Jump to content

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


Recommended Posts

16 minutes ago, Svm420 said:

@blowfish

 

Would ambient temperate affect rotor thrust as well? I am getting only ~83kN and ~33Khp from the f135 lift fan even though it is 90kN and 35kHP in the config. Though the thrust I am getting is pretty spot on for RL. Just curious as I am trying to get a functional f-35b. Thanks!

Indirectly, I think.  Rotor thrust is proportional to density, with a reference density of 1.2 kg/m3.

Link to post
Share on other sites

@blowfish

I am working on making davon throttle control compatible with AJE specifically the rotor and jet engine modules. I am having some weird results when using it with rotors, and I am guess it has something to do with how things are done with ModuleEnginesAJERotor. Is the throttle linked to power with rotors as opposed to "thrust" with other engine modules? It just seems like DTC is not reading how much thrust a rotor will have a a set throttle level. Hope that make sense otherwise I am really lost as to why I don't get the expected behavior when using rotors.

 

 

Edited by Svm420
Link to post
Share on other sites
4 minutes ago, Svm420 said:

@blowfish

I am working on making davon throttle control compatible with AJE specifically the rotor and jet engine modules. I am having some weird results when using it with rotors, and I am guess it has something to do with how things are done with ModuleEnginesAJERotor. Is the throttle linked to power with rotors as opposed to "thrust" with other engine modules? It just seems like DTC is not reading how much thrust a rotor will have a a set throttle level. Hope that make sense otherwise I am really lost as to why I don't get the expected behavior when using rotors.

That's more or less correct, rotor thrust depends on not just throttle but density, vertical and horizontal speed, and maybe some other stuff that I'm forgetting.  In general with AJE there's no way to know how much thrust you'll get at a particular throttle level and flight condition without actually running the solver there.  So calculating backwards from thrust to get a particular throttle is impossible - if you want this kind of control it needs to be done with a PID.

Link to post
Share on other sites
9 minutes ago, blowfish said:

That's more or less correct, rotor thrust depends on not just throttle but density, vertical and horizontal speed, and maybe some other stuff that I'm forgetting.  In general with AJE there's no way to know how much thrust you'll get at a particular throttle level and flight condition without actually running the solver there.  So calculating backwards from thrust to get a particular throttle is impossible - if you want this kind of control it needs to be done with a PID.

Dang was hoping it maybe some math/conversions. I was able to get expected performance from the AJEJet module and it works great my harrier has never been more stable. My f-35 suffers from this issue though. If I set the throttle to full DTC properly balanced the jet thrust, but didn't seems to not do anything to the rotor throttle which is what led me to believe if I had it read the variable throttle controls for rotors I would be able to run the math to convert that to thrust for the mod to properly work. So close to a great VTOL system :/

Link to post
Share on other sites

General update: After a couple of minor changes, AJE seems to be generally working in 1.1.  No release version is available yet, but the changes are pushed to Github.

@Ser I added a field to jets: throttleResponseMultiplier, which should affect how fast the jets respond.  I still want to implement a long-term solution, but hopefully this will satisfy your needs in the interim.

Link to post
Share on other sites

Hi,

I just tried to launch an SSTO using the SABRE M engine with Realism Overhaul and RSS. I'm a bit suprised of the "poor" performances of that engine, I have no problem for take-off with my small test space plane, but I can't even get to mach 1, and the thrust quickly lowers as I get more altitude, with only 100 or 200 kN at 15km of altitude. I would have expected something closer to 2000kN as is indicated on the wikipedia page of the SABRE engine? Am I missing something, or is there a configuration issue for the SABRE in RSS?

My space plane is very lightweight for now, but if I were to follow the Skylon design, it should weight 300T at launch, I'm pretty sure two SABRE engines, with the current configuration, can't even get that off the runway, not talking about getting to mach 5 at 25km of altitude...

Link to post
Share on other sites

@Youen A few points there

  1. AJE doesn't do any scaling of that engine to bring it to to the indended size and performance of the real SABRE.  The part in question has a diameter of 2.5m, the real one is about 3.8m at the intake and 5m at the nozzle.  This sort of scaling is RO's domain, not AJE's.
  2. Static thrust is going to be much less than 2 Mn.  Depending on which of REL's sources you use, it's somewhere between 600 and 800 kN per nacelle.  Thrust will of course drop off if you climb, but it will increase if you gain speed.
  3. If you're hitting 15 km and haven't even passed mach 1 yet, you're doing the ascent wrong.  If you look at REL's simulated ascent (xls file), you will notice that the SKYLON barely climbs at all before going supersonic.  It takes a very shallow ascent path which increases drag but also thrust, meaning that the ascent happens quicker and probably results in net fuel savings.
  4. Also note that the SKYLON is intended to have a very long takeoff roll, longer than the 2.5km stock runway allows (I think the intended runway length is 4 km).  For a shorter runway you need a higher TWR to achieve takeoff speed before you run out of space.
Link to post
Share on other sites
7 minutes ago, blowfish said:

Also note that the SKYLON is intended to have a very long takeoff roll, longer than the 2.5km stock runway allows (I think the intended runway length is 4 km).  For a shorter runway you need a higher TWR to achieve takeoff speed before you run out of space.

@Youen, That is more/less main reason why you can't fully recreate SKYLON look a like craft (same weight, length and wingspan) in KSP and have the same result. You need, either, craft with less weight or larger wing surface area. For KSP runway, lower limit is aproximate TWR of 0.3 and enough wingspan so you can takeoff at ~120 m/s @10 AoA on the end of runway.

If TWR is better on takeoff, you can have higher speed at the end of runway, meaning you don't need large wing surface as you can have same required lift for same AoA with smaller wings. Smaller wings means less dead weight once craft is in space and this give more dV.

For SSTO in KSP however, you need to find right spot between wing surface area, number of required engines, fuel and payload weight for successfull SSTO. In KSP 1.0.5. I didn't have any issue with SABRE engines underpowered, quite opposite, you can easy achieve high speed in too dense atmosphere that brings different issue - overheating.

In KSP 1.0.5. even with stock heating system, you will melt down craft much sooner than SABRE is going to run out of thrust due to high speed/thin atmosphere.

Link to post
Share on other sites

@blowfish OK, thanks for the info, I didn't know thrust would increase with speed. Is that true for all jet engines ? Turbo fans, etc? I should probably read more about how these things work in the first place ^^ I'll try with a shallower ascent.

I also thought the SABRE M looked a bit small to get such a big thing into orbit, though that doesn't have to prevent having the real performances. I suppose it should be easy enough to make a new part with the same mesh and just scale it, I'll give that a try. But I don't know what to change on the AJE parameters to get the performances to scale as well...

@kcs123 If the runway length is the only problem, I can take off from the flat grass around it :D. About your other comments, I'm not sure if you're playing with realism overhaul? It changes pretty much everything in the game... Suddenly you need really big rockets just to get to orbit, and getting to the moon requires a rocket just as big as the Saturn V and its 3 000 000kg.

Link to post
Share on other sites
8 minutes ago, Youen said:

@blowfishOK, thanks for the info, I didn't know thrust would increase with speed. Is that true for all jet engines ? Turbo fans, etc? I should probably read more about how these things work in the first place ^^ I'll try with a shallower ascent.

It's somewhat true in general, but the exact extent varies by engine - in general, the higher the bypass ratio, the less thrust growth you'll see and the sooner it will start to fall off.  High bypass turbofans won't see much thrust growth at all, whereas turbojets will keep gaining thrust until the compressor overheats.

13 minutes ago, Youen said:

@blowfishI also thought the SABRE M looked a bit small to get such a big thing into orbit, though that doesn't have to prevent having the real performances. I suppose it should be easy enough to make a new part with the same mesh and just scale it, I'll give that a try. But I don't know what to change on the AJE parameters to get the performances to scale as well...

Well, a lot of the trouble here is that the exact performance parameters are not known (since it's still a mostly theoretical engine).  But if you want to increase thrust, modifying the Area parameter is probably the way to go (at least on the SABRE - for other engines there might be other things going on).

Link to post
Share on other sites

I hope this is not too much off topic. Here is my first Skylon attempt with a scaled SABRE engine. I just changed the rescale factor, and since dimensions are multplied by two I also multiplied the Area parameter by 4 and also set mass to 60 tons, don't know how that's supposed to scale.

I can take off with it, but it's too instable and I lost control shortly after that (looks like my aerodynamic center is ahead of the center of mass...). Engine performance seems more like what I would expect from a vessel capable of Mach 5 in air breathing mode but I'm no expert.

 

EDIT: If TWR of the SABRE is 14 (according to wikipedia), and max thrust is 1960kN, the mass should be more like 14 tons?

Edited by Youen
Link to post
Share on other sites

@Youen, I have to admit, I didn't tried RO in KSP 1.0.5, for various resons,  mostly due to lack of stability of some mods that I like in this game and that works only in Kerbol system. But, atmosphere of Kerbin is simulated like as Earth, gravity is same, but Earth radius is much larger. I know that required dV to achieve low earth orbit is much larger, >9000 m/s, IIRC. While I didn't created SSTO for RO, I have created space planes that have much more dV in low Kerbin orbit to spare, so those crafts should be able to achieve earth low orbit too.

Probably, I would have to adjust my crafts slightly - already mentioned ratios of fuel/payload mass and wing span, but realy it is all about messing with those things. I don't like SKYLON concept at all, in my opinion, there is much better design aproach, but if you realy want to create something similar, I would put much larger vertical tail surface, 2-3 times larger than already is on your craft and also put some elevators at tail too.

Roll stability will be much harder to accomplish with SKYLON a like design, but try with slightly dihedral main wing and slightly more dihedral angle on canrads and elevators. It is also better idea to use all moving wings (I assume you are using B9 procedural wings) for canrads than scale stock pre-made parts. You have much more control how to shape PW than stock parts. All this tips are just from eyeballing on provided screenshots.
For exact shape/positions I need much more info than provided, but I hope that those tips will help you. Have fun.

Link to post
Share on other sites
On 04/13/2016 at 0:58 AM, Youen said:

I hope this is not too much off topic. Here is my first Skylon attempt with a scaled SABRE engine. I just changed the rescale factor, and since dimensions are multplied by two I also multiplied the Area parameter by 4 and also set mass to 60 tons, don't know how that's supposed to scale.

I can take off with it, but it's too instable and I lost control shortly after that (looks like my aerodynamic center is ahead of the center of mass...). Engine performance seems more like what I would expect from a vessel capable of Mach 5 in air breathing mode but I'm no expert.

EDIT: If TWR of the SABRE is 14 (according to wikipedia), and max thrust is 1960kN, the mass should be more like 14 tons?

Mass scales cubically, but you're better off just checking it for the real one.

Also it's a canard configuration, so don't be afraid to move the wing back. However also take note of the pitching moment at high angle of attack (30-50) at hypersonic velocity, ie at reentry. If you cannot maintain absolute static stability through all stages of flight (you probably can't) you should design for near neutral stability instead and use the Atmosphere Autopilot mod in FBW mode (turn moderation off in all axis but yaw during reentry) to stabilize it.

Link to post
Share on other sites

@kcs123 @m4ti140 Thanks for the tips :) I've actually managed to get it stable (pushing the wings back, even if this also moves the center of gravity, and adding another vertical tail aileron under the body even if that reduces clearance for take off and landing ; I've updated the above imgur album). I also managed to get it to orbit (though without payload yet), now I'm just struggling for reentry + staying in the mass budget (RO shielded tanks are too heavy, the Skylon design is supposed to weight only 53t empty, but there is no fuel tank currently in RO that would allow that, so I'm messing a bit more with parts to create a new tank type that would be closer to the target design, i.e. carbon-fiber frame, aluminum internal tanks, ceramic skin for heat shielding). I'm not sure the Skylon specifications are so realistic anyway (I've read some criticism saying they're too optimistic) but I suppose they still know better than me :)
 I also have issues with FAR aerodynamics/heat calculations ; my craft experiences way less drag than it should according to the Skylon reentry spreadsheet (even though it's approximately the same dimensions), so it doesn't slow down enough. And also, even if I cheat with hyperedit to keep at the same velocity (depending on altitude) than the spreadsheet, temperatures rise higher than they predict (so even my modded ceramic skin rated for 1500K is not enough).

I hope I'll be able to contribute to RO, or maybe make an independent mod, with all the parts needed for a Skylon-like design (which I think could be classified as "near-future" technology).

Now back in the AJE topic: my modified SABRE engines (just increased Area compared to the RO config) seem to have a huge Isp under 2/3 throttle (i.e. when post-combustion cuts off), like 10000s or so. I suppose this isn't realistic? But I don't know why it behaves like this.

Link to post
Share on other sites

@Youen The Isp is higher under 2/3 throttle because it's modeled as an afterburning turbojet with some funky parameters.  Is it accurate?  No.  Have I come up with a reasonable alternative model?  I've tried, but alas also no.  I might add a parameter in to unify the throttles at some point though.

Link to post
Share on other sites
14 hours ago, blowfish said:

@Youen The Isp is higher under 2/3 throttle because it's modeled as an afterburning turbojet with some funky parameters.  Is it accurate?  No.  Have I come up with a reasonable alternative model?  I've tried, but alas also no.  I might add a parameter in to unify the throttles at some point though.

OK, makes sense. So, basically, throttling down this engine in air-breathing mode is cheating, good to know, thanks :)

I've moved to this thread to not pollute here with things not related to AJE.

Link to post
Share on other sites

AJE Version 2.7 for KSP 1.1 is now Available

  • Updated for KSP 1.1 and SolverEngines 2.0
  • Fix some effects on B9 engines
  • Reduce default idle nozzle pressure ratio from 1.25 to 1.1 for jets, make it configurable through the field idleNPR
  • Rotors now respect thrust limiter
  • Add throttleResponseMultiplier field to jets, not modified by default anywhere
  • Fix some effects being active unexpectedly when the flight scene first loads
Link to post
Share on other sites
18 minutes ago, mcirish3 said:

OP update?  if ok to ask.  

camlost was not loged in sice november, several months ago. While he is awol, blowfish and other moders maintain this, but they can't change OP.

Link to post
Share on other sites

Would it be possible to apply this mod without adding the different engines (maintaining the original engines and their masses, but altering the way they work)? It's kind of a pain to install this and realise that my otherwise functional designs are now scrap piles because of the change in the CoM of the crafts.

Link to post
Share on other sites
22 minutes ago, DownHereInChile said:

Would it be possible to apply this mod without adding the different engines (maintaining the original engines and their masses, but altering the way they work)? It's kind of a pain to install this and realise that my otherwise functional designs are now scrap piles because of the change in the CoM of the crafts.

It would be rather complicated to set up like that.  At any rate, the performance itself can vary quite a lot from the stock equivalent - even if the mass and CoM didn't change, the plane still might not behave the same way.

Link to post
Share on other sites
1 hour ago, blowfish said:

At any rate, the performance itself can vary quite a lot from the stock equivalent - even if the mass and CoM didn't change, the plane still might not behave the same way.

I figure it'd be like going back to stock aero in terms of top speed, but a change in performance is needed. My issue is more in the line of CoM displacement and the way it changes the aircraft's maneuvering. Having to move all of my crafts wings in order for them to move like they did is too much of a hassle, that's why I'm asking if there's an option to keep the engines' mass, while altering their performance.

One thing I really like though, is the addition of afterburning turbofans with no thrust vectoring.

Link to post
Share on other sites
3 minutes ago, DownHereInChile said:

I figure it'd be like going back to stock aero in terms of top speed, but a change in performance is needed. My issue is more in the line of CoM displacement and the way it changes the aircraft's maneuvering. Having to move all of my crafts wings in order for them to move like they did is too much of a hassle, that's why I'm asking if there's an option to keep the engines' mass, while altering their performance.

One thing I really like though, is the addition of afterburning turbofans with no thrust vectoring.

Ok.  It's pretty easy to change only for specific engines - just go into GameData/AJE/Squad.cfg and comment out/remove all the @mass = x or %CoMOffset = a,b,c. That will put the stock engines back the way they were.

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