with a few mods and did everything what the instruction said. Over and over again. But I have still just liquid fuel and oxidizer. But the engines require real fuels. What do i have to do?

I am running on 64bit btw.

[NathanKell]

Hallebumba: On what engines? Not all engines are supported by Realism Overhaul. Also, please follow the sticky HERE.

TweakScale will be fixed for the next RF release.

[Hallebumba]

Hallebumba: On what engines? Not all engines are supported by Realism Overhaul.

The engines are totally fine, but i cant put the real fuels in the tanks (AIES,Stock,KW). I'll try another install from scratch and tell you if it worked.

Edit: It worked this time. But I cant tell what was different.

Edited September 3, 2014 by Hallebumba

[dreadicon]

Just a quick question, are B9 Aerospace v5 parts going to be supported in the next release? Seeing as the mod is out now, I mean. I know engines are the responsibility of the pack maintainers, but it seems the tank configs are handled here. I'd ask for when they'll be out too, but I know that's not kosher

My only complaint, if any, would be that I have yet to find a use for hydrox or pure H2; the weight of the tank itself kills total dV in most cases despite better isp on the engines, and has low thrust, not to mention boil-off. Even with the NERVA trimodal variant's insane thrust and decent ISP...

Also, why are there MON10, MON20, etc fuel types? What do they do?

Thanks for a great mod! I love it!

Edited September 3, 2014 by dreadicon

[NathanKell]

dreadicon: Yup, as I've said a couple times (on the previous page of the thread at the most recent) the next version will support B9 v5. Turns out Raptor831 was already working on that too.

Sounds like you're using tanks that are far too small. Make sure you use *enough* tanks. Check the FAQ in the third post for a longer explanation; it's written in re NTRs, but applies equally well to chemical hydrolox engines.

MONxx are Mixed Oxides of Nitrogen (with an N2O concentration in the NTO of [given number]). They're slightly-worse-performing but more-stable storable oxidizers (compared to pure dinitrogen tetroxide).

And--thanks so much!

[dreadicon]

Ah, seems I missed your posts about B9, sorry about that!

As to Hydrogen, I used a very large tank for the Hydrolox; 3.75m part, at least twice as long as it was wide in my testing. I tried pretty much all of the engines I had (stock, KW, and KM), and the only useful scenario I was able to find was for low altitude stages; however, SRBs are so cheap, and liquid fuel tanks so expensive, SRBs have universally outshined them as an early launch solution in practical situations. Might just be SRBs are under-priced, or IRL factors ignored (vibration, reliability, etc) which would normally counter this. The lack of a sizable Aerospike has also hindered my preference towards Hydrolox, as it seems it's best use would be for early stages where the Aerospike shines, but only a 1.25m Aerospike engine is available baring obscure/outdated, unsupported mods.

Random suggestion; Argon is used by a number of mods, if convenient it would be nice to have as a fuel in the core pack, though I understand if you'd rather not include it (the stock ion engine could use it, you know ).

Also, I am fiddling with MM comfigs for tank weight based on surface area and mass, rather than volume. Moreover, different tanks having different attributes; for example, Unpressurized tanks only hold liquids, while pressurized tanks can hold either, but tend to weigh slightly more, and hold less volume (rounded or spheroid in shape). Spherical tanks in general are the lightest for their volume, having the least surface area. Then there is the matter of cryogenics; they are all heavier due to insulation, but sometimes extreme cryogenics which are actively cooled are required to avoid boil-off, and alternatively, maybe the player could sacrifice insulation to save weight when the stage will be immediately used. These are things I am fiddling with on the side, currently.

Lastly, would there be any way to have tech level default to the highest level available, or alternatively just update to the higher level and not show up in the GUI? Just a thought

Thanks again for your responses, I look forward to the next release!

[Rothank]

Quick workaround for those who REALLY need to use new B9 cockpits, IVAs and all that. Use old B9 parts, but replace cockpits with 5.0 B9 cockpits + add(/replace old) all 5.0 plugins. Another hint: for some reason the new HL cockpit in RO is downscaled to 2.5m, some other parts are of strange sizes too. I simply got rid of b9 RO cfg not the most elegant way to deal with it, but hey, works for me!

[NathanKell]

dreadicon:

1. What set of engine configs are you using?

2. What size engine are you testing with? And payload? If you're using a 3.75 medium-height tank (which is what, 40 tons or so wet?), then you probably want about 300-400kN of engine to go with it (call it 0.5t dry mass) at ~450s Isp, and maybe 4t of payload. Assuming ~5% for tank mass, that means a mass ratio of 7.55 and a total delta V of just shy of 9km/sec. Even if you quadruple the payload (to 16 tons) you still get over 5km/sec of delta V.

If you're testing with more engine or more payload, it's not going to work well; hydrolox stages are best for orbital applications where burn length matters less so dry mass can be minimized (in this case it's a burn time of 7+ minutes).

*All* tanks are pressure vessels, which means tank dry mass does scale with volume (although insulation does scale with surface area so it's more like vol^.95 or something). Even pump-fed engines need a fair amount of pressure in the tank. Obviously highly-pressurized tanks will have a higher structural fraction, as you point out--look at the mass ratio of the Able/Delta upper stage if you want to weep (22% or so, vs. ~5-6% for a pump-fed stage; now, some of that's RCS and avionics, of course, which are absent or a lower relative amount for a larger lower stage). The ServiceModule tank type is for that type of stage; Default is for storables or mildly-cryogenic liquids (or very early cryogenic tanks); Cryogenic is for 1970s+ cryogenic tanks. Balloon and BalloonCryo are as they say, balloon tanks like Atlas or Centaur.

Nope, there's no way to set default tech level right now. It's actually non-trivial to be based on what's available (I'd have to catch when tech nodes are updated and then update everything in the editor parts list). And certainly I don't want to prevent changing tech level; there are many reasons you might want a lower TL engine in career (especially cost-related ones if you're using the sane costs MCE creates).

[dreadicon]

Hey NathanKell, thanks for your patience with my various theories, experiences, and speculation.

I'm using Stock-alike engine configs. After some fiddling, I was able to figure out some cases where Hydrogen could be useful with the LV-N, but launching the requisite tank into orbit with a mod like FAR is easier said than done. the tanks are absolutely massive (using procedural parts, the tank was like 3.75 x 10 by the time I got decent dV for a ~6t payload), and the payloads proportionally tiny. Combined with the huge cost of tanks in general, I've found Ion and Plasma solutions to be almost universally preferable, but that's if you have something like Interstellar or Near Future. Without larger scale ion propulsion, LV-Ns would be a solid solution. And I've found a few other useful cases here and there, but overall, the bulky nature of them goes against my style of engineering in KSP, so it is more or less personal taste, but it's definitely not a bread & butter solution for anything. I personally use SRBs for heavy lifters + Kerox for orbital insertion, and either ion/electric or hypergolic bi-propellant for vacuum propulsion after initial orbital insertion. Bulky crafts can be very hard to land, after all

Interesting about the tank weight being based on volume, but I suppose that makes sense. Still, a sphere should have lower mass than a cylinder for the same volume, and comparing a tank with similar dimensions to a tank with highly dissimilar dimensions, the latter should have much higher mass (a wide, flat tank vs a tank with roughly similar height and width); more surface area for the same volume means higher mass, unless there's some trick I don't know about, that's just math/physics. My thought was to give a point over aesthetics to things like exposed tanks in trusses vs. tanks in aerodynamic shells. Crafts intended only for operation in vacuum look like structural skeletons with parts attached for a reason (crewed areas and advanced components excepting). They also often use spherical tanks, as those can optimize the weight-volume ratio of the tanks. After more careful inspection of your post and the RF configs, I realized this more or less is already done, though Procedural Parts tanks don't seem to actually enforce your configs based on shape at all.

I understand about the tech levels; I kindof suspected as much, but it's tripped me up in my designing more than a few times (especially when the right-click menu doesn't update until a reselect, which happens on about 80% of my parts). Might write up myself a MM config to do what Mechjeb does (replacing/re-factoring parts as unlocks happen; if that will work even), as I have never needed to downgrade the tech level on an engine.

Thanks for your replies!

[Sandworm]

I understand about the tech levels; I kindof suspected as much, but it's tripped me up in my designing more than a few times (especially when the right-click menu doesn't update until a reselect, which happens on about 80% of my parts).

Are you using RemoteTech2? I've found that the non-updating menu problem only occurs with unmanned probes using the RemoteTech module. Mannded craft, or non-Remotetech-enabled probe cores, don't have the problem.

[NathanKell]

dreadicon: sure!

1. With a 6 ton payload and a 4.3 ton NTR, you're looking at a dry mass of over ten tons *before* you start adding fuel. To even get a mass ratio of 2 (i.e. as much fuel as dry mass) you're going to need something like 11 tons of hydrogen. And, given how not-dense hydrogen is, you'll need something like 155,000 liters of it. That's 155 cubic meters, which is probably more like a 16-17 meter 3.75m diameter tank. You have to look at the *mass* of fuel you're carrying, not the volume.

Of course, the other thing to remember is that 3.75 x 16 is actually...not that big a payload for a real life payload fairing. It would easily fit in a Delta IV or Atlas V 5m PLF. You'd need the Heavies to get it into LEO, though, with it being 20+ tons all up.

YMMV if you're on stock Kerbin and your payload is larger than your launcher.

2. While spheres are slightly stronger (relatively) than capsule tanks, they're also more expensive and not as useful. So a purely spherical tank should mass something like 70% of a capsule tank, but it will also lead to poorer volume utilization (and thus a higher structure:tank mass ratio) than a capsule tank, so that evens out some. However, comparing apples to apples (a 1m radius sphere vs a 2m radius sphere) the mass scales linearly with volume.

Since RF treats a part not as a tank but as a stage filled with user-definable tanks, I have not thought it worthwhile to try to exactly model the dry masses of the tanks involved; heck, even those "you can actually see the tank" parts get *simulated* by RF as a modular volume into which can be placed any number of "sub-tanks." For that reason I default to a volume-utilization of about 86% (somewhere between the Titan II first stage and the S-IC), although that is customizable if you want to replicate some other (set of) tank shapes.

3. Ah, probably the best approach would be to add a button to the toolbar in the VAB/SPH that will, when clicked, auto-upgrade all engines to their highest-allowed TL. Because even if the parts in the editor part list default to highest available TL, when you load a craft all engines will have the TL that they had when you saved the craft.

[Comet Tail]

Here's something I've come across. I'm planning a very long mission, and the hydrogen boiloff, even with cryo tanks, leaves my tanks completely empty. However, after doing some research, I found that Zero Boil-Off (ZBO) technology does exist. I've modified my own copy of the mod so that the "service module" tank has ZBO for LH2 and LOX, since that's the heaviest tankage and thus would probably realistically reflect that ZBO tanks would probably be quite a bit heavier. Just something to consider.

Here's a link to an article proving it's possible.

[...]

2. While spheres are slightly stronger (relatively) than capsule tanks, they're also more expensive and not as useful. So a purely spherical tank should mass something like 70% of a capsule tank, but it will also lead to poorer volume utilization (and thus a higher structure:tank mass ratio) than a capsule tank, so that evens out some. However, comparing apples to apples (a 1m radius sphere vs a 2m radius sphere) the mass scales linearly with volume.

[...]

That's actually something that bothers me a bit. Larger tanks should get greater fuel/tank mass ratios, since if you keep the thickness of the walls constant, and make the radius larger, the volume increases more quickly than the mass of the walls. So shouldn't larger tanks have better tankage ratios?

[NathanKell]

1. ZBO is modeled by the use of cooling fins (they're in the Science tab). Add them to your tanks to keep your tanks cool. The paper itself talks about how it's not passive cooling (unless I'm misreading it).

2. Nope, because you don't keep the thickness of the walls constant. Tanks are pressure vessels; all tanks of the same shape will have the same mass fraction (as a function of volume). Spheres and pills are different shapes.

[The Pink Ranger]

1. ZBO is modeled by the use of cooling fins (they're in the Science tab). Add them to your tanks to keep your tanks cool. The paper itself talks about how it's not passive cooling (unless I'm misreading it).

Any chance you could make tanks cross feed through the pumps? I like using the radial tanks from NFP but it's kind of impractical to stick a fin on every individual tank, especially the smaller ones.

A ship like this for example looks kind of funny when every tank has a fin sticking out of it and the part count increases quite a bit too.

[Comet Tail]

1. ZBO is modeled by the use of cooling fins (they're in the Science tab). Add them to your tanks to keep your tanks cool. The paper itself talks about how it's not passive cooling (unless I'm misreading it).

2. Nope, because you don't keep the thickness of the walls constant. Tanks are pressure vessels; all tanks of the same shape will have the same mass fraction (as a function of volume). Spheres and pills are different shapes.

1. Oh, okay! I was wondering what those were. I reviewed the readme, and there's some brief mention of the thermal fins but not really enough to explain their purpose for me: I got that they cool parts, but for some reason my brain didn't connect that that would mean cryogens wouldn't boil off (I remembered reading that hydrogen will leak out of any container made of matter, somewhere, so I'd assumed H2 boiloff occurs due to an intrinsic property of the substance, rather than its temperature. Now I'm guessing what I read was regarding gas-state hydrogen and not LH2? I guess its pretty apparent now, though, that liquid cryogens boiloff because they're not kept cold enough).

And yeah, when I was poking around the configs, I was wondering what function the data value for the temperature of the fuels served. That makes a lot of sense, now, thanks!

~will be resetting the configs to the backups since I now know how to pack my cryogens for the long voyage...

As for the paper, I'm not sure. The whole thing seems to go on as though it's an active system, but then it mentions some passive systems, which I guess are subsystems? But yeah, I dunno. No matter, though, the thermal fins got this.

2. Whelp, that's a surprise. And here'd I'd gone so long believing a misconception. Drat, I wish I'd learned better sooner.

[NathanKell]

Sorry, I'm not going to revisit heating/cooling until RealHeat's done...at which point I'll have to *entirely* revisit it.

Comet Tail: It's true for non-pressurized objects, of course; it's just that even non "pressurized" (read: highly pressurized, for pressure-fed engines) tanks are still *somewhat* pressurized, and thus obey pressure vessel rules (and can be only a limited number of shapes, if you care about having decent dry mass fractions).

AFAIK you can't get ZBO on passives alone; you can reduce boiloff a shedload (and get *LOX* boiloff practically near zero) but you can't get LH2 ZBO on passive.

Edited September 5, 2014 by NathanKell

[The Pink Ranger]

༼ 㤠◕_â—• ༽㤠Give 7.4

[NathanKell]

Due to a packaging error, Custom Biomes was not included. It is now. Please re-grab 7.3 if you have already grabbed it.

The Pink Ranger: here you go! Pretty fast, eh?

[The Pink Ranger]

Oh thanks.

[Comet Tail]

I'll have to follow your posts more. Lots of great information

So, I've done some poking around, though, and I can't seem to find much on how to use these thermal fins. I tested them a bit, too, and looked at what it says in the VAB, and I'm a bit confused...

Heat Pump: 50/s

Requirements: Electric Charge: 2F/h

So... it pumps 50 Watts per second? What does the F/h on the requirements mean? It uses 2 electric charge an hour for every *F it removes from the tank? Is that how this works? So how do I go about finding out how many thermal fins I need for X cryo tank to stay at Y temperature, and how much charge this will require?

[NathanKell]

Heh. :]

Right now, KSP doesn't model heat *at all*. It models temperature. So at best that heat pump will...literally "pump out" 50 degrees C from the part it's attached to, at the cost of (it looks like) 0.833 kW.

The 2F appears to be a bug in the display; that's supposed to be setting a format. I'll check.

[Armchair Rocket Scientist]

I can't remember, does the Roll Your Own configs spreadsheet use some sort of improved gimballing mod? I ask because I have a vague memory of using one in previous versions, and I can't seem to get the thrust vectoring working on any of my engines (tip: even without RO, a single 2.5m reaction wheel is not sufficient to control a 400+ ton rocket).

[Kitspace]

How to make the mod work with B9?