Realism Overhaul

[NathanKell]

I knocked it down a lot for the next release, but I'm happy to add that.

[Guest]

Why don't you just take BobCat's Soyuz-U/FG and change parameters to real ones from Wiki or some other source? The beauty of realism is that you don't need made-up number - use real ones

Well, I did that, using data from Astronautix. I couldn't get the ratios though, and MFT ones resulted in an overweight rocket. Gonna try a more efficient ascent profile, but I'm afraid I'll have to do some calculating to get mass in line.

EDIT: I've found a nice website with rocket specifications, this includes mixture ratios of some engines. Some interesting stuff can be found there. For example, Kerolox rockets usually have mixture ratio around 2.5, contrary to MFT's 1.5 (0.61 units of LOX per 0.39 units of LF), which means the latter are running ridiculously fuel-rich.

Edited October 27, 2013 by Guest

[asmi]

Well, I did that, using data from Astronautix. I couldn't get the ratios though, and MFT ones resulted in an overweight rocket. Gonna try a more efficient ascent profile, but I'm afraid I'll have to do some calculating to get mass in line.

EDIT: I've found a nice website with rocket specifications, this includes mixture ratios of some engines. Some interesting stuff can be found there. For example, Kerolox rockets usually have mixture ratio around 2.5, contrary to MFT's 1.5 (0.61 units of LOX per 0.39 units of LF), which means the latter are running ridiculously fuel-rich.

Remember that sometimes mixture ratio is defined as mass ratio, and other times - as volume ratio. These are different things, so be careful with that.

[Guest]

I know. I didn't find any info about which one is that, so I'll go with volume ratio and switch to mass if it doesn't get off the pad or makes it to the Mun with those settings.

[Lucchese]

My Orbital Research Stationin high orbit:

This mod is awesome!

Feel free to use these to showcase your mod if you like them ^^

Edited October 27, 2013 by Lucchese

[NathanKell]

Dragon01, ratio is almost always given as mass ratio. KSP is the loner, in requiring volume ratios.

If you look at the Volumes page I use for MFS here: https://docs.google.com/spreadsheet/ccc?key=0AvHneDAy4k99dDZYOHJraGpQc2VuMUxxU0FjSUp6NXc&usp=drive_web#gid=7

you'll see the conversions between mass ratio and volume ratio.

For Kerolox MFS is set to use a mass ratio of 2.2:1 for kerolox, which converts to a volume ratio of .61/.39

That's the ratio S-IB used; S-V used a bit more oxidizer rich. I'm thinking of changing it to 2.5 all-round, since that seems more like what modern engines use. Ideally it would differ per engine (which MFS supports) but that takes a lot of work.

I've finished Gemini-Titan II GLV, but interestingly despite having spot-on masses, Isps, and mass ratios, it's running out of fuel slightly before a stable orbit.

Lucchese, thanks--will do!

[p3asant]

My Orbital Research Stationin high orbit:

Awesome! How many lauches did that take?

[Guest]

Oddly enough, when I used the ratio given for a Soyuz ISM (dunno if volume or mass), it gave me the proper fuel weight. So maybe those were volume ratios afterall. I'll try it out and see.

I've finished Gemini-Titan II GLV, but interestingly despite having spot-on masses, Isps, and mass ratios, it's running out of fuel slightly before a stable orbit.

Maybe your ascent profile is sub-optimal. I know that with a low-tech craft like Gemini, you absolutely have to nail it. Does circularizing with the Gemini main engines help? If it's not a big error, that might the correct behavior.

Edited October 27, 2013 by Guest

[Starwaster]

Dragon01, ratio is almost always given as mass ratio. KSP is the loner, in requiring volume ratios.

If you look at the Volumes page I use for MFS here: https://docs.google.com/spreadsheet/ccc?key=0AvHneDAy4k99dDZYOHJraGpQc2VuMUxxU0FjSUp6NXc&usp=drive_web#gid=7

you'll see the conversions between mass ratio and volume ratio.

For Kerolox MFS is set to use a mass ratio of 2.2:1 for kerolox, which converts to a volume ratio of .61/.39

That's the ratio S-IB used; S-V used a bit more oxidizer rich. I'm thinking of changing it to 2.5 all-round, since that seems more like what modern engines use. Ideally it would differ per engine (which MFS supports) but that takes a lot of work.

I've finished Gemini-Titan II GLV, but interestingly despite having spot-on masses, Isps, and mass ratios, it's running out of fuel slightly before a stable orbit.

Lucchese, thanks--will do!

I was getting that a lot when I was trying to get in LKO for Earth-Sized-Kerbin. (ESK?)

I had only just got my second rocket in stable orbit when that awful clicking sound started coming from my drive. You know the rest.

[Guest]

That's the ratio S-IB used; S-V used a bit more oxidizer rich. I'm thinking of changing it to 2.5 all-round, since that seems more like what modern engines use. Ideally it would differ per engine (which MFS supports) but that takes a lot of work.

What kind of volume ratio does 2.5 mass ratio give? It seems that all Soyuz engines use this. It might be useful to have a more straightforward converter for those things.

[NathanKell]

Here is the conversion formula.

For a given mass ratio Rm (mass of oxidizer to mass of fuel), and given Fuel density Df and Oxidizer density Do:

Volume percent fuel: (1/Df) / ((1/Df)+(Rm/Do))

Volume percent Oxidizer: (Rm/Df) / ((1/Df)+(Rm/Do))

For RP-1 and Liquid oxygen, 2.5:1 yields .36 fuel, .64 oxidizer

[SFJackBauer]

Something regarding the launch clamps - I dont know if it can help, but they behave perfectly if you launch from the runway instead of the launch pad.

[NathanKell]

(I should point out that for KSP as long as the Isp is accurate for the engine, all that matters is that the total fuel mass is correct and the ratio of the fuel volumes in the tank matches the fuel ratios in the engines. KSP doesn't actually care about propellant ratios, whereas in reality that would help determine Isp.)

[Starwaster]

Something regarding the launch clamps - I dont know if it can help, but they behave perfectly if you launch from the runway instead of the launch pad.

Show me 100 consecutive successful launches from the runway without even a single UPD and I'll think you're on to something.

Otherwise I'm calling it coincidence

[SFJackBauer]

Show me 100 consecutive successful launches from the runway without even a single UPD and I'll think you're on to something.

Otherwise I'm calling it coincidence

I have tried 20 times on launch pad and 20 times on runway - none of the times in the runway the launch clamps detached. 15 of 20 times on launch pad it detached.

And this with three KW heavy struts between each clamp and the craft.

[Guest]

(I should point out that for KSP as long as the Isp is accurate for the engine, all that matters is that the total fuel mass is correct and the ratio of the fuel volumes in the tank matches the fuel ratios in the engines. KSP doesn't actually care about propellant ratios, whereas in reality that would help determine Isp.)

Actually, LOX is much heavier per unit than RP-1, which means that if you've got a higher LOX percentage, you get less fuel units overall out of a particular mass, which means less dV.

EDIT: Also, Ferram, could you extend your FAR control surface assigner to RCS somehow? The real Soyuz turned out to have a rather complex, double RCS system (fine and coarse), which can't be balanced with normal KSP settings. I'm running into issues trying to keep the RCS balanced on a Soyuz with real masses for each module and non-everywhere RCS fuel, so it's possible that a closer replication of the system would be needed.

Edited October 27, 2013 by Guest

[Starwaster]

I have tried 20 times on launch pad and 20 times on runway - none of the times in the runway the launch clamps detached. 15 of 20 times on launch pad it detached.

And this with three KW heavy struts between each clamp and the craft.

Sounds great! 80 more times to go

[NathanKell]

OK, got Gemini working.

I had like 1-200m/s extra from an equatorial launch, now to try the 31 degree launch.

Note that the vertical dimension hasn't been touched, and the FASA Gemini / Titan II GLV is only about 85-90% as tall as it should be given it's diameter.

(Rescale factor was a uniform 1.22, or 2.5->3.05)

The base is in lieu of launch clamps. Use a giant pFairings adapter--you can make it as wide as you want, almost.

In orbit. Good gracious the OAMS has low thrust! 20 minutes to exhaust the fuel supply when thrusting forwards, for just 222m/s!

Splashdown.

[Guest]

Vertical-only scaling KSP parts is though, but doable. If you use MODEL calls, you can scale it independently, but this plays merry hell with the rescale factor, meaning it's hard to get the exact total size you want.

[NathanKell]

Yeah, I was planning to do that next. Didn't realize they were off-scale until I used the ruler to check and came up ~4m short.

It's actually not hard at all as long as all your nodes are on the centerline; you can just use the vertical scale factor as scale = (the scale outside MODEL{}), and then the nodes will adjust. I always use rescaleFactor = 1 when using MODEL nodes, the interaction is buggy.

Interestingly, the first stage burns 10s longer than it should. Maybe the real TIIGLV doesn't burn all its first-stage propellants?

[Starwaster]

Yeah, I was planning to do that next. Didn't realize they were off-scale until I used the ruler to check and came up ~4m short.

It's actually not hard at all as long as all your nodes are on the centerline; you can just use the vertical scale factor as scale = (the scale outside MODEL{}), and then the nodes will adjust. I always use rescaleFactor = 1 when using MODEL nodes, the interaction is buggy.

Interestingly, the first stage burns 10s longer than it should. Maybe the real TIIGLV doesn't burn all its first-stage propellants?

Are you aware that scaling isn't really linear when using MODEL nodes? There's a thread around that talks about it.

[DresCroffgrin]

Yeah, I was planning to do that next. Didn't realize they were off-scale until I used the ruler to check and came up ~4m short.

It's actually not hard at all as long as all your nodes are on the centerline; you can just use the vertical scale factor as scale = (the scale outside MODEL{}), and then the nodes will adjust. I always use rescaleFactor = 1 when using MODEL nodes, the interaction is buggy.

Interestingly, the first stage burns 10s longer than it should. Maybe the real TIIGLV doesn't burn all its first-stage propellants?

If I recall correctly, it is standard procedure to put extra fuel in all launch vehicles, to be used for recovering from malfunctions/other unexpected events.

[NathanKell]

Actually, LOX is much heavier per unit than RP-1, which means that if you've got a higher LOX percentage, you get less fuel units overall out of a particular mass, which means less dV.

Dragon, sorry, missed this when you posted (top of page). Actually, it works out, because KSP cares only about the ratio of the numbers given in PROPELLANT blocks, not the actual numbers, and will dray however much total mass of fuel it needs to (given Isp) to produce the thrust you want. So if you, say, had a ratio of 9 LOX 1 RP-1, it wouldn't draw any more total mass; just more of the, say. 5kgs of mass it needed to draw that tick, would be LOX (it certanly wouldn't draw 9 units of LOX!)

DresCroffgrin: yeah, but that shouldn't change the total burn time of the lower stage; the lower stage will always burn the maximum amount, the upper stage just might be cutoff early (when desired apogee achieved). All the sources I've found for the Titan 2-1 say 139s burn time, and I'm getting 148s.

I'm guessing it's propellant residuals, since the upper stage is 4 seconds too long too.

[Guest]

Perhaps the flameout threshold is higher for the real Titan. The last bits of propellant might not be able to be properly burned in the engine and just get vented. Remember, it's an old, low-tech rocket.

BTW, could you post some kind of "neutral" config for the MFT customizer? I've found that the thrust corrector doesn't kick in if MFT modules are not present, but I don't want to allow any customization for the tanks, since it's pointless. I don't want any adjustments, either, because the rocket is already realistic. The only thing I want it to do is to set the 2nd stage as vac-rated and 1st one+boosters as sl-rated.

Also, how do tech levels work? Would it be possible to change the engines and tanks to emulate Vostok/Voskhod/Molniya/Soyuz-2 in addition to Soyuz-U? The early variants had slightly worse stats, while Soyuz-2 is somewhat better. That could come in handy for various things, the R-7 went through a lot of uprates and modifications.

[NathanKell]

Yeah, that must be it.

You don't need a tank for thrust correction, only a modular engine. (All your RESOURCE blocks can stay as they are.) Here's what I'm using for the LR-91. Just set thrust always to Vac thrust, and MFS will properly scale down (by Isp ratio) to SL thrust. With type/TL customization disabled (as it is without any techLevel key/value pairs etc) no values below will be overwritten.

MODULE
{
	name = ModuleEngines
	thrustVectorTransformName = thrustTransform
	exhaustDamage = True
	ignitionThreshold = 0.1
	minThrust = 444.8
	maxThrust = 444.8
	heatProduction = 100
	fxOffset = 0, 0, 0.51
	PROPELLANT
	{
		name = Aerozine
		ratio = 0.47
		DrawGauge = True
	}
	PROPELLANT
	{
		name = N2O4
		ratio = 0.53
	}
	atmosphereCurve
	{
	 key = 0 316
	 key = 1 160
	}

}
MODULE
{
	name = ModuleEngineConfigs
	configuration = Aerozine+N2O4
	modded = false
	CONFIG
	{
		name = Aerozine+N2O4
		minThrust = 444.8
		maxThrust = 444.8
		heatProduction = 100

		PROPELLANT
		{
			name = Aerozine
			ratio = 0.47
			DrawGauge = True
		}
		PROPELLANT
		{
			name = N2O4
			ratio = 0.53
		}
		atmosphereCurve
		{
			key = 0 316
			key = 1 160
		}
		IspSL = 1.0
		IspV = 1.0
	}
}

Note minThrust = maxThrust, because it isn't throttleable.

Just replace the thrusts, propellant names and ratios, and the Isps with the correct values in both places. That will prevent any customization. With that config, tech-level will be permanently -1 and no techlevel customization will occur.

(Yeah, I finally added Aerozine. Note that UDMH is already in, as MonoPropellant (the density is 99% identical) so you can use NTO/MP for all those Russian NTO/UDMH engnies)

As to tech levels...right now I draw from a universal set of tech level TWRs and Isps. But I could easily add the ability for local override, so you could have custom tech-level progression per engine. Let me add the appropriate override to MFSC.

And finally, going back to bipropellant RCS (I had to make Gemini's engine use monopropellant, for it to draw from the same fuel reserve as the thrusters, since the OAMS is just 16 NTO/MMS thrusters, it's not like there's a separate engine and fuel supply).

I've been thinking: maybe I can derive an RCS class from ModuleRCS and just override the resource use functions? Hopefully they'll still function as RCS, and if m4v makes a two-line change to the Build Aid, they'll be recognized by Build Aid.

Edited October 28, 2013 by NathanKell