Jump to content

[1.3.1] Ferram Aerospace Research: v0.15.9.1 "Liepmann" 4/2/18

ferram4

By ferram4
in KSP1 Mod Releases

 Share

Recommended Posts

silvermistshadow

silvermistshadow

Posted
Posted (edited)
  On 4/30/2016 at 1:29 AM, blowfish said:

Not sure what's causing that but it's definitely not FAR.  What other mods do you have installed?

Expand  

That looks like a resource mod of some kind. I think I've had it before, but I have no idea what it is anymore. The hexagons that are green had good resources in them or something. Or maybe this is an entirely different mod from what I'm trying to remember.

 

Also, random screenshot tip if you're using printscreen: press alt+printscreen... wait, was it shift or alt? Anyways, one of those plus printscreen copies just the active window, so then we wouldn't also see your desktop.

Edited by silvermistshadow
Link to comment
Share on other sites
blowfish

blowfish

Posted
Posted
  On 4/30/2016 at 1:12 AM, Ilya said:

http://sendvid.com/wgku7t19
Could you tell me what I'm doing wrong here? I really feel like a rocket like this should be able to fly, even without fins. I'm pretty sure that in 1.0.5 FAR it would have worked, too. I never had problems back then. Now it seems that to fly a rocket, I absolutely need to put fins that have control, or to only start my gravity turn over 15km altitude. Note that this happens whether I have a fairing or not, and that the payload isn't particularly heavy.

Expand  

A few things:

  1. The fairing is really big compared to the rest of the rocket.  That's going to make it unstable.
  2. The fins are only providing stability along one axis.  You need stability along both.
  3.  I don't think an updated MechJeb-FAR compatibility mod has been released.  Mechjeb has no idea how the rocket is going to respond to control input without it.

 

  On 4/30/2016 at 1:35 AM, silvermistshadow said:

That looks like a resource mod of some kind. I think I've had it before, but I have no idea what it is anymore.

Expand  

Yeah, it definitely looks like something intentional and not a bug.  Either way, not FAR.

Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted
  On 4/30/2016 at 1:41 AM, blowfish said:

A few things:

  1. The fairing is really big compared to the rest of the rocket.  That's going to make it unstable.
  2. The fins are only providing stability along one axis.  You need stability along both.
  3.  I don't think an updated MechJeb-FAR compatibility mod has been released.  Mechjeb has no idea how the rocket is going to respond to control input without it.
Expand  

1. It's really not that big. It's as small as it can be to contain the payload, and there are rockets in real life that have fairings that are about proportionately as large. Example: https://en.wikipedia.org/wiki/Titan_IV . This is especially true when you consider that I have 2 liquid fuel boosters, both with gimballed engines.
2. This is true,  but I don't think that I should need fins at all for a rocket like this, especially not such large ones. It's got 3 gimballed engines...
3. It works perfectly if I put delta deluxe winglets. It also fails if I control it manually without winglets.

Link to comment
Share on other sites
danfarnsy

danfarnsy

Posted
Posted (edited)
  On 4/30/2016 at 1:48 AM, Ilya said:

1. It's really not that big. It's as small as it can be to contain the payload, and there are rockets in real life that have fairings that are about proportionately as large. Example: https://en.wikipedia.org/wiki/Titan_IV . This is especially true when you consider that I have 2 liquid fuel boosters, both with gimballed engines.
2. This is true,  but I don't think that I should need fins at all for a rocket like this, especially not such large ones. It's got 3 gimballed engines...
3. It works perfectly if I put delta deluxe winglets. It also fails if I control it manually without winglets.

Expand  

Eh, proportionally the width of your payload is (by the Mk1 danfarnsy eyeball) is quite a bit wider than the Titan IV example, and the Titan IV is also much longer in the booster stage. I'm fairly confident any rockets I built like that needed a lot of control authority in 1.0.5. I could go back and try it. But, seriously, instead of trying to defend your design as a point of pride, why not try to add fins stabilizing both pitch and yaw, like blowfish suggested? He's a talented dude. As long as you make sure your center of mass is forward of your aerodynamic center (or center of lift, loosely), your rocket will try to align itself with your velocity vector rather than squirreling away from it.

Edited by danfarnsy
Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted
  On 4/30/2016 at 1:59 AM, danfarnsy said:

Eh, proportionally the width of your payload is (by the Mk1 danfarnsy eyeball) is quite a bit wider than the Titan IV example, and the Titan IV is also much longer in the booster stage. I'm fairly confident any rockets I built like that needed a lot of control authority in 1.0.5. I could go back and try it. But, seriously, instead of trying to defend your design as a point of pride, why not try to add fins stabilizing both pitch and yaw, like @blowfish suggested? He's a talented dude. As long as you make sure your center of mass is forward of your aerodynamic center (or center of lift, loosely), your rocket will try to align itself with your velocity vector rather than squirreling away from it.

Expand  

I'm really not defending my design because of pride. There's absolutely nothing special about it, it's a very basic rocket. I already specified in both my last posts that adding fins (specifically delta deluxe winglets) does fix the problem. The thing is it really doesn't seem to me like it should be necessary to put fins on a pretty aerodynamic rocket with 3 gimballed engines. Real rockets don't usually have these huge fins. Removing the fairing doesn't fix the problem either, it still fails at about 5km altitude.

Link to comment
Share on other sites
danfarnsy

danfarnsy

Posted
Posted
  On 4/30/2016 at 2:06 AM, Ilya said:

The thing is it really doesn't seem to me like it should be necessary to put fins on a pretty aerodynamic rocket with 3 gimballed engines. Real rockets don't usually have these huge fins.

Expand  

I agree with this. There are a number of things different in real rockets, but FAR makes it a lot closer to real life. FAR doesn't adjust the awfully heavy dry masses of the stages (realistic mass fractions would shift your center of mass forward, but they're also game-breakingly cheaty in anything but realism overhaul). There's a mod which adjusts this, SMURFF, but I haven't kept up with it. I'm willing to be wrong on this, but I don't think the aerodynamics of FAR is the issue so much as the rocket designs available to us. We can optimize what we've got, but it's not the same thing as having realistic rockets.

Link to comment
Share on other sites
tetryds

tetryds

Posted
Posted

@Ilya

1 - Yes, they are enormous relative to your craft, even ignoring length, that is a huge bump.

2 - 3 gimballed engines mean nearly nothing at those speeds, but that is not the point, notice that they are forcing you to try to get you off course, you have almost 5 degrees of AoA during that ascent, whereas real life rockets go up at a very minimal AoA, performing an actual gravity turn.

3 - It fails if you pilot manually because manual piloting is not as smooth.

So, in resume, you have a lot of drag on the tip and push the rocket away from it's course, that is a good way to blow it up on purpose.
If you want to make an ascent like that you have no choice, you really need the winglets, or change your design, that is not how real life rockets do it and you cannot compare yours to them.
Disregarding the fact that real life rockets have complex control systems to keep that kind of stuff from happening, of course.

Edit: ah, they also have their fairings very close to the edge of what would be considered safe for a fairing with a higher radius than the rocket itself.

Edit2: No, there is absolutely nothing wrong with FAR there, don't try to insist on it, it's doing its job much better than you imagine.

Link to comment
Share on other sites
ferram4

ferram4

Posted
  • Author
Posted

Also, your fairing is stupidly blunt, which means that it will make a ton of drag, especially as you approach the transonic regime.  Add that to the body lift effects, and your entire design is basically an experiment in "how to make a rocket as unstable as possible."  Also, MJ does not appear to be as aggressive in keeping it on target as it really needs to be, which might be part of the problem.

And don't sit there and tell me that body lift is too high.  Body lift is actually probably lower than it should be right now, tbh, but that's the result of not having very good models for how partial flow separation affects body lift over the cross-section-reducing sections of vehicles, so I assume that the entire result basically cancels out to no potential flow body lift effects there.

Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted

Okay, I took all your criticism seriously and tried to make another rocket for the same payload with your specifications in mind (apart from the fins, since we already know they do work). I've reduced the AoA from 3 to 2, made the fairing smaller proportionally to the rocket, and even removed the second stage to allow for a smaller fairing and larger first stage. I've also made the fairing pointier, even though I disagree that it was too blunt. This is the result: http://sendvid.com/a83z394s . Note that flying manually gave a slightly better result, but still worse than the previous rocket. 

And by the way, since you guys seem to be getting defensive and angry at me, I'd like to stress that I'm not trying to poodle about the mod or any of you. I'm just very confused as to why I used to be able to play with FAR perfectly well in 1.0.5, while now I absolutely need fins on all my rockets.

Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted
  On 4/30/2016 at 3:18 AM, tetryds said:

@Ilya It's simply the AoA, try starting with angle right out of the launchpad and that will not happen.

Expand  

What do you mean by "starting with angle right out of the launchpad" ? I've tried everything from AoA 1 to 5, as well as no limit to AoA, and every time this rocket starts to fail at around the same altitude.

Link to comment
Share on other sites
tetryds

tetryds

Posted
Posted
  On 4/30/2016 at 3:22 AM, Ilya said:

What do you mean by "starting with angle right out of the launchpad" ? I've tried everything from AoA 1 to 5, as well as no limit to AoA, and every time this rocket starts to fail at around the same altitude.

Expand  

At around 30 seconds on the video the rocket starts to tilt sideways, that is adding angle of attack to it, this will certainly cause problems.

The second you leave the launchpad, angle it three to five degrees, since it's slow the thrust will quickly kill your AoA and you will have a sideways trajectory, then just leave it going at zero AoA and gravity will slowly turn you while you ascend, just like real rockets do.

Not sure if MechJeb can do that, but it should.

If you pick up speed going straight up it will be much harder to incline later, your ascent looks more like something that would work on stock pre 1.0 KSP, it's not going to work here.

Link to comment
Share on other sites
Skystorm

Skystorm

Posted
Posted

@Ilya I notice that you are also at nearly Mach 1 (~300 m/s) by the time you hit 6km, and that is probably quite a bit faster than you should be going at that altitude.  I would suggest you stay under 250 m/s until you are at least over 10km and keep your starting TWR around 1.25 - 1.50 instead of  the 3.0 I saw in the video.

From my experience, the more akward or draggy the vessel is the more important it becomes to keep your velocity from climbing too high until you get up into the thinner atmosphere.

Link to comment
Share on other sites
ferram4

ferram4

Posted
  • Author
Posted (edited)

No, get out of here with that pre-1.0-stock-aero advice.  There is no reason to limit speed and he's already got a TWR of 1.2 at SL.  Throttling down near Max-Q isn't even an option until initial TWR is already at 1.6, and he's nowhere near that.

So, that fairing is better, but it's still too blunt.  You'd be hard-pressed to find rockets with fairings that are that blunt, most go for much, much sharper cones instead.  Honestly, to me it just looks like insufficiently aggressive control.  The rocket needs to be kept closer to prograde the entire time and MJ is just being too timid to do the job; prograde SAS should do it quite nicely.

Edit: I tried rebuilding your design, though slightly modified; I'm pretty sure that I gave it a longer (but just as blunt) fairing and didn't give it a payload, so it should be much less stable.  This is how far it got before prograde SAS lost control, not much higher up but going a lot faster and clearly not going to space with that shallow of a trajectory.  I think your control system just isn't aggressive enough to control it.

Edited by ferram4
Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted
  On 4/30/2016 at 4:08 AM, tetryds said:

At around 30 seconds on the video the rocket starts to tilt sideways, that is adding angle of attack to it, this will certainly cause problems.

The second you leave the launchpad, angle it three to five degrees, since it's slow the thrust will quickly kill your AoA and you will have a sideways trajectory, then just leave it going at zero AoA and gravity will slowly turn you while you ascend, just like real rockets do.

Not sure if MechJeb can do that, but it should.

If you pick up speed going straight up it will be much harder to incline later, your ascent looks more like something that would work on stock pre 1.0 KSP, it's not going to work here.

Expand  

I've just tried your idea, and even put basic fins just to test it fairly. For the first few seconds, it starts angling down on the right, then eventually it stops, and it starts going backward and flips. I've tried several angles and throttles, and it's always the same result. Mechjeb can't do real gravity turns, no.

  On 4/30/2016 at 5:38 AM, Skystorm said:

@Ilya I notice that you are also at nearly Mach 1 (~300 m/s) by the time you hit 6km, and that is probably quite a bit faster than you should be going at that altitude.  I would suggest you stay under 250 m/s until you are at least over 10km and keep your starting TWR around 1.25 - 1.50 instead of  the 3.0 I saw in the video.

From my experience, the more akward or draggy the vessel is the more important it becomes to keep your velocity from climbing too high until you get up into the thinner atmosphere.

Expand  

You've misunderstood the delta v stats. The 3.0 TWR was for the second stage. The first one has a TWR of 1.56 on launch. I've tried throttling down to 80%, but the result is exactly the same.

 

 

  On 4/30/2016 at 6:00 AM, ferram4 said:

No, get out of here with that pre-1.0-stock-aero advice.  There is no reason to limit speed and he's already got a TWR of 1.2 at SL.  Throttling down near Max-Q isn't even an option until initial TWR is already at 1.6, and he's nowhere near that.

So, that fairing is better, but it's still too blunt.  You'd be hard-pressed to find rockets with fairings that are that blunt, most go for much, much sharper cones instead.  Honestly, to me it just looks like insufficiently aggressive control.  The rocket needs to be kept closer to prograde the entire time and MJ is just being too timid to do the job; prograde SAS should do it quite nicely.

Expand  

Well, doing a slight turn and putting SAS on prograde does work at not flipping the rocket, on both models that I've made in this thread. The angle changes a bit every time, though. 

Unfortunately, it's not going to be a very viable option for me, because it depends too much on luck, and I'm doing a career on hard with no reloads. Thanks for your help, though. I just wish there was a way to reliably automate this.

Link to comment
Share on other sites
ferram4

ferram4

Posted
  • Author
Posted

Go up, start your pitch over at a certain speed (probably ~100 m/s, with an initial angle of 5 degrees), switch SAS mode to prograde once you've got that, and then make adjustments once out of the atmosphere doesn't exactly depend on luck.  Look, I'm sorry that MJ isn't aggressive enough for this, but... FAR isn't gonna sacrifice realism because MJ is being timid in controlling things.

Also, btw, last FAR version for 1.0.5 had the same exact lifting body physics, so any difference in control-ability between 1.0.5 and 1.1/1.1.1 isn't on FAR's side.

Link to comment
Share on other sites
Ilya

Ilya

Posted
Posted
  On 4/30/2016 at 6:39 AM, ferram4 said:

Go up, start your pitch over at a certain speed (probably ~100 m/s, with an initial angle of 5 degrees), switch SAS mode to prograde once you've got that, and then make adjustments once out of the atmosphere doesn't exactly depend on luck.  Look, I'm sorry that MJ isn't aggressive enough for this, but... FAR isn't gonna sacrifice realism because MJ is being timid in controlling things.

Also, btw, last FAR version for 1.0.5 had the same exact lifting body physics, so any difference in control-ability between 1.0.5 and 1.1/1.1.1 isn't on FAR's side.

Expand  

Technically it doesn't depend on luck, but since there's such a tiny difference between not pitching enough, pitching perfectly, and pitching too much, and the fact that it's going to be different for every rockets, and that I'm not allowed to reload, then it pretty much is going to be luck whether or not I succeed, since I'm not a perfect machine.

I'm not asking you to change anything, mate. I was just asking questions, giving my opinions, doing tests, and trying to understand. As I said earlier, it was nothing against you or your mod. I think it's great that you made this and keep developing it.

As for having made no change since 1.0.5, I'll take your word for it, but I can also swear I didn't have nearly that much difficulty before, so I thought maybe something did, if not in FAR, then perhaps because of KSP itself.

Link to comment
Share on other sites
Somgosomgo

Somgosomgo

Posted
Posted (edited)

So after 1.1 came out I dashed to my usual starter modpack - FAR and KER. When I got well into the game with a dozen more mods I noticed that no matter where I do it, EVAing Kerbals drop my framerate to absolutely unplayable levels (I consider 25 playable for KSP).  I narrowed the problem down to FAR. Any ideas on why it might be happening? I run the 32 bit version because it loads faster and stutters less. I tried running 64 bit and one EVAing Kerbal crashed the game.

Crash Log:

  Reveal hidden contents

Edited by Somgosomgo
Link to comment
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
 Share
Go to topic listing
×
×
  • Create New...