Modular Fuel System Continued v3.3 (OBSOLETE)

[HoneyFox]

Why am I having flashbacks to that video with the indian guy screaming, "You die, I die, everybody DIE!!! Understand???"

ROFL

Well this might be inevitable when several plugins have lots of interactions...

[HoneyFox]

HoneyFox: Weird. Delete your MFT and ST folders and (re)grab the latest pre (should be v4p10 and v8p1 respectively), http://forum.kerbalspaceprogram.com/threads/52780-Modular-Fuel-System-Continued-v3-3?p=865815&viewfull=1#post865815

Sounds like dual modules on the tanks, from an older prerelease of Stretchy (like, the original v8p0 I posted).

I'll try tonight. Hope that it's because of using out-dated versions.

[Spartan125]

Actually, it wasn't - you're thinking of the J-2S, a limited-run experimental engine which only had 6 units produced. The production version J-2 wasn't throttleable.

http://en.wikipedia.org/wiki/Rocketdyne_J-2

He was referring to the descent propulsion system on the lunar module.

[evilphish]

evilphish: I asked for output_log.txt. It's in ksp/KSP_Data. ksp.log doesn't really tell us anything. Also, please zip before you upload.

My bad, here is a zip of the two files, one with and one without MF.

http://phishtank.de/~evilphish/output_logs.zip

Regarding your inquiry further down, I am not using stretchy tanks.

I use the following mods:

Kethane

KSPInterstellar

InfernalRobotics

PreciseNode

RemoteTech2

SCANsat

TextureCompressor

EditorExtensions

EnhancedNavBall

B9Aerospace

KWRocketry

DockingPortAlignment

Engineer

FAR

KAS

KerbalAlarmClock

KJR

AlternateResourcePanel

MJ2

ProceduralFairings

TacFuelBalancer

ModularFuels (the basic one without real fuels)

I also spent some time (about a day in fact) removing mods, restarting KSP, re-adding them to find the mod that actually does this. No matter which mods I have installed, as soon as I add ModularFuels and I go to space towards the Mün things go haywire. As soon as I remove it I can play for hours without any hicup. I can reliably reproduce the erratic behaviour and as soon as I load up KSP, load the save and enter the ship that has just entered Müns SOI things go ballistic.

As I am still very early in the TechTree, the vessel that started acting weird did not have a lot of mod parts on it. Most was stock appart from a KW Fuel Tank and the odd antenna. Also I did not alter the fuel inside the tanks using MF at that point. I used the parts as they come straight off the part list.

If it is of any help I can offer to show the behaviour using twitch.tv or I can zip up my whole KSP installation including the save and put it on a secure part of my webspace for you to download (if that is ok with SQUAD, I assume since you have bought it and I have bought it this should be doable in the name of bugfixing).

[HoneyFox]

HoneyFox: Weird. Delete your MFT and ST folders and (re)grab the latest pre (should be v4p10 and v8p1 respectively), http://forum.kerbalspaceprogram.com/threads/52780-Modular-Fuel-System-Continued-v3-3?p=865815&viewfull=1#post865815

Sounds like dual modules on the tanks, from an older prerelease of Stretchy (like, the original v8p0 I posted).

OK by checking the stack-trace of the error and the source code of ST plugin, I guess I found the issue.

In the tech.cfg of the ST plugin, there is a key-value pair named "name = Default", which will throw an exception when updateMaxRFactor() is running and calls "float r = float.Parse(value.value);".

By removing that "name = Default" line in the tech.cfg, the issue got fixed.

[eggrobin]

You can even do it with a MM patch

@PART[*]:HAS[@MODULE[StretchyTanks]]
{
	@MODULE[StretchyTanks]
	{
		%keyWhatever = desiredKey
		// and the others
	}
}
@PART[*]:HAS[@MODULE[StretchyConicTank]]
{
	@MODULE[StretchyConicTank]
	{
		%keyWhatever = desiredKey
		// and the others
	}
}

Thanks! You might want to add this snippet to the documentation, lest somebody else complain you can't rebind keys.

[Deredere]

evilphish: WEIRD. Ok, does anyone else get this? Can I have output_log.txt?

Are you sure it's MFS's fault?

I am getting the same thing as evilphish, yes. I'm not at all certain it's MFS's fault, but given EF's research and the fact it involves strange mass distributions and COM rocketing off into space sometimes, it seems likely.

It seems to happen whenever a craft is re-loaded from a quicksave or a revert to launch. Going back into the VAB fixes the launch issue but a quicksaved/loaded craft seems boned.

Edited December 26, 2013 by Deredere

[Starwaster]

I am getting the same thing as evilphish, yes. I'm not at all certain it's MFS's fault, but given EF's research and the fact it involves strange mass distributions and COM rocketing off into space sometimes, it seems likely.

It seems to happen whenever a craft is re-loaded from a quicksave or a revert to launch. Going back into the VAB fixes the launch issue but a quicksaved/loaded craft seems boned.

You might fix it by editing the vessel in the quick save and swapping module orders

[Deredere]

You might fix it by editing the vessel in the quick save and swapping module orders

Unfortunately I'm not very familiar with the KSP file system; there's a lot of modules here and I'm not sure what most of it means. Also not sure what to swap them with... the .craft file?

But I was able to rescue this mun lander by decoupling the offending part, which was a fuel drop tank. Lost some DV but Jeb will make it home.

[Starwaster]

Unfortunately I'm not very familiar with the KSP file system; there's a lot of modules here and I'm not sure what most of it means. Also not sure what to swap them with... the .craft file?

But I was able to rescue this mun lander by decoupling the offending part, which was a fuel drop tank. Lost some DV but Jeb will make it home.

See, Jeb knows what to do.

But no, regarding editing the save files, it's a little trickier than the craft files because modules there have extra persistent data stored in them and it can be a pain figuring them out. (MechJeb or MFT modules can contain pages worth of data)

What you would do is cut one of the modules (StretchTank for example) and then paste it back in so that it has swapped order with MFT.

It is a pain to do though

[ferram4]

Just to chime in, it seems like the MFS and StretchySRB pre-releases are fine, except for the solids-of-the-same-type-all-making-the-same-thrust issue. I tried making a Delta 7925 and the PAM solid stage ended up picking up the characteristics of the 3 air-lit vacuum-optimized SRBs. So instead of making 67 kN (as it said in the editor) it made ~450 kN and crushed the probe. So that bug's still a thing, but otherwise everything looks fine.

[NathanKell]

Thanks for the logs. I'll go through them tomorrow (I hope).

Ferram: Darnit, I thought I fixed that bug! Tested and got no repeats of it since I switched to invoking the method (per your help) rather than sendmessaging.

.craft?

Glad the rest seems solid enough.

[ferram4]

Okay, this Delta II 7295 is the problem craft. In the editor, the upper stage SRB says that it is set to make 67 kN, but in the flight scene it makes 450 kN.

Requires AEIS, KW, PFairings, StretchySRB, MFS and using SFJackBauer's Real Engines, with a minor change to the real_engines_rescale.cfg to make the AJ10-118K look better on the Delta K; just look for the KW1mengineVestaVR1 entry and replace it with this:

@PART[KW1mengineVestaVR1]
{
    %title = AJ10-118K
    %manufacturer = Aerojet
    description = Upper stage, pressure-fed engine of the Delta II vehicle, burns hypergolic propellants and is optimized for vacuum operation.
    %attachRules = 1,1,1,0,0
    !MODEL {}
    !MODEL {}
    MODEL
    {
        model = KWRocketry/Parts/Engines/1mVestaVR1/KW_1mEngine_VestaVR1_M
        scale = 0.5, 0.5, 0.5
    }
    %rescaleFactor = 2
    %attachRules = 1,1,1,0,0
    %node_stack_top = 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 2
}

That just controls looks, so you shouldn't have to do that to cause the bug, but just in-case. If you're testing in an install with RemoteTech you'll need to add an antenna to it because I didn't add one yet (haven't installed RemoteTech yet).

Just get it up to a point where you can ditch everything below the 3rd stage and fire it. Stupid amounts of thrust from that tiny thing even though it shouldn't be able to produce it.

Oh, an unrelated Stretchy SRB request:

Could you try detecting whether there is a part attached to the stretchy SRB's nozzle and change the nozzle size based on that? I understand not making them larger than the SRB body if you're dealing with upper stage SRMs, but for the vacuum-optimized SRBs attached radially it would be cool to have the nozzles have equal throat areas, not equal exit areas. Should be easier to do now that you're not hiding an RT-10 in each one.

[jrandom]

Not sure if this is fixable on the MFS side, but if you have two different kind of RCS thrusters, say one monopropellant and one hydrazine, only the monopropellant ones will fire. I discovered this when using the Gemini pod (built-in monopropellant RCS) on top of a custom service module with hydrazine RCS.

[Zander]

Hey,

I already asked this in the RO thread but that one seems to be a little inactive now so i'll ask it here

anyways

I've read through the last few pages of the mods threads and I was a little confused on the most recent stable experimental versions for MFS, RO, and stretchy for use together. Can someone point me in the right direction? I want them to be the realistic ones (real engines, real mass).

thanks

[AbeS]

Here they are:

MFT v4 pre10: https://www.dropbox.com/s/ch8sw3iqtzv9dt7/ModularFuelTanks_v4_pre10.zip

OR

RF v4 pre10: https://www.dropbox.com/s/g49soc45iraz5ji/RealFuels_v4_pre10.zip

unchanged since pre9:

RO: https://www.dropbox.com/s/yt5f90v47cqpwfx/RealismOverhaul_v3pre2.zip

Note that in RF/RO engine throttles are limited (by type, techlevel, and the individual engine). Some can't throttle at all, some can throttle deeply; most U and L don't get any throttling until TL4 or so. As before, note that stock RCS is (a) modular and ( defaults to hydrazine, which is *not* ALL_VESSEL.

Link to stretchy pre for use with these: https://www.dropbox.com/s/rcemfqzq1ci4tnw/StretchySRB_v8pre1.zip

(posted in both forums in case anyone else needs it)

[Zander]

Awesome thanks.

Question for Nathan, I heard you were planning on making some engines only able to ignite once. But are you planning to implement the need for the fuel to be settled in the bottom of the tank for ignition (either via gravity or ullage motors). Because if you are that would be so awesome. Maybe like at least .1 gs down along the ships long axis are needed or the engine won't work?

[DasBananenbrot]

Check out Engine Ignitor Mod by HoneyFox this covers limited ignitions and the need for the fuel to be at the bottom. That is what he wants to integrate in RO as far as I know. Can't link it because I'm on my phone but shouldn't be hard to find

[Buddy431]

Maybe I'm just missing something, but with the latest install of real fuels, I can change the fuel type in the tank, but all of the engines (just stock engines used) look like they just want to use liquid fuel/oxidizer. Did the way of changing engine fuel type change?

[jrandom]

Maybe I'm just missing something, but with the latest install of real fuels, I can change the fuel type in the tank, but all of the engines (just stock engines used) look like they just want to use liquid fuel/oxidizer. Did the way of changing engine fuel type change?

The pre-releases of MFS/RF don't come with an engine config. Grab RftSEngines.cfg from Realism Overhaul or use the RealEngines config file.

[TomatoSoup]

I can't seem to get this to work. I grab a stock fuel tank, go to the action group manager, click the tank, and nothing happens. Incidentally, what's the word on changing that over to use the new tweakables?

Edited December 28, 2013 by TomatoSoup

[Buddy431]

When I'm using just modular fuels (even with no other mods), the base masses of the fuel tanks are too high by a factor of about 4.079. Looking in the Tank Types.cfg appears to give the appropriate mass (0.000625*volume), but in game they are not this value.

[Starwaster]

I can't seem to get this to work. I grab a stock fuel tank, go to the action group manager, click the tank, and nothing happens. Incidentally, what's the word on changing that over to use the new tweakables?

MFT 3.3 does not work with KSP 0.23

If you have KSP 0.23 then grab the pre-release. The link was posted just a page ago

(actually see below for another link)

http://forum.kerbalspaceprogram.com/threads/52780-Modular-Fuel-System-Continued-v3-3/page76?p=865815#post865815

You can currently use the tweakable system so far as changing fuel amount. but not changing fuel type

Edited December 28, 2013 by Starwaster

[TomatoSoup]

MFT 3.3 does not work with KSP 0.23

If you have KSP 0.23 then grab the pre-release. The link was posted just a page ago

(actually see below for another link)

http://forum.kerbalspaceprogram.com/threads/52780-Modular-Fuel-System-Continued-v3-3/page76?p=865815#post865815

You can currently use the tweakable system so far as changing fuel amount. but not changing fuel type

Oh hell, really? I saw that post for prerelease stuff but I didn't realize that 3.3 wasn't compatible yet! Thanks for straightening me out.

However, I must now list a bug where the Toroidal and Spherical tanks (and only those, as the KW Rocketry and Stock tanks behave correctly) show way too little fuel. On the largest toroidal tank I can cram in only ~47 units of fuel and on a medium spherical tank I can fit only nine units. Is this a Prerelease bug or is something else afoot?

Edited December 28, 2013 by TomatoSoup

[NathanKell]

Buddy431, thanks for the catch. It's supposed to do the multiplication only if the useRealisticMass line is both present and false. However it now defaults to doing the multiplication. Will fix.