Changing the way air-breathing jet engines are constructed.

[S4qFBxkFFg]

(I originally posted this in a thread about 2.5m jets, but it probably makes more sense for it to be here.)

However, before expanding the jets available I would be much happier if Squad improved the existing ones in several ways. This would include placement of parts - at the moment we have what appear to be nozzles (possibly including the turbines - who knows?) and intakes with no need for any combustion chambers, and barely any compressors; really, it should be enforced that these parts have to be arranged vaguely in-line -- S-ducts and gearing can give some room for manoeuvre here, but not remotely to the degree possible in the game.

What I would suggest is that the air-breathing engines are altered slightly to make designs a bit more sane - have a selection like so:

1. One part (we could call this the engine core) consisting of (front to back) second+ stage compressor(s)->combustor->turbine(s) for each size of turbine engine you want to have - this could look something like the existing nacelles or radial engine bodies.

2. Various first stage compressors that must be attached to the front of the engine core to function - these could include:

- first stage turbojet compressors which are pretty much the same diameter as the rest of engine and look like the existing circular intake (even though it says it's a turbofan...)

- low-bypass fans - slightly wider than the rest of the engine

- high-bypass fans - significantly wider than the rest of the engine

- geared propeller, to make a turboprop engine

3. Various nozzles, which must be attached to the back of the engine core to function, which may:

- possess thrust vectoring capability

- include an afterburner

4. Various intakes - this is where we'd see things like the fancy multi-shock intakes like on the F-15, precoolers like on the Sabre engines, s-ducts for when there's an inconvenient fuselage right in front of your compressor, etc.

This doesn't even consider things like ramjets/scramjets which should probably be discrete parts.

I would hope that the above could be done without making the design/build process too difficult - for the existing system, we'd usually have intake+fuselage/nacelle+engine, and for the alternative suggested above it would be intake+compressor+engine core+nozzle. Admittedly it does make the minimum part count 4 instead of 3, but in compensation I think it's a more elegant solution while still allowing quite a bit of flexibility.

Thoughts?

[Dodgey]

It's a very thought out and detailed idea I'll give you that. From a gameplay point of view I think that I will be too restricting for the player, just my two bits.

[Camacha]

It's a very thought out and detailed idea I'll give you that. From a gameplay point of view I think that I will be too restricting for the player, just my two bits.

Why would it be too restrictive? Other much more complex mechanisms have not proven to be so either. I certainly would not mind.

[Dodgey]

From what I gathered it would turn the jet engine into a rather large construction of four parts that all need to be inline with each other. Basically a long engine. Though honestly I have no idea how big you plan for these parts to be, say comparable to the pre cooler or even longer. Maybe a diagram could help, if I'm wrong by all means let me know. It sounds like this idea has great potential.

[Anton P. Nym]

It's not a bad idea, as such, but I think it's out of step with the rest of the game. With KSP rocket engines we don't select fuel mixes, turbo pumps, combustion chambers, bell profiles, and the like; we get pre-built, standardised units. Mix-your-own jet engines would be radically different unless we also extended this to rockets, making this a very large task. Add in that it would increase part counts (potentially by a large fraction for multi-engined craft) when we still have part-count based performance constraints, and I can't see it as part of KSP.

I think it would be an ideal mechanic for a game that concentrated on high-performance aircraft, though. Might be worth exploring from that perspective.

-- Steve

[S4qFBxkFFg]

Firstly, I've thought about what I proposed, and now think that it's slightly too complex for the benefit it gives - and the idea of mixing and matching compressors with engine cores similarly compromises realism without much gain.

As a compromise with the current system, how about this amendment?

1. Engines are not split up so much, one (long-ish) part represents the compressor(s), combustor, turbine(s), and the nacelle containing it. (This would look something like a stretched nacelle or radial engine body)

2. Intakes are still separate, but must attach to the front of the engine; there can still be a variety of intakes, to allow for variable geometry intakes, s-ducts, and the like (for example, a "duct" part could accept intakes radially attached to it, as long as the duct is attached to the front of an engine).

3. Nozzles are separate, but must attach to the back of the engine; again, this is where afterburners or vectored thrust could be incorporated OR to further reduce part count without losing that much functionality, the main part could include the nozzles and if things like afterburners or thrust vectoring are desired, they can appear in separate engines.

Here, we'd have a minimum of two or three parts per engine - which is as good, or better, than what we have right now, just arranged a little differently.

It's a very thought out and detailed idea I'll give you that. From a gameplay point of view I think that I will be too restricting for the player, just my two bits.

Agreed - see above.

From what I gathered it would turn the jet engine into a rather large construction of four parts that all need to be inline with each other. Basically a long engine. Though honestly I have no idea how big you plan for these parts to be, say comparable to the pre cooler or even longer. Maybe a diagram could help, if I'm wrong by all means let me know. It sounds like this idea has great potential.

For a basic turbojet, I'd suggest that the length be about 4x to 5x its diameter, for example:

For a high-bypass turbofan, or turboprop, there would be parts of the engine significantly wider - I'd suggest that in this case, the part is categorised (e.g. as a 1.25m part) based on its core diameter:

It's not a bad idea, as such, but I think it's out of step with the rest of the game. With KSP rocket engines we don't select fuel mixes, turbo pumps, combustion chambers, bell profiles, and the like; we get pre-built, standardised units. Mix-your-own jet engines would be radically different unless we also extended this to rockets, making this a very large task. Add in that it would increase part counts (potentially by a large fraction for multi-engined craft) when we still have part-count based performance constraints, and I can't see it as part of KSP.

I think it would be an ideal mechanic for a game that concentrated on high-performance aircraft, though. Might be worth exploring from that perspective.

-- Steve

My reason for suggesting this is that, to a noticeable degree, we do have a different game mechanic for constructing air-breathing jets as opposed to rockets - it is not required to have nacelles and intakes for rockets, so the part count for jets is probably going to be higher anyway. However, as stated at the beginning of this post I agree that splitting up engines to the extent initially described would be taking it too far.

[Red Iron Crown]

Jet engines should definitely be longer, the current models only include the nozzle and not enough room for the various stages of the engine itself. I would prefer for them to remain a single part for performance reasons.

Requiring air intakes to be attached to the front of the engine is neither realistic nor gameplay enhancing.

[S4qFBxkFFg]

Requiring air intakes to be attached to the front of the engine is neither realistic ...

I may be completely missing your point here, but I simply don't understand what you mean -- if it doesn't go in through the front of the engine, then where exactly do you think the intake air should go?

[Red Iron Crown]

I may be completely missing your point here, but I simply don't understand what you mean -- if it doesn't go in through the front of the engine, then where exactly do you think the intake air should go?

The intake itself should go wherever it is convenient on the airframe. Just a few examples (far from exhaustive):

http://en.wikipedia.org/wiki/F-100_Super_Sabre

http://en.wikipedia.org/wiki/F-105_Thunderchief

http://en.wikipedia.org/wiki/MiG_21

It is not necessary or even desirable to have the intake attached to the front of the engine. Basically that would limit jet engines to airliner-style pods.

Edit: For clarity, of course intake air has to eventually reach the front of the engine, but the intake itself needn't be there. The internal ductwork connecting intake to engine is something I'm happy is abstracted away.

Edited March 18, 2014 by Red Iron Crown
Clarity

[S4qFBxkFFg]

Ah - it's a semantics issue - I was considering the intake to include all the components that take the air from outside until it reaches the compressor, therefore including all the ducts - apologies if this isn't the correct terminology.

Taking the examples you posted, they have long internal ducts which are wide enough to comprise a non-negligible proportion of the cross-sectional area of the aircraft - this could be abstracted away by assuming that certain parts (like the structural fuselage or Mk2 cockpit) can contain them, but when considering fuelled fuselages, a significant part of their capacity would be lost. If the intervening parts are girders or I-beams then it becomes absurd, almost verging on an exploit.

I admit it wouldn't be trivial to calculate whether a given route from intake to compressor is feasible (perhaps within a cone of a certain angle from the front of the compressor...?) but that's up to devs/modders.

The question is, which is preferable? A system that allows bolting a cluster of intakes onto a girder mounted a few metres behind the engine nozzle, but satisfies the assumption that internal intake ducts can exist; or vice versa.

[Dodgey]

The issue of the air ducting is the same as it is for fuel flow, I say have magical ducting through steel grinders, otherwise it could just become more annoying.

[Red Iron Crown]

Taking the examples you posted, they have long internal ducts which are wide enough to comprise a non-negligible proportion of the cross-sectional area of the aircraft - this could be abstracted away by assuming that certain parts (like the structural fuselage or Mk2 cockpit) can contain them, but when considering fuelled fuselages, a significant part of their capacity would be lost. If the intervening parts are girders or I-beams then it becomes absurd, almost verging on an exploit.

I admit it wouldn't be trivial to calculate whether a given route from intake to compressor is feasible (perhaps within a cone of a certain angle from the front of the compressor...?) but that's up to devs/modders.

This would require treating the intake air resource differently than it is now. Currently, it works like electrical charge or monopropellant; as long as there is some anywhere in the craft it can be used by any part without requiring piping or ductwork to connect it. To work the way you're describing would require it to work more like liquid fuel does; an explicit route between a source and an engine must be created using fuel crossfeed capable parts and fuel lines.

This does add a significant layer of complexity. All parts, stock or mod, would need an "intake air crossfeed capable" parameter added, along with (perhaps) a fuel line analog for intake air.

I'm mostly a rocket guy, so all this complexity adds nothing for me as I only use jets occasionally as boosters. I think this sort of functionality belongs in a mod aimed at aircraft-focused players. Even among them, I think you'll find some resistance to the idea as it limits the ways a jet powered craft can be assembled.

[Capt Snuggler]

This would require treating the intake air resource differently than it is now. Currently, it works like electrical charge or monopropellant; as long as there is some anywhere in the craft it can be used by any part without requiring piping or ductwork to connect it. To work the way you're describing would require it to work more like liquid fuel does; an explicit route between a source and an engine must be created using fuel crossfeed capable parts and fuel lines.

I actually think this is a nice idea. (even if you don't). surface mounted ducts that are placed similarly to fuel lines (maybe wrapping around the outside of parts?).

it would also help to combat "airhogging", which I have been known to do, but I feel like I'm cheating.

"airhogging" wouldn't be imposable but with the added weight of all the ducts, it would be significantly less attractive.

[Jouni]

Taking the examples you posted, they have long internal ducts which are wide enough to comprise a non-negligible proportion of the cross-sectional area of the aircraft - this could be abstracted away by assuming that certain parts (like the structural fuselage or Mk2 cockpit) can contain them, but when considering fuelled fuselages, a significant part of their capacity would be lost.

A significant part of their capacity is already lost. The Mk1 Fuselage is the same size as the FL-T400 Fuel Tank, but its capacity is only 37.5% of the FL-T400.

[rkman]

I support any idea that would eliminate air intake spam.

[Red Iron Crown]

I support any idea that would eliminate air intake spam.

There's some truth here, airhogging make SSTOs too easy.

[Kasuha]

There's some truth here, airhogging make SSTOs too easy.

Airhogging makes SSTOs easy. Is that a bad thing? Will the game be more fun if they'll be hard?

[Red Iron Crown]

Airhogging makes SSTOs easy. Is that a bad thing? Will the game be more fun if they'll be hard?

Maybe for some? SSTO should be challenging to design, in my opinion. It certainly is in real life, and adding more intakes is not a realistic solution. It's not easily discoverable, I don't think most new players would stumble upon putting 5+ intakes per engine. I consider airhogging almost as much of an exploit as infiniglide.

But that's just me.

[Kasuha]

To be honest, I think useful SSTOs are hard enough already. Yes, you can build a plane with two or three intakes and one engine (notice: many people don't consider three intakes to one engine airhogging) and it will get to orbit like a breeze. But when you start moving things, building old classic rocket lifter is way easier than building an SSTO for the same payload. So I don't think this area is completely free of challenge, even with airhogging possible. Plus you can always make it a challenge for yourself, making an SSTO with less and less intakes.

[Red Iron Crown]

That's a good point. I don't do much in the way of SSTOs, I've only ever built them for crew transfer payloads to see if I could do it. Maybe that will change when economics become a factor, assuming discarding stages is penalized somehow.

I still think airhogging is unrealistic, though. Adding more intake area just doesn't have the effect on real jets that it does on KSP's. I know that it's just a game and that some things are simplified or abstracted away, but this is messing with some of the fundamentals of aircraft design. I think a greater modicum of realism is desirable in a game that has spacecraft design as one of its primary modes of play.

[Jouni]

To be honest, I think useful SSTOs are hard enough already. Yes, you can build a plane with two or three intakes and one engine (notice: many people don't consider three intakes to one engine airhogging) and it will get to orbit like a breeze. But when you start moving things, building old classic rocket lifter is way easier than building an SSTO for the same payload. So I don't think this area is completely free of challenge, even with airhogging possible. Plus you can always make it a challenge for yourself, making an SSTO with less and less intakes.

Useful SSTOs are easy, once you learn the proper ascent profile for turbojets. Start with the payload. Add a rocket stage that is roughly the same size. Then add enough turbojets to comfortably lift the rocket and the payload from launchpad, and some jet fuel for them. It only becomes hard, if you want to add wings.

Those guidelines work with one intake per turbojet, and you don't even have to manually reduce throttle.

[Kasuha]

Useful SSTOs are easy, once you learn the proper ascent profile for turbojets. Start with the payload. Add a rocket stage that is roughly the same size. Then add enough turbojets to comfortably lift the rocket and the payload from launchpad, and some jet fuel for them. It only becomes hard, if you want to add wings.

Those guidelines work with one intake per turbojet, and you don't even have to manually reduce throttle.

I think we're running into the "what's an SSTO" question here. Because if we understand SSTO as anything that gets payload to orbit in single stage then you don't even need turbojets (rocket engines don't flameout so it's even easier to steer). So I think what we're talking about in this thread is rather an "SSTO airplane", something powered mainly by jets and launching/landing horizontally.

[Capt Snuggler]

I think we're running into the "what's an SSTO" question here. Because if we understand SSTO as anything that gets payload to orbit in single stage then you don't even need turbojets (rocket engines don't flameout so it's even easier to steer). So I think what we're talking about in this thread is rather an "SSTO airplane", something powered mainly by jets and launching/landing horizontally.

also re-useable.

[Jouni]

I think we're running into the "what's an SSTO" question here. Because if we understand SSTO as anything that gets payload to orbit in single stage then you don't even need turbojets (rocket engines don't flameout so it's even easier to steer). So I think what we're talking about in this thread is rather an "SSTO airplane", something powered mainly by jets and launching/landing horizontally.

I was talking about jet-boosted rockets that are basically spaceplanes without wings. As this thread is about the way jet engines should work, it concerns them in the same way as it concerns proper spaceplanes. It's easier to build and fly a jet-boosted rocket than a proper spaceplane, and the ascent profile and the engine configuration are similar. Vertical launching allows the form to follow function better, while vertical landing basically requires you to use a few parachutes to make the landing less risky.

Spaceplanes are more fun to fly, but for practical purposes, it's just better to remove the wings and add a bit more thrust. On the other hand, I would prefer removing turbojets from the game, as they give SSTOs unreasonably high payload fractions. A basic jet/rocket combo feels more like a real-world spaceplane, while RAPIERs would now be exotic and efficient high-tech engines.

[bartekkru99]

Let's revive this thread!!!