NEAR is killing me.

[iFlyAllTheTime]

...I also know that I need to follow a shallower ascent path than I normally would. Knowing all that, however, I still can't get into orbit consistently...

Isn't it quite the opposite?

The fact that one shouldn't point the ship to far off of the velocity vector would mean that by the time one hits 45 deg the altitude, in most cases, is higher than 10 km.

When I started with NEAR, the first thing I noticed was that my pitch over had to be very slow, which meant it was a rather steep climb through the first two layers of the atmosphere.

[NathanKell]

No. Waiting until you are high up, then turning sharply, is neither efficient nor a gravity turn. A gravity turn, by definition, is letting gravity turn you. That means you make a little turn at about 100m/s and then *don't touch anything.* If your rocket is statically stable, as it should be, you will fly an actual gravity turn: gravity will decrease your vertical velocity while leaving your horizontal velocity unchanged, and the airflow will keep you aligned with your actual velocity vector as it slowly drifts towards the horizon.

If you ever place your hands on the controls again after your initial turn off "straight up," you are not flying a gravity turn. You are flying a manually or automaticallly-controlled pitch program.

If you have your stage TWRs right, you should be able to fly a perfect zero-lift ascent (a perfect gravity turn) by kicking to about 85 degrees at 100m/s and then doing nothing other than staging until about 60km. Indeed, you should also be thrusting all the way to apokerb, and circularizing right at apokerb.

[KerikBalm]

It's because, unlike in stock, where low ascent is limited by the souposphere and you design to hit terminal velocity, in FAR/NEAR, you are limited by your rocket experiencing an unplanned departure from controlled flight regime,

That is a piloting issue. You can perfectly well control a rocket with a high TWR in near, in fact its easier, as a stable rocket will want to point where you are going (ie at your velocity vector), and a higher TWR brings your velocity vector closer to where you are pointing.

I submit that this is the OP's issue.

Consider a 1.4:1 TWR rocket, launching at 45 degrees:

The vertical component of lift matches the acceleration of gravity, the rocket neither appreciably falls nor climbs, but it does build up horizontal velocity, the velocity vector moves toward the horizon, not where the rocket is pointing -> the rocket nose drops to the horizon, the vertical component of lift gets smaller, the rocket crashes.

If the same rocket had a 5:1 TWR, it would simply... welllll... skyrocket off at about 45 degrees

In order to keep your rocket under control in low atmosphere, you've got to limit your launch TWR. Also, because aero drag is so low, you build up speed a lot faster than in stock, even with much lower TWR, reducing gravity losses.

I find th TWR limit to absoultely not be the case. Also, while you do accelerate faster due to reduced drag, thus reducing gravity drag, accelerating even faster reduces gravity drag even further.

It is well known that FAR and NEAR make it easier to get into orbit (where easier = less dV)

Good luck turning it. If you go too fast the low atmosphere makes it hard to turn because the faster are, basically the slower you need to turn to avoid control issues.

You just start turning sooner, and a bit sharper (see my earlier point about how a high TWR makes the rocket tend to fly towards where it is pointing)

A low TWR rocket may need to climb near vertical to about 100 m/s before turning, while as I already said, a high TWR rocket can basically pitch to 45 degrees from a standstill.

[Albert VDS]

The main problem with the OP's design is that his 2nd stage (4th stage) is smaller than his 1st (2nd stage). It should be the other way around.

Before the 1st stage is done the CoM will be mostly in the top.

Also the control surfaces are not needed if you can get the rocket high enough up before staging.

Or in other word; make your rocket look like a real rocket.

[Streetwind]

If you have your stage TWRs right, you should be able to fly a perfect zero-lift ascent (a perfect gravity turn) by kicking to about 85 degrees at 100m/s and then doing nothing other than staging until about 60km.

What are the right stage TWRs, incidentally?

Remarks like this are relevant to my interests because I currently also play with NEAR, and I have constant issues with too high TWR. I aim to be around 1.4 off the pad, and I turn fairly soon - well before 100 m/s. And yet I still have to manually wrestle the rocket around all the way to orbit because it doesn't want to pitch over fast enough due to the rocket gaining speed too quickly too low.

In many cases I can't throttle down due to using SRBs (playing on a high difficulty setting with limited funds). In the one case where I got to use a liquid fuel engine, just yesterday, I still had to practise, trying to find the right compromise between acceleration and pitchover rate. Mind you, that launch was the closest to a self-controlled turn that I've managed yet, but I was getting (minor) reentry effects on the way up so I was definitely still going too fast. So, what would you recommend I should be aiming for, especially with SRBs?

[Empiro]

I use FAR, and I almost always launch around 1.5, and my later ascent stages are around 1.5-2.0 (right when I stage). If your ascent is too steep, then you'll want to pitch over earlier -- say at 50 m/s and possibly at a larger angle. It's different for every rocket configuration and TWR. It's all about trial and error before building up an intuition about exactly how much and when you should pitch.

I rarely have a "perfect" launch where I don't need to touch anything after the turn. Most of the time I need to use the trim settings to make minor adjustments.

[KerikBalm]

Mind you, that launch was the closest to a self-controlled turn that I've managed yet, but I was getting (minor) reentry effects on the way up so I was definitely still going too fast. So, what would you recommend I should be aiming for, especially with SRBs?

I don't think minor reentry effects means that you were going too fast. It means that in stock, but not in NEAR. (as far as optimal velocity is concerned, I don't know about when the DRE mod is included)

[MarvinKitFox]

My 2c.

With FAR(NEAR is same,really), you want a nice high TWR.

With one caution: Display and keep an eye on your dynamic pressure (Q). Things get wobbly at q=18000, and most things shred by 30000.

So, launch at monster TWR, then throttle down while transiting the max-q regime, then throttle up again when the air is thin enough.

Kinda similar to what every single realworld rocket does, you know?

As for the tipping-over problem you have...

You have a bunch of fins/wings right at the bottom. Fine.

You are pointing a bit off from vertical, say 85*, and thrusting at 1.3g

so, you are burning 1g fighting gravity, and 0.3 for vertical acceleration. (actually 1.3*sin85 degrees minus 1.0)

at the same time, you are accelerating horizontally at 1.3*cos(5 degrees), with only friction working against you.

You are accelerating about equally as much SIDEWAYS as UP.

As you are now sideslipping through the wind, you have a stiff side breeze blowing on those hefty fins right at the bottom of your rocket.

This will, of course, tip your rocket more. Which increases the ratio of horizontal to vertical acceleration, which increased the horizontal wind on your fins, which......

feedback loop right into the ground!

To fix this:

Increase your TWR a bit. Double should be good...

Don't start tipping until you are moving fast enough UP so that the sideways movement from tipping is less than your control authority.

For most of us, that is around 100-120m/s

When you do start tipping, you need to so so at a rate that will keep your rate-of-tipping constant, all the way up to space!

Yes, if you plan to be in air for 2 minutes, you want to tip 90 degrees in 120 seconds, thus WELL UNDER 1 degree per second.

[Mystique]

What's most interesting, I have similar problem with small stock rockets like OP has. What's even more strange, some huge multistage rockets with lots of boosters and stuff fly nicely O_O

I also noticed yesterday that attaching MJ radial thingy to one side of my payload (which is totally symmetrical) causes CoL to shift a bit in direction of MJ part and then causes instability during ascent, especially on supersonic speed. Is it normal behavior?

[Streetwind]

I also noticed yesterday that attaching MJ radial thingy to one side of my payload (which is totally symmetrical) causes CoL to shift a bit in direction of MJ part and then causes instability during ascent, especially on supersonic speed. Is it normal behavior?

Well, you just answered your own question, didn't you?

Having a part only on one side of an otherwise symmetrical object makes it unsymmetrical. The CoM shifts, so trying to push from below will tend to turn you slightly; also, the extra part will be subject to drag, but the other side will be perfectly smooth.

What did you THINk would happen?

[robopilot99]

What's most interesting, I have similar problem with small stock rockets like OP has. What's even more strange, some huge multistage rockets with lots of boosters and stuff fly nicely O_O

I also noticed yesterday that attaching MJ radial thingy to one side of my payload (which is totally symmetrical) causes CoL to shift a bit in direction of MJ part and then causes instability during ascent, especially on supersonic speed. Is it normal behavior?

Try these:

http://forum.kerbalspaceprogram.com/threads/61040-0-23-6S-Service-Compartment-Tubes-Design-smooth!

They still work in the current version of KSP but you may need to update firespitter.

[Mystique]

Well, you just answered your own question, didn't you?

Having a part only on one side of an otherwise symmetrical object makes it unsymmetrical. The CoM shifts, so trying to push from below will tend to turn you slightly; also, the extra part will be subject to drag, but the other side will be perfectly smooth.

What did you THINk would happen?

Taking into account MJ's part weight, CoM seems not to be shifting anywhere, moreover, RCS build mod doesn't show any torque (and it's quite precise), so cause of problem is drag and shifting of CoL. Though I'm still surprised it's so noticeable, since, IIRC, drag in KSP is function of mass, which is too low to be significant in this case. Generally it's logical, but c'mon, entire XX-ton rocket swaying because of weightless thing with 2 lights and tiny antenna? I hope they'll never add birds, because one of them landing on rocket could cause instant FUBAR with such aerodynamics

Try these:

http://forum.kerbalspaceprogram.com/threads/61040-0-23-6S-Service-Compartment-Tubes-Design-smooth!

They still work in the current version of KSP but you may need to update firespitter.

Good looking mod, though I prefer procedural fairings for this

[Wanderfound]

My 2c.

With FAR(NEAR is same,really), you want a nice high TWR.

With one caution: Display and keep an eye on your dynamic pressure (Q). Things get wobbly at q=18000, and most things shred by 30000.

Sure you're not missing a zero there?

[iFlyAllTheTime]

No. Waiting until you are high up, then turning sharply, is neither efficient nor a gravity turn. A gravity turn, by definition, is letting gravity turn you. That means you make a little turn at about 100m/s and then *don't touch anything.* If your rocket is statically stable, as it should be, you will fly an actual gravity turn: gravity will decrease your vertical velocity while leaving your horizontal velocity unchanged, and the airflow will keep you aligned with your actual velocity vector as it slowly drifts towards the horizon.

If you ever place your hands on the controls again after your initial turn off "straight up," you are not flying a gravity turn. You are flying a manually or automaticallly-controlled pitch program.

If you have your stage TWRs right, you should be able to fly a perfect zero-lift ascent (a perfect gravity turn) by kicking to about 85 degrees at 100m/s and then doing nothing other than staging until about 60km. Indeed, you should also be thrusting all the way to apokerb, and circularizing right at apokerb.

Oh, I'm guessing then that SAS is off when you launch?

[dzikakulka]

Oh, I'm guessing then that SAS is off when you launch?

I think it should be enabled on launch and then disabled once airflow is strong enough to stabilise you.

[MarvinKitFox]

Sure you're not missing a zero there?

Nope.

But then, I fly with FAR, and without using Unobtanium parts.

Just for your info:

Your jet there should correctly be colored somewhere between cherry red and brilliant yellow.

At 1017C, it is quite hot enough to boil aluminum, or turn ordinary steel into a nice red taffy.

Now, take this red-hot taffy and put a concrete block weighing 2200 tons on it.

*that* is what you are exposing your craft to.

I venture to bet you have "aerodynamic failures" switched off.

Might even need to have alt-f12 "unbreakable structures" on.

[Wanderfound]

Nope.

But then, I fly with FAR, and without using Unobtanium parts.

Just for your info:

Your jet there should correctly be colored somewhere between cherry red and brilliant yellow.

At 1017C, it is quite hot enough to boil aluminum, or turn ordinary steel into a nice red taffy.

Now, take this red-hot taffy and put a concrete block weighing 2200 tons on it.

*that* is what you are exposing your craft to.

I venture to bet you have "aerodynamic failures" switched off.

Might even need to have alt-f12 "unbreakable structures" on.

FAR, DRE, all stock, failures on.

Try it for yourself: http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1515356&viewfull=1#post1515356

[NathanKell]

What are the right stage TWRs, incidentally?

Remarks like this are relevant to my interests because I currently also play with NEAR, and I have constant issues with too high TWR. I aim to be around 1.4 off the pad, and I turn fairly soon - well before 100 m/s. And yet I still have to manually wrestle the rocket around all the way to orbit because it doesn't want to pitch over fast enough due to the rocket gaining speed too quickly too low.

In many cases I can't throttle down due to using SRBs (playing on a high difficulty setting with limited funds). In the one case where I got to use a liquid fuel engine, just yesterday, I still had to practise, trying to find the right compromise between acceleration and pitchover rate. Mind you, that launch was the closest to a self-controlled turn that I've managed yet, but I was getting (minor) reentry effects on the way up so I was definitely still going too fast. So, what would you recommend I should be aiming for, especially with SRBs?

Well, it's complex because a zero-lift ascent (aka perfect gravity turn) is not necessarily the most efficient, as I understand it. Also, there's the question of "efficient in terms of what" -- an ascent that optimizes for delta V expended will not be mass-optimal, for example, and neither will be cost-optimal.

It's almost certainly more efficient to start with a TWR around 1.6-1.8 with fair-sized fins on the first stage and turn sharper than gravity allows, and use a lower-thrust upper stage that points below prograde once at about 40km, for example. It's just that's not a gravity turn.

The key things you need to worry about are not just starting TWR but ending TWR and the stage burn time; a starting TWR of 1.4 is meaningless if you end up at 3.0 in 30 seconds...

Basically, you will want to launch such that your apokerb rises "enough" (whatever that means), then try to keep time to apokerb as low as possible while still leaving time to expend all your propellants. I.e. on launch kick it out to 30 seconds or so, then by turning and/or throttling let it crawl back and try to keep it constant at about 10 seconds after you stage to your upper stage. (I pulled those numbers out of my...hat, but they might serve; the main point is you want to keep your time-to-apoapsis as low as possible always, with the exception of right at launch).

[Panzerbeard]

Well, I can now get things into orbit well enough, thanks all. The problem I'm having now is launching to an inclined orbit. Equatorial orbits work fine, but when I try to tilt over at a slight inclination for, say, a Minmus launch, it just wants to keep drifting until my inclination is way off, to the point that It'd be more efficient to launch into an equatorial orbit and do my plane change there. I'm not sure why it's stable for equatorial ascents but not for even slightly inclined ascents.. Any thoughts?

The main problem with the OP's design is that his 2nd stage (4th stage) is smaller than his 1st (2nd stage). It should be the other way around.

Before the 1st stage is done the CoM will be mostly in the top.

Also the control surfaces are not needed if you can get the rocket high enough up before staging.

Or in other word; make your rocket look like a real rocket.

I make the upper stages larger because otherwise the CoM is too low and it keeps wanting to flip out. Since the payload is really light I can't see any way of keeping the CoM high without using larger upper stages. Unless I'm missing something?

[Leszek]

Inclined orbits change their heading constantly. Point in the right direction at launch and just let it drift after the initial 10 KM's. You will have to do an orbital incline correction regardless but from 0 inclination it is about 250ish DV so you should be more than able to get better than that.

[Empiro]

Well, I can now get things into orbit well enough, thanks all. The problem I'm having now is launching to an inclined orbit. Equatorial orbits work fine, but when I try to tilt over at a slight inclination for, say, a Minmus launch, it just wants to keep drifting until my inclination is way off, to the point that It'd be more efficient to launch into an equatorial orbit and do my plane change there. I'm not sure why it's stable for equatorial ascents but not for even slightly inclined ascents.. Any thoughts?

I'm not sure exactly what you're describing. However, you'll need to take into account your initial 100 m/s eastward velocity thanks to Kerbin's rotation. Also, you need to make sure that you are launching right when you are at the ascending or descending node for Minmus.

I make the upper stages larger because otherwise the CoM is too low and it keeps wanting to flip out. Since the payload is really light I can't see any way of keeping the CoM high without using larger upper stages. Unless I'm missing something?

You shouldn't have to do that. I've never had an issue with just putting the stabilizers near the bottom of my rockets, and I've launched quite a few different designs. Your CoM actually tends to move up as you use the fuel in the first stage, because the mass at the top of your rocket stays the same while you are burning away the fuel near the bottom.

[Panzerbeard]

Thanks again for the replies, all, I think I know everything I need now, cheers

I'm not sure exactly what you're describing. However, you'll need to take into account your initial 100 m/s eastward velocity thanks to Kerbin's rotation. Also, you need to make sure that you are launching right when you are at the ascending or descending node for Minmus.

Yeah, I kinda didn't explain it clearly, my apologies. I've actually resolved the problem now, for some reason launching using launch clamps was causing some phantom forces and making my rocket want to keep pulling northwards. I assume it was a bug, although I can't seem to reproduce it anymore.

You shouldn't have to do that. I've never had an issue with just putting the stabilizers near the bottom of my rockets, and I've launched quite a few different designs. Your CoM actually tends to move up as you use the fuel in the first stage, because the mass at the top of your rocket stays the same while you are burning away the fuel near the bottom.

I guess so, but if you've got multiple tanks on the bottom stage it's going to want to burn from the highest tank first, and when I simulate it by emptying the tanks with tweakables, the CoM shifts further down until the bottom tank starts emptying. I kinda assumed it'd be best for the CoM to continue rising all the time. I may just use TAC Fuel Balancer, it'd make this whole thing a lot easier.