Jet engines are way overpowered

[Red Iron Crown]

Specific impulse is defined in a way that only takes the change of vehicle mass into account. The numbers actually go higher than 6000 s: typical turbofans have Isp 12000 s at sea level. Of course, people usually talk about specific fuel usage (grams of fuel per kilonewton-second) instead of specific impulse, because it makes more sense with jet engines.

I'm not 100% sure about this, but it was my understanding that specific impulse is another way of expressing exhaust velocity, whether the mass expelled comes from the ship or is taken in whole or in part from the atmosphere. If it works the way you describe, would a thermal turbojet (like this) have a specific impulse of zero?

[KerikBalm]

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

For a rocket, specific impulse and exhaust velocity are perfectly correlated.

For a jet engine, you can talk about "effective exhaust velocity"

http://en.wikipedia.org/wiki/Specific_impulse#Specific_impulse_as_a_speed_.28effective_exhaust_velocity.29

For air-breathing jet engines, particularly turbofans, the actual exhaust velocity and the effective exhaust velocity are different by orders of magnitude. This is because a good deal of additional momentum is obtained by using air as reaction mass. This allows for a better match between the airspeed and the exhaust speed which saves energy/propellant and enormously increases the effective exhaust velocity while reducing the actual exhaust velocity.

While the actual exhaust velocity is lower for air-breathing engines, the effective exhaust velocity is very high for jet engines. This is because the effective exhaust velocity calculation essentially assumes that the propellant is providing all the thrust, and hence is not physically meaningful for air-breathing engines; nevertheless, it is useful for comparison with other types of engines

In such a case, you only consider the fuel, and not the air that serves as reaction mass and oxidizer.

That makes it easier for calculating, as you don't need the fuel:air ratio

The exhaust velocity of a jet engine is quite low relative to a rocket engine. However, as you get "free" reaction mass, you can have a lower real exhaust velocity, but a higher effective exhaust velocity. For the same amount of energy (since in a jet engine, your fuel is basically only providing energy, very little ReMass comes from fuel), you can propell 4x as much mass at half the velocity, increasing thrust per unit energy by a factor of two. Jet engines are limited by energy, not reaction mass.

You can pretend its just jet fuel that is the ReMass, and give it a very high effective exhaust velocity, which is where figures of 2500s come from and are quite reasonable.

But using this effective ISP already takes into account the fuel:air ratio... KSP then factors that fuel:air ratio in again (at 15:1), giving an engine that is roughly 15x as efficient as it should be.

Edited June 2, 2014 by KerikBalm

[Red Iron Crown]

KerikBalm, thanks for that clarification, looks like Jouni is correct. Something new everyday.

[rtxoff]

Even if jet engines are 15x as efficient as they are in real life i do not see problem with that in the game. It's not like they gonna help you much in space. And for flying aroung Kerbin they are just good like they are in the moment. Somehow i would not feel comfortably with the game if they start to consume more fuel then they already do. They purpose is to build fun to fly planes in Kerbin's atmosphere and for this they are perfect.

[Dodgey]

They are fine as they are for the moment. They aren't stupidlly overpowered and they sort of compensate for the soup that is kerbins aerodynamic model. Once the aerodynamic system is given a once over fix them then. I don't think they are too much of a worry to, well, worry about.

[KerikBalm]

Basically:

* You can count air as propellant in your equations, and then use the real exhaust velocity/ISP (which will be lower than that of a rocket)

or

* You can only count the fuel as propellant in your equations, and then use the effective exhaust velocity/ISP (which will be much higher than that of a rocket)

If you do both, you will end up with something way too powerful.

KSP does both, and ends up with something way too powerful.

Of course, KSP has signficant deviations from reality.

Most obvious of which is the scale, and the "soupy" atmosphere and lift equations.

If the lift and drag calculations are "fixed" (made more realistic), jet engines will be way way way too powerful.

Meanwhile, if the scale is fixed - as I found out by installing the RSS mod - most KSP parts are way way way too underpowered.

IRL, fuel tanks have wet:dry ratios of 13-17:1 In KSP its 8:1 or 9:1 for rocket fuel, and about half that for the jet fuel tanks

IRL, rocket engines mostly have TWR of over 80:1 -> most in KSP are under 20, and only one acheives 30:1

IRL, the best chemical rockets (Liquid hydrogen, Liquid oxygen) get an ISP of 421 s, in KSP, the best are only 390s

IRL, the orbital velocity is much higher (just under 8 km/s in LEO instead of just over 2km/s for LKO)

So, for the whole combination to be realistic, does not require realistic values all the time.

Due to the small scale, most of the time, things are heavier than they should be, and less efficient than they should be, with some exceptions:

* LV-N: ISP is perfectly realistic, I think the TWR is a little high, but its not way out there. So its not weaker than RL engines

* Ions: thrust is orders of magntude too high relative to energy consumption. IRL burns would be days, weeks, months, or even years. As the engine doesn't support high time warp factors while thrusting, this is an acceptable concession for gameplay purposes (i'd much rather have the engine support ion thrusting while in time warp)

* Jet Engines: ISP is way way too high. This is somewhat mitigated by the "soupy atmosphere". However, their OPness also manifests in another way: They can easily propel a craft to 95% of orbital speed.

There is nothing inherently unrealistic about a ramjet getting to 2,000 m/s in the atmosphere. Its a bit high... Hybrid turbojet/Ramejt engines (as in the SR-71) only get to about half that, but in theory a Scramjet coul get to double that.

The problem is that getting to 2,000 m/s in Kerbin's upper atmosphere basically puts you in orbit, and getting to 4,000 m/s would put you way past escape velocity... whereas on earth, getting to 4,000 m/s only gets you to half of orbital speed.

To keep "proportional realism", one shouldn't be able to get going much faster than 1,000 m/s in Kerbin's atmosphere using jets.

There is a reason spaceplanes are not a reality, but are so easy in KSP, and its primarily due to Kerbin's low orbital speed.

The ISP of jets is too high, but that just means you'd carry more fuel, and ideally the jet tanks would be nearly empty and not weighing you down that much when you switch to rockets anyway (and IRL, the tanks don't weigh nearly as much for how much fuel they hold).

Fixing the lift equation would only make it worse, fixing the drag equations worse still.

To have realistic aerodynamics, while still making it a challenge to get into orbit with spaceplanes, requires lowering their maximum speed. The fuel usage isn't so important, as this is a space game after all.

[Clockwork_werewolf]

I do have a fix without needing to update the atmosphere equations (which I would still like to see but it's not a priority).

Up the mass of the normal intakes from 0.01 to 0.1 without out any increase in intake air. This means to get enough air for an exit should add about 0.8 to 1.5 tons per jet engine making the extra drag much higher and exit very hard.

Second up the mass of ram intakes from 0.01 to 0.5, increase their intake surface from 0.01 to 0.08 and decrease the airintake storage from 0.2 to 0.15. This means one or two rams per engine would work like 8 or 16 intakes BUT the mass means one would add 0.5 tons and 2 would be an extra ton PER engine and extra air friction. Lastly the reduced air storage means one ram might run one engine but it couldn't reliably run 2 engines.

Normal intakes would be lightest and fine for low to medium atmosphere but useless in upper atmosphere. Rams would be heavier with no benefit for low atmo but much better for high atmo. You could still build SSTO's and in fact people wouldn't need to Air hog or build unrealistic looking craft like the primary picture. The extra weight would also mean those pictured craft would have a very low TWR but Vtols with normal intake would still be light enough.

[rtxoff]

Well every time i tried to get to near orbital speeds with jet engines i failed. It is not possible if you are not spamming intakes and such.

Stay with one intake per jet and build planes that actually look like planes then you won't get the orbital speeds you are talking about because you won't get high enough and atmosphere will prevent you from reaching this speeds.

[Kasuha]

Well every time i tried to get to near orbital speeds with jet engines i failed. It is not possible if you are not spamming intakes and such.

Stay with one intake per jet and build planes that actually look like planes then you won't get the orbital speeds you are talking about because you won't get high enough and atmosphere will prevent you from reaching this speeds.

Let me prove you wrong here. The design below uses one intake per engine. And that was not the highest speed I achieved.

Moreover, I don't see intake spam as a problem. Real planes use intake area corresponding to the plane's intended flight conditions. In KSP intakes have fixed area so the equivalent of reality is scaling their numbers.

Edited June 2, 2014 by Kasuha

[Jouni]

Even if jet engines are 15x as efficient as they are in real life i do not see problem with that in the game. It's not like they gonna help you much in space. And for flying aroung Kerbin they are just good like they are in the moment. Somehow i would not feel comfortably with the game if they start to consume more fuel then they already do. They purpose is to build fun to fly planes in Kerbin's atmosphere and for this they are perfect.

Try increasing the Isp of the LFB and the KS-25x4 engine cluster to 5000 s / 6000 s. That's about how ridiculous the current jet engines are when used in rockets. When part costs are introduced, there will probably be no point in building traditional rockets for payloads smaller than 100-200 tonnes, when you can just hack quickly together a small jet-boosted rocket to do the job.

Squad basically has two choices: nerf spaceplanes or make rockets pointless.

[Kasuha]

Try increasing the Isp of the LFB and the KS-25x4 engine cluster to 5000 s / 6000 s. That's about how ridiculous the current jet engines are when used in rockets. When part costs are introduced, there will probably be no point in building traditional rockets for payloads smaller than 100-200 tonnes, when you can just hack quickly together a small jet-boosted rocket to do the job.

Squad basically has two choices: nerf spaceplanes or make rockets pointless.

We all know that current part prices are placeholders. HarvesteR has mentioned that in his latest update as well. So let's just wait for how parts will be priced in 0.24.

If the jet engine is expensive, wou'll think twice before you slap it on rocket without any changes in physics.

[BudgetHedgehog]

Squad basically has two choices: nerf spaceplanes or make rockets pointless.

Or they could just not do either. Spaceplanes already have their limitations - it's pretty hard to get a sizable payload into orbit with one. For some things, you simply have to use a rocket. For everything else, there's Mastercard spaceplanes.

And speculating about how the cost and budget will be implemented is pointless right now, because we don't know. For smaller payloads, it might even be cheaper to send up an SSTO rocket than a whole plane (think about it - enough fuel to get into orbit, one strong enough engine to lift it and one decoupler could well be cheaper than intakes, jets, rockets, wings, cockpits, control surfaces etc etc)

[Jouni]

We all know that current part prices are placeholders. HarvesteR has mentioned that in his latest update as well. So let's just wait for how parts will be priced in 0.24.

If the jet engine is expensive, wou'll think twice before you slap it on rocket without any changes in physics.

Part prices don't really matter when we're talking about 100x differences in fuel efficiency. If you make jet engines cheap enough for spaceplanes, they'll be cheap enough for rockets as well. If you make them too expensive for rockets, it'll be cheaper to build spaceplanes around SSTO rockets.

Or they could just not do either. Spaceplanes already have their limitations - it's pretty hard to get a sizable payload into orbit with one. For some things, you simply have to use a rocket. For everything else, there's Mastercard spaceplanes.

Spaceplanes are the hard mode, where you intentionally add wings, control surfaces, and other unwieldy decorative parts to a jet-boosted rocket. Without those decorations, jet-boosted rockets scale up easily for 100-tonne payloads.

[rkman]

Though OP: try and re-create those little 'rockets' on your post with FAR installed and without using Alt + F12... You'll find it quite impossible!

I did not make those craft, and the suggestion to use FAR to make it more difficult confirms my point that stock jet engines are overpowered.

[Kasuha]

Part prices don't really matter when we're talking about 100x differences in fuel efficiency. If you make jet engines cheap enough for spaceplanes, they'll be cheap enough for rockets as well. If you make them too expensive for rockets, it'll be cheaper to build spaceplanes around SSTO rockets.

Do you have any numbers to prove your point with 100x efficiency? My results don't correspond to that.

Though OP: try and re-create those little 'rockets' on your post with FAR installed and without using Alt + F12... You'll find it quite impossible!

I did not make those craft, and the suggestion to use FAR to make it more difficult confirms my point that stock jet engines are overpowered.

FAR makes it easier to get Jets to orbit. Eventual aerodynamic problems of these ships can be fixed by rearranging parts. And you don't need debug menu to build them.

[KerikBalm]

Try increasing the Isp of the LFB and the KS-25x4 engine cluster to 5000 s / 6000 s. That's about how ridiculous the current jet engines are when used in rockets. When part costs are introduced, there will probably be no point in building traditional rockets for payloads smaller than 100-200 tonnes, when you can just hack quickly together a small jet-boosted rocket to do the job.

Squad basically has two choices: nerf spaceplanes or make rockets pointless.

I fully acknowledge that they are OP'd.

But since I do most of my flying outside atmospheres with oxygen... meh....

But I do get your point, while it was fun to make SSTOs that hauled 100 ton payloads to orbit, it was also very unrealistic, and would obsolete rockets when part costs are introduced.

However, I like to focus on the payload, how I get it to orbit is the boring part, IMO. If I haul it up with my 100 ton payload lifter, or fire it up on a SLS-theme stack, it doesn't matter much for the rest of the mission, so I don't care much if the first few minutes are on a rocket or on a spaceplane.

OFC, this sort of argument would also work for just letting people start with what they built already in orbit... so.. yea....

They are OP'd as I said, but I think the ratio of max speed to orbital speed is a bigger "OP" factor than their 15x too high ISP.

[NWM]

at once - the drag is usually much higher in KSP due to its dump aerodynamic model.

there is no afterburner for turbojets - these should be solved before more giving realism of thrust.

The jet engines are short, so they ruins the center of mass more - the would be attached in an engine body or nacelle - this also must solved before increasing their mass.

[Kasuha]

I think you're forgetting one important factor about Turbojets.

They're fun. It's fun to build a spaceplane that can reach orbit and fly it.

If you nerf them for "realism", they won't be by far as fun as they are now.

Now, if you want to play a realistic game, you definitely have a choice: Orbiter.

I don't think we need two Orbiter games.

I fully agree that (space-)planes and classical rockets need to be balanced for the Career game. But I don't think nerfing substantially any flight characteristics of Turbojets is the right way to go.

[TeeGee]

Cmon guys the problem isn't the stats or the thrust, it's the fact that we can operate these jets at ridiculous altitudes at stupid temperatures. Jets LOSE thrust RAPIDLY when air gets thinner, while these jets can continue operating at super high levels in comparison to air thickness on Kerbal You CANNOT reach mach 5 in kerbal's soupy lower atmosphere. When our planes reach 15 km the turbojet goes into overdrive pushing our craft into well over Mach and CONTINUES to do so the higher we get.

YES the thrust does decrease BUT AT VERY VERY SLOW AMOUNTS per km vertical altitude gained.

The second point is, turbojets require fans in their engine nacelle. The fans MELT at high temperatures, like the kind we find at higher Mach speeds. Simply put, the engines both LOSE thrust at higher altidues regardless of air intakes AND the turbofan blades MELT because the speed to waaaay too high to tolerate. The HOTTER the air, the harder airbreathing engines have in turning that superheated air into meaningful thrust because engine parts start turning to goo.

How can we balance these jet engines in stock ksp?

1) Have intake air INDEPENDENT on the number of ram intake/circular intakes we put on our craft

2) Heating needs to be applied to engines

3) Engines should lose thrust dramatically at higher than max ceiling height levels.

Stock KSP will be rebalanced to nerf these engines before its release because as of right now, they render EVERY other engine in the game USELESS in space travel. I've gotten 200+ tons of payload INTO ORBIT via SSTO's using ONLY turbojet engines as boosters. It's ridiculous and I agree that needs to change.

[BudgetHedgehog]

The second point is, turbojets require fans in their engine nacelle. The fans MELT at high temperatures, like the kind we find at higher Mach speeds. Simply put, the engines both LOSE thrust at higher altidues regardless of air intakes AND the turbofan blades MELT because the speed to waaaay too high to tolerate. The HOTTER the air, the harder airbreathing engines have in turning that superheated air into meaningful thrust because engine parts start turning to goo.

Then make the intercooler actually do something. Interstellar models this - if you don't have one attached to a RAPIER or SABRE, it'll overheat at high speeds and explode.

[KerikBalm]

I think you're forgetting one important factor about Turbojets.

They're fun. It's fun to build a spaceplane that can reach orbit and fly it.

If you nerf them for "realism", they won't be by far as fun as they are now.

Actually, I found it quite fun to make spaceplanes before optimizing them for flight at near orbital speed/abusing them.

My earliest SSTO designs were barely getting over 10 km and 700 m/s before I'd cut jets and fire rockets.

It may also add some fun to overcome the challenge of having to use two separate types of jet propulsion systems.

Lets say basic jets have a velocity curve like:

velocityCurve

{

key = 750 0 0 0

key = 500 0.6 0 0

key = 0 1 0 0

}

And we rename turbojets to "Ram Jets" (or SCRAM jets), and give them a velocity curve like

velocityCurve

{

key = 2000 0 0 0

key = 1500 0.5 0 0

key = 750 1 0 0

key = 500 0.4 0 0

key = 300 0 0 0

}

This would lower the maximum speed, so you can't simply thrust in atmosphere, get your apoapsis way out there, and then raise your peri with Ions.

I have quite a few designs that use 0 oxidizer or monoprop to get to orbit....

It would also require two types of jets, as the ramjets would be non functional at the start (or like some real designs, launch it with SRBs or LRBs to get it up to speed)

The fun is in the challenge. Right now, the challenge is in making the most preposterous/unbelievable SSTOs you can.

SS to Duna and back.... sure...

I had a lot of fun with my early SSTOs that just barely got to orbit... more fun than making my SSTOs carrying 100 ton payloads actually.

I guess a good interim solution is to only use basic jet engines for SSTO designs

[rtxoff]

I think you're forgetting one important factor about Turbojets.

They're fun. It's fun to build a spaceplane that can reach orbit and fly it.

If you nerf them for "realism", they won't be by far as fun as they are now.

Now, if you want to play a realistic game, you definitely have a choice: Orbiter.

I don't think we need two Orbiter games.

I fully agree that (space-)planes and classical rockets need to be balanced for the Career game. But I don't think nerfing substantially any flight characteristics of Turbojets is the right way to go.

Fully agreed and btw. i think i know now why my jets do not work that good as yours, it's the interstellar mod. It's giving me hard time to get into orbit with them, at least without precoolers.

[Jouni]

I think you're forgetting one important factor about Turbojets.

They're fun. It's fun to build a spaceplane that can reach orbit and fly it.

If you nerf them for "realism", they won't be by far as fun as they are now.

Now, if you want to play a realistic game, you definitely have a choice: Orbiter.

I don't think we need two Orbiter games.

They're about as fun as LFBs with Isp 5000 s / 6000 s: good for building utterly ridiculous rockets, but nothing else. I gave up on using turbojets some time after I learned the proper ascent path, because they didn't make any sense. At first, it was amusing that small rockets could lift large payloads, but eventually turbojets started feeling more like some magic Star Wars engines than anything that should be in KSP.

I fully agree that (space-)planes and classical rockets need to be balanced for the Career game. But I don't think nerfing substantially any flight characteristics of Turbojets is the right way to go.

Is there any other way to balance them?

Do you have any numbers to prove your point with 100x efficiency? My results don't correspond to that.

The nominal Isp of jet engines is in low thousands, while the actual Isp in in tens of thousands. This can be easily tested by burning a tonne of fuel on launchpad with different engines generating the same thrust.

[wizzlebippi]

How to balance jet engines? Keep Isp constant and scale thrust with dynamic pressure/Mach number. The biggest catch is Mach number involves actually computing airspeed and not basing everything off ground speed.

[rkman]

I was inspired to toy around with micro airplanes and jet assisted rockets. It's a bit awkward to design because 99% of the mass of the craft is in the engine, needs extenders to get wings behind the center of mass - and then i noticed the mass of the ram air intake is 0.01.

10kg - that must be a typo by the devs, right?