ajburges

Excuse me as I bang my head into a wall... Of course! Your fuel cost is directly proportionate to drag. Constant velocity implies thrust=drag and Isp (fuel per thrust for this aplication) is constant. Since distance is the integral of velocity WRT time, your peak fuel efficiency is the minima of drag over velocity. Said function is roughly constant until it ramps up for the Mach wall. Any sub critical Mach velocity will offer similar drag efficiency, but faster also offers better lift!

I must confess. I have never found the ion drive terribly useful. It never seemed to offer much more dV than NERV engines and was more expensive for the craft. Since I have operational, recoverable, heavy lifters by the time I get ion engines, craft mass is a non-issue. There's also the additional issue that you can't refine more Xeon. I started playing with numbers and derived the following equation to determine the asymptotic dV value for an engine: lim dV as m[payload]->0 = ln( f[engine] * m[full tank] / (f[engine] * m[empty tank] + 9.81* m[engine] * (m[full tank] - m[empty tank]) * TWR) ) * Isp * 9.81 With this equation in hand we can plot different engine dV limits as a function of minimum TWR: White: Poodle with orange tanks Purple: NERV with mk1 tanks (note: mk0 is more mass efficent) Red: Dawn engine with no EC Green: Dawn engine with 6 OX-4* panels an engine to supply EC (conveniently .4 T/engine total) Lo and behold the answer to my question. For a minimum TWR of .3, NERVs offer better asymptotic behavior! Since most my designs target .3 TWR for ejection burns, NERVs will offer better single stage dV limits. Another interesting result of this exercise is the intercept between NERV and Poodle. A TWR of .6 matches up fairly well with the observations that NERVs can be dubious lander engines for the larger planetoids (Moho, Val, and Mun). I found this exercise useful and thought I would share these findings.

Obviously max aerobic velocity is the most efficient. I'm more concerned about short distances; when you realize you are short of the runway, but breaking Mach 1 again may not be sensible. I expect that either TheKorbinger is correct and gunning the engine is always best or that the drag wall that starts around 300 m/s is significant enough that there is a noteable distance where you are better off governing your speed to minimize drag losses.

Something I've wondered is what subsonic velocity offers the best fuel efficency (WRT distance) if you drop out of Mach without enough energy for a dead stick landing. Assume RAPIER and Whiplashes for jet Isp Just looking for rule of thumb here. Is it better to fly near stall speed, hold a high fraction of Mach 1, or gun the engines?

4 static panels in a tetrahedron arrangement (just need the relative angles, normal vertices are not required) are the minimum to guarantee solar exposure. I you are willing to keep an axis near solar normal, you can get by with just 3. 2 keeps one near exposure and is mass optimal, but can't guarantee coverage. Time warping on-rails avoids battery drain (so I rarely encounter the issue since I always run 5+ flights at once) Unless all your panels are facing sun normal or you are suborbital, the panels will eventually face the sun. This happens because you have no angular velocity in the Kerbol reference frame. If you can't revert, putting a little EC into a battery is an easy save edit. BTW: any panel can be docked (only one at a time) by clicking on it.

First thing: as mentioned, you are exposing 1200 skin tolerance parts to the shockwave. That makes it hard to preserve them. If you are flying an efficient profile on RAPIERs or Whiplashes, they shouldn't even make orbit! Second thing: it's not widely recirculated, but space planes do not do well with compromise reentries. Skimming at 45 km is the most hazardous entry because you maximise time going fast in low drag high temp regions. This allows the heat the skin takes to soak into the body and destroy stuff. Some designs can handle the compromise solution and gain some flight efficency, but it is the most demanding rentry thermally. There are two common safer re-entry profiles: spend some orbits above 55 km to bleed energy in the cool region or set Pe ASL and race to the cooler regions in a cobra manuver. I only have experience flying the second. The race down works well because the thermal profile you see in it is very similar to an efficent SSTO spaceplane launch. It does cost more dV to initiate though.

Actually, trajectories is pretty good prediction these days. You just need to program and follow your flight path. I use horizonal AoA for entry and high atmo to plan burn. By the time you hit low atmo you have lower cross-range anyway.

I think that is a little low for optimal speed. SSTO anerobic dV is a function of mass ratio, but max aerobic speed is a function of thrust vs drag (assuming sufficient wing loading). The fuel mass ratio as a function of dV for rocket mode is exponential. Assuming 1000 m/S dV for a 1600 m/s switchover, switching 100 m/s earlier requires an 12.0% increase (4.7% of total mass) in vessel mass dedicated to rocket fuel when you toggle! That ignores the extra drag losses making up that 100 m/s as well Conversely, matching force of drag to force of aerobic thrust scales better for fuel because of the combo of high Isp and low thrust. This assumes you can enter the low thrust region of your engines at all though. RAPIER designs typically can due to the requirements of braking the sound barrier which is also a thrust vs drag comparison. High aerobic speed is the goal. You get there with low drag to thrust ratio and flight plan. I don't know where peak efficiency lies, but I believe you are overlooking relevant trade-offs.

I agree! I assume Slashy meant 16 seats since that is what his craft shows. The biggest difficulty flying this was dealing with pitch caused by poorly placed wings. In case He did indeed mean 18. This one I may refine a bit more for a crew carrier. It made orbit with more dV than my current 7 seat ferry (though it does use 33% more Ox and have less surplus LF). Given the ease it made Mach 1, I could theoretically squeeze some more efficiency out of the RAPIER. Again this is a case of the Mk2 parts have no advantage in gaining kinetic energy over Mk1. The Cabins have the same seats per mass as Mk1 but Mk1 wins on drag without a thought. Mk2 wins on part count especially with the included mono-prop and reaction wheel options, but I have yet to hear of single engine craft needing part count reductions. My best outcomes with Mk2 always leverage either their cargo bays or utilize their crazy high drag and improved heat tolerance for better, more aggressive aerobraking. Ah, That explains why a Mk3 Cockpit + Aerodynamic Nose Cone has similar drag than a series of adapters from Mk3 to a nose cone. Before comparison I thought the adapter series would win on account of being more "pointy."

Slash, I don't claim to have a deep knowledge of the drag system, but I would think values via debug options would be a more reliable source than interpreting configuration files. I put together a test craft to compare two obvious nose solutions (Shock Cone + adapter and NCS + nose cone) against each other. Since they are on the same craft, they should be close enough to equivalent conditions for evaluation. Note: any craft I make in 1 min has interesting flight characteristics; I wanted to capture a speed run, but wound up on a ballistic trajectory. The speed is still over mach 5 and climbing. I also forgot to auto-strut the radial stacks together so that could also throw values slightly. Unless I read the values wrong, the Mk1 nose configuration totals at .62 drag while the Mk2 nose configuration totals 5.20 drag! These results collaborate experiments I tried with SRBs to compare Mk2 fronts against paired Mk1 frronts. Don't get me wrong: well thought out design can take Mk2 far, but it just does not hold up against the aerodynamic efficiency of Mk1 or the "volumetric" efficiency of Mk3.

1. Requires relay I believe. 1a. Irelevant given answer 1. 2. In networking, a hop is a connection; thus direct connect would be 1 hop. I have seen nothing that suggests KSP uses a different definition.

Slash, Can you please elaborate on how Mk2 can be aerodynamically cleaner? Both Mk1 and Mk2 have the same minimal set of radial stacks. One fuselage, wings, gear, and control surfaces/stabilizers. However Mk2 only offers up to 2 times the Mk1 engine nodes for 3 times the frontal drag. Sure the volume inefficiency can be translated into smaller control surfaces due to the length, but that does not make up for the drag increase. Also that longer fuselage means that you may want to go up a size for landing gear to avoid tail strikes. Honestly the only time I seriously consider Mk2 is for the cargo bay. Service bays are uncomfortably small to launch payloads and can't be extended. On a less serious note: my super-cruise Goliath concept did handle better terminating to a Mk2 cockpit and I was never able to pin-point why quickly.

Without demo pics/files I can only offer a few of my observations. Pitch up effect often comes from contacting first with gear forward of CoM and rebounding. The only way to address this is to reduce the strength of the rebound or move it: Contact with wheels closer (and behind) the CoM first. The rebound from their contact behind CoM will reduce AoA instead of increasing it. Also, the closer to CoM the rebound is, the less torque it applies so you bounce instead of pulling up. You can also reduce forward gear spring coefficient to reduce the strength of their rebound. I think you still want to keep them critically dampened. You can also play with the main gear as well.

Precise Node gives this info. I use it to plan insertion of scansats. I like polar orbits that fly a ground track orthogonal to the equator.

The design that started me on this investigation was a Mk3 Lifter. Termination to 1.25 m was obvious (fuel adapters), The 1.25 m termination was not obvious because of the greater variety of choices. At the Mk3 scale, 80 units of liquid fuel is less impressive sounding because it is only 1% or less of fuel. Currently I have nothing but hate for the Mk2 parts. Their drag and low wet mass puts them at too much of a disadvantage against Mk1 stacks. Mk3 just makes for a more flexible cargo ship once you get Rapiers. The best results I've gotten in pursuit of a Mk2 lifter is longer than my Mk3 lifter because I've limited it to one stack. That ship uses a Shock Cone to feed a pair of Rapiers. Tail connector with a single size 0 stack clipped on the end as a fat aerospike. 2 Mk0 tanks make for an elegant transition and a nose cone to fill the end node. It does add an additional stack, but the minimal drag from the tail cone almost completely offsets the penalty. Experimental observations were that there was more side drift between the Shock Cone and NCS than lagging of the mentioned nose. So it's not the best, but it is pretty good (and is more aesthetically pleasing than the NCS on the 2.5 m adapter tank). Altitude is somewhere between 18 and 21 km. I likely don't have enough wing (2 pairs of big-s + 1 pair big-s "tail" @ 4 degrees I believe) but I have trouble slapping on more wing in a style I like. The beast is probably overpowered; it uses 8 Rapiers because 6 Rapiers was insufficient for take-off. It always wound up in the water after 1 km. Its cargo bay can only hold 1.5 orange tanks and has over 600 m/s after circurlarization. I could probably extend the cargo bay, but anything I make longer than 5 Mk3 segments tends to be too large for a cargo bay regardless. I will likely use the air frame for a fuel tanker by replacing the cargo bay with 1-2 4 segment tank and a short bay for the 2 reaction wheels, EC, and probe core. I have yet to actually configure a stick to play, so I still fly using the keyboard. This leads to difficulty adding fine input and I find I have the most success doing the speed run at whatever altitude the craft can maintain while following prograde. It will rise some during the speed run so I use occasional inversions to create negative lift avoid entering a yo-yo. At the end (when horizontal speed gain stalls out), I allow the increasing lift and pitch at the end to feed into the zoom when I enter closed cycle mode. If I find myself in a (shallow) dive, I engage flaps for more lift/pitch and ride it out. Such a dive can take me below 18 km, but tends to feed into a better horizontal angle for the zoom which helps the closed cycle burn. I must admit, I never thought of using aerodynamically shielded radiators to manage heat; I thought shielding prevented radiators from working. I think I'll pass though; seeing physics cry in the corner makes me feel bad. I don't plan on going fast enough to need measures that extreme. I imagine milking air breathing for that long just wastes fuel and mass budget for an SSTO.

I finally got Rapiers in my career game so I'm now playing with SSTO designs flying level past Mach 5. A new problem emerged due to this: any nose with a heat tolerance less than 2300 K will fail during the speed run! This eliminates both the Adv. Nose Cones and Tail Connectors as nose pieces. A quick (and moderately good looking) workaround was to conically shield the Tail Connector with a Small Nose Cone Stack (Radially attached stack of Small Nose Cones and Mk0 tanks). Modeling experiments off of Luch Kot's videos I compared my workaround against bot a Shock Cone Intake and a NCS adapter on mass controlled SRBs (0 mass difference). It was slightly worse (a couple rocket lengths) than the other two which were tied. Does anyone know of a less draggy hyper-sonic nose cone (1.25 m) than the Shock Cone or NCS + Small Nose Cone? I know from a previous experiment that the Ramp Intake is higher drag. I guess a fairing might work, but I don't like the thought of abusing the fairing as a nose cone.

Late game SSTOs seem to all be either be a Skylon mimic or a thrust plate. And they all use Rapiers and maybe Nukes. It just feels formulaic. OTOH I don't see as many novel low tech designs. Meet the Lance (an unintentionally ironic class name). It does require a few building upgrades and some 300 level nodes, but it serves as a fine SSTO platform until you get the Dart. It is very efficient for what does: less than 211 funds a passenger (4) to orbit. It is trivial to upgrade the jet engine and intake as you get better tiers. Power-plant isn't that original, but I like the aesthetics of my spin. This is actually an evolution of another, lower tech design I've since expunged. Flight profile is easy but doesn't have too much margin. Reach peak velocity (~200 m/s) ASL under military thrust. Be careful the wing incidence makes it a little squirrely here. Engage Afterburners, Pitch nose to 10 degrees above horizon. Again the wings will fight to go higher. Don't let them! If you manage pitch right, you should wind up at the base of a shallow zoom climb at angels 12 with prograde 10 degrees above the horizon and ~800 m/s. Light the boosters (stage) and pitch up. Keep heading 10 degrees above prograde (touch the v to the prograde circle) during climb for thermal management. Cockpit will not easily survive atmospheric circulation. You should have 300+ m/s in a 75 x 75 orbit. Deorbit burn PE below angels 20 and use cobra reentry.You can level out at 1400 m/s. Glide AoA is somewhere near prograde. That excess lift now allows for a nice slow landing.

Third. My favorite launchers are SRB first stages that launch at 1-2° and 60-80% thrust with no SAS. I only exercise control to correct heading/inclinàtion. Then engage prograde follow for stage 2.

The biggest tip I can give is to use a retrograde Mun orbit. You can't free return from a prograde Mun orbit in a practical manner.

Actually plane correction optimization is frequently different than that. If you can cause the desired and actual orbital planes to intersect at Pe during mid course correction you can include plane alignment during capture burn. Low to moderate inclination corrections are cheaper as a normal component of a already large capture burn than a separate normal burn at higher altitude thanks to the nonlinear nature of vector addition. Getting that Pe on the intersection at mid course burn only requires prograde and normal vector adjustments. Throw in radial though and you now control insertion time. At this point you can also rendevous on capture moderately efficently! The only challenge is polar orbits where you want a highly different AoP. That relies on rendevous time more than inclination. However the only time I've cared about AoP is satillite constellation construction.

Mk2 parts for atmospheric planes. Their drag is conspicuously worse than Mk1. Additionally that drag offers no mass savings or capacity upgrades. The only reason I still try to design with them is 1100 K is far too low a heat tollerance for an exposed part to sustain velocity past Mach 3. They do make great aerobrake capable taxis though. So I guess they technically don't count. Speaking of too low a heat tollerance: FAT wings. By the time I unlock these, it's high-Mach or bust! Also, an obligatory mention of struts and fuel lines.

Further food for thought on canards vs. tail planes: Tail plane control surfaces decrease in AoA as you pitch; max deflection produces weaker lift the higher your craft AoA Canards are the opposite: AoA increases the further your craft is from prograde. Stalls of canards still create a retrograde force as drag increases and now you are pulling back on the nose! I currently believe the secret to stable canard designs are enough passive stability that the canard cannot overcome it even in the worst case (radial ship orientation with control plane perpendicular to airflow). A sufficiently large, tailless, delta wing helps here.

I use the same constellation for my career save. Injector releases from high, equitorial, circular orbit. Sats were placed in three burns each: inclination, phase correction, final orbit. I used a Google sheet to calculate parameters (I wanted the tetrahedron orbit to have even resonance with a polar sacansat for science so I used various SMA to radius ratios). Wonder if I should share it.

That thread is actually what got me started on this. For my Minmus network I inserted a carrier into high (final apsis) equatorial orbit and delivered relays into inclined orbits without modifying orbital period. Then as they reached the intended final apsis after (up to) one go around I performed burns to place the first into final orbit and the others into individual phase correction orbits (to both correct the different periods between release and swap the phase of one pair). The remaining three burned into final orbit after one more orbit. My Mun network was used to debug this procedure. My polar scan-sat will act as a pace keeper if I need to correct phase later. This system doesn't work efficiently for Kerbin because there is a greater dV cost between low and high orbits. Inclination changes in low orbit are also painfully expensive. If I could have a high degree of confidence in launch efficiency, I could launch strait from Kerbin into an inclined intermediate orbit, thus my question. I am starting to think the best approach would be a LKO seed ship, and have the relays boost themselves up to a HKO intermediate orbit and perform the plane change during the second Hohmann burn.

I'm planning on launching a Kerbin based Draim Tetrahedron relay network before turning off DSN in my career settings. Given the circumstances of the location, 4 separate launches would be the most efficient way to deliver the satellites (as no orbital planes are shared between relays). Timing my launches for correct orbital planes sounds easy enough, but I don't know how I can keep the launches consistent enough for synchronized relay placement. Does anybody know of anything that can (or help) launch payloads into consistent orbits or should I just synchronize orbits with a T/(2*n) reference orbit before final placement maneuvers (including phase correction orbits)? Alternatively, I could just throw money at the problem and use the same system I used to seed the Mun and Minmus with their networks.