Jump to content

Rockets with Ferram: Where to place control surfaces?


Recommended Posts

Hi all,

So I've been playing with the Ferram Aerospace mod for quite a while, and I've generally got no problem with making rockets that are stable, i.e. they don't flip around and deconstruct in the air. (Admittedly, I haven't designed many planes, but I've done a couple so far.)

Sometimes, though, I end up with a rocket that's just too darn stable  too much drag at the back, I guess  and it's incredibly difficult or impossible to achieve a decent ascent profile. It just keeps going straight, even if I start trying to push it down to the desired pitch (e.g. 70º) right from the launchpad (since SAS is nerfed and all I have is thrust vectoring).

I've tried adding control surfaces like the AV-R8 pivoting winglet, but almost universally, what I end up with is not control, but a rocket that's oscallating wildly instead whenever I try to use SAS or MechJeb SmartASS. I've tried various arrangements  near the CoM (laterally and/or longitudinally), or far from it; I've tried horizontal with only pitch enabled plus vertical with only yaw enabled; etc.

Almost inevitably, I end up giving up and going back to just thrust vectoring and/or just settling for a sub-optimal ascent.

So I'm wondering, is there a general rule of thumb for how to place these (or what settings to use, or how to use them in flight) in a way that won't cause a rocket to freak out?

Link to comment
Share on other sites

That oscallation you're noticing tends to happen when the rocket is simply traveling too fast trough the lower atmosphere. I found that reducing the speed of your launches works much better and gives you more control over your ascent path.

When using SRB's I usually make sure my TWR is around 1.5 to 1.6 to make sure that I'm not going too fast. If you can keep your speed around 140m/s bellow 10km you should be just fine on that regard. As for where to put control surfaces... Well in making sure your ascent speed is optimal I find that control fins on the bottom of the rocket work just fine. I occasionally have to fight the SAS a bit since it always wants to re-align with the prograde vector but if you can manage to get to a lower speed the effect isn't nearly so bad.

Ideally, Ferram actually allows for the realistic gravity turn where you tip the craft over about 5 degrees and let gravity do the rest. This takes a lot of practice on the throtle though.

Link to comment
Share on other sites

That oscallation you're noticing tends to happen when the rocket is simply traveling too fast trough the lower atmosphere. I found that reducing the speed of your launches works much better and gives you more control over your ascent path.

Okay, interesting. I don't seem to have any problem with going at high speed without the control surfaces. Since FAR has such high terminal velocity and low drag, I generally aim for a high-thrust ascent in order to get to orbit ASAP and minimise losses to gravity.

Things I need to try: Reducing control deflection (less control at low speed but more stability at high speed); maybe enabling FAR DCA flight assist (but that might not affect SAS/MechJeb control).

Ideally, Ferram actually allows for the realistic gravity turn where you tip the craft over about 5 degrees and let gravity do the rest. This takes a lot of practice on the throtle though.

Yes, this is the part I love. In my previous game, I recall I would often launch straight up until about 300 m/s (to have enough velocity to hold my vector), then pitch them to about 80º, 70º, sometimes even 60º (depending on their flight characteristics), and then just let go of all controls. They would slowly, slowly tip closer to the horizon (as their prograde vector changes and pulls their nose onto the new heading) and I got true gravity turns (as far as I know). But I guess the difference is, I was using a much higher TWR, so I needed a much lower starting pitch in order to properly gravity turn.

Edit: Ah, no, I checked my notes. Turns out I would go straight up only to clear the tower (if any), then pitch over to ~70º using SmartASS, maintain that until 350 m/s, then set SmartASS to "SVEL+" or just turn it off and the nose would follow my velocity vector. My actual vector would drop below 70º because I was too slow for the atmosphere to hold me on a steady path over my (top-heavy) rocket wanting to pitch over, but that's fine, since I was really aiming for a 60º or lower pitch anyway.

- - - Updated - - -

Okay, I think I answered my own question: Enabling FAR DCA assist will kill the oscillations because it dampens control input at high velocity, which is where the oscillations occur. So, I'll consider this answered.

Edited by Wisq
Link to comment
Share on other sites

Neither SAS nor Mechjeb (even with the optional FAR-compatibility dll) deal well with FAR control surfaces. You would be best served by making your rocket statically stable and then, instead of selecting +SVEL, just leave it uncontrolled: statically stable means the air itself will align it with +SVEL.

Link to comment
Share on other sites

Oscillation is caused by an under-damped response. Basically your feedback is too great for the desired input. For the winglets, try placing them closer to the centre of mass as this will reduce their torque. If placed directly over the CoM they should basically have minimal effect.

Another cause of oscillation is the rocket flexing. If your control surfaces are pointing at a different direction to your SAS module, that's not good either. That can also be mitigated by placing control surfaces and reaction wheels closer to the CoM, or by using more struts.

I try and aim for one or two calibre stability with my rocket designs - CoL 1-2 diameters behind the CoM.

Link to comment
Share on other sites

Yeah, there wasn't much flexing that I could see; I'm using Kerbal Joint Reinforcement and that seems to keep it pretty stable.

At least part of my problem was that I misunderstood the new FAR control surface sliders. My pitch-only surfaces for example were +100% pitch and -100% yaw and roll, when they should've been 0% yaw and roll.

Enabling DCA assist helped, as did reducing the max deflection down from 15º to ~5º.

Part of the problem I was having was that, at this stage in the game, I've got crap for SAS  just the command pod (5 strength) and the 0.625m SAS module (also 5 strength). I realise that SAS doesn't do a lot for FAR once you reach speed, but my problem was that I couldn't tip over to the desired ascent angle before I reached an airspeed that would lock me to my current course.

I think the "can't tip over early" problem was a mix of not having enough control at low speed (SAS+vectoring) and also really having too high a TWR. I started out with a crazy 3.61 TWR which meant that I reached the aerodynamically stable speed way too fast before I could tip over at all.

If anyone's interested, all this is documented in yesterday's post in my modded Kerbal playthrough blog series: http://wisq.net/2015/01/04/kadet-3/

I don't know what made the final design so incredibly stable yet controllable, but my guess is that

a) the much longer length (and more gradual fattening with length) contributed to stability;

B) the centre of mass was moved even lower, such that those control fins at the top had even more direction authority at the start; and,

c) the TWR was much more controlled, due to the extra weight and due to delaying the liquid main engine burn.

Edited by Wisq
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.

×
×
  • Create New...