Realism Overhaul

[Guest]

MFT should take care of that, actually. It could adopt various different effects for various fuels. Also, exhaust should change appearance much more when exiting atmosphere.

[NathanKell]

That's planned, yes.

At least the bulking up of exhaust, and it changing based on propellants used. Won't have a chance to actually make new FX for a while though.

[mossman]

At the very least someone could edit the particles to have relative velocities to the engine. That and them colliding with the launch pad would be a leap forward.

[Guest]

Even just a static animation of fire filling the blast deflector would be nice.

[Camacha]

I guess a proper particle simulation PhysX style would be pretty neat, but unfortunately that would only work for a part of the KSP players (with Nvidia cards) - if it even works at all in Unity. Time for some decent OpenCL implementation, but I guess that is a bit beyond the scope of KSP modders.

That's planned, yes.

At least the bulking up of exhaust, and it changing based on propellants used. Won't have a chance to actually make new FX for a while though.

Great! Although it is not a major factor gameplay wise, I think it could eventually really add to the proper feeling of a rocket launch.

Edited November 5, 2013 by Camacha

[Guest]

Success! You know what was the problem with Soyuz? All the bloody time I was running with an overweight 3rd stage! And why's that? I forgot to account for the cargo decoupler when setting stage mass. It turned out that it had mass of 1.5, which made the whole rocket weight one ton too much. Strangely, I didn't forgot about this when setting Soyuz ST decoupler mass (it massed 0.5). With this removed, you actually need to shut 3rd stage down early, because the equatorial orbit is so much more effecient. If you burn it to depletion, you might end up with something silly like 200x800 orbit.

Also, Asmi, disregard the PM I sent you. Everything's fine now, though you're still free to play around with the ascent.

So, without much further ado: Soyuz-TMA v.1.0, real scale. The configs are a drop-in replacement for BobCat's configs and require the latest Soviet Pack to work. The Soyuz uses MFT, Ioncross Crew Support, Engine Ignitor, Deadly Reentry, MechJeb2 (looking to switch to kOS in the near future), FAR and KJR.

http://www./download/5n8zad99915c6d1/SovietPack.zip

Also, here are the modified configs for Ioncross. They include basic food and water support. Drop-in replacement, like before. Mind you, this wasn't really tested on a prolonged mission.

http://www./download/98titbm0yihj89c/IoncrossForReal.zip

To launch the Soyuz, set Ap to about 250km and turn end to the same altitude. Turn start should be at 0.5km, with the grade being 35%. Disable autostaging. Jettison boosters when empty, jettison SAS (press "1") and fairings after passing the edge of second thickest atmosphere layer. Just before 2nd stage burnout (Ap should be 200-220km, you could tweak it), disable autopilot and drop the nose to the horizon. Engage 3rd ("2") stage just before MECO. After jettisoning the 2nd stage, remember to stage again to activate 3rd stage gimbal and jettison it's engine fairing. After you pass the Ap (stage still burning), pitch up slightly (5 or so degrees) so that Ap stops to rise. Cut off the 3rd stage when Pe is at the altitude you want it on. After that, stage and deploy the Soyuz ("3"). Congratulations, you made orbit. Fun, wasn't it?

Future plans (in no particular order):

kOS implementation. I kind of hope to write multiple ascent programs for various orbits and maybe rendezvous programs to simulate the Kurs docking system. The real Soyuz works that way.

Turbopump simulation. This Soyuz is slightly overpowered, because I didn't simulate the H2O2 load that drives the turbopump on the real thing nor the N2 for tank pressurization. I'll try to implement that in a future release.

Spacecraft improvements. Currently, Soyuz spacecraft needs a few tweaks, most notably to it's attitude control and electrical systems. I'll get to that eventually, but I need a few additional models for this. Also, it's main engine isn't really throttleable, but instead has 3 modes with different thrusts and Isps. MFT can simulate that, so I'll try to do it.

Soyuz-U2. I'll need to talk to Nathan about that, but since the only difference here is propellant (Synthin vs. RP-1), it should be easy to make the Soyuz configurable as either U or U2. We'll see about that.

Fregat. I wanted to release this version without further delay, so Fregat is out for the time being. It'll be back.

Pre-flight. It takes about 20 seconds to spool up the engines, but I've got mixed feelings about it. It requires a fueling tower anyway, so unless we can simulate that, it's pointless.

Far-future plans:

Other Soyuz variants. Possible, although I'll need to master kOS first, since Igla is a very complex docking system that requires the target to be cooperative, because early Soyuzes had very unbalanced RCS.

Rest of the Soviet pack+MrTheBull's additions. Proton, N1, MIR, Luna... There's a few things I'd like to do. The N1, an overengineered, half-classified beast that it is, will probably take forever, so don't hold your breath dor that. But Proton's possible, along with Luna-17. For Buran and Kliper, I'll probably need Ferram to do anything at all.

Edited November 6, 2013 by Guest

[NathanKell]

Hey awesome!

Yeah, pretty amazing how much difference a ton will make now...

Regarding Synthin, I can add it, sure. Also, regarding fueling towers: MFS makes the launch clamps into fueling pumps.

[AbeS]

where is the craft file Dragon01?

[AbeS]

Awesome Dragon01! I barely made it though in the last second before burnout I got my Pe from negative to 134km, can't wait for v1.1

EDIT: Tried landing it but neither the Orbital Module or Instrument SM decoupled, the pod would have made it back but the crew died because they reached their g-force limit, any thoughts? (also the solar panels didn't break until the ISM blew up)

Edited November 6, 2013 by AbeS

[Guest]

They didn't? Did you use KJR? It's been known to bug decouplers, though I never encountered this. They definitely should've decoupled, did you try right-clicking on them and decoupling from the menu? Also, try placing your Pe at about 9km. This should work for a LKO descent. Oh, and don't forget to decouple the umbilical from the OM, it can obstruct decoupling of both OM and ISM.

[AbeS]

after I decoupled, I right clicked and the option was unavailable, I made a quicksave before though, will try a third time! And yes I'm using KJR and my Pe was at 35km and Ap was like 170km, shouldn't have died I guess.

[Guest]

It was KJR causing this. Talk to Ferram, it's a known bug in his plugin. Oh, and try with a higher Pe then. It should give you lower Gs in exchange for greater heating. I might beef up the heatshield in the next version, but I think it should work.

[asmi]

Yea, I've got the same problem with fairings - they stuck after decoupling, and only detached once I released the spacecraft in orbit.

Oh, and the rocket is very overpowered now - it inserted the Soyuz to 250x1200 km orbit.

[Guest]

What kind of ascent profile did you use? Anyway, I guess I better implement those turbopumps. To think I couldn't get it to orbit with just on more ton... Anyway, the next update will be delayed a bit, I've ran into some minor issues. Maybe tommorow I'll do it, if not, this weekend is the deadline. Most of the stuff is done, but it needs some final tweaks and checks (mostly on how it reenters) and I'm gonna have a busy week. As a compensation of sorts, I'll include Fregat and Progress, the latter with a configurable fuel module. No spoolup or turbopumps yet, but I'll probably change dry mass to account for N2 that pressurizes the tanks (since it's not expanded during flight, I'll count it into dry mass).

[asmi]

What kind of ascent profile did you use?

I throttled down a bit after 0.4M until boosters sep. Gradual gravity turn right after liftoff, up to around 50°, once Ap reached ~180 km went horizontal and stayed that way all the way until 3rd stage cutoff. MJ sucks big time

[NathanKell]

Of interest:

http://www.kepu.dicp.ac.cn/photo/07sl02/Handbook%20of%20Photovoltaic%20Science%20and%20Engineering/10.%20Space%20Solar%20Cells%20and%20Arrays.pdf

Medieval Nerd and I were discussing solar power. He found references that power per m^2 varied from about 50W in the earliest satellites to ~160W today (ATV was 142, and I think the panels were probably finalized in the early 2000s).

I found that PDF which gives approximate kg / m^2 figures. Looks like for the solar cells alone, we're talking up to 0.5 for the early ones and up to 1.0 for recent ones. (Efficiency doubles, so power/kg stays approximately constant).

I'm guessing that the structural materials to hold the cells mass about the same as the cells for a simple panel, and, say, twice that for an extensible one?

I also calculated some areas: It looks like the OX-STAT is approx 0.2m^2, and the extensible ones 1m^2 (0.168 each, minus the little bit of edge).

That means we should get anywhere from 10 to 32W for the OX-STAT, at somewhere between 0.0002 to 0.0004 tons each. For the extendable ones, call it 3x the mass for the regular and 4x for the shielded one, so: 50 to 160W, 0.0015 to 0.003 for the unshielded and 0.002 to 0.004 for the shielded. (Well, maybe 0.0025 to 0.0045? Shielding shouldn't change mass when the panels do)

Note that probe EC usage is pretty much OK: 28W to 50W sounds about right to me. RT2 antenna, if they had a factor of 0.1 applied to their power draw, are probably also OK?

The issue is right now instruments don't draw any power--they should be the major power draw IIRC.

Also, the probe cores themselves are way too heavy of course, at least for the early ones.

Regarding batteries, it looks like in the early-mid sixties it's approx 320,000kW-s [i.e. EC] per ton. (Based off Ranger 5's batteries)

[Autochton]

Regarding batteries, it looks like in the early-mid sixties it's approx 320,000kW-s [i.e. EC] per ton. (Based off Ranger 5's batteries)

*grinds teeth* Okay, so I know engineers (and, really, most people) often use derived units instead of base units, but the physicist in me just has a fit whenever I see 'kilowatt-seconds' or the like. They're called kilojoules! *goes and yells at a cloud or something* *shakes walking cane*

[Guest]

I throttled down a bit after 0.4M until boosters sep. Gradual gravity turn right after liftoff, up to around 50°, once Ap reached ~180 km went horizontal and stayed that way all the way until 3rd stage cutoff. MJ sucks big time

You can fly that with MJ. It's clear that you're running a much more efficient profile than I do. My ascent profile might be poorly optimized, I'll try out yours.

[asmi]

You can fly that with MJ. It's clear that you're running a much more efficient profile than I do. My ascent profile might be poorly optimized, I'll try out yours.

No you can't, because just about avery control system looks after velocity as well, not just altitude like MJ does.

[Guest]

Well, TBH, I don't think you should throttle Soyuz engines in flight. They're only throttleable in order to be able to simulate preflight.

[asmi]

Well, TBH, I don't think you should throttle Soyuz engines in flight. They're only throttleable in order to be able to simulate preflight.

Actually it does towards end of boosters operation to keep acceleration under 4G - this is a part of Russian man-rating requirements.

Soyuz-U rocket puts up to 5G at the end of boosters burn.

[deadshot462]

This reminds me of Orbiter, which was really cool. Gonna give this a try.

[Guest]

Actually it does towards end of boosters operation to keep acceleration under 4G - this is a part of Russian man-rating requirements.

Soyuz-U rocket puts up to 5G at the end of boosters burn.

I'll keep that in mind. BTW, I'm not letting MechJeb do throttle control nor the staging. So I can do that part manually. And I'll program it into kOS when I wrap my head around it.

[asmi]

I'll keep that in mind. BTW, I'm not letting MechJeb do throttle control nor the staging. So I can do that part manually. And I'll program it into kOS when I wrap my head around it.

It's not that I'm against MJ or anything, but it's ascent AP is just too simplistic for real things

As for kOS - I've tried that some time ago, and I've noticed that kOS doesn't play nice with SAS. Without SAS it has a hard time maintaining direction when FAR is installed, which trashes orbital insertion accuracy and can lead to LoM.

Also the whole concept of kOS is more of sequential commands, while in reality most control systems are event based (like "jettison fairing one reached X altitude", "start pitch over once reached Y velocity" and so forth). While this totally could be done, the code is going to be rather large.

[SFJackBauer]

I'll keep that in mind. BTW, I'm not letting MechJeb do throttle control nor the staging. So I can do that part manually. And I'll program it into kOS when I wrap my head around it.

Does not compute... How MJ can be configured to pitch the aircraft through ascent but not touch the throttle?