Basic jet engine vs. R.A.P.I.E.R jet engine

[Magma]

0.23 came with the rapier jet engine, a hybrid engine of a jet engine and a rocket engine.

I tested the new rapier on one of my aircraft (project Hawk). The results were not as spectacular as i'd hope.

Javascript is disabled. View full album

The first 2 pictures shows Hawk with jet engines.

At the end of the runway, this version reached a speed of 130m/s, a thrust of 138.6kN and a specific impulse of 1995.8s.

the version with rapier engines only reached a speed of 87.7m/s, a thrust of 97kN and a specific impulse of 832.5s.

in both instances the breaks were off and both dropped their tail from the runway.

though Jet-Hawk preformed much better then Rapier-Hawk, who nearly crashed because of a stall.

My question:

Why is the specific impulse of a rapier jet engine so low?

[Murph]

Yeah, I'm somewhat underwhelmed by the thrust and impulse of the RAPIER. In fact, I think I'd go as far as to say that I'm actually significantly disappointed by it. I'm very glad that they have added the engine concept, just not very happy with the current performance of it.

[FenrirWolf]

It's the trade-off you get for the convenience of essentially having two kinds of engine in one neat little package. If its values were too high it would push the other jets into obscurity instead of properly filling its niche.

[Murph]

It's the trade-off you get for the convenience of essentially having two kinds of engine in one neat little package. If its values were too high it would push the other jets into obscurity instead of properly filling its niche.

Well, to be honest, I'm kinda feeling that it falls short of its niche, when twin turbojets combined with an Aerospike outperform twin RAPIERs. I love the concept, but it gives significantly lower performance and worse fuel economy, so it's hard to justify using it.

[Hyomoto]

I believe this was discussed during EVERY live stream. How many people played with the RAPIER engine? Tanuki Chau responded to some comments of whether she thought it was OP by saying that the ISP balances it out. Of course, some changes may have been made between KerbalKon and 0.23 release but it seemed covered. I'm a bit relieved, every engine in KSP fills a niche right now and I think it is better to have unique parts than better ones right now. It's sort of why I've stayed away from Interstellar, and why KW bugs me. Sure, all of KW's engines balance against one another, but every single one is better than the stock alternatives.

If nothing else it serves as a template to the obligatory better one someone will mod in.

[Murph]

That's the thing. What I'm saying is that I feel that the RAPIER actually doesn't fill a niche, unless there's a niche for a fuel-hungry, underpowered engine. It's early days, I'll grant you, but the testing I've done so far tells me that I'm better off ignoring the RAPIER and sticking to turbojets combined with Aerospikes. I'll gladly put up with having to use a couple of action groups compared with poor power and excessive fuel consumption (and those 2 attributes combine horribly, since needing to carry more fuel with less power isn't exactly great).

[FenrirWolf]

I'm a bit relieved, every engine in KSP fills a niche right now and I think it is better to have unique parts than better ones right now.

I tend to agree. Jets work great in the air, rockets work great in space, this mainly aims for that hairy bit between the two. Depending on your design or preference the hybrid RAPIER might be just what the doctor ordered, or maybe the more traditional rocket/jet combo will still be better. Making those decisions as you create your craft is a part of the appeal.

Maybe its numbers will be altered if there arises a general consensus that it's underperforming overall, but I'd give it some time to sink in and see what situations people will use it for.

[ravener]

if it is a poor ascent engine, poor jet engine and a poor orbital engine, what is it really good for? if it had the power that other engines of the size lacked you could do a better ascent, but right now it is underpowered. it was meant to be the jack of all trades, but when you can bring multiple specialists it needs to be moderatly ok at something, it is now the jack of nothing.

you dont have to ditch a set of jet engines to keep mass down to make an ssto, so it has that going for it.

[Tidus Klein]

look at the code!

 atmosphereCurve
 	{
		key = 0 1200
		key = 0.3 2500
		key = 1 800
 	}
	velocityCurve
	{
		key = 0 0.5 0 0
		key = 1000 1 0 0
		key = 2000 0.5 0 0
		key = 2200 0 0 0
	}

In other words the rapier reaches max isp at maybe.... 7,000 or 12,000 meters

and reaches max trust at 1000 m/s

hope this answers any questions.

[Bartybum]

Give that the R.A.P.I.E.R engines are a new technology IRL in development AFAIK, I'm disappointed that they're not significantly better ISP and TWR-wise. I mean, is new tech not supposed to be much better than older tech?

[Psycix]

Tidus Klein is correct. The ISP of basic jets, turbojets and rapiers depend on air pressure (altitude) and the thrust depends on speed.

This is why turbojets are worse than basic jets when flying slow and low, but basic jets don't help at all for supersonic high altitude flight.

The rapier is like the turbojet, and thus, sucks at takeoff.

I think it's nicely balanced. It's supposed to perform a bit worse because it is so light (compared to a turbojet + aerospike)

[Murph]

you dont have to ditch a set of jet engines to keep mass down to make an ssto, so it has that going for it.

Yeah, that's the concept, but it doesn't even have that going for it when you have to add an extra fuel tank for the same overall capability, negating the mass saved by having fewer engines.

For reference, most of the testing I've done with it so far has been on a close-to-stock Aeris 4A. Comparing the Aeris with an Aerospike in place of the standard LV-T30, and the Aeris with twin RAPIERs replacing the standard turbojets (and no middle engine). The Aerospike Aeris hits a 100x100 orbit with plenty of spare fuel. The RAPIER Aeris hits 100x100 with much less spare fuel.

[Atoning Unifex]

I agree the RAPIER looks underpowered. Looking at the details for the SABRE engine in development in the UK (according to wikipedia anyway!) it has a design TWR of 14, compared to 5 for conventional jets, and 2 for scramjets. It has a peak ISP of 3200, way above anything else in existance, although it's not specific on what the min ISP is. It also states that the design is "able to provide high thrust from zero speed up to Mach 5.5" so the comment about it sucking on take off is incorrect.

Basically once this SABRE engine exists IRL it should be revolutionary. The RAPIER engine, in comparison, is like a cheap, badly designed knock off!!!

I'd suggest trying the SABRE engines in the B9 aerospace mod if you want real dual mode engines.

[Magma]

-snip-

The rapier is like the turbojet, and thus, sucks at takeoff.

-snip-

Okay, I've taken this to the test with the following results:

Turbo-hawk left the runway at 122.3 m/s, with a thrust of 117.0kN and a specific impulse of 840.4s

no problems occurred during take-off and handling was comparable to Basic jets.

the specific impulse of the Turbojet and RAPIER are more or less equal, though the thrust generated by the RAPIER during takeoff is way less that comfortable

Edited December 18, 2013 by Magma

[SofusRud]

The reason the RAPIER is underpowered compared to the real-world SABRE is the same as why the LV-N has much lower TWR compared to the real-world NERVA, game-play balance. And I for one am ok with it. The advantages RAPIER engine are that can automatically switch, negating the whole "flame-out flat-spin" problem that plagued older SSTOs, and it is not only lighter than having more engines but also means you use fewer parts. A slightly lower thrust and ISP is worth it in my opinion.

[Nao]

I agree the RAPIER looks underpowered. Looking at the details for the SABRE engine in development in the UK (according to wikipedia anyway!) it has a design TWR of 14, compared to 5 for conventional jets, and 2 for scramjets. It has a peak ISP of 3200, way above anything else in existance, although it's not specific on what the min ISP is. It also states that the design is "able to provide high thrust from zero speed up to Mach 5.5" so the comment about it sucking on take off is incorrect.

Basically once this SABRE engine exists IRL it should be revolutionary. The RAPIER engine, in comparison, is like a cheap, badly designed knock off!!!

I'd suggest trying the SABRE engines in the B9 aerospace mod if you want real dual mode engines.

You cant compare KSP engines to RL as the whole scale of simulation is skewed, and no engine in KSP is even close to the performance of RL counterparts. That is by game design and it works pretty well.

Also for jets in KSP, Isp is counted including air, so actual liquid fuel Isp is 16 times more. So even at lowest Isp, the RAPIER has as much Isp as highly efficient real turbofan engine.

I for one like the RAPIER, as it is not blatantly overpowered as some other new additions (48-7S). It has its use and comes very close to alternatives (turbojet+several 48-7S). Just a well designed engine fit for its purpose, just like Skipper.

[Murph]

…and it is not only lighter than having more engines but also means you use fewer parts. A slightly lower thrust and ISP is worth it in my opinion.

That would be great if it was true, but it's not true if you want to achieve the same overall capability. For example, changing from 2 turbojets + 1 aerospike to 2 RAPIERs requires additional fuel for the same capability, negating the weight saving, and possibly negating the part saving. In some cases, more RAPIERs will be required than you would need with turbojets due to the very lacklustre performance in air breathing mode.

It should be great, the concept is great, the current implementation is mediocre at best.

[Nao]

That would be great if it was true, but it's not true if you want to achieve the same overall capability. For example, changing from 2 turbojets + 1 aerospike to 2 RAPIERs requires additional fuel for the same capability, negating the weight saving, and possibly negating the part saving. In some cases, more RAPIERs will be required than you would need with turbojets due to the very lacklustre performance in air breathing mode.

It should be great, the concept is great, the current implementation is mediocre at best.

In the above comparison RAPIERs have more than twice the power of the aerospike, and only 25% less thrust than turbojets. So I would say, that better comparison would be 2 turbojets + 2 aerospikes to two RAPIERs (where the latter would win).

Of course we don't need as much rocket thrust for getting to orbit. But the problem is that RAPIERs are modeled after real life engines (SABRE have 50% more rocket thrust than jet) but KSP ascent profiles are quite different form real life ones.

That said while it would be nice to have some small tweaks like getting more jet power and less rocket power on it. The RAPAIER as it is, is quite well balanced. It Has its niche, and more importantly helps KSP in maintaining the easy to learn hard to master trait, that is VERY important from game design standpoint. It's very much ok that people will fly their first SSTO on RAPIERS, and then as they learn discover that it's possible to be a little more efficient by using more complicated and harder to execute design.

Also having every new thing be straight better then previous works only with DLC or microtransaction stuff. This is 0.23 patch, it's not the first nor the last engine added or tweaked. Let's be happy that they didn't make another 48-7S disaster.

[Murph]

In the above comparison RAPIERs have more than twice the power of the aerospike, and only 25% less thrust than turbojets. So I would say, that better comparison would be 2 turbojets + 2 aerospikes to two RAPIERs (where the latter would win).

Nope, just tried twin turbojets + twin aerospikes, and the twin RAPIERs still lose by a significant margin. I even used a bi-coupler when doubling up on the aerospikes, to try to make it as fair a test as possible. I also don't agree that it's a better comparison, as it's not about having the same number of engines during both ascent phases, but delivering everything in front of the engines into orbit with the most fuel remaining (using that as a metric for payload delivery capability).

I guess the RAPIER is ok for newbies, for lazy inefficiency, and possibly micro-planes, but that's about it from my point of view.

I'm just disappointed that it's basically useless for me. I strongly disagree that it's well balanced  well balanced would be providing an alternative option which gave roughly comparable results, not significantly worse results which push it close to being relatively useless for any practical purpose. To me, the RAPIER has been crippled due to excessive concern about preventing it from being overpowered, which is actually worse than if it had been slightly overpowered. It's meant to be state of the art technology, but what we have today certainly doesn't live up to that for me.

As for the 48-7S, that's not a disaster to me, it's a perfectly reasonable little engine.

[NWM]

You are arguing aout SABRE and RAPIER forgeting one important paramether: SABRE uses hydrogen as fuel. The correct peak specific impulse of the RAPIER would be ~1000s due to hydrogen has over 3x more MJ/kg than liquid fuels!

[Atoning Unifex]

Sorry if this is an obvious question but isn't hydrogen fuel used in rockets stored as a liquid fuel!?

[Murph]

Sorry if this is an obvious question but isn't hydrogen fuel used in rockets stored as a liquid fuel!?

Yes, the O2 and H used in real world rocketry is typically liquid in cryogenic tanks.

It's not explicitly defined in KSP, as far as I'm aware, but I'd guess our liquid fuel is mostly meant to be kerosine, as it's the same used for both jets and rockets (which both use kerosine in the real world, just different grades of it and different additives).

[antbin]

look at the code!

 atmosphereCurve
 	{
		key = 0 1200
		key = 0.3 2500
		key = 1 800
 	}
	velocityCurve
	{
		key = 0 0.5 0 0
		key = 1000 1 0 0
		key = 2000 0.5 0 0
		key = 2200 0 0 0
	}

In other words the rapier reaches max isp at maybe.... 7,000 or 12,000 meters

and reaches max trust at 1000 m/s

hope this answers any questions.

This is great - I'm not at home right now, so I can't check it myself, but has anyone compared the RAPIER atmosphere and velocity curves against the turbojet?

atmosphereCurve
{
key = 0 1200
key = 0.3 2500
key = 1 800
}
velocityCurve
{
key = 0 0.5 0 0
key = 1000 1 0 0
key = 2000 0.5 0 0
key = 2400 0 0 0

So it looks like the RAPIER engine has pretty much identical atmosphere / ISP curves as the turbojet - so it's just a directly-scaled ISP and thrust. Too bad, I was hoping there might be some altitude/speed ranges where Turbojets and RAPIER engines could complement each other.

Edited December 18, 2013 by antbin

[Nao]

Nope, just tried twin turbojets + twin aerospikes, and the twin RAPIERs still lose by a significant margin. I even used a bi-coupler when doubling up on the aerospikes, to try to make it as fair a test as possible. I also don't agree that it's a better comparison, as it's not about having the same number of engines during both ascent phases, but delivering everything in front of the engines into orbit with the most fuel remaining (using that as a metric for payload delivery capability).

I guess the RAPIER is ok for newbies, for lazy inefficiency, and possibly micro-planes, but that's about it from my point of view.

I'm just disappointed that it's basically useless for me. I strongly disagree that it's well balanced  well balanced would be providing an alternative option which gave roughly comparable results, not significantly worse results which push it close to being relatively useless for any practical purpose. To me, the RAPIER has been crippled due to excessive concern about preventing it from being overpowered, which is actually worse than if it had been slightly overpowered. It's meant to be state of the art technology, but what we have today certainly doesn't live up to that for me.

As for the 48-7S, that's not a disaster to me, it's a perfectly reasonable little engine.

Welp, i'll try to not get into big argument about it, but 48-7S is better than 90% of the stock engines, except partcount per thrust of course. Notably its almost always better than: Mainsail, LV-T30, RT-10, LV-909, and for ascents up to ~4500m/s aerospike engine. And this are the "best for the job engines" so i'm not even talking about others.

This also makes any performance flight quite dull and taxing on partcount as its always either 48-7S or LV-N nothing else. It also is quite counter intuitive as IRL efficiency tends to increases with size, while in case of 48-7S is the other way.

As for RAPIER vs jet-spike. It does give comparable performance, i didn't have much time to design new planes yet so i used stock aeris 4A and 2 rapiers have slightly better performance than 2x(jet+spike).

And even 2x rapiers vs 2xjet + 1x aerospike, have quite similar performance. Using the aeris example 2jet-1spike i've got 1350m/s in 85x88km orbit (using the most efficient ascent profile) and 1340m/s at 85x88km orbit for 2x RAPIERs

[WhiteWeasel]

Seriously? How is this better than a mainsail? This has less ISP than a lv 909. It's just a non radial version of the 24-77.

On-topic I feel the RAPIER engine is like the jack of all trades - master of none vibe.

Edited December 18, 2013 by WhiteWeasel