totm apr 2023 Engine plumes (Split from AMA questions)

[Gargamel]

1 hour ago, Smegsy said:

Thread = mind blown. Can't think of another game forum where you would see this sort of stuff. Really interesting.

Should have joined up sooner.       I’ve planned vacations around info taken from these forums.  

[RocketRockington]

Why don't we just ask the guy who did the research, before he was hired by Intercept to add something similar to his mod to KSP2.  Clearly one of the best decisions they made.

@Nertea  Do you want to chime in here, what sort of resources did you find, for why a rocket engine in a vacuum showed a concave plume past the nozzle, rather than linear or convex?

[j12sfgd23]

  

2 hours ago, Periple said:

Oh right, got it, thanks! 

If you do find out more about this please post what you discovered, this is interesting stuff (and way above my pay grade )

definitely plan to, I 100% plan to try and figure out how to run a sim to test (I have literally no idea how to do optics in this situation so no promises but I will try), so if I do get interesting results I will definitely post em!

1 hour ago, Smegsy said:

Thread = mind blown. Can't think of another game forum where you would see this sort of stuff. Really interesting.

Yeah, honestly it blows my mind how such a seemingly simple game can be as thought provoking and brilliant as KSP is. squad was a pioneer, and I really cant wait to see what intercept does with the amazing groundwork theyve laid. 

28 minutes ago, i dont know how to forum said:

This is almost entirely over my head, but what little I do understand is really interesting. Thank you for the elaboration! It seems clear this is a lot more complicated than I had thought, but of course it is rocket science. As I understand it the devs are always on the lookout for more in-depth information on things like this, so I hope they see your AMA question and/or this thread. While I'm sure it's near the very bottom of the priority list at the moment, I'd love to see them continue to develop the plumes to make them even more realistic.

If there is anything I can do to clear anything up please let me know! I would love to make this as comprehensive as possible, if for no other reason than to make this a resource people can find in the future if they are curious. And yeah I hope they do, I am really curious to hear their feedback on it, because i know they've done a ton of research and id love to know the results of that. Honestly, this topic is especially interesting because afaik this is kind of a new discussion. From what ive seen, engineers dont care one way or another what the plume *looks* like,  we're way more concerned with the actual parameters like thrust and pressure, so im really curious to find out the answer to this.

Edited March 22, 2023 by j12sfgd23

[rjbvre]

Really interesting read. Not surprisingly it's hard to find pics of larger engines firing in a vacuum, but rcs thrusters can give some examples. Looks like they can appear either cone shaped or more parabolic. The difference could come down to camera sensitivity and lighting conditions since the expanded part of the plume cools down quickly, and when there is a parabolic edge it's a lot fainter.

Spoiler

Spoiler

Spoiler

Spoiler

 

[Periple]

The second and last ones almost look like they fans out although it's hard to tell. Super interesting!

[LameLefty]

On 3/21/2023 at 3:45 AM, j12sfgd23 said:

Plume expansion is almost solely a function of nozzle geometry (fixed) and ambient pressure (dynamic), and so for any given engine …

This thread is giving me flashbacks to my own undergraduate aerospace engineering degree (mumble mumble) years ago. 

 

Good job!

 

Edited March 22, 2023 by LameLefty

[Bej Kerman]

12 hours ago, darthgently said:

Once free of the bell it will not accelerate any faster radially than at that point where it exits the bell.

Won't they? They still coexist with the other particles. The idea the outer particles do not experience any force despite having gaseous turmoil on one side and vacuum on the other confuses me.

[darthgently]

My thought experiment must be broken.  I just can't see how any remaining pressure in the plume would somehow favor radial acceleration over axial acceleration and it seems like the hyperboloid requires any remaining pressure to somehow to be preferring radial over axial acceleration.  And that is not how pressure works.  What the heck am I missing?

Finally, watching vids of F9 stage 2 MVAC at work, it looks like a straight walled cone.  Granted, it isn't in full vacuum at that point, but dang near 

[rjbvre]

25 minutes ago, darthgently said:

My thought experiment must be broken.  I just can't see how any remaining pressure in the plume would somehow favor radial acceleration over axial acceleration and it seems like the hyperboloid requires any remaining pressure to somehow to be preferring radial over axial acceleration.  And that is not how pressure works.  What the heck am I missing?

Just my uneducated guess, but other than at the leading edge of the plume, wouldn't there be resistance axially from the exhaust that left the nozzle earlier and is directly in front?

[darthgently]

30 minutes ago, rjbvre said:

Just my uneducated guess, but other than at the leading edge of the plume, wouldn't there be resistance axially from the exhaust that left the nozzle earlier and is directly in front?

But assuming that other previous exhaust had the same acceleration, why would it slow down?  If it doesn't slow down, how is it "in the way"?

Edited March 22, 2023 by darthgently

[Bej Kerman]

51 minutes ago, darthgently said:

My thought experiment must be broken.  I just can't see how any remaining pressure in the plume would somehow favor radial acceleration over axial acceleration

Does it have to favour either?

[rjbvre]

16 minutes ago, darthgently said:

But assuming that other previous exhaust had the same acceleration, why would it slow down?  If it doesn't slow down, hours is it "in the way"?

Because while the exhaust has net linear momentum away from the nozzle, the individual molecules are still all moving randomly and providing pressure in all directions. Basically some of the molecules randomly get pushed more backward and provide that resistance

[j12sfgd23]

Just now, darthgently said:

My thought experiment must be broken.  I just can't see how any remaining pressure in the plume would somehow favor radial acceleration over axial acceleration and it seems like the hyperboloid requires any remaining pressure to somehow to be preferring radial over axial acceleration.  And that is not how pressure works.  What the heck am I missing?

Finally, watching vids of F9 stage 2 MVAC at work, it looks like a straight walled cone.  Granted, it isn't in full vacuum at that point, but dang near 

Just now, Bej Kerman said:

Won't they? They still coexist with the other particles. The idea the outer particles do not experience any force despite having gaseous turmoil on one side and vacuum on the other confuses me.

Thats a really good question, but as strange as it sounds, there is a point where the radial pressure is more of a dominating force than the axial, and that is immediately after it leaves the engine bell. The reason this is the case is because its not just about total pressure, its also about the differential on either side of each particle. 

At the very beginning of the post-nozzle flow, the flow is not expanding much, and so the axial pressure change (and therefore axial acceleration) is extremely small. On the flip side, the radial pressure drop is VERY high, as it goes from the high pressure exhaust to the vacuum of space, which causes the initial expansion of the plume. As it expands though, the radial pressure drop quickly becomes almost zero, and so the radial acceleration of the fuel slows to a crawl, while the axial acceleration starts to increase with the decreasing plume pressure (hence the bell shape). You can sort of visualize it in the image I attached here!

Hopefully this helps!

Edited March 22, 2023 by j12sfgd23

[j12sfgd23]

Just now, Periple said:

The second and last ones almost look like they fans out although it's hard to tell. Super interesting!

  Interesting, yeah you are right. In the fourth one though the exposure is actually high enough to see a very faint outline of the greater bell, so I am curious as to what's actually happening. I wonder if maybe the plume does still grow in the bell shape, but the hottest and densest part grows in a much smaller hyperbolic shape like that which is why it looks the way it does? I hope @Nertea can answer this.... that is my best guess for those images, 

Edited March 22, 2023 by j12sfgd23

[lemon cup]

Assuming that the system is anything like Waterfall mod for KSP1, that means any rocket plume you see in the game now can (and very likely will) be made better as time goes on with very little impact on performance, which is awesome.
 

I’ve done a decent amount of tinkering with plumes in Waterfall, and you can achieve some really spectacular effects in no time.

 

 

[Aaron VFX]

Great discussion everyone!

Although I'm not an expert on rocket exhaust, I wanted to share that the overall shape and composition of the plumes, from atmospheric pressure to vacuum, was partially based on research from the book "Rocket Propulsion Elements" by George P. Sutton.

We held several meetings to discuss key information and to hear what Nate wanted to achieve in KSP2. From this, we devised a plan to create exhaust plumes that would respond to atmospheric pressure (and a host of other variables) in a  way that aligns with science while also being performant in the overall context of the game.

While we have very clear work in front of us, it would be great to circle back around to plumes in the future and see what we can adjust to make them even belter.

[LameLefty]

13 minutes ago, Aaron VFX said:

I wanted to share that the overall shape and composition of the plumes, from atmospheric pressure to vacuum, was partially based on research from the book "Rocket Propulsion Elements" by George P. Sutton.

Fantastic! Pretty sure I have my old copy of that one around here in a box somewhere, too.

[darthgently]

5 hours ago, Bej Kerman said:

Does it have to favour either?

Uh, no.  Why are you going there?That is the point.  I don't think I'm seeing this right and the explanations are helping, but it is really unintuitive to me at this point. 

Edited March 22, 2023 by darthgently

[asmi]

This is what operating ion engine (specifically, Hall Effect Thruster) looks like in real life in space (from 0:54 if timecode won't work):

 

Edited March 23, 2023 by asmi

[j12sfgd23]

5 hours ago, Aaron VFX said:

Great discussion everyone!

Although I'm not an expert on rocket exhaust, I wanted to share that the overall shape and composition of the plumes, from atmospheric pressure to vacuum, was partially based on research from the book "Rocket Propulsion Elements" by George P. Sutton.

We held several meetings to discuss key information and to hear what Nate wanted to achieve in KSP2. From this, we devised a plan to create exhaust plumes that would respond to atmospheric pressure (and a host of other variables) in a  way that aligns with science while also being performant in the overall context of the game.

While we have very clear work in front of us, it would be great to circle back around to plumes in the future and see what we can adjust to make them even belter.

I am definitely gonna be reading into that book now, thank you so much for the reference!! To be honest, I really do hope I am wrong about the way the plume looks now, its a very unique shape that ive never really seen before, plus it means theres even more stuff for to learn lol. Thanks for taking the time to answer this question!

[Periple]

6 hours ago, darthgently said:

Uh, no.  Why are you going there?That is the point.  I don't think I'm seeing this right and the explanations are helping, but it is really unintuitive to me at this point. 

I think the initial expansion is intuitive — the pressure differential at the edge of the plume favors radial expansion. I assumed that wouldn’t change at all, which would make the plume a hyperboloid converging on a cone. But as we’ve now learned there’s a lot more going on: the hypersonic flow causes all kinds of internal shock waves and pressure differentials that produce surprising results — notably, a low-pressure region inside the plume where axial acceleration increases, at the same time causing the plume to “fall in” on itself, slowing down radial expansion and creating the bell shape.

Come to think of it, these must be the same dynamics that create the underexpanded “pinched” plume under certain conditions in the atmosphere, right? For that to happen, radial pressure in the plume must be less than ambient pressure, otherwise it wouldn’t “pinch?”

Is this more or less correct or am I misunderstanding something again?

Edited March 23, 2023 by Periple

[Periple]

I've been still thinking... those pictures from the Space Shuttle do make the exhaust plumes look conical or even hyperboloidal. From what we've learned here, it sounds very much like that must be the hot, dense, central portion of the plume,  with the full bell-shaped plume invisible.

If that's how the plume changes as the rocket goes up, we should see the underexpanded "pinched" plume gradually expand to a bell, then the high-altitude "jellyfish," at which point the outer bell would fade out of view; by the time we're in space, we'd only see the conical/hyperboloidal inner plume. 

In this scenario, the plumes already look "right"  – only the transition between the bell- or jellyfish-shaped atmospheric plume and the conical/hyperboloidal outer space plume is incorrect; the bell shouldn't transform into a cone, instead it should fade out and leave only the inner cone visible. 

Would this make sense?

[rjbvre]

5 hours ago, Periple said:

I've been still thinking... those pictures from the Space Shuttle do make the exhaust plumes look conical or even hyperboloidal. From what we've learned here, it sounds very much like that must be the hot, dense, central portion of the plume,  with the full bell-shaped plume invisible.

If that's how the plume changes as the rocket goes up, we should see the underexpanded "pinched" plume gradually expand to a bell, then the high-altitude "jellyfish," at which point the outer bell would fade out of view; by the time we're in space, we'd only see the conical/hyperboloidal inner plume. 

In this scenario, the plumes already look "right"  – only the transition between the bell- or jellyfish-shaped atmospheric plume and the conical/hyperboloidal outer space plume is incorrect; the bell shouldn't transform into a cone, instead it should fade out and leave only the inner cone visible. 

Would this make sense?

It definitely sounds reasonable. I've been wondering too just how much of the plumes in those photos isn't visible, and what the shape of those parts would be. Just shows how complicated it would be to accurately animate: not only do you have to model the plume, but you have to take into account what the eye would see

[LameLefty]

So noted real-life rocket launch photographer extraordinaire John Kraus captured this wonderful image of last night’s Relativity Space launch of their Terra 1 mission. The second stage failed to ignite and so the mission failed to achieve orbit, but the first stage seemed to perform nominally.

To the point, this is what 9 small methalox plumes actually look like at sea level.

[j12sfgd23]

To anyone still curious about this, sorry for disappearing for so long, life gets in the way sometimes lol. I have been reading through the various articles and books ive found online, and from what I can tell online it seems pretty unclear. From what I can tell, it seems like regardless of pressure (assuming the engine is still working properly), the fuel flow has *some* degree of a hyperbolic streamline (generally the hyperbolic part is just so unbelievably small that its beyond negligible), which bends towards a more parabolic shape in the direction of the engine exhaust. The amount this streamline bends (and therefore shape of the FLOW) seems to really depend on both the engine itself, and the surrounding pressure.

Now the fun part though: what shape would it look like? Well, the unfortunate answer is.... I still have no idea. I am working my way through a few articles about light emissions from gasses, but all I have found so far is grey body emissions, which wont really tell me how much visible light is being emitted. That being said, I ran a super janky fluid sim, and I did actually see a temperature curve similar to what we see in the game and in the other books, so unless im misunderstanding this I think its pretty safe to say that I think I was wrong on my initial thoughts! I am happy to say that more likely than not, it should be hyperbolic like it is in the game! If I find anything else I will make sure to report it here, but thank you all for discussing this with me, I really appreciate being exposed to all of this new knowledge, and I really enjoyed sharing my knowledge with you all, so thank you!!