Jump to content

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

camlost

By camlost
in KSP1 Mod Releases

 Share

Recommended Posts

NathanKell

NathanKell

Posted
Posted
Wat? show me the J58 config pleeease

And what's that diamond shock? Don't seem to recall such an effect in HotRockets?

Here's the repo. For the J58 you'll need both the part cfg and my retexture of the B9 turbojet. AJEcode:


IspMultiplier=1
defaultentype=2
defineenparm= true
acore=6.75
byprat=0
tinlt = 1700
tfan = 2200
tcomp = 2600
tt4 = 2900
tt7 = 3600
prat2 = 1
prat13 = 1
prat3 = 7.5
prat4 = 1
eta2 = 1
eta13 = 1
eta3 = .968
eta4 = .99
eta5 = .982
eta7 = .96
abflag=1
fhv=20000

For the afterburner, I sped it up a bit and increased emission slightly. I've also now made it a bit larger (since the pic was taken).

MODEL_MULTI_PARTICLE_PERSIST
            {
                name = bninejet
                modelName = MP_Nazari/FX/flamejet3
                transformName = thrust_transform
                emission = 0.0 0.0
                emission = 0.6666 0.0
                emission = 0.6667 0.4
                emission = 1.0 3.2
                speed = 0.0 3.96
                speed = 1.0 3.28
                scale = 0.0 1.2 // Rescale the emitters to +0%
                scale = 1.0 1.8 // Rescale the emitters to +0%
                energy = 0.0 0.05 // Same for energy
                energy = 0.7 0.21 // Same for energy
                energy = 1.0 0.99 // Same for energy
                fixedEmissions = false
            }

Link to comment
Share on other sites
camlost

camlost

Posted
  • Author
Posted

Yours is basically a turbojet with very low CPR. Not exactly realistic, but to think about it, most flight simulators are at about this level of realism anyway

Link to comment
Share on other sites
Guest

Guest

Posted
Posted

I have a proposition. Would that be possible to require an engine to be attached to a structural part? Most jet engines in KSP are just nozzles, and on most planes, the rest goes into the hull. Now, the entire mass of the engine is concentrated within the nozzle. Needless to say, this is very wrong, and screws up COM on replicas. To combat this, I propose adding something like "ModuleEngineBody" to structural parts that could house and engine, and upon node-attaching an engine, increasing their mass by an appropriate amount. All fuelless fuselages could get this module added to them, making plane construction a bit more realistic.

Also, what's the status on intake simulation? It just doesn't feel right that engines don't depend on intakes at all. There's a lot of depth to be added there.

Link to comment
Share on other sites
MAKC

MAKC

Posted
Posted
I have a proposition. Would that be possible to require an engine to be attached to a structural part? Most jet engines in KSP are just nozzles, and on most planes, the rest goes into the hull. Now, the entire mass of the engine is concentrated within the nozzle. Needless to say, this is very wrong, and screws up COM on replicas. To combat this, I propose adding something like "ModuleEngineBody" to structural parts that could house and engine, and upon node-attaching an engine, increasing their mass by an appropriate amount. All fuelless fuselages could get this module added to them, making plane construction a bit more realistic.

Also, what's the status on intake simulation? It just doesn't feel right that engines don't depend on intakes at all. There's a lot of depth to be added there.

I believe camlost has already accounted for this by using CoMOffset.

Link to comment
Share on other sites
NathanKell

NathanKell

Posted
Posted
Yours is basically a turbojet with very low CPR. Not exactly realistic, but to think about it, most flight simulators are at about this level of realism anyway

It's actually higher than some of the overall pressure ratios I've seen quoted for the J58 (everything from 5 to 8.5; I went with 7.5 because it let me match SFC better). As you pointed out, there's no way to account for the bleed in EngineSim, so I just upped the max temperatures some.

Link to comment
Share on other sites
Phredward

Phredward

Posted
Posted

I tried to get gimbaling working, but no dice so far. Can anyone spot something I'm doing wrong?

Here's what I changed in AJE.cfg.

Old:

!MODULE[ModuleGimbal]

{

}

New:

!MODULE[ModuleGimbal] {}

!MODULE[ModuleTweakableGimbal] {}

!MODULE[KM_Gimbal] {}

!MODULE[smarterGimbal] {}

MODULE

{

name = KM_Gimbal

gimbalTransformName = NozzleTransform

yawGimbalRange = 17

pitchGimbalRange = 17

responseSpeed = 100

enableRoll = true

}

The gimbaling shows up when I right click on the engine, but seems to do nothing in flight. I started with 7 degrees, and pumped it up to 17 for the second try.

Link to comment
Share on other sites
NathanKell

NathanKell

Posted
Posted

Well, you need to set the gimbalTransformName the same as what the ModuleEngines uses as the thrust transform. (Or, if the model is built such that there is a separate gimbal transform, use that; but few are.)

Link to comment
Share on other sites
Phredward

Phredward

Posted
Posted
Well, you need to set the gimbalTransformName the same as what the ModuleEngines uses as the thrust transform. (Or, if the model is built such that there is a separate gimbal transform, use that; but few are.)

Thanks so much, that was it.

Link to comment
Share on other sites
NathanKell

NathanKell

Posted
Posted

camlost: noticed a few more problems in my J58 entry. Updated as follows:

	MODULE
	{
		name = AJEModule
		IspMultiplier=1
		defaultentype=2
		defineenparm= true
		acore=7.31
		byprat=0
		tinlt = 1700
		tfan = 2200
		tcomp = 2600
		tt4 = 2700
		tt7 = 5300
		prat2 = 1
		prat13 = 1
		prat3 = 6.0
		prat4 = 1
		eta2 = 1
		eta13 = 1
		eta3 = .98
		eta4 = .99
		eta5 = .99
		eta7 = .83
		abflag=1
		fhv=20000
	}

This is still to try to match the dry and wet SFC.

Speaking of, I've run into issues with your J93. What source did you use for the stats? I have found the turbine inlet temperature is 2200F, and I've found one source that states, at least early in the YJ93's development it had an overall pressure ratio of 8.8. But on other stats I've come up short. However, I do have sources for dry SFC of 0.7 and wet of 1.8 which I have been utterly unable to match in AJETester with the rated 85 dry and 128kN wet thrust. In effect it appears to require eta3/4 > 1 and eta7 < 0.7 (although values < 0.7 seem to have no effect).

Link to comment
Share on other sites
camlost

camlost

Posted
  • Author
Posted (edited)
camlost: noticed a few more problems in my J58 entry. Updated as follows:

	MODULE
	{
		name = AJEModule
		IspMultiplier=1
		defaultentype=2
		defineenparm= true
		acore=7.31
		byprat=0
		tinlt = 1700
		tfan = 2200
		tcomp = 2600
		tt4 = 2700
		tt7 = 5300
		prat2 = 1
		prat13 = 1
		prat3 = 6.0
		prat4 = 1
		eta2 = 1
		eta13 = 1
		eta3 = .98
		eta4 = .99
		eta5 = .99
		eta7 = .83
		abflag=1
		fhv=20000
	}

This is still to try to match the dry and wet SFC.

Speaking of, I've run into issues with your J93. What source did you use for the stats? I have found the turbine inlet temperature is 2200F, and I've found one source that states, at least early in the YJ93's development it had an overall pressure ratio of 8.8. But on other stats I've come up short. However, I do have sources for dry SFC of 0.7 and wet of 1.8 which I have been utterly unable to match in AJETester with the rated 85 dry and 128kN wet thrust. In effect it appears to require eta3/4 > 1 and eta7 < 0.7 (although values < 0.7 seem to have no effect).

I couldn't find its CPR so I guessed based on that Wikipedia says it's 11-stage compressor.

According to Wikipedia, GE4 is a civilian design based on YJ93, with 9-stage compressor and CPR=12.5

Edited by camlost
Link to comment
Share on other sites
NathanKell

NathanKell

Posted
Posted

I will have to try the other Firespitter engines! I've tried a couple of the prop engines for some WW2 fun...

With an OPR (not CPR) of 16, and the given compressor temperature, it seems to overheat very much too quickly (it's supposed to be Mach 4-rated in the XB-70, even if they never pushed her that far...), and the SFCs seem way off.

*is* there some way, in AJE as it currently stands, to get low dry SFC and high wet SFC, as engines in that era appeared to have? I keep having too-high dry SFC and too-low wet SFC, even with very low eta7 and >1.0 eta3/4!

Link to comment
Share on other sites
camlost

camlost

Posted
  • Author
Posted

Usually when I get something off I just take a compromise between parameters. The entype means 0=turbojet, 1=turbojet with afterburner 2=turbofan 3=ramjet. If entype=3 the algorithm will neglect compressor. The coding of EngineSim is very obscure and lacks extensibility IMO

Link to comment
Share on other sites
Dirt_Merchant

Dirt_Merchant

Posted
Posted

@camlost: if its not too big of a pain, could you lay out descriptions for the AJEpropellor module quick and dirty style? It would make it way easier to get propellor engines dialed in for my various funzies projects, namely Reno race planes, and various other rowdy design types.

Link to comment
Share on other sites
NathanKell

NathanKell

Posted
Posted

camlost (and others): here's a slight edit of AJE to support variable compressor pressure ratios, keyed to mach.

https://www.dropbox.com/s/z0f72hoozzlcdkh/AJE.zip

Source in zip.

With this, and with a bit more J58 info (I found a source that with bleed fully engaged, TPR lowers to 2.93) I get much better-fitting performance.

	MODULE
	{
		name = AJEModule
		IspMultiplier=1
		defaultentype=2
		defineenparm= true
		acore=9.0
		byprat=0
		tinlt = 1800
		tfan = 2000
		tcomp = 2300
		tt4 = 2450
		tt7 = 4000
		prat2 = 1
		prat13 = 1
		prat3 = 8.5
		prat4 = 1
		eta2 = 1
		eta13 = 1
		eta3 = .99
		eta4 = .99
		eta5 = .75
		eta7 = .75
		abflag=1
		fhv=20000
		usePrat3Curve = True
		prat3Curve
		{
			key = 0.0	8.5	0	0
			key = 2.0	8.5	0	0
			key = 3.0	2.93	0	0
		}
	}

Link to comment
Share on other sites
camlost

camlost

Posted
  • Author
Posted

@Dirt_Merchant, I have made AJE propeller tester and AJE rotor tester already. Will upload in the evening

@NathenKell, very good! Definitely using that code in the next version. Probably will make the stock turbojet J58, and add a .625m ramjet.

I don't have the program with me right now, how much did your performance improve in the high mach region?

Link to comment
Share on other sites
Phredward

Phredward

Posted
Posted

Thanks for the help with the engine gambling! With that I was able to use the F100 + Ramjet as a first stage lifter of a very light craft. As you predicted it made a poor first stage. The thrust is lower than stock jets, and the weight is higher, so the multiple engines needed masses more than a rocket stage. Still fun, and great to see realism advance on all fronts!

Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...