[Stock] [1.0.5] Nova IIB - 230 tons to LKO

[Temstar]

So a while ago I created the Nova II Ultra Heavy Lift Launch Vehicle. But there was always a bit of nagging thought in my mind that it was a compromised solution. The main reason why the original Nova II had a reusable upper stage was that at the time I didn't think it was possible to recover the entire core stage of the asparagus. The reusable upper stage solved the problem of cost, but also introduced a structural weakness with its thin probe core / reaction wheel section that weakened the rocket and was only partially fixed by strut spam.

Then, while working on my fully reusable launch vehicle I came upon a way to recover large rocket stages horizontally:


Of course this is a relative small stage, a Nova II core will be more than 3 times as heavy as this, but nevertheless it showed the way forward.

Nova IIB Ultra Heavy Lift Launch Vehicle
Craft file: http://kerbalx.com/Temstar/Nova-IIB-UHLLV
(Craft file contains proofing payload)

Specifications:

• dry weight: 156 tons
• wet weight: 866 tons
• cost: √304,039 (including fairing)
• part count: 163
• payload: 230 tons to 75km x 75km orbit
• payload fraction: 20.58%
• core stage value (without fuel): √110,299

Cost per ton to LKO:

• without recovery: √1320
• 95% average recovery rate: √865
• 90% average recovery rate: √889

Nova II's original cost per ton was calculated incorrectly because the RHEUS value was double counted, cost per ton to orbit with 95% recover for Nova II should be √1080 per ton. Comparing the two vehicle the Nova IIB has cheaper per ton to orbit, slightly higher payload to orbit, lower part count and greater structural strength by getting rid of the upper stage weak spot. When not using a fairing the Nova IIB can launch a 240+ ton payload, the same payload cannot be launched at all on the Nova II because it will snap the upper stage at its structural weak point. So it's cheaper, stronger, more powerful and less complex.

Typical mission profile:
 

Spoiler

Turn on SAS, throttle up to max and stage. Once the booster has reached altitude of 800m, tilt over 5 degrees to east, toggle SAS to prograde and wait for separation of first booster pair at T+ 31 seconds.


Keep following the prograde and leave the throttle on full power for the entire boost phase. Second booster pair separation should happen around T+ 1 minute 13 seconds, and the last pair at T+ 2 minutes 17 seconds. You want the Time to Apoapsis to hang around 30-60 seconds during the core stage burn. Since the core stage has relative low TWR you may need to pitch up to above the prograde marker to keep Time to Apoasis to within acceptable range. Don't be alarmed by engine overheat warning - the extremely long burn utilised by asparagus staging causes this and its well within the engine's capability to withstand.


Discard the fairing at 40,000m. Be careful with steering the rocket once fairing is jettisoned - the fairing is actually load bearing with the payload tip strutted to the inside of the fairing. With the fairing gone the rocket is more wobbly so steer gently


Payload release to 75km x 75km orbit. Drop out of warp 30 seconds before the AP and turn on RCS to steer to prograde. Nova IIB uses the very powerful bipropellant RCS thrusters to give plenty of steering authority during coasting. There's also a forward facing RCS thruster inside the payload decoupler so you can back away from the payload.


Reentry of the core stage after one orbit around Kerbin. Peak-heat load is generally at about 81% of the maximum core temperature with the hottest part being the airbrakes, so plenty of thermal head space should you want to send lighter payloads to higher orbit. You do need to deploy the airbrakes at the start of the reentry, if airbrakes are not deployed throughout the reentry the Mammoth engine will eventually overheat.


Approaching KSC.


Core stage shortly before touch down on KSC ground. Remember to deploy the landing gears! The parachutes have been tuned so that as long as you've emptied all the LFO it will land the core stage horizontally without any reaction wheel input at all.


Core stage safely landed at KSC for high recovery value.


Splashdown of the core stage - as with water landing no steering input is needed. Remember to deploy the landing gear for water landing as well as the gears act to cushion the water landing, if they are not deployed there's a chance that the forward Kerbodyne S3-7200 tank could break.

Enjoy

Edited January 31, 2016 by Temstar

[Rune]

6 hours ago, Temstar said:

Nova II's original cost per ton was calculated incorrectly because the RHEUS value was double counted, cost per ton to orbit with 95% recover for Nova II should be √1080 per ton. Comparing the two vehicle the Nova IIB has cheaper per ton to orbit, slightly higher payload to orbit, lower part count and greater structural strength by getting rid of the upper stage weak spot. When not using a fairing the Nova IIB can launch a 240+ ton payload, the same payload cannot be launched at all on the Nova II because it will snap the upper stage at its structural weak point. So it's cheaper, stronger, more powerful and less complex.

So basically, going simpler and reusing more ended up being cheaper. Big surprise!

Anyhow, I got an idea for the Nova III or whatever: reuse/recover all expensive chemical engines, and use less in the first place. Yeah, SSTO is hard on payload, but solids and tankage are cheap. Take out the Twin Boars, put 4xKickback bundles in their place, and feed the fuel in the upper tanks to the core so when the SRB's go they are empty and the core is comfortably above TWR 1 (without them). The core might get a bit bigger, and you are looking at a higher takeoff weight and lower payload fraction but I doubt you need more thrust in the core for a similar payload, so price/kg should be way better.

At least a 5-minute hastily put together draft seems to have a promising 3,400m/s dV with a 225mT payload, sans recovery equipment, with a single Mammoth on the core and six SRB bundles to help it until TWR>1, bringing pad TWR to 1.67 and, most likely, giving a beautiful thrust curve/ascent path. 220k√ on the pad, BTW, but I didn't even bolt on a probe, so take with a grain of salt. SRB are sooo cheap. I want a 2.5m one!

Edit: also, I don't get why landing them horizontal was such a hard-set requirement... you can put pretty tall things down in one piece if you have sturdy legs!

 

Rune. I might have gotten a bit carried away as I wrote, having KSP open helps.

Edited January 31, 2016 by Rune

[Giggleplex777]

2 hours ago, Rune said:

So basically, going simpler and reusing more ended up being cheaper. Big surprise!

Anyhow, I got an idea for the Nova III or whatever: reuse/recover all expensive chemical engines, and use less in the first place. Yeah, SSTO is hard on payload, but solids and tankage are cheap. Take out the Twin Boars, put 4xKickback bundles in their place, and feed the fuel in the upper tanks to the core so when the SRB's go they are empty and the core is comfortably above TWR 1 (without them). The core might get a bit bigger, and you are looking at a higher takeoff weight and lower payload fraction but I doubt you need more thrust in the core for a similar payload, so price/kg should be way better.

At least a 5-minute hastily put together draft seems to have a promising 3,400m/s dV with a 225mT payload, sans recovery equipment, with a single Mammoth on the core and six SRB bundles to help it until TWR>1, bringing pad TWR to 1.67 and, most likely, giving a beautiful thrust curve/ascent path. 220k√ on the pad, BTW, but I didn't even bolt on a probe, so take with a grain of salt. SRB are sooo cheap. I want a 2.5m one.

Rune. I might have gotten a bit carried away as I wrote, having KSP open helps.

Woah. Can you post pics of the entire rocket?

Edited February 1, 2016 by Giggleplex777

[Temstar]

2 hours ago, Rune said:

So basically, going simpler and reusing more ended up being cheaper. Big surprise!

The thing is, dividing the centre state into two should improve the dry weight to orbit and thus gain you some delta-V. The thing to offset this is that the upper stage engine is a dead weight until staging. It just happened that Nova II's staging have these two effects just about cancelling each other out so combining the two stages back into one neither gained or lost performance. I think rule of thumb for rocket design is that if all your stages use the same fuel, then each of them should contribute about equal delta-V to be optimum. Nova II's upper stage is a lot smaller than the equal delta-V divide for other reasons so it wasn't very delta-V optimum.

I know about the all solid first stage then liquid 2nd stage trick and in fact I use it on my go to medium left launch vehicle. It won't really work on 200+ton sized rockets though. By my count a Nova II using all solid first stage will need to fire 30+ kickbacks! Until they give us 2.5m SRBs that's probably too many parts to be worth it.

As for tail sitter landing, I find it's mostly water landing that's giving me grief. I can get it to land fine without smashing the engine but inevitable it will tip over and if the stage is tall enough the tanks at the top will be smashed when they hit the water. In any case horizontal landing uses the same number of parachutes as tail sitter and probably need less parts since the landing gears are so robust can you can get away with just three of them, rather than many landing legs.

[Rune]

33 minutes ago, Temstar said:

The thing is, dividing the centre state into two should improve the dry weight to orbit and thus gain you some delta-V. The thing to offset this is that the upper stage engine is a dead weight until staging. It just happened that Nova II's staging have these two effects just about cancelling each other out so combining the two stages back into one neither gained or lost performance. I think rule of thumb for rocket design is that if all your stages use the same fuel, then each of them should contribute about equal delta-V to be optimum. Nova II's upper stage is a lot smaller than the equal delta-V divide for other reasons so it wasn't very delta-V optimum.

I know about the all solid first stage then liquid 2nd stage trick and in fact I use it on my go to medium left launch vehicle. It won't really work on 200+ton sized rockets though. By my count a Nova II using all solid first stage will need to fire 30+ kickbacks! Until they give us 2.5m SRBs that's probably too many parts to be worth it.

As for tail sitter landing, I find it's mostly water landing that's giving me grief. I can get it to land fine without smashing the engine but inevitable it will tip over and if the stage is tall enough the tanks at the top will be smashed when they hit the water. In any case horizontal landing uses the same number of parachutes as tail sitter and probably need less parts since the landing gears are so robust can you can get away with just three of them, rather than many landing legs.

Yup, all true. couple that with KSP's TWR and tankage ratios, and you can plot stuff to give you an optimum mass ratio per stage, optimized either for weight, or cost. Once youhave that, you can get the "right" number of stages for a given trip, and LKO is about 1.5-2 as you are finding out . In any case, I don't think you got my comment right... 3.5km/s might be a bad ratio to get a decent payload, but 1.5+1.5, in a stage-and-a-half config is great, since you can take big hits to performance and still make it, and solids make great boosters since they are cheap. This is what I meant:

Not quite 30 kickbacks, but rather 24. >3.4km/s with a 200mT shrouded payload and nice TWR, thanks to dropping part of the tankage (the Big Red + a nosecap tank) along the way. Part count before adding recovery stuff (and struttign between teh boosters and the core!) is 101, so not that bad. And cost is great, you gotta give me that, not to mention the core is quite a bigger chunk of the initial cost (60k√ out of 220k√). Would work out to ~160k√ per 200mT, or ~800√/mT with recovery, but even better, only a bit over 1100√/mT without. Not so bad, percentage-wise... if the cost of the recovery gear is kept down.

37 minutes ago, Giggleplex777 said:

Woah. Can you post pics of the entire rocket?

You meant the landed thing or my suggestion for solid boosters for the Nova? If it's the former, it's easy, that's a 50mT SSTO that can loft a fueled LackLuster to orbit, so just picture one on top. Short of my take on the MCT concept, still WiP. If it's the second, I already did.

 

Rune. Vectors would also be the ultimate powerplant if you want to take things bigger... or smaller. As long as you recover them!

Edited February 1, 2016 by Rune