[1.0.5] Advanced Jet Engine v2.6.1 - Feb 1

[blowfish]

Out of curiosity, what are the displayed Isp values for the B9 Sabres ASL for you? I was a little surprised about 20k today

Surely i have something installed incorrectly, but i would like to know for sure before i delve in the mysteries of my gamedata folder...

It's not too unreasonable. Remember that the SABRE uses hydrogen fuel and AJE assumes this energy density regardless of whether you have RF installed.

[PTNLemay]

Realism Overhaul has this listed as a dependent mod (which I take to mean it's really important for the proper function over the final modded game overall). I worry that, from my understanding all this will do really is... nerf the engines. Does this mean it will be impossible (or just near-impossibly difficult) to create spaceplanes that can reach orbit? Because that's plenty difficult already.

[Tellion]

It's not too unreasonable. Remember that the SABRE uses hydrogen fuel and AJE assumes this energy density regardless of whether you have RF installed.

So I can get away with that? Nice! Though I am not sure how we get to those numbers. It drops to something like 5k Isp at 10km, which still seems like alot... care to elaborate? Oh and yes, I am using realfuels.

Realism Overhaul has this listed as a dependent mod (which I take to mean it's really important for the proper function over the final modded game overall). I worry that, from my understanding all this will do really is... nerf the engines. Does this mean it will be impossible (or just near-impossibly difficult) to create spaceplanes that can reach orbit? Because that's plenty difficult already.

Looking at all the spaceplanes we have got now, I would say yes

No seriously, it will nerf them into being realistic - and that makes building your own Spaceplanes really damn hard if doable at all. (If you manage it, I bow to you^^)

I am currently playing a 6.4x rescale game and have not made a single one yet, and its easier here than in RSS. I was a little sad about that initially, but trust me, building space shuttles is just as fun! Also you start appreciating the ingenuity behind it all.

Edited January 8, 2015 by Tellion

[blowfish]

So I can get away with that? Nice! Though I am not sure how we get to those numbers. It drops to something like 5k Isp at 10km, which still seems like alot... care to elaborate? Oh and yes, I am using realfuels.

The energy density of hydrogen is about 3 times that of kerosene, so one would expect a comparable increase in Isp compared to regular jets. Of course energy density isn't the only factor determining Isp but it's still a useful comparison.

[camlost]

Realism Overhaul has this listed as a dependent mod (which I take to mean it's really important for the proper function over the final modded game overall). I worry that, from my understanding all this will do really is... nerf the engines. Does this mean it will be impossible (or just near-impossibly difficult) to create spaceplanes that can reach orbit? Because that's plenty difficult already.

1.Since you're already using RSS, the increase in SSTO difficulties is trivial, since you'll spend most of the fuel in rocket mode anyway. Someone posted a working skylon pages back. What really is affected is aircraft designs

2. It's not always a nerf. At certain points it is more powerful than without AJE. It's not my intention to nerf or buff anything, I just want realism and realism is all I want. I never care about disabling or enabling certain kinds of designs.

3. I appreciate the endorsement from RO guys, but RO still works without AJE, just missing some realism.

@Tellion, As said, the LH2 has much higher combustion heat than kerosene. But it also has very low density, so you'll need big tanks. What's really bad is when you don't have RF.

Also keep in mind that Isp is heavily dependent on throttle, speed, temperature.

Also, SABRE is an imaginary engine with little data, that is to some extent different from a turbojet. So don't expect too much realism, I can only say it makes more sense than the stock.

[DivisionByZero]

Camlost,

Do you have a post in this thread with the details of how you currently handle pre-coolers and the Sabre/Rapier engines? I'd think you could adjust the thermodynamic cycle calculated with the AJE code by adding a serious intercooler on the compressor stage. Do you have a write-up on handling these parts somewhere in the thread or do I have to wade through the code?

[camlost]

Camlost,

Do you have a post in this thread with the details of how you currently handle pre-coolers and the Sabre/Rapier engines? I'd think you could adjust the thermodynamic cycle calculated with the AJE code by adding a serious intercooler on the compressor stage. Do you have a write-up on handling these parts somewhere in the thread or do I have to wade through the code?

Currently there isn't an intercooler. I just pretend that SABRE is a turbojet with super-high temperature tolerant compressor.

The reason is I don't think it makes too much difference, a cooler merely transfers energy from before the compressor to after the compressor.

[DivisionByZero]

Currently there isn't an intercooler. I just pretend that SABRE is a turbojet with super-high temperature tolerant compressor.

The reason is I don't think it makes too much difference, a cooler merely transfers energy from before the compressor to after the compressor.

Ok, good to know. Given your penchant for realism, I figured you might have had something for the precooler.

[blowfish]

The SABRE really operates differently from normal jets - in a normal jet the turbine and compressor exchange work directly, whereas in the SABRE the helium loop acts as a heat engine, taking heat from the incoming air, doing work in the compressor, and dumping waste heat into the fuel. There's some data on how it operates but it may not be enough to create a full model.

[DivisionByZero]

huh... I'll have to take a second look now because I could have sworn it was preheating the cryogenic hydrogen to cool the compressed intake air and there wasn't any helium at all...

[blowfish]

huh... I'll have to take a second look now because I could have sworn it was preheating the cryogenic hydrogen to cool the compressed intake air and there wasn't any helium at all...

You'll definitely see the helium loop mentioned if you look around. There are two basic ideas involved: 1) Running fuel directly through the precook is problematic for various reasons so you want to have some inert intermediary and 2) If you're running that cycle, why not extract useful work from it?

[camlost]

The SABRE really operates differently from normal jets - in a normal jet the turbine and compressor exchange work directly, whereas in the SABRE the helium loop acts as a heat engine, taking heat from the incoming air, doing work in the compressor, and dumping waste heat into the fuel. There's some data on how it operates but it may not be enough to create a full model.

How's that different from what I said?

OK a little different, but that is on another level of simulation, we need to reach that level first, by making the compressor more real with performance maps

[DivisionByZero]

You'll definitely see the helium loop mentioned if you look around. There are two basic ideas involved: 1) Running fuel directly through the precook is problematic for various reasons so you want to have some inert intermediary and 2) If you're running that cycle, why not extract useful work from it?

You are right. Hydrogen entitlement is the reason for the loop.

[Hyomoto]

I tired out a propeller from KAX, but had several problems with it. I'm sure at least one is my lack of understanding but I want to know if any of this is intended behavior or something else is conflicting. First off, when I started the prop at zero throttle it started up, then died out and wouldn't restart without reverting to launch. So, I tried pushing the throttle up to full and as soon as I pressed space the place started moving forwards, became uncontrollable and veered off to the left. I thought maybe the landing gear was positioned badly so I moved it to the fuselage from the wings and it proceeded to perform the same maneuver.

As far as I can tell the hard pulling to the left and the instant application of thrust are probably not intended, but it could obviously just be a limitation in how the propellers are handled. Like I said, I'm trying to figure out if this is expected or if something might be conflicting, any thoughts?

[Genius Evil]

Here's some pictures of some piston engine models (to replace the rescaled Firespitter parts) that I've been working on:

Javascript is disabled. View full album

The spinner and cowl textures are tweakable with FSTextureSwitcher, and the cowl style can be tweaked via FSMeshSwitcher. Through the magic of extreme cfg welding, the engines are patched together from several different meshes -- blades, spinners, the cowl and the cylinders (one cylinder, copied many times) are all separate meshes. This means I don't have to model every single engine. The entire mod takes up about 3MB so far (thanks to cfg welding), and probably won't go above 5MB.

Edited January 11, 2015 by Genius Evil

[blowfish]

Ok, here's a question for everyone: What's the smallest SSTO you've built with AJE (with or without RF)? What ascent path did you use? The smallest I've managed is about 23t, which is significantly heavier than the smallest pure rocket SSTO I've built (about 14t).

[Tellion]

Suppose I want to buff the thrust of the AJE modified Sabres, would editing it (and it only) in the config file mess with AJE's function or would it be fine?

Edited January 12, 2015 by Tellion

[Genius Evil]

Tellion: If you mean thrust in rocket mode, it should not affect AJE's function, as far as I know. Just be sure to change it in the "B9_Sabre.cfg" file in the AJE folder. If you want to buff atmospheric thrust, you will have to go into "B9_Sabre.cfg" and increase the "acore" value for the engine you want to change. If you're worried about breaking AJE, make a copy of the cfg file. If editing it breaks AJE, just replace it with the copy.

[Tellion]

I cannot seem to find an "acore" value in B9_Sabre.cfg.

This

@PART[b9_Engine_SABRE_*]:FOR[AJE]:NEEDS[RealFuels]{

@MODULE[ModuleEnginesFX],*

{

@heatProduction = 225

}

@MODULE[ModuleEnginesFX],0

{

@PROPELLANT[LiquidFuel]

{

@name = LqdHydrogen

@ratio = 1

@DrawGauge = True

}

!PROPELLANT[intakeAir]{}

}

@MODULE[ModuleEnginesFX],1

{

@PROPELLANT[LiquidFuel]

{

@name = LqdHydrogen

@ratio = 0.73

@DrawGauge = True

}

@PROPELLANT[Oxidizer]

{

@name = LqdOxygen

@ratio = 0.27

%DrawGauge = True

}

}

}

@PART[b9_Engine_SABRE_S]:FOR[AJE]

{

@title = SABRE S

MODULE

{

name = AJEModule

Area = 0.6

BPR = 0

CPR = 12

FPR = 0

Mdes = 1

Tdes = 280

eta_c = 0.95

eta_t = 0.98

eta_n = 0.9

FHV =92000000

TIT = 900

TAB = 3800

exhaustMixer = True

maxThrust = 400

maxT3 = 2000

}

}

@PART[b9_Engine_SABRE_M]:FOR[AJE]

{

@title = SABRE M

MODULE

{

name = AJEModule

Area = 2.4

BPR = 0

CPR = 12

FPR = 0

Mdes = 1

Tdes = 280

eta_c = 0.95

eta_t = 0.98

eta_n = 0.9

FHV = 92000000

TIT = 900

TAB = 3800

exhaustMixer = True

maxThrust = 400

maxT3 = 2000

}

}

@PART[b9_Engine_SABRE_M]:FOR[AJE]:NEEDS[RealSolarSystem]

{

@MODULE[ModuleEnginesFX],1

{

@maxThrust = 1350

}

}

is the relevant section.

[blowfish]

It used to be acore, now it's just Area. Thrust should be proportional so just scale it how you want.

[camlost]

Area is proportional to thrust, but you'll need more inlet area as well. For more stuff about the parameters and physics look for my other repo on github.

Nonetheless, I don't see why you'll want to buff the thrust of SABRE.

[blowfish]

Nonetheless, I don't see why you'll want to buff the thrust of SABRE.

The static thrust got nerfed significantly between versions 1 and 2. I believe that it makes more sense now but I can understand why some people might be unhappy.

Also, I've been through the code and made a few notes:

• AJESolver.cs, line 256: Isn't mdot the total mass flux of fuel+air? Doesn't that mean that ram drag is being subtracted for fuel too and not just air?
  
• AJESolver.cs, line 249-251: I'm pretty sure 250-251 repeats the calculation on 249
  
• AJESolver.cs (AJESolverTester only), line 275: 3600 should be changed to 101972 to reflect the change from imperial to SI units.
  
• For the compressor calculation iteration, wouldn't it be more computationally efficient to start with the prat3 of the previous thrust calculation (which is likely to be under similar conditions) rather than starting from the design CPR every time? Of course this might create an issue if the engine is shut down and then reactivated...
  

I'm also thinking about how to implement variable bypass, but it seems to get more complicated the more I think about it. The basic idea would be to modulate the core/bypass areas in order to simulate bypass doors opening and closing, which is simple enough. The trouble is how to control it. You might want to close bypass doors in order to create more thrust for takeoff or at high altitudes to allow supercruise. On the other hand, you might want more bypass at very high speeds in order to create a turboramjet effect (since this is basically how the J58 operates). The only way I can think of to give the right amount of control is via action groups, but it might still be difficult to operate in practice.

[NathanKell]

Genius Evil, those are so pretty you're tempting me to redo all my planes!

[Tellion]

Thanks for the replies, they were very helpful!

As for why I am curious about buffing them, their initial TWR in air breathing mode is around 4 per engine with a thrust of ~425kN and an Isp of 4000s, coming down from 20000 at engine start (the engine+precooler+intake assembly weighs about 10 tons without any propellant). Playing a 6.4x rescale install, anything that can even remotely dream of having the label "SSTO" on it will not make it off the runway by far, which is also due to the runway being a bit to short for the task at hand, but I digress...

I also found a document by R.E.L stating that the thrust of a sabre in air breathing mode is estimated to be between 0.8 and 2 mN, which did at first give me confidence in the idea... After some more reading i am slowly arriving at the conclusion that I may have been a little shortsighted and need to do even more reading.

[blowfish]

Thanks for the replies, they were very helpful!

As for why I am curious about buffing them, their initial TWR in air breathing mode is around 4 per engine with a thrust of ~425kN and an Isp of 4000s, coming down from 20000 at engine start (the engine+precooler+intake assembly weighs about 10 tons without any propellant). Playing a 6.4x rescale install, anything that can even remotely dream of having the label "SSTO" on it will not make it off the runway by far, which is also due to the runway being a bit to short for the task at hand, but I digress...

I also found a document by R.E.L stating that the thrust of a sabre in air breathing mode is estimated to be between 0.8 and 2 mN, which did at first give me confidence in the idea... After some more reading i am slowly arriving at the conclusion that I may have been a little shortsighted and need to do even more reading.

The RL SABRE is a lot bigger than B9's version - about 3.8m in diameter at the intake and closer to 5m at the nozzle - it's a much bigger engine than B9's, which are really scaled for stock Kerbin.