PSA: Aerospike is the new best 1.25m engine!

[Red Iron Crown]

Isn't the entire point of the aerospike to work well in atmosphere?

No, the entire point of an aerospike is to work well across a broad range of atmospheric conditions. It will always lose out to a bell nozzle at the bell's ideal ambient pressure.

[AfailingHORSE]

But it's stackable now ?

With its return to being stackable, the aerospike has truly become one of the best engines again

[Hysterrics]

I used it for a spaceplane and actually got suborbital.
Aerospike is pretty kewl.

[NuclearNut]

The only thing that changed was the model and the fact that it can now be used with fairings. It is a good chemical engine, but weighs a ton and does not vect- uh.. gimbal.

[Agost]

AFAIK real aerospikes can have thrust vectoring by varying the propellant distribution inside the combustion chamber... and in some other ways


Hope it gets added Edited November 16, 2015 by Agost

[r_rolo1]

[quote name='Agost']AFAIK real aerospikes can have thrust vectoring by varying the propellant distribution inside the combustion chamber... and in some other ways


Hope it gets added[/QUOTE]
Don't wait for it :/ The devs aren't even willing to give it back more realistic Isp values ( once upon a time the aerospike actually had them, but it was nerfed due to gameplay considerations ( basically it was too good compared with the LV-30 and 45 ( hum , where did I heard that one recently ? ;) ) and everyone was only using aerospikes nd nukes. IMHO there are other ways to balance the engine nowadays , like for a example, aerospikes in RL are harder to keep cool than bell engines, but who am I to say such things ? :P ) and better not talk about other examples, like how there isn't a single SRB in game that has thrust vectoring ( or even how historically the devs always make them underpowered compared with RL .... ) Edited November 16, 2015 by r_rolo1

[sal_vager]

[quote name='r_rolo1']but who am I to say such things ?[/QUOTE]

Who indeed, [U][URL="http://www.aerospaceweb.org/design/aerospike/losses.shtml"]when aerospikes do not maintain ISP throughout the flight envelope.[/URL][/U]

Theory didn't match reality, and no amount of talking will change that.

[Niemand303]

[quote name='r_rolo1']Don't wait for it :/ The devs aren't even willing to give it back more realistic Isp values ( once upon a time the aerospike actually had them, but it was nerfed due to gameplay considerations ( basically it was too good compared with the LV-30 and 45 ( hum , where did I heard that one recently ? ;) ) and everyone was only using aerospikes nd nukes. IMHO there are other ways to balance the engine nowadays , like for a example, aerospikes in RL are harder to keep cool than bell engines, but who am I to say such things ? :P ) and better not talk about other examples, like how there isn't a single SRB in game that has thrust vectoring ( or even how historically the devs always make them underpowered compared with RL .... )[/QUOTE]

*ahem-ahem* Realism Overhaul *ahem-ahem*

[r_rolo1]

@sal_vager

Well, I never said that the ingame aerospike should have the Isp values of the teoretical ideal aerospike ( that, among other things, has infinite lenght ;) ) or that the aerospikes didn't had flaws in RL ;) I just pointed out that the in game T-1 Isp was nerfed back in the days basically because the LV-30 and 45 were so underpowered compared with it ( you were here at the time, sal, so you must remember it ;) ) and it was never brought up to more realistical values based on RL tests. I also said that IMHO this should had been reverted a long time ago, especially since nowadays we can actually replicate in game some of the RL aerospike weak spots, like the cooling ( hence the fuzz there was about the aerodynamic aerospike ( that, BTW is not the case of the ingame T-1 ) ) or the non ideal performance in atmosphere ( that one is actually in game already , btw ;) ) Edited November 16, 2015 by r_rolo1

[Temeter]

[quote name='NuclearNut']The only thing that changed was the model and the fact that it can now be used with fairings. It is a good chemical engine, but weighs a ton and does not vect- uh.. gimbal.[/QUOTE]

Nah, it's actually the lightest 1.25m engine in terms of thrust-to-weight if my calculations aren't completely off. That's the one thing that surprised me. Edited November 16, 2015 by Temeter

[Alias72]

[quote name='sal_vager']Who indeed, [U][URL="http://www.aerospaceweb.org/design/aerospike/losses.shtml"]when aerospikes do not maintain ISP throughout the flight envelope.[/URL][/U]

Theory didn't match reality, and no amount of talking will change that.[/QUOTE]

ummmm. The paper you linked to included performance enhancing suggestions that compensate, almost entirely, for the truncated aerospike. Namely sufficient bleed and sidewalls, with sufficient bleed being the most important.

so actually your link proves that it is possible to closely match theoretical altitude compensating performance by using certain design tricks.

[selfish_meme]

I think the major problem with it was good vacuum ISP, that's the most unrealistic thing and had it competing with poodles and other dedicated vacuum engines. 3x Terrier give the same vacuum thrust as the aerospike but weigh 1.5 compared to 1.0 and have 345 ISP compared to 340 in vacuum!

[Tourist]

I do like the aerospike. Its the only engine I have used that has allowed me to reach orbit from the runway with spaceplane without needing air-breathing engines to reach the thin air. I get there with just enough fuel to de-orbit (landing again is unfortunately another thing... it likes flipping and entering through the atmosphere butt first or spinning uncontrollably... back to the drawing board).

Not sure how I feel about it being stackable though. Sure, I will probably use it.... but I think there was some game-play benefit with some equipment having quirks that create design complications. I think it actually encourages creativity by forcing you to problem-solve around the quirks.

[NathanKell]

Guess what guys: the aersopike ingame _does_ match theoretical aerospike performance. It removes [I]nozzle losses[/I]. It does not magic away [I]physical laws[/I], like when the atmosphere pushes back at you, you generate less thrust. Period. End.

The aerospike is modeled, ingame, to have very nearly the chamber pressure of the Mammoth (and now the Vector). That means it has nearly as good sea level Isp as them. But it [I]also[/I] has nearly as good vacuum Isp as the Poodle. And it loses very little Isp, in comparative terms, at Eve's surface (5 atmospheres of pressure), something that can be said for no other engine.

[selfish_meme]

[quote name='NathanKell']Guess what guys: the aersopike ingame _does_ match theoretical aerospike performance. It removes [I]nozzle losses[/I]. It does not magic away [I]physical laws[/I], like when the atmosphere pushes back at you, you generate less thrust. Period. End.

The aerospike is modeled, ingame, to have very nearly the chamber pressure of the Mammoth (and now the Vector). That means it has nearly as good sea level Isp as them. But it [I]also[/I] has nearly as good vacuum Isp as the Poodle. And it loses very little Isp, in comparative terms, at Eve's surface (5 atmospheres of pressure), something that can be said for no other engine.[/QUOTE]

Isn't the whole schtick of the aerospike that atmospheric pressure is what makes it eficient so IRL they have to be going fast. (mach*) before their efficiency kicks in. From the way I understand it they are more efficient over a 'range' of atmospheres but can't compete with a tuned bell at individual atmospheres.

[NathanKell]

That is correct, they are, in effect, [I]nearly as efficient as the optimum bell for a given pressure, at any pressure[/I]. Which is precisely what KSP does, and apparently is enough to offend the "no! that means equal efficiency everywhere!!!" crowd. :]

[problemecium]

Uhhh, guys?
Did we all forget about the LV-N?

I mean sure the aerospike is much improved, but it's still got nothing on that 800 vacuum Isp. Low TWR hardly matters when the efficiency is that high, unless we're specifically talking about landers. I guess you could say "best 1.25m lifter engine" or "...chemical engine" maybe, but in general the LV-N is still way ahead of the pack.

[KerikBalm]

Lets not talk about "realistics Isp" values in a game where things take about 1/3 the dV to reach as real life.
Getting to Jupiter from low Earth orbit takes about 3x as much dV as it takes to get to Jool from LKO.
Just getting to low earth orbit takes about 3x as much dV as it takes to get to LKO....
Similar values are found for Earth->Moon and Kerbin->Mun

Considering that, it would be more "realistic" if engine Isp was 1/3 their real life values... enjoy your LV-N with 266 Isp...
But then it would also get a better TWR (as would all rocket engines), and a lot of the mass of parts would come down, tanks would have much better mass ratios.

Forget about realistic Isps (but 340 Isp is quite realistic already as long as your not using lH2... which we don't seem to be using given the liquid fuel has none of the drawbacks of lH2).

[selfish_meme]

[quote name='NathanKell']That is correct, they are, in effect, [I]nearly as efficient as the optimum bell for a given pressure, at any pressure[/I]. Which is precisely what KSP does, and apparently is enough to offend the "no! that means equal efficiency everywhere!!!" crowd. :][/QUOTE]

I must admit my understanding was deficient, I expected a worse vacuum ISP when it is not, also a worse sea level ISP when it wasn't, I guess I expected a curvier curve ;) also greater weight, but it seems they really are all KSP makes them out to be (at least theoretically), the extra weight of the nozzle is compensated for by having greater inherent structural strength, meaning less trusses.

There is a great site here that goes into experimental and theoretical aspects of aerospikes [url]http://www.aerospaceweb.org/design/aerospike/compensation.shtml[/url]

[Red Iron Crown]

[quote name='KerikBalm']Lets not talk about "realistics Isp" values in a game where things take about 1/3 the dV to reach as real life.
Getting to Jupiter from low Earth orbit takes about 3x as much dV as it takes to get to Jool from LKO.
Just getting to low earth orbit takes about 3x as much dV as it takes to get to LKO....
Similar values are found for Earth->Moon and Kerbin->Mun

Considering that, it would be more "realistic" if engine Isp was 1/3 their real life values... enjoy your LV-N with 266 Isp...
But then it would also get a better TWR (as would all rocket engines), and a lot of the mass of parts would come down, tanks would have much better mass ratios.

Forget about realistic Isps (but 340 Isp is quite realistic already as long as your not using lH2... which we don't seem to be using given the liquid fuel has none of the drawbacks of lH2).[/QUOTE]
The smaller size of the KSP solar system is already compensated for with fuel tanks and engines that are very heavy for their capacity/thrust, nerfing Isps would make things more difficult still.

[More Boosters]

[quote name='KerikBalm']Lets not talk about "realistics Isp" values in a game where things take about 1/3 the dV to reach as real life.
Getting to Jupiter from low Earth orbit takes about 3x as much dV as it takes to get to Jool from LKO.
Just getting to low earth orbit takes about 3x as much dV as it takes to get to LKO....
Similar values are found for Earth->Moon and Kerbin->Mun

Considering that, it would be more "realistic" if engine Isp was 1/3 their real life values... enjoy your LV-N with 266 Isp...
But then it would also get a better TWR (as would all rocket engines), and a lot of the mass of parts would come down, tanks would have much better mass ratios.

Forget about realistic Isps (but 340 Isp is quite realistic already as long as your not using lH2... which we don't seem to be using given the liquid fuel has none of the drawbacks of lH2).[/QUOTE]

If this were to be done, then fuel would need to be much cheaper and most everything would need to be upscaled. Better wet/dry mass ratios on fuel tanks too. I mean, Saturn V with no payload had 17k Delta-V, how do you even reach that in KSP without ions or nukes?

[Temeter]

[quote name='parameciumkid']Uhhh, guys?
Did we all forget about the LV-N?

I mean sure the aerospike is much improved, but it's still got nothing on that 800 vacuum Isp. Low TWR hardly matters when the efficiency is that high, unless we're specifically talking about landers. I guess you could say "best 1.25m lifter engine" or "...chemical engine" maybe, but in general the LV-N is still way ahead of the pack.[/QUOTE]

True, didn't even think about the good old nuke. To me that thing is a category of its own.^^

[KerikBalm]

[quote name='Red Iron Crown']The smaller size of the KSP solar system is already compensated for with fuel tanks and engines that are very heavy for their capacity/thrust, nerfing Isps would make things more difficult still.[/QUOTE]

#1) you're pointing out what I already stated, as if I hadn't said it (ditto to MoreBoosters)
As I said if it were to be made more realistic "then it would also get a better TWR (as would all rocket engines), and a lot of the mass of parts would come down, tanks would have much better mass ratios."

#2) Its only partly compensated by that.
Consider a craft with a 35:1 mass ratio, vs a craft with a 9:1 mass ratio (about the best you can get in KSP), ln 35 = 3.55 ln 9 = 2.19
3.55/2.19= 1.61 which is less than 3
The craft could get roughly 61% more dV out of a given stage with the same Isp if we had more realistic mass ratios... If we take this metric, 1.61/3 *800 = 430...
How about giving LV-Ns 430 Isp...
At this point, you may start to notice that the poodle in game is almost as good as a NTR in real life.

#3) The basic argument about using real Isp values is the same as the argument for using real TWRs, or real tank mass ratio. Parts with "real" stats would be very OP in a system where everything takes 1/3 the delta V to do.
Imagine a Saturn V cubed... ie a Saturn "CXXV" that can lift a Saturn "XXV" to orbit, that can lift a Saturn V to orbit....
If we had a rocket that could lift a rocket that could lift an entire Saturn V to orbit, that rocket would be as powerful as a Saturn V would be in the Kerbal system...
So I don't give any credit at all to arguments about "realistic Isp values" precisely because this is not a real system.
The part mass, thrust, Isp, all have to be balanced, and a change in one affects the balance required for the others.
"Real" numbers are simply irrelevant in the Kerbin system.

[Red Iron Crown]

Ah, the old "realism is all or nothing" argument. I don't lend that one much credence, it's entirely possible for a game to be realistic in one or more aspects while unrealistic in others.

[Pecan]

:-)
"Essentially, all models are wrong, but some are useful." [URL="https://en.wikiquote.org/wiki/George_E._P._Box"]George EP Box[/URL]

(The longer version of this quote can also be applied to KSP as, "Since all models are wrong the scientist cannot obtain a "correct" one by excessive elaboration. On the contrary following William of Occam he should seek an economical description of natural phenomena. Just as the ability to devise simple but evocative models is the signature of the great scientist so overelaboration and overparameterization is often the mark of mediocrity.", = it might not be right for your particular playstyle, but it's a damn good approximation appealing to a wide range of audiences)