Project Sigma rocket

[T.Neo]

I'm not sure about the SSME exhaust that you have here, from the pictures I've seen. The SSME has relatively minimal visible exhaust, and as far as I know does not emit any visible smoke or mist (of course, it's pretty difficult to tell with those boosters firing as well ), and late in ascent they don't have any visible exhaust plume at all- the visible sign of engine operation is a bright blue glow inside the engine bell.

There simply isn't anything for exhaust products to react with up there. The same sort of thing is clear in Apollo videos too; the famous video of first stage seperation shows no visible exhaust from the J-2 engines on the second stage while the video of S-IVB seperation shows (after engine startup) a visual effect which is quite similar to that seen in Shuttle videos: a bright blue glow in the engine bell.

The RS-68 emits a very spectacular flame early on during launch (presumably from unburnt fuel reacting in the lower atmosphere). That thing practically sets itself on fire during startup!

But this flame also diminishes as the vehicle climbs into the upper atmosphere.

Of course in Orbiter, rocket engines regularly emit flames of all shapes and sizes... which is not necessarily adherant to reality. :shifty:

But it differs for different propellants and different engines. Solid motors, for example, will likely utter just about as much visible flame and smoke in a vacuum as they would in the lower atmosphere.

 

[N_Molson]

Yes, I know all that, and that's precisely why I want to try to improve this.

The problem is that Orbiter offers limited options for exhausts rendering, and that if you want to keep realistic, you have to make them change in function of the external pressure, and switch between 2D textures / particlestreams in a smooth way.

For the vacuum operation, I wonder since a few weeks if a blue semi-transparent emissive mesh inside the nozzle bell could simulate the glow of the exhaust. Will give it a try. With, again, the constraint to make it visible / invisible depending of the engine state.

Also, I think that the exhaust lights should be defined as spotlights, not pointlights. Except for, let's say, the first 100 seconds of the ascent, where, with the SRB releasing significant quantity of smoke, aluminium particles and so, that dust cloud reflects a part of the light.

Like here, we see clearly that the SRB flame lights the rear of the Orbiter :

After that, except the three SSME, the Shuttle becomes a "dark" object. This is clear from the tank cam, where you see the orange glow disappear after separation. Still, a "white star" exhaust could be appropriated for the SSME, it produces a comparable effect on the video.

It seems that the violet/blue flames of the SSME are related to ambient pressure and quickly fade.

---------- Post added at 12:55 PM ---------- Previous post was at 09:26 AM ----------

So, maybe something like that outside the dense atmosphere :

Here I supressed the local pointlight, my plan is to make a spotlight casting light backwards from inside the nozzle (usefulness : mostly at boosters separation), else there is little to illuminate behind the rocket)



It will avoid having the rocket body illuminated by the SSME after booster separation, which is hard to believe, especially in space where there is nothing to diffuse light...

For the texture, the default Atlantis one is quite realistic, but deserves some "cleaning", as it have some ugly "edges". My plan would be to make it fade as the pressure decreases (I could use a dummy thruster for that effect). In the same time, a "white star" glow effect could grow at the exit of the nozzle (it actually shows on the videos, not sure if it would be observable from space though, might be an artefact caused by viewing the Shuttle through the atmosphere).

 

[T.Neo]

I don't think the internal engine glow should "phase in" as the vehicle climbs; this glow inside the engine bell is present from launch:

I think the "white star" effect has more to do with camera characteristics and comparison with a dark background, than actual brightness of the engine. In a vacuum it looks pretty similar, with the same bright whitish glow in the engine bell:

That glow just looks so cool! :thumbup:

 

[N_Molson]

This is a picture from a SRB (what else ?) ? Where did you find it ? Very interesting, I wasn't aware there were pictures of the Shuttle flying (with SSME on) in space.

The glow will be present as soon as the engine ignites. Before that I'd like to add a red-orange exhaust that seems to last 1-2 seconds before the "blue-ish glow" stabilizes.

 

[T.Neo]

This is a picture from a SRB (what else ?) ? Where did you find it ? Very interesting, I wasn't aware there were pictures of the Shuttle flying (with SSME on) in space.

I found the image on Wikipedia (see image page here). The description on that page states:

Discovery and her external fuel tank as she climbs into orbit 3 seconds after solid rocket booster separation on STS-121, July 4, 2006. Photo from video from the right aft SRB camera.

 

[N_Molson]

OK, I managed to make the "diamond exhaust" we can see at liftoff vanish as atmospheric pressure drops with altitude.

Here, the rocket is in vacuum, so no exhaust and only the glow. I added a condition so that when the SSME is off, the glow mesh is hidden, and when the SSME is on, the glow mesh shows up.

The glow mesh, in fact, is just a scaled-down copy of the nozzle, with an emissive semi-transparent material.



OK, now, let's work further on the ignition itself...

 

[Scruce]

OK, I managed to make the "diamond exhaust" we can see at liftoff vanish as the atmospheric pressure drops with altitude.

What altitude does the exhaust cut off completely at?

 

[N_Molson]

Roughly 20 kilometers, I could easily add a multplier/divider to change this, as I don't know how realistic it is.

So, in the ignition sequence there are :

- igniters sparks.
- vapor that comes from the pad (sound suppression system) => ParticleStream
- red/orange flames that blast the vapor (mostly oxygen burning, I guess) => ParticleStream or maybe exhaust texture, have to try.
- blue-ish glow (equilibrium of the LOX/LH2 burn ratio, SSME get their nominal 67% idle thrust). => glow mesh + exhaust

That one is interesting :

Added a ParticleStream for the "orange" phase of the ignition that lasts 1 second, looks rather convincing. OK, from the video 2 seconds is more realistic. Well, 1.5 seconds.

OK, it works, I put a strong spotlight into the nozzle and adjusted the angle so that it doesnt illuminate it. The vapor stuff will require a launchpad .dll that I have not started. I'll apply the same treatement to the second stage, no more light blazing through the nozzle.

 

[N_Molson]

Again hours of work on the sounds today, quite unrewarding for now.

Well, the "normal" soundtrack (for the first stage) is over and works properly. Now I'm focusing on "alternate tracks" that have to trigger if there is a failure. Finding videos where you have the right stuff, extracting the track from them, working it and cutting the interesting bits on Audacity, converting it, that takes a lot of time. I just mounted a "last second abort track" by mixing sound extracts of 2 videos. Doesn't sound that unnatural once in Orbiter.

Other than that, I think that the first stage is nearly finished... A few HUD displays bugs to fix.

 

[DaveS]

Roughly 20 kilometers, I could easily add a multplier/divider to change this, as I don't know how realistic it is.

So, in the ignition sequence there are :

- igniters sparks.

The sparks are just the Hydrogen Burn Off Igniters(HBOIs). They don't actually ignite the SSMEs, they just burn off any residual hydrogen gas that may be present.

 

[N_Molson]

Started the developpement of a third stage, for GTO or interplanetary missions. Basically, it will be a big Centaur, a little over 60 tons.

 

[T.Neo]

Shouldn't you add more engines? Maybe the extra thrust is not that necessary for an upper stage, after all.

What engines are those supposed to be?

Have you considered the similarity with the ACES (which is also effectively a larger Centaur)?

What payload to GTO would you expect with the Sigma rocket and this upper stage?

 

[N_Molson]

I know where I'm going.

Well, more or less

Edit : if someone has the dimensions (max & min diameter, length) of the RL-10-4-2 engine nozzle, I'm (very) interested. Can't find that on Google.

Edit 2 : That works. The key is a "diving trajectory", much like the Ariane 5. The 3-stage version should be able to deliver 15-20 tons to GTO (including the 28° orbital plane correction). And probably around 15 tons to any interplanetary destination. The rocket looks a bit strange tough. Not sure about the aerodynamics/structural integrity. Will post pics tomorrow.

 

[kuddel]

...if someone has the dimensions (max & min diameter, length) of the RL-10-4-2 engine nozzle, I'm (very) interested. Can't find that on Google.

Maybe this @ Astronautix can help
(I don't know how reliable the information is, but at least it is some information )

/Kuddel

 

[N_Molson]

Yes, I looked at this. I guess that the 1.53 m diameter is the exit of the nozzle, seems correct in comparison of the scientist.

 

[kuddel]

Hi again,

does the 'Area Ratio' of 84 result in a min. radius of 0.083468m[1] ?
Or can't I use the 'Area Ratio' for that calculation[2]?
/Kuddel

[1] Diameter: 0.1669m
[2] My guess : r2 := sqr(r2^2/ratio) ; // with r2 := dia2/2 and ratio := 84
Don't quote me on that Unless you found out it's correct :lol:

 

[N_Molson]

Actually, I think you are not far from the truth. Good idea. :thumbup:

 

[T.Neo]

Ah, modelling rocket engines. It would be nice if there were actually collections of decent images to work from.

I'm still battling with the RS-68... :dry:

 

[N_Molson]

Yup, and the worst part is when it comes to model the pumps, hydrolics and plumbing at the top. :compbash2:

There is a technique I used for some engines, to get the correct nozzle curve :

1) Find an image where you have an almost perfect side view of the nozzle. Schematics are of course the best source.
2) Print it. No matter the resolution, full sheet.
3) With a good ol' paper pen, trace a "median line" that is the Z axis.
4) Re-draw (with hand) the curve of one side.
5) Take a good dozen of points on that curve, more if you have the patience.
6) With a rule, you mesure the distance from these points to the Z axis.
7) Get the size of the nozzle at exit on the paper, then compare it to the real number.
8) Using proportionality, you convert (back) the mesures taken on paper to SI units.
9) Enter the stuff in Animator, if you were precise enough the result can be quite realistic.

Edit : actually, I did a RS-68 nozzle that way for a DG-Launcher. If interested, you can get the Anim8tor file...

That takes time, I admit it. :lol:

Well, working on the RL-10-4-2, still a long way to go :



---------- Post added at 10:25 PM ---------- Previous post was at 03:49 PM ----------

Several pics of the 3-staged version (D3D7) :













Work in progress, of course.

Edit : Just sent 15 tons of payload into GTO. 2 burns for the third stage : the first immediately after second stage separation (orbit circularization and GTO injection), the second 12 hours after (resolution of the 28° Inc at the Ascending Node, very fuel efficient at that precise moment). Then the payload(s) can be realeased, they make an Earth flyby at 300 km or so, and then go to the GTO apoapsis, where they use their engines to get to GSO. The 3rd stage trajectory can be adjusted so that its Periapsis make it burn in a controlled fashion. I had even 5.5 tons of propellant left, so its probably possible to achieve 16-17 tons. So the LV performs as expected, with those numbers 15 tons payload capacity to interplanetary trajectories is acquired :jj:

Edit 2 : 17 tons to GTO achieved easily. Next try : 20 tons, after some realism adjustements on the third stage (cryo boiloff, ignition sequences, RCS, progress on the mesh...)

 

[N_Molson]

Spending an outrageous amount of time on modeling the engines, it's not easy to find good pics (the under one being a rare exception).

A brand new RL-10-4-2 engine (without the frail nozzle extension)



Also, I'm not good at making the "turns" in the pipes, because Anim8tor uses a script for that (bending things), and Urumpwe an8->msh exporter doesn't support it (as far as I tried).

Still, it is slowly taking shape, mostly working from those pics (its horizontally mirrored).

---------- Post added at 09:51 PM ---------- Previous post was at 09:22 PM ----------

Also found interesting data about LOX/LH2 boiloff rates in space, they seems a little optimistic to me, what do you think ?

http://www.dunnspace.com/cryogen_space_storage.htm

This statement seems important too :

As an additional data point, a graph of boiloff vs. layers of insulation shows that if only 10 rather than 50 layers of insulation are used, the hydrogen boiloff rate is about 2.2 kg/hr rather than 0.45, and the oxygen boiloff rate is about 1.6 kg/hr rather than 0.36.