Jump to content

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


camlost

Recommended Posts

They don't actually have a "jet mode" as such - there is a turbocompressor but the combustion side of the engine is pure rocket, so it really should behave like a rocket. There's a pretty old doc by Alan Bond showing peak TWR at around 2.0M with a big falloff either side which I've been trying to replicate but failing miserably through lack of understanding atm - also I don't quite understand the reasons behind that shape - although perhaps to do it properly is a bit out of the scope of AJE anyway ( given it isn't a jet engine ). The later Reaction Engines docs have the SABRE 4's thrust varying from 0.5 to 2MN in airbreathing mode for whatever reason, so you can expect some difficulty using them even if they were perfectly modelled.

Well, the cycle is different, but the SABRE should have many of the same characteristics as a conventional jet. Thrust will grow with speed, as ram pressure increases, top speed will be limited by compressor heating (though when this happens will of course be very different because of the precooling). One of the major differences is that the rising intake temperature actually increases the amount of energy available to the helium loop and thus the compressor, so compression ratio (and thus efficiency) will not drop in the same way it would in a normal jet engine.

- - - Updated - - -

Since I didn't realize that you already had the Saturn engine under quiztech, which was simply to clone as another VSR TRJ for a nice single nozzle variant, here is another engine to add. The SNECMA M88-2 from the Rafale. Sourced here with all the stats :)

And a bit more russian love with the R-15-300 from the MiG-25

Maybe the m53-P2 from the mirage 2000 as well? Source

It's pretty easy to add configs, but I don't want to add a bunch of engines to AJE that I don't have time to test. If you like, I can walk you through creating new engine configs.

- - - Updated - - -

What mods give proper afterburner effects to go with this mod? I like being able to see the afterburner flame when it's on rather than just seeing the engine's insides turn blueish.

Depends on the engine. The J-58 has stock effects. All of the B9 engines have effects included in B9. Most of the others require HotRockets or RealPlume, though I'm not sure the afterburner thresholds are set properly.

Link to comment
Share on other sites

It's pretty easy to add configs, but I don't want to add a bunch of engines to AJE that I don't have time to test. If you like, I can walk you through creating new engine configs.

That'd be great I was going to ask, but wanted to see what you had to say first. Thanks!

Link to comment
Share on other sites

I would have presumed the limits on the SABRE compressor are simple capacity issues - there is a bypass to controllably shed excess intake air rather than have it spill uncontrollably back out the front so one might assume the intake can take more air in than they actually need, and there's going to be hard limits to how much oxidizer the rocket engine can use anyway. I'm not sure why the decrease in thrust after Mach 2 though ( there wasn't anything relating to air density in that graph so it doesn't seem like an ascent profile ), I guess I'll have to try and decypher the paper a bit.

The current RAPIER thrust runs away until it overheats though, which is convenient but doesn't seem right ( plus it ends up using massively more fuel than in vac mode which is definitely not right ).

Link to comment
Share on other sites

I would have presumed the limits on the SABRE compressor are simple capacity issues - there is a bypass to controllably shed excess intake air rather than have it spill uncontrollably back out the front so one might assume the intake can take more air in than they actually need, and there's going to be hard limits to how much oxidizer the rocket engine can use anyway.

Matching intake flow and engine flow in supersonic engines is a nontrivial problem. At subsonic speeds, excess air can just spill around the intake, but at supersonic speeds, there's no way for conditions downstream of the intake to affect what the intake is doing. Flow through the core of the engine is usually limited by choking in the turbine (and the nozzle throat, but that's usually adjustable in supersonic engines). The SABRE doesn't have a turbine to choke, but presumably the nozzle throat is fixed.

Theoretically, there's no limit to how much air you can force through a jet engine - you just need to increase the total pressure at the compressor. In practice though, inefficiencies of the intake begin to reduce the total pressure at higher mach numbers, and of course eventually the engine would be unable to handle that amount of pressure and explode (though it'd be rare that this would actually matter in reality).

I'm not sure why the decrease in thrust after Mach 2 though ( there wasn't anything relating to air density in that graph so it doesn't seem like an ascent profile ), I guess I'll have to try and decypher the paper a bit.

Yeah, I'm not quite sure why that would be. If you figure it out let me know :)

The current RAPIER thrust runs away until it overheats though, which is convenient but doesn't seem right ( plus it ends up using massively more fuel than in vac mode which is definitely not right ).

Link to comment
Share on other sites

The SABRE doesn't have a turbine to choke, but presumably the nozzle throat is fixed.

Theoretically, there's no limit to how much air you can force through a jet engine - you just need to increase the total pressure at the compressor. In practice though, inefficiencies of the intake begin to reduce the total pressure at higher mach numbers, and of course eventually the engine would be unable to handle that amount of pressure and explode (though it'd be rare that this would actually matter in reality).

Well there's a bit more than the compressor throat, there's the rocket combustion chamber also & the connections between that & the bells so that would quite likely destroy anything resembling a jetpipe flow. The rocket design seems a bit wierd too, why the four bells? but I know even less about rocket design...

Haven't worked out the M2+ thrust loss yet but it did occur to me that slowing the airflow down at that point might start overwhelming the cooler, so that is my current point of investigation.

Overheating is normal. The only reason why the SABRE (or an afterburning jet) would loose thrust at high mach numbers is because the intake causes total pressure loss, and that only becomes large enough to overcome the increasing ram pressure past mach 6 or 7. The rocket mode thrust might be lower than it should be, but camlost didn't want to change that.

The rocket shouldn't overheat though - from what I understand any imbalance in propellant mix is going to reduce temps. At it's best it'd seem the precooler/compressor is only able to supply a large fraction of the max demand of the rocket motor, although the later SABRE docs seem to contradict that a bit. I think I might try setting it up as non-afterburning, if that gets rid of any pretenses of being a ramjet, although I'm a bit curious if the bypass contributes any thrust - been reading a lot about Concorde recently, the intake arrangement on that was responsible for something like 80% of the thrust at M2.

Link to comment
Share on other sites

The rocket shouldn't overheat though

Well, it's not the rocket that's overheating, but the compressor. Turbocompressors are limited to about 1000-1200K in practice.

At it's best it'd seem the precooler/compressor is only able to supply a large fraction of the max demand of the rocket motor

Well "max demand" is slightly complicated here, since the maximum mass flow you can put through a given unit area depends on pressure and temperature (see this). In jet mode the chamber pressure is going to vary a lot. The compressor needs to be able to push enough to choke the nozzle throat at all flight conditions.

Link to comment
Share on other sites

Well, it's not the rocket that's overheating, but the compressor. Turbocompressors are limited to about 1000-1200K in practice.

Which is even more complicated in this case because of the cooler - although you can treat the compressor & cooler as one system. But I see what the angle was there. Will probably be more clear after I dig up the M2+ issue.

Well "max demand" is slightly complicated here, since the maximum mass flow you can put through a given unit area depends on pressure and temperature (see this). In jet mode the chamber pressure is going to vary a lot. The compressor needs to be able to push enough to choke the nozzle throat at all flight conditions.

Yes, but there's also mass flow limits through the rocket pipework ( and well, you have to completely mix H2 in too ) so the choke isn't so much the back of the compressor as *everything* behind the compressor. "Max demand" in this case is pretty simple, it's the oxidizer use of the rocket at full throttle in rocket mode, which would be near ideal combustion ratio. Would like to find a paper where they talk about compromises for mass flow in different modes, actually.

Link to comment
Share on other sites

Which is even more complicated in this case because of the cooler - although you can treat the compressor & cooler as one system. But I see what the angle was there. Will probably be more clear after I dig up the M2+ issue.

Well, theoretically at high speeds you could start throttling the compressor to keep it cool - i.e. dumping more of the incoming air's heat into the fuel rather than extracting useful work (sacrificing thrust and some efficiency). But this shouldn't really matter until you're approaching mach 5 already.

Yes, but there's also mass flow limits through the rocket pipework ( and well, you have to completely mix H2 in too ) so the choke isn't so much the back of the compressor as *everything* behind the compressor. "Max demand" in this case is pretty simple, it's the oxidizer use of the rocket at full throttle in rocket mode, which would be near ideal combustion ratio. Would like to find a paper where they talk about compromises for mass flow in different modes, actually.

I should clarify - by nozzle throat I mean the throat of the exit nozzle, not the compressor. Generally you don't want anything else to choke - if something is restricting mass flow more than the nozzle then the nozzle might not choke and thus would fail to produce supersonic thrust. The turbine also limits mass flow in conventional jet engines, but for all supersonic designs the nozzle throat area is variable so that doesn't matter.

Link to comment
Share on other sites

I should clarify - by nozzle throat I mean the throat of the exit nozzle, not the compressor. Generally you don't want anything else to choke - if something is restricting mass flow more than the nozzle then the nozzle might not choke and thus would fail to produce supersonic thrust. The turbine also limits mass flow in conventional jet engines, but for all supersonic designs the nozzle throat area is variable so that doesn't matter.

Well, this isn't really much like a jet nozzle at all - the end of the compressor feeds the ( I guess fairly conventional ) rocket combustion chamber and then that's off to seperate rocket nozzles, so the gas flow there is at all sorts of different speeds and angles. Having reread one of the papers yet again:

"Also unlike a simple turbojet the engine does not suffer from a reduction in gross thrust with increasing Mach number since the precooler ‘irons out’ the intake recovery air temperature variation allowing the compressor to operate with a nearly constant inlet temperature." - also - "The SABRE engine operates over the whole trajectory with an inlet pressure of only around 1.3 bar, which enables maximum chamber pressure to be achieved with minimal variations in the turbomachinery operating point." And *then* there's that graph with the M2 hump which seems to completely contradict what they just said. I did note there *is* a ramjet component ( or rather there's burners in the inlet bypass ) in the latest design which I missed from the earlier ones I was looking at, which makes that graph even more wierd.

So - the inlet to the turbocompressor stays at the same pressure ( once it's reached that pressure anyway ), the temperature is roughly constant too, so the thing is presumably working optimally once it has enough air coming in. There's a ramjet component of completely unknown efficiency - haven't found any details of that. Latest estimates they give for SABRE 4 are 0.5-2MN thrust on air at ~4000s SFC, and 2MN on LH/LO at ~460s, 6:1 ratio. What I've not found is the LH/air ratio which would help work out how much the ramjet/bypass gives. And that graph is still confusing.

Link to comment
Share on other sites

Well, this isn't really much like a jet nozzle at all - the end of the compressor feeds the ( I guess fairly conventional ) rocket combustion chamber and then that's off to seperate rocket nozzles

And why is that so different from a jet nozzle? In both cases you've got a subsonic fluid accelerated to sonic through an area constriction (the nozzle throat), and then to supersonic speed through a diverging section. Rockets typically have much more extreme shapes because they operate at higher pressure, but the concept is exactly the same.

Link to comment
Share on other sites

And why is that so different from a jet nozzle? In both cases you've got a subsonic fluid accelerated to sonic through an area constriction (the nozzle throat), and then to supersonic speed through a diverging section. Rockets typically have much more extreme shapes because they operate at higher pressure, but the concept is exactly the same.

Yep you're completely right for this situation, I'm just overthinking.

At this point I might just try emailing them & attempting to get a non-contradictory set of up to date answers. Anything you'd like answered?

Link to comment
Share on other sites

anyone else havent been able to get this to run with Active Texture Manager? my loader goes on an infinite hang at squad/parts/basic jet engine somethingorother i can provide the ksp log thing but im not sure how to upload it

There should be no conflict with ATM. If you're trying to run on 1.0.5 though, it's expected to not work. An update is in progress but there are a lot of changes that need to be made.

If you're on 1.0.4, then we can investigate it. Dropbox or Google Drive make good places to upload an output log. Make sure it's publicly viewable.

Link to comment
Share on other sites

1.0.5 support is being worked on, but there are a lot of config changes to make, as well as several under-the-hood changes to both AJE and SolverEngines. It'll take some time.

The new afterburner mechanic doesn't change anything in AJE. The secondary engine module will be removed and the effects folded into the main engine module. Afterburners will work as they always have in AJE. We will have to devise a system to make the nozzle animations look right but that'll probably wait for another version.

Link to comment
Share on other sites

Is a 1.0.5 release being developed? Also, how is the new afterburner mechanic going to change the way the pack works perhaps? Would be nice to be able to have realistic propulsion with the new models.
There should be no conflict with ATM. If you're trying to run on 1.0.5 though, it's expected to not work. An update is in progress but there are a lot of changes that need to be made.

If you're on 1.0.4, then we can investigate it. Dropbox or Google Drive make good places to upload an output log. Make sure it's publicly viewable.

Literally the post above your so you didn't even read a thing.

Link to comment
Share on other sites

If I can make a request, right now RealFuels integration assumes the standard RealFuels resource list and adds resources as appropriate.

This makes it very difficult for a modder who wants to use the RealFuels.dll, but roll their own configs for resources.

Is there any way that specific references to resources (such as LqdMethane) can be removed from AJE, so that RealFuels itself is responsible for adding them?

Link to comment
Share on other sites

If I can make a request, right now RealFuels integration assumes the standard RealFuels resource list and adds resources as appropriate.

This makes it very difficult for a modder who wants to use the RealFuels.dll, but roll their own configs for resources.

Is there any way that specific references to resources (such as LqdMethane) can be removed from AJE, so that RealFuels itself is responsible for adding them?

You're right about this I think - the appropriate place for setting the fuel type in a particular set of RF engine configs and not in AJE. I'll try to get these removed from AJE and moved into RF Stockalike and RO.

Link to comment
Share on other sites

Yep you're completely right for this situation, I'm just overthinking.

At this point I might just try emailing them & attempting to get a non-contradictory set of up to date answers. Anything you'd like answered?

If you haven't sent it yet, there are a few things I was wondering

  • Does the engine run at stoichiometric fuel-air ratio in air-breathing mode?
  • Is the helium loop also used to cool the engine nozzle?
  • Was a more complex intake setup similar to the J58 (boundary layer bleed, etc) considered for the SABRE? Was the mass penalty deemed to just be too great?

Link to comment
Share on other sites

If you haven't sent it yet, there are a few things I was wondering

  • Does the engine run at stoichiometric fuel-air ratio in air-breathing mode?
  • Is the helium loop also used to cool the engine nozzle?
  • Was a more complex intake setup similar to the J58 (boundary layer bleed, etc) considered for the SABRE? Was the mass penalty deemed to just be too great?

I can see where some of the confusion about numbers comes from: apparently the SABRE4 design has a very different ratio to SABRE3... seen a quote of 2:1 for airbreathing ( air:H2, not airborne O2:H2 ) but whether that's both the rocket motor and the bypass ramjet or just the core rocket I don't know. Also saw a quote directly from RE about smaller SABRE designs probably using centrifugal compressors. The intake is an interesting one given considerable experience over here with clever intakes ( Concorde et al ).

Haven't sent it yet, will add those to the list.

Link to comment
Share on other sites

[b][url=https://github.com/camlost2/AJE/releases/tag/2.5]Version 2.5[/url][/b] is now available on Github.

This release is mainly to facilitate KSP 1.0.5 compatibility, and includes all the new engines Squad added to the game. Please not that it is potentially craft-breaking - as a result of the new engine models in 1.0.5, many engines have been shuffled around.
Link to comment
Share on other sites

Would you be a bit more specific about squad engines having bad effects with RealPlume or HotRockets and Non-squad engines may have effects looking wrong due to change? I made the configs for rf stockalike for aje and I want to ensure compatiblity. Thanks!
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...