What are the baiscs of rocket design?

[Harry Rhodan]

You might want to use angle snap (press c in the VAB or press the circle thingy next to the symmetry selector) to keep your attachments from being off center.

And if this is really 0.18.3, you don't need all those fins and winglets. They add a bit control authority but don't make your craft more stable. If anything the added mass adds useless drag to your rocket.

[John Doe]

9 hours ago, cicatrix said:

I may be late for the discussion, but have you used struts to tie the three stocks together at the bottom / middle? I think they wobble and this is what causing you trouble. If they're only attached to the tri-couple adapter it's no wonder that you're having difficulty with steering it.

If that were the case, I'd have seen the problems in the Stock Kerbal 5 from which I derived it from. I did not substantially modify the bottom stage. The major modification was adding the middle stage. The top and bottom stages are stock, and have been pretty much left alone.

 

8 hours ago, Harry Rhodan said:

You might want to use angle snap (press c in the VAB or press the circle thingy next to the symmetry selector) to keep your attachments from being off center.

And if this is really 0.18.3, you don't need all those fins and winglets. They add a bit control authority but don't make your craft more stable. If anything the added mass adds useless drag to your rocket.

Yep. this is the 0.18.3 demo.  Given the velocity figures I'm seeing, I don't think that drag is a problem. I'm seeing at least several hundred m/sec velocity inside the space of the first minute. The major problem is that If I execute the turn at the same point as with the stock version, I get a marginal and barely acceptable Ap, but my velocity is so fast that I don't have enough time raise the Ap enough to get an acceptable Pe for orbit. My Ap will raise up to a certain point, and then actually Lower right in the middle of the prograde burn, which is during the back half or so of the fuel capacity of the second stage. I also find that the second stage will burn out and have to be decoupled right in the middle of the insertion burn, assuming that initial Ap is high enough. If I try to correct this by keeping vertical and doing the turn on starting up the middle stage, I get an extremely elliptical suborbital path, which results in a really high orbital Ap, ending up at around 3,060,000 m, buit with a relatively LOW Pe by comparison, at maybe 200,000 m at the most, though the Pe here tends to be just over 100,000 m, and mind you this burns alot of fuel, leaving me just enough fuel to do an antegrade burn exactly timed with Ap in order to start re-entry. (i.e. lowering the Pe to around 40,000 or so at engine burnout. )

[Harry Rhodan]

23 minutes ago, John Doe said:

The major problem is that If I execute the turn at the same point as with the stock version, I get a marginal and barely acceptable Ap, but my velocity is so fast that I don't have enough time raise the Ap enough to get an acceptable Pe for orbit. My Ap will raise up to a certain point, and then actually Lower right in the middle of the prograde burn, which is during the back half or so of the fuel capacity of the second stage. I also find that the second stage will burn out and have to be decoupled right in the middle of the insertion burn, assuming that initial Ap is high enough. If I try to correct this by keeping vertical and doing the turn on starting up the middle stage, I get an extremely elliptical suborbital path, which results in a really high orbital Ap, ending up at around 3,060,000 m

The idea is that you already build most of the horizontal speed you need to raise your PE while getting up to AP. That's why you are turning and why you should look for your time to AP after initiating that 45° turn. If that time is more than one minute you should lower your nose towards the horizon and if it is getting lower than, say, half a minute you should raise it. That way you slowly raise the AP while keeping it in front of you and while building the speed to raise PE. If your AP is already sufficiently high there's really no need to push it it even farther.

If you wait for the turn longer you loose a lot of fuel to gravity and need longer to build the orbital speed you already should have at that point.

 

You can also watch the Scott Manley tutorials for the older versions:

 

[John Doe]

32 minutes ago, Harry Rhodan said:

The idea is that you already build most of the horizontal speed you need to raise your PE while getting up to AP. That's why you are turning and why you should look for your time to AP after initiating that 45° turn. If that time is more than one minute you should lower your nose towards the horizon and if it is getting lower than, say, half a minute you should raise it. That way you slowly raise the AP while keeping it in front of you and while building the speed to raise PE. If your AP is already sufficiently high there's really no need to push it it even farther.

If you wait for the turn longer you loose a lot of fuel to gravity and need longer to build the orbital speed you already should have at that point.

 

You can also watch the Scott Manley tutorials for the older versions:

 

Ok, so then basically we want to be working with a relatively LOW suborbital Ap, say at around 60,000 m? if i'm understanding that right? And I would also understand that when firing prograde, we are essentially pointing the rocket directly at the ground, with the nose cone pointed into space, and thus increasing velocity in what would be a vertical position from the ground. I would think that increasing velocity horizontally is just going to increase delta-v and thus cause you to reach Ap faster, if I understand things right....

[reapersms]

Not quite.

You want a suborbital Ap close to what you want your final orbit to be, but certainly above the atmosphere. Pushing it higher means either spending dv later to lower the Ap back down to your intended orbit, or burning with a radial component to twist the orbit.

Prograde exactly at Ap is parallel to the ground, the nose points at the horizon, and all of your thrust is (essentially) horizontal. Prograde along the way up will smoothly transition from nearly vertical, to horizontal, as you ascend. Depending on how much thrust you're providing at a particular point, one of three things will be happening: (assuming you are thrusting prograde)

Ap increases slowly, while time to Ap decreases -- low thrust, or you are very close to Ap already. Pe should be increasing, and the eccentricity of the orbit is decreasing (becoming more circular)

Ap increases slowly, time to Ap holds constant, Pe climbing moderately -- thrust is well balanced with your position on the ascent. This lets you spend more time to get to orbital velocity, and spend more of that time in the thin portions of the atmosphere, and as you continue, you spend more of the thrust towards the horizon rather than fighting gravity. 

Ap increases quickly, time to Ap increases quickly -- thrust vastly exceeds what you need for your target orbit. chances are you have a highly eccentric orbit, with a Pe way below ground level (so, still sub-orbital) and if you cut the engines to coast to Ap here, you will lose a lot of velocity to gravity on the way up. this will result in a long, hard burn at or near Ap to circularize. If you hit this point early, you also lose velocity to the atmosphere, especially for 0.18's atmosphere.

I don't quite recall the details of the 0.18 atmo, other than a lack of heating and it being exceedingly soupy at lower altitudes.

 

The suggestion to pitch the nose above or below the prograde marker is fine, though I think it's more efficient in KSP to adjust with throttle -- dv spent away from prograde is being spent to steer your orbit, not get into orbit, if that makes any sense.

 

The only case I can think of for Ap reducing while burning prograde would be right around the time of the turn, after a steep, fast ascent. At the point you turn, your surface trajectory will be a long, thin ellipse. Burning towards orbital prograde will widen the ellipse, at the expense of lowering the Ap for a bit. If you keep pointing at orbital prograde it will start increasing again.

[John Doe]

The following orbit is what I tend to get , and this is when I'm lucky, leaving little fuel left for re-entry or anything else. Ap is 7,198, 200 m and Pe is 104,820. I raised suborbital  Ap to my figure of 100,000 m as the intended Orbital Pe. The initial suborbital path had a nice circular curve, but when I did the prograde burn at ap, my Ap rose like wildfire, while taking quite a bit to raise Pe.

Edited April 26, 2016 by John Doe

[Harry Rhodan]

AP and PE are relative terms. The highest point will always be AP and the lowest will be PE. The moment you raise the PE above AP the AP marker will run away from you to the other end and trade places with PE which will in turn go to your end. If you keep burning you will just raise the "new" AP at the opposite side of your orbit while you won't raise the PE because you are basically sitting on it. What you should do is kill the engine the moment both are high enough (at least 70km).

If you want to change your orbital altitude you basically do the same: burn prograde or retrograde at some point (ideally AP or PE) to raise or lower the opposite end of your orbit and then burn at that opposite end to recircularize your orbit at the new altitude.

[reapersms]

Could you provide a screenshot of the map screen, with the navball, right as you're about to start your prograde burn at Ap? That orbit looks like you started burning prograde around 100km, went circular, and then just kept burning prograde after Ap and Pe switched places until you ran out of gas.

or what he said.

[John Doe]

I figured out what happened. I was accelerating with about 8 times too much velocity during the initial phases of the launch, causing me to accelerate to Ap so fast that I ended up doing my burn too late, and thus have to fight both my fall toward the suborbital Pe AND at the same time fight to Raise the Ap,  so in, effect I was doing two manuevers  at once, in a single really LONG burn.

[mikegarrison]

The basics of rocket design:

Throw stuff out the back fast, and it will push you forward.

Edited April 26, 2016 by mikegarrison

[John Doe]

58 minutes ago, mikegarrison said:

The basics of rocket design:

Throw stuff out the back fast, and it will push you forward.

If your trying to be funny, don't quit your day job :-P

[Harry Rhodan]

8 hours ago, John Doe said:

I figured out what happened. I was accelerating with about 8 times too much velocity during the initial phases of the launch,

No. You simply keept on burning when you already reached your goal.

And if it takes you too much time to circularize at AP, you are not too fast but still too slow. Ideally you already built most of your horizontal speed already by that point and only need 100-200m/s to circularize.

Edited April 27, 2016 by Harry Rhodan

[cicatrix]

Ok, step by step:

1) Liftoff vertically

2) At ~100 m/s start turning slightly to the east (if you want a prograde orbit) - a few degrees at a time. Watch your prograde marker on the navball - you steer east and then wait till the prograde marker catches you. Do not go too far from the marker because your rocket will flip or disintegrate because of the aerodynamic stress,

3) Stage as your fuel is burnt.

4) At the altitude about ~30 km your trajectory should almost horizontal (almost, but not quite, keep your prograde marker slightly above the horizon).

5) Watch the Ap reading all the time. When it hits the desired orbit altitude, cut the engine. For example, if you want a 80x80 orbit - turn off your engine as soon as your Apoapsis is 80 km. If you are losing speed due to the drag - use engine to correct your Ap and then shut them off again.

6) Put a maneuver node at your Apoapsis, drag the prograde vector until the resulting orbit is circular.

7) Orient your craft to point at the maneuver node vector. Note the estimated burn time. Divide it by 2 (half burn time).

8) Watch the time remaining till the Ap. When it shows T - (half burn time) start your engines again. *

9) Complete the maneuver.

* If you need precision, try to design your rockets so that you won't have to stage during the circularization burn because when you stage, the max acceleration of your ship changes and the resulting orbit might slightly deviate from what you've been aiming for. Normally it's not a big deal though.

Edited April 27, 2016 by cicatrix

[John Doe]

11 hours ago, Harry Rhodan said:

No. You simply keept on burning when you already reached your goal.

And if it takes you too much time to circularize at AP, you are not too fast but still too slow. Ideally you already built most of your horizontal speed already by that point and only need 100-200m/s to circularize.

What seemed to do the job was throttling down to about 75 percent or so after reaching a velocity of around 150 m/sec or so, before 10,000 m, executing the 45 turn , and Then opening up the throttle and burning all out, which in turn, caused my suborbital path to broaden out quite a bit. I have two possible schools of thought on this-  The first one, the obvious one, is by keeping my velocity relatively low before the  10,000 m mark slowed down the overall max velocity proportionately. The second is that it saved the fuel until a point in the flight with significantly lower gravity / drag effects, and thus burning the engines at the same rate does more work, thus increasing velocity that you might not get otherwise.

[Alshain]

I didn't read the whole thread, I'm sure someone out there mentioned TWR.  What is often overlooked is, assuming you are playing career, you want a cost efficient design.   The simplest most cost efficient designs are 2 stages with radial SRB's.  You should NOT be using more than 2 liquid engines, and never use them radially (i.e. Onion/Asparagus staging).  The Onion and Asparagus staging certainly work, but since 1.0 they simply aren't necessary and are far too costly. 

The lower stage will be your power house, SRB's will augment it's fuel and power capabilities.  It will typically use engines like the Reliant, Swivel, Skipper, Mainsail, Twin Boar and Mammoth.

The upper stage is your kicker, it will take you the final way into orbit.  It's purpose is to use a much more fuel efficient engine in the high atmosphere, this makes the overall rocket a lot smaller and more manageable, and cheaper.  It typically uses engines like the Spark, the Terrier, Poodle, Skipper (yes, it is versatile enough to be either 1st or 2nd stage, you can pair it with a Poodle for 1st stage or a Mainsail for 2nd depending on your payload mass), or Rhino.

Using the two stage design + SRB's you should be able to lift over 100 tons, making asparagus and onion staging completely superfluous and costly.

Edited April 27, 2016 by Alshain

[John Doe]

1 hour ago, Alshain said:

I didn't read the whole thread, I'm sure someone out there mentioned TWR.  What is often overlooked is, assuming you are playing career, you want a cost efficient design.   The simplest most cost efficient designs are 2 stages with radial SRB's.  You should NOT be using more than 2 liquid engines, and never use them radially (i.e. Onion/Asparagus staging).  The Onion and Asparagus staging certainly work, but since 1.0 they simply aren't necessary and are far too costly. 

The lower stage will be your power house, SRB's will augment it's fuel and power capabilities.  It will typically use engines like the Reliant, Swivel, Skipper, Mainsail, Twin Boar and Mammoth.

The upper stage is your kicker, it will take you the final way into orbit.  It's purpose is to use a much more fuel efficient engine in the high atmosphere, this makes the overall rocket a lot smaller and more manageable, and cheaper.  It typically uses engines like the Spark, the Terrier, Poodle, Skipper (yes, it is versatile enough to be either 1st or 2nd stage, you can pair it with a Poodle for 1st stage or a Mainsail for 2nd depending on your payload mass), or Rhino.

Using the two stage design + SRB's you should be able to lift over 100 tons, making asparagus and onion staging completely superfluous and costly.

I'm using the old 0 series demo. So all I have are the three stock liquid engines, two sizes of fuel tanks, and the small SRB to work with. and none of the basic tutorials mentioned the concept of TWR.  Basically what I'm after is a twist on the Saturn V design.(  i.e. a massive tri-coupled lower stage with a moderate middle stage, and a smaller final stage, to where what we get is more or less something like a Saturn II-B sitting on top of a massive bottom stage, which is is quite large because of all the weight sitting on top of it.

[Harry Rhodan]

13 hours ago, John Doe said:

The second is that it saved the fuel until a point in the flight with significantly lower gravity / drag effects, and thus burning the engines at the same rate does more work, thus increasing velocity that you might not get otherwise.

The old souposphere had a really low terminal velocity. You basically want to follow that terminal velocity curve to minimize gravity losses because any second you spend burning against the gravity vector it will steal 10m/s from your potential speed. But at the same time you don't want to waste fuel by pushing against the soup. You can follow the old terminal velocity chart:

Forget that table the moment you ever play 1.0 or 1.1.

[WeirdCulture]

At launch: pointing the top of the rocket towards the sky  *SCNR* 

[John Doe]

18 hours ago, Harry Rhodan said:

The old souposphere had a really low terminal velocity. You basically want to follow that terminal velocity curve to minimize gravity losses because any second you spend burning against the gravity vector it will steal 10m/s from your potential speed. But at the same time you don't want to waste fuel by pushing against the soup. You can follow the old terminal velocity chart:

Forget that table the moment you ever play 1.0 or 1.1.

Great so that explains WHY accelerating to a velocity of around 800 m/sec. or better (which my intuition told me was a good idea, in the sense of faster as early as possible would seem to afford one the best chance of orbit.) was just NOT working... lol.  I  was getting a velocity of around 900 m/sec. by 1 minute to 1:30 minutes into the mission. The question then becomes, just so I'm getting the concept right, why then, NASA's Saturn V rocket was so BIG. ( at around a little less than a 100m stack), If the idea that faster- sooner is a bad thing for establishing orbit. I mean, even with throttle control, the bottom of the Saturn V was a MONSTER as compared to the Redstone or Atlas rockets.  and I would therefore at least think, that the velocity between the two (even given the differences between the weight of the payload) would be as such that the velocity encountered with the saturn V, but for the roll and  turn which was standard in the Apollo launch sequence, would result in an escape path.

[Harry Rhodan]

3 hours ago, John Doe said:

If the idea that faster- sooner is a bad thing for establishing orbit.

Because you keep forgetting that in your version you are not flying through air but through pea soup. In the newer versions you would just look for a sane launch accelaration, begin your turn shortly after launch and fly to orbit without ever having to throttle down.

[Hodari]

19 hours ago, John Doe said:

The question then becomes, just so I'm getting the concept right, why then, NASA's Saturn V rocket was so BIG. ( at around a little less than a 100m stack), If the idea that faster- sooner is a bad thing for establishing orbit. I mean, even with throttle control, the bottom of the Saturn V was a MONSTER as compared to the Redstone or Atlas rockets.

One reason why it was so big was that it had to lift a payload which itself was probably the size of those smaller rockets.  The other rockets just had to deliver a payload to low earth orbit.  Saturn V had to deliver something into orbit which was then capable of going to the Moon, entering orbit there, deploying a lander which could actually land on the Moon and then return to lunar orbit, redocking, and finally having the whole thing come back to land safely on Earth. 

And after playing KSP at all, you should learn just how quickly the size of each subsequent stage needs to grow when you increase the size of the payload.  So yeah, just to get off the pad at all with that large a payload, the Saturn V NEEDED to be huge.

[John Doe]

50 minutes ago, Hodari said:

One reason why it was so big was that it had to lift a payload which itself was probably the size of those smaller rockets.  The other rockets just had to deliver a payload to low earth orbit.  Saturn V had to deliver something into orbit which was then capable of going to the Moon, entering orbit there, deploying a lander which could actually land on the Moon and then return to lunar orbit, redocking, and finally having the whole thing come back to land safely on Earth. 

And after playing KSP at all, you should learn just how quickly the size of each subsequent stage needs to grow when you increase the size of the payload.  So yeah, just to get off the pad at all with that large a payload, the Saturn V NEEDED to be huge.

 But for the fact that the STS payload (the  orbiter )  was ALOT heavier than the Apollo CSM / lM payload. YET the STS stack was quite a bit smaller than the Saturn V stack. If the Saturn V had to be big due to the Payload, one would think the STS stack would have had to have been more around the size of the current SLS / Orion, given the fact that the orbiter itself was much larger. (the Apollo craft weighed at around 37,000 lbs, where as the STS orbiter (without the two SRBs and External Fuel Tank) weighed in at around 151,419. IF more weight should necessarily mean a bigger stack, then it stands by reason that the STS stack should have been somewhere around twice as big as the Saturn V.  ( This also accounts for the fact that the STS  External tank,  was roughly the same diameter, though considerably shorter than the Sat V stack,) However, the external tank, when the stack was pad-ready, was the largest and heaviest component of the stack, (66,000 Lbs Empty and 1,667,667 Lbs in launch-ready state) with most of the space / weight being nothing but fuel, as opposed to the Saturn V's weight being influenced a bit more hardware-dependent.  This would tend to suggest that the size of a stack is dictated by the ratio of how much mass within the stack is actual fuel Vs. payload and hardware, which may not necessarily require an increase the size of the stack itself.

 

Edited April 30, 2016 by John Doe

[FullMetalMachinist]

18 hours ago, John Doe said:

But for the fact that the STS payload (the  orbiter )  was ALOT heavier than the Apollo CSM / lM payload.

I'm afraid that you have this backwards. 

18 hours ago, John Doe said:

(the Apollo craft weighed at around 37,000 lbs, where as the STS orbiter (without the two SRBs and External Fuel Tank) weighed in at around 151,419.

The reason this is so wrong is because you're comparing the Apollo craft mass to the STS pad mass. Those are two different things. If you're going to compare the two different vehicles, you have to look at the same goal. Let's take payload to LEO. Yes the Apollo vehicle itself (CSM and LM) was only  45,200kg, but the whole trans-lunar payload that the Saturn V lifter delivered to LEO was 140,000kg. Compare that to the STS payload to LEO of 60,600kg. 

Now, to get back on topic. The reason that you're having trouble with going too fast (and Apollo didn't) is because, as @Harry Rhodan said, you're playing an outdated version of the demo that has you flying through "pea soup". In the newer demo and since the 1.0 version of the full game, the aerodynamics are much more realistic, and it's very hard to actually hit terminal velocity on the way up.