Interstellar Engines for In-System? (Far distant speculation)

[mattihase]

5 hours ago, NexusHelium said:

Well... You can't exactly expect to travel to a place like Debdeb with, say, an RS-25. So you will be forced to use the tech if you want to travel to the other systems.

Someone is going to try and get there using one RAPIER and 1000 years of gravity assists and I look forward to celebrating it.

[mcwaffles2003]

15 hours ago, Periple said:

A citation would be nice!

I don't really know anything about this but most of the stuff I've looked at that seems relatively well grounded in science seems to put the theoretical top speed at around 10% of c, give or take a few %, if you also need to decelerate. Even an antimatter drive would only get you to maybe 40% of c, and that if you're able to somehow deal with the gamma radiation it produces. But maybe I've missed something big!

Depends on the enrichment, see: https://www.projectrho.com/public_html/rocket/enginelist2.php

20% give just under 7000s

90% gives nearly 500,000s

[The Aziz]

13 hours ago, Bej Kerman said:

Yes, there will be mid-tier torch drives made specifically for smaller in-system vessels,

Ah ah. The torch drive is the endgame engine. (Source: one of the very first dev videos) Tier higher than the fusion drive. Which in turn, is a tier higher than MH engine which everyone here seemed to forget about.

Really, nothing else than the basic physics and resource requirements/availability is stopping anyone from using anything anywhere they want - the question is, is it worth it when there are better (faster, smaller, cheaper, more maneuverable) alternatives.

[Periple]

I don’t think the game would be much fun with torch drives. It seems to me that they would remove most of  the challenge.

[stephensmat]

1 hour ago, Periple said:

I don’t think the game would be much fun with torch drives. It seems to me that they would remove most of  the challenge.

Well, that's the point, isn't it? The game gets easier as you move on to the more advanced stuff, like travelling to another System.

And if you want the challenge, all you have to do is not use them.

[Periple]

1 hour ago, stephensmat said:

Well, that's the point, isn't it? The game gets easier as you move on to the more advanced stuff, like travelling to another System.

I think it stops being fun if the challenge goes completely away, and Epstein drives would do that.

1 hour ago, stephensmat said:

And if you want the challenge, all you have to do is not use them.

I did this a lot with KSP1 -- for example, I would play without MPLs, without RAPIERs, or various other limitations. I would like it if KSP2 was better balanced so that I didn't feel like I had to do it to keep it fun!

[The Aziz]

4 hours ago, Periple said:

I don’t think the game would be much fun with torch drives. It seems to me that they would remove most of  the challenge.

Probably one of the last nodes on the tech tree. If you've been nearly everywhere without it, you had your challenge. Treat it as the super duper OP armor/weapon you take off from beaten final boss.

[NexusHelium]

13 hours ago, mattihase said:

Someone is going to try and get there using one RAPIER and 1000 years of gravity assists and I look forward to celebrating it.

Oh yeah, I would never dare put it past the KSP community.

[Periple]

21 minutes ago, The Aziz said:

Probably one of the last nodes on the tech tree. If you've been nearly everywhere without it, you had your challenge. Treat it as the super duper OP armor/weapon you take off from beaten final boss.

I just hope the tech tree will be balanced strictly enough that it won't be unlocked too soon!

[regex]

3 hours ago, Periple said:

I think it stops being fun if the challenge goes completely away, and Epstein drives would do that.

The Epstein drive is a literary device, it's not even a theoretical engine. I'm fairly certain it's not going to be in KSP2 and if there are plans for it I'll consider the game an absolute failure.

Real talk though, if you want to get to another solar system in a reasonable amount of time you're going to need something which has an isp which is a significant fraction of C. If you use that in-system it will likely trivialize travel but it's worth noting, again, that engines can usually only prioritize thrust or isp so you'll still need landers and utility craft, and you're probably not going to be using something like a nuclear salt water rocket for a lander either because that would constitute landing on a continuously detonating nuclear explosion. The challenge is not going to go completely away.

[Bej Kerman]

10 hours ago, Periple said:

I don’t think the game would be much fun with torch drives. It seems to me that they would remove most of  the challenge.

What makes you think that? A. They will be hard to aim (and if your aim too well, directly at the destination, you can't rely on atmospheric drag to help you), B. You need the tech and resources to make them in the first place, and C. Any malfunction that makes it difficult or impossible to control will leave it on a trajectory that's difficult to rendezvous with, especially if you need a more resource-expensive ship to catch up and reach the crew.

7 hours ago, Periple said:

I did this a lot with KSP1 -- for example, I would play without MPLs, without RAPIERs, or various other limitations

I can say RAPIERS definitely do not make things easier.

[Periple]

6 minutes ago, Bej Kerman said:

I can say RAPIERS definitely do not make things easier.

They do if you make spaceplanes!

[Bej Kerman]

41 minutes ago, Periple said:

48 minutes ago, Bej Kerman said:

I can say RAPIERS definitely do not make things easier.

They do if you make spaceplanes!

Which I do, yeah. But you also had to get used to using them before you could call them easy.

[chefsbrian]

Definitely should be using interstellar drives in-system, and I can actually defend that mechanically and via gameplay.

We know that resources and resource gathering is going to be exceedingly important to expanding the footprint of what you can build and fuel in the game. We also know that Interstellar drives that are anywhere near as capable as their predicted IRL counterparts make travel vastly easier - A metallic hydrogen drive or fusion/torch drive turns an IRL mars mission into weeks rather than months. It stands to reason that these drives are going to be extremely expensive as a result - Aside from accurately requiring a lot of valuable, rare materials manufactured to exacting standards IRL, it just makes sense as a balance and progression model.

Which means, in my opinion, you'd be wasting it to use your first handful of drives to yeet yourself out of the kerbol system. These drives trivialize moving around the home system, but you probably want to be doing a lot of that now as you will want to scale resource extraction and production with your colonies. Using a torchship to ferry materials or goods sounds wasteful initially, but it'll let you move more goods, faster, creating much more responsive supply chains. Using such a beast of a ship to deploy a mining operation would let you bring way more equipment for every setup trip you do, vastly increasing the rate that you'll grow your available resources. And the player will be doing all of this with a ship that makes launch windows far less of a concern, and with a Delta-V thats sufficiently generous so that even the more relaxed, "Make it look cool" engineer can be confident of making all their destinations, even oblique and off-plane ones they might have struggled with.

These engines would provide an easy option for players to scale up and widely engage with the other mechanics of the game in the kerbol system, which I see as an obvious reward for having set up all the infrastructure to build that first engine. Players will still have to play with conventional gear up to that point, and players who want to scale up on the conventional stuff can still absolutely do so - You'll be flying more missions, and they'll be a little harder, but that sort of self-restriction is already the realm of the 'serious' players anyway so I doubt it'll be a big deal breaker.

To me, it just seems like a system that automatically provides a bunch of different options on how to have fun, without requiring you to take a specific path - for in system, anyway, I'm sure interstellar drives will be the kings of actual interstellar. I think it'd be naïve to try and force them to only be good at that though.

As for stuff like Epstein drives (Fusion torches with incredibly high efficiency and thrust) I wouldn't blame them if there was one or two at the end of the technology tree, maybe using alien system resources. But I'd prefer if it was left to mods, gotta have something worthwhile to add to the near-future Intergalactic mod

[K^2]

I don't see a way for Intercept to make interstellar drives anything but exceptionally useful for in-system travel.

It's been stated directly that the interstellar distances are meant to be realistic. I don't think we still got any confirmation how that correlates to the 1/10th scale of the game and potential relativistic limits of the real world, but we're still talking distances in billions of km under the most generous interpretation.

It takes 1 year accelerating at 1g to reach light speed.

I know we're all going to have warp set to absolute max the game allows when traveling between the stars, but there are inherent limits to how much warp that can be,  and if it takes hours of real time, that just simply won't do for a game. Interstellar ships have to accelerate at a pretty good clip to make the game playable. Maybe not quite 1g, but a sensible fraction of it even when fully loaded with fuel and cargo.

 

So what we learn from all of the above is that the interstellar drives are going to be very efficient, having I_SP several orders of magnitude above anything we had in the original game or have access to now. And that to be pushing all that fuel at a reasonable acceleration, these engines will have to come with very good TWR. Far better than ions, likely better than the NTR engines have been, and possibly approaching TWR of the chemical engines in the game, depending on just how big the interstellar gaps are.

There might be ships you want to build that are just too small to make use of an interstellar drive in the end game, or ships that have to go in the atmosphere. So I don't mean to say that there will be no uses for other engines. But I cannot think of any sensible barrier Intercept can put in the game that will allow interstellar gameplay to be fun and not make interstellar drives hands down the absolute top choice for in-system hauling.

[stephensmat]

7 hours ago, chefsbrian said:

These engines would provide an easy option for players to scale up and widely engage with the other mechanics of the game in the kerbol system, which I see as an obvious reward for having set up all the infrastructure to build that first engine.

Having gotten a better look at the new Tech Tree, this is not only what I'm expecting, it's my absolute hope. Consider it a NewGame+ mode.

Remember, when we get the 'Interstellar' Expansion, we'll be arriving in a new, untouched system, with Interstellar-Level craft. One way or another, we'll be using these engines 'In-System', if only in the *new* systems. 

38 minutes ago, K^2 said:

There might be ships you want to build that are just too small to make use of an interstellar drive in the end game, or ships that have to go in the atmosphere.

There's the obvious balance. First, there's the question of just how *big* an Interstellar Drive has to be. I can't imagine the modders won't handle that, one way or another.  The other is a viable 'lore' reason. An Interstellar Drive in atmo could be easily declared 'too dangerous' or 'too destructive'. By the time the Interstellar expansion comes, we'll have Colonies first, and that includes an Orbital VAB.

Either way, I can't wait to find out.

[The Aziz]

7 minutes ago, stephensmat said:

First, there's the question of just how *big* an Interstellar Drive has to be.

The Crucible? Bigger than VAB. But if you want to match its stats to even have a chance of Interstellar travel with smaller engines not designed for such task, it's gonna be tricky if not nearly impossible. Depending on what array of engines we'll have.

11 minutes ago, stephensmat said:

An Interstellar Drive in atmo could be easily declared 'too dangerous' or 'too destructive'

I believe that's the plan for anything high tech - either too hot (vacuum type MH), too explosive (Orion), or simply with not enough thrust to counteract the gravity of most bodies. You're going to crash your ship or destroy the colony you'd be launching from/landing on. And that still is just physics, not some arbitrary limitations.

[regex]

14 hours ago, chefsbrian said:

- A metallic hydrogen drive or fusion/torch drive turns an IRL mars mission into weeks rather than months.

Minor nitpick here, the metallic hydrogen engine stats they previewed aren't all that much better than "conventional" nuclear thermal engines. Something like a nuclear salt water engine would absolutely allow for near-brachistochrone transfers but that sort of drive has a six-digit aspirational isp. Metallic hydrogen is still in the lower end of four-digit range; it'll be great for utility craft but quickly outpaced by better transfer craft.

[Wheehaw Kerman]

14 hours ago, The Aziz said:

I believe that's the plan for anything high tech - either too hot (vacuum type MH), too explosive (Orion), or simply with not enough thrust to counteract the gravity of most bodies. You're going to crash your ship or destroy the colony you'd be launching from/landing on. And that still is just physics, not some arbitrary limitations.

Too explosive?  This is KSP!  No such thing.  Jebediah wept…

[K^2]

17 hours ago, regex said:

nuclear salt water engine would absolutely allow for near-brachistochrone transfers but that sort of drive has a six-digit aspirational isp

Do you have a quote on that? Six-digit I_SP would put the core temperature in a 100MK ballpark, which cools basically instantly by emission of hard gamma radiation. I'm wondering how that's addressed.

[regex]

4 hours ago, K^2 said:

Do you have a quote on that? Six-digit I_SP would put the core temperature in a 100MK ballpark, which cools basically instantly by emission of hard gamma radiation. I'm wondering how that's addressed.

Apparently speculation by Robert Zubrin about an NSWR using 90% enriched U-235, which would achieve an isp of around 480,000. Link, source also included. Note that it requires better efficiency over the "regular" NSWR, which itself is kind of dubious.

[K^2]

10 hours ago, regex said:

Apparently speculation by Robert Zubrin about an NSWR using 90% enriched U-235, which would achieve an isp of around 480,000. Link, source also included. Note that it requires better efficiency over the "regular" NSWR, which itself is kind of dubious.

Yeah, I'm going to just say that nozzle efficiencies quoted in that section are impossible due to black body radiation losses at the temperatures in the detonation region that this design implies.

Estimates in spoiler.

Spoiler

Even completely ignoring oxygen in the exhaust, the kinetic energy of hydrogen would have to be 50,000 times higher than that of water vapor in the LH2 rocket exhaust. So ~250 million Kelvin, give or take, in the detonation region? The radiation intensity is fourth power in temperature. If we were to reach these temperatures, the photon energy at spectral peak would be about 100keV, so on threshold of pair production. Absolutely everything's a black body in these energy ranges, so we don't need to consider albedo. The flux will then be some silly number, like 2.2x10²⁶ W/m². So that 3cm proposed reaction plenum would output 6.3x10²³ W of radiation. For comparison, our Sun's total is 3.8x10²⁶ W. This engine would have to glow like 0.2% of the Sun, except in hard X-Ray.  But more importantly, we have 200kg/s of fuel mixture going through this engine, which has a total mass energy flow of only 1.8x10¹⁹ W. Given that only 2% is fuel, and Uranium's mass defect is not spectacular... Energy in is short by many orders of magnitude of just the losses, meaning that kind of temperature cannot be sustained. It's not even close.

Point is, if you were to build a NSWR to that section's spec, almost all of the energy from the detonation region would be escaping as X-rays, drastically reducing the thermal energy that can be converted into the exhaust velocity by the nozzle.

 

It's still a cool concept, if you dial it down a bit. The previous section building up a more realistic proposal for 6,730s version seems a lot more plausible. Again, that 4th power in temperature does a lot of damage to rockets that convert thermal energy into propulsion, but only on the high end. There's a lot of room to grow beyond what we can squeeze out of chemical rockets.

But if we want to get these 6-figure I_SP engines, we have to have a different way of accelerating the exhaust. It has to be some sort of an electromagnetic drive.

[regex]

1 hour ago, K^2 said:

But if we want to get these 6-figure I_SP engines, we have to have a different way of accelerating the exhaust. It has to be some sort of an electromagnetic drive.

Zubrin's theory-crafting aside, the point still stands that you'll need an engine with much higher isp than what they've previewed for metallic hydrogen in order to really cut travel time down.

[MarcAbaddon]

The biggest issue with a torch drive right now would be that the maneuver planner does not support constant burn and flip trajectories well. But while it would remove a lot of the planning and delta-v conservation aspects, I think working with those kinds of trajectories would be an interesting change of pace. So I don't think it completely takes the challenge away, if you have to learn about new kinds of trajectories. Other than that with high ISP/high thrust engines you really have heat management issues to take care of as well... I remember that in the Far Future Mod my vessels all had radiatiors extending out at all points to somehow manage to get rid of all that heat.

[The Aziz]

19 minutes ago, MarcAbaddon said:

the maneuver planner does not support constant burn and flip trajectories well

*Yet.