sevenperforce

I suppose there is some degree of difficulty in precisely defining "amateur". For reference, this is what the world-record-holding amateur hybrid rocket launch looks like: Dry mass of 75 kg, wet mass of 163 kg, single 10 kN N2O+paraffin hybrid rocket motor. Max velocity was Mach 2.3. Some analysis here. Specific impulse was 198-217 seconds, probably giving an effective dV of around 1500 m/s, but Mach 2.3 is only about 720 m/s. Gravity drag on a 15-second burn period comes to about 150 m/s, so for this particular design we're looking at roughly 600 m/s of aerodynamic drag losses. That's with a launch TWR of just over 6:1. This is instructive, since at least the initial launch phase will be very similar for our vehicle. We may go with a slightly lower TWR to decrease drag losses, though this in turn increases gravity drag. There's a whole optimization problem there. A useful exercise would be to use the parameters of the HEROS-3 hybrid rocket and determine how many serial stages it would take to get this vehicle into orbit, using the OTRAG bundled-stage approach. Then compare to the Lambda 4S vehicle for reference. Air-augmentation shrouds do need to be integrated, but if you need a fin mounting place anyway, then you're already planning for some structural mass. Modeling is a lot cheaper than building, that's for sure. Note that air augmentation at these specific impulses will be useless above 2.5 km/s, but by that time I hope we'd be well out of the atmosphere anyway. We cannot come near the energy density of orbital-grade solid-rocket fuel, but we can get the impulse density quite high with jellied petrol and HTP. Higher than kerolox for sure. I'll try to put together a dV check on the Lambda.

They didn't use a drone.

You could put them in paralell. Which is probably what @sevenperforce meant. Like, poke a lot of holes into a tank and put a valve into each hole. And you don´t need that many valves. 6 would give you the ability to throttle in steps of ~1.5%... Precisely. The base of the HTP tank would have, say, 7 openings (one in the center and the others arranged in a hexagonal pattern) with an on-off valve in each one. Although the steps wouldn't be as fine as 1.5%; it would be in 17% increments. And it's very possible that 17% might be below minimum possible throttle for stable combustion, in which case you might only have five throttle settings: 33%, 50%, 67%, 83%, and 100%. But, still, that's a reasonable degree of control. The point of having multiple binary valves rather than finely throttleable valves is, again, cost reduction. The HTP tank valves need much more flow capacity than, say, RCS thrusters. But it will almost definitely be cheaper to purchase a dozen small binary valves than it would be to purchase small binary valves for the RCS thrusters and one big throttleable valve for the HTP tank. Plus, that's only one wiring configuration to worry about. Another thought I had last night: we don't actually need a separate pressurant tank. We can just take a page from OTRAG's playbook and fill the HTP tank 75% full, then top off the rest with air at 600 psi or so. Actually really ridiculously simple: The vertical spacing between the catalyst bed opening and the nozzle throat is very important. The chamber pressure is a direct function of the burn surface area. As the jellied petrol burns away, the surface area must decrease so that chamber pressure decreases and the jellied petrol can flow down. This ensures the correct feedback loop; otherwise you are trying to compensate by constantly varying pressure and you end up with runaway problems. Another issue, particularly if we are considering air-start of hybrid upper stages, is ullage. Ullage is never a problem in KSP, but since we are dealing with liquid HTP rather than compressed nitrous or another gas, there must be an acceleration vector when the throttle valves open or it will have very big problems. The RCS thrusters need to be angled down, like Verniers. If we end up air-starting, then we would need to start firing the upper-stage RCS thrusters before separation, while lower-stage acceleration is still holding the HTP at the bottom of the tank. This will ensure that HTP is forced up through the uptake lines. Note that there would also be a smaller catalyst bed in front of each of the RCS valves, though it's not shown here. With core-throttling asparagus staging (like Delta IV Heavy), you have the benefit of moderately high TWR off the pad AND moderately high TWR at staging because you've already burned a lot of your core's fuel. So TWR issues (whether we have too much or too little) can always be handled by MOAR BOOSTERS, of the hybrid-rocket variety. SRBs are fine for a serial kick stage, but I am hesitant to use them as parallel boosters, because of combustion inconsistencies. A hybrid rocket can be throttled in real-time, but SRBs cannot, and amateurs don't have the resources to cast perfectly identical SRBs that ignite at exactly the same instant and burn at exactly the same rate and burn out at exactly the same altitude. One thing we absolutely need to model (probably iteratively) is the ideal TWR curve and ascent profile for very small orbital rocket. It's not a simple problem. Instantaneous drag forces are hard enough to calculate without wind tunnel testing, but we are dealing with parallel stages, which add an entirely new set of parameters. I think our best option is to look at something like the Japanese SS-520-4, which was a serially-staged, pure-solid-fueled attempt at launching a single 3-kg cubesat. We could also look at the successful Lambda 4S, which was similar but had two strap-on SRBs. The L4S first stage is a solid-fueled sounding rocket motor pushing 215 seconds of specific impulse at sea level, a little better than White Lightning. Unfortunately, it's still very big for being the smallest successful orbital vehicle: Despite its size, this represents the minimum size for a solid-fueled orbital ELV. So this is what our hybrid parallel-and-serial rocket needs to beat. As a starting point, we should really look at the dV budget of each stage for the Lambda 4S, since it will give us a good idea of what we need to be able to field for a hybrid-fueled PRLV. From our position, we'd need to include a survey and analysis of launch site options. If you go international then you have to worry about exporting what are essentially munitions. We should field a design which is capable of reaching orbit even from sea level, though; that way we can simply note increased payload margins with alternate launch sites. EDIT: Another consideration is the potential for boosting our first stage performance with air augmentation: If the lower stage(s) need(s) a mounting point for stabilizing fins, then it might make sense to make that mounting point a shroud that can help increase thrust and specific impulse, at least on the parallel boosters.

High-test peroxide isn't corrosive, per se; it's just extremely likely to decompose explosively if there are any heavy metals or imperfections in its container. It'll be entirely benign if encased in simple aluminum. And it's only hypergolic if you decompose it; you can freely mix it with without exploding, as long as the petrol doesn't have any suspended palladium salts or something. Granted, I certainly don't want to be anywhere near a mixture of HTP and petrol, but that's beside the point. The high-test peroxide is definitely the most dangerous part of the whole apparatus, I'll give you that. But it's not as bad as it might be. IIRC, Black Arrow got over 300 seconds of specific impulse with mere 85% peroxide. I mean, I'd like to be able to put a Cubesat in orbit. But yeah, we can target our design as merely getting our terminal stage up there. That's the first step, anyway. Hmmm, very good point. A naked Falcon 9 FT first stage comes in with a pad TWR of only 1.8, which is why I was originally thinking more conservatively, but I guess I really shouldn't be making that comparison. We may have to factor single-stick pad TWR into our ultimate optimization equation, unfortunately. Sorry, I guess I wasn't clear. I didn't mean simple on-off for each of the cores; I meant that our valve choice should be uniform across the system. So each RCS thruster would have a single valve, each head pressure opening would have 3 or 4 valves, and each HTP injector would have 6 or 7 valves. So, that way you can open a discrete number of valves to give a finite but stepwise throttle range. Remember that even though decomposed HTP is hypergolic with petrol, the design I'm proposing wouldn't be readily restartable, since shutting down oxidizer flow completely would likely result in fuel extrusion. I think one of the initial investigations should be precisely how large a White Lightning orbital rocket would need to be. We need to be able to propose a design that outperforms an SRB-based system. I'm much more inclined to put additional strap-on hybrid rockets on the first stage and just throttle down the sustainer after launch. But an SRB kick stage is probably fine. All the more reason to make the strap-ons the same cores as the sustainer. Drag separation maybe, though that only works on lower stages, unless you're dealing with a LOT of stages. Spin stabilization on the terminal stage, maybe, but I prefer a vernier-oriented RCS attitude control system.

I only lawyered on that one because I've got a 17-lawn-chair RAPIER-and-nuke SSTO that I'm gonna use to get the maximum possible score....

70 km is the ceiling. The way I'm reading the rules, you're not supposed to exceed 70 km at any point in the mission.

Thought: to simplify, we could have binary (on/off, no in-between) valves and just have multiple ones in parallel for the things that need to be throttled. Then you only have one actuated control component and you can buy in bulk.

Unless I recall incorrectly, HTP can be synthesized and distilled with a fairly small lab setup. Also, to an earlier point -- we should be thinking less about "increasing payload" and more about actually ensuring we can get our rocket into orbit. What single-stick target TWR should we have? 2? 3? White Lightning can go as high as 214 seconds. Definitely a good kick-stage choice.

Well, with the requirement that each individual stage MUST be small enough to test, handle, and transport easily, I don't think we can avoid clustering at the very least. From there it's a small step to parallel staging. I think assuming a parallel+serial approach like Falcon Heavy is a safe bet. Maybe even something like a four-core sustainer with 8 strap-ons, and a single-stick second stage. I'd be interested in getting a good estimate of stage mass fraction. How heavy will each stage need to be? One advantage is that HTP is very thrusty.

I was thinking about it, and the separation system could be even simpler than I thought. You've got a high-pressure nitrogen (or other pressurant) tank, after all. So, once you reach stage burnout, you just vent remaining pressure into the separation module, forcing open the clamps holding the stages together. The venting then serves to push the spent stage away without the need for a separation motor. You can do both serial and parallel staging. Common cores ARE what I'm proposing, after all; just with throttle control to allow the core to act as a sustainer. We can look at the range of hobby rocket systems to see what possibilities there are for serial staging arrangements.

Do you mean nitrous oxide? Nitrous has higher specific impulse but much lower impulse density. Since we need a head pressurant for the fuel, its self-pressurizing capabilities don't help us much. One major advantage to HTP that I didn't note before is that once it is decomposed, it is hypergolic with kerosene and gasoline, and likely jellied petrol as well. So we need no ignition system. In fact, we could even consider an air-start if all you have to do is open a valve. For stage separation (both parallel and serial), what about using the RCS monoprop for a pneumatic sep?

Added/clarifying thought: In terms of engine design, the goal is to have stable, quasi-steady-state combustion (like a bipropellant liquid engine) while still having the fuel surface serve as a flameholder and chamber wall+casing insulator (like a hybrid rocket). This avoids the complexity of active cooling on the combustion chamber. As discussed above, the nozzle can be ablatively cooled with relatively inexpensive materials. A secondary goal for engine design is reasonably rapid reusability, another advantage of napalm over paraffin or rubber. There are solid+solid rockets (standard SRBs), liquid+solid rockets (LOX hybrids), vapor-solid rockets (nitrous hybrids), liquid-liquid rockets (standard liquid rockets), vapor-liquid rockets (FRSC and ORSC), and vapor-vapor rockets (FFSC), but I don't think I've ever seen a vapor-gel rocket. In terms of staging design, the goal is to have individual rocket cores which are small enough to be handled, filled, and transported easily but large enough that you don't need dozens and dozens of them in parallel. It may be that we end up with a stage size where a single-stick can do high-altitude atmospheric sounding, a 1+3 parallel arrangement can do suborbital flight, and a two- or three-layer asparagus system with a sugar-rocket kick stage can reach orbit. For those unfamiliar with OTRAG, it's a modular pipe-based pressure-fed liquid rocket system with an ablative nozzle, using differential thrust for pitch and yaw control.

The chamber would be annular, that is, cylindrical. So the pressure and flow rate would be constant around the cross-section. There would definitely be a minimum throttle settling, though that's typical of any real-world throttleable rocket engine. It probably would not be restartable (though restarting would be outside the range of what we would try to do anyway). Cabling could run to the pressurant chamber without difficulty since it would not be a high-temperature region; the pressurant would necessarily come out quite cold due to expansion. The valve leading to the catalyst bed could be passively pressure-actuated; all throttling could be controlled from the pressurant tank. You could even vary the mixture ratio, within certain limits.

At the very least, we could show what the minimum threshold for an amateur shot at orbit would be. The primary concern I have about using jellied petrol is fuel stability. In this video of a LOX+wax hybrid rocket, imperfections in the wax casting cause a chunk of the wax to break off and be forced out through the nozzle; the LOX burns through the exposed motor casing a second or so later. From what I've read, the biggest challenge with hybrid motors is getting fuel that is "wet" enough to vaporize consistently, but still solid enough to retain its shape throughout the burn. Napalm is, of course, much closer to a liquid than to a solid, so the main problem is making sure the fuel stays inside the chamber. It's not an insurmountable problem -- after all, we're just designing, not building, so we can skip over some of the engineering challenges. But we do need to make sure we have a viable approach. It might be worth proposing/investigating some alternate shapes that would ensure combustion pressure keeps the fuel inside rather than forcing it out (e.g., blurring the line between a hybrid engine and a pressure-fed liquid engine). The traditional cylindrical fuel casting has some advantages, like thermally insulating the motor casing, so those issues would need to be considered. For example, here's a possibility: In this design, the pressurant (which could be liquid nitrogen, as shown here, or even really high-pressure air for additional oxidization potential) and the HTP tank are stacked inside the motor where you'd typically have the combustion area. A large lower valve on the pressurant tank forces the HTP down, through a catalyst bed, while a smaller upper valve produces head pressure on the napalm. In this way, the napalm column burns from the bottom, but undergoes plastic deformation and flow under head pressure and acceleration, so that the burn surface remains at the same point, and combustion pressures prevent it from flowing out the nozzle. So it's basically a pressure-fed liquid bipropellant rocket, but with the fuel being stored and "injected" in gel phase rather than in liquid phase. Shouldn't be substantially harder to build than a standard hybrid rocket. Agreed. Clustering and parallel staging is inherently more challenging than serial staging due to the potential for differences in combustion rates, but if you are using throttled hybrid rockets then you can compensate. It may be that we find a 10 or 15-core cluster is what's required.

Well, again, it comes down to the difference between amateur-level rocketry and professional rocketry. Solid-fueled rockets with reasonably high specific impulse are possible -- the Shuttle SRBs have a vacuum isp of 269 seconds -- but high-energy solid rocket fuel is ridiculously dangerous, difficult to handle, hard to cast, and chemically complex. Solid fuel available to amateurs is much less energetic; commercial Estes hobby rockets have isp ranging from 60 to 90 seconds, and the sugar+KNO3 rockets I've made range from 100 to 130 seconds. In contrast, a HTP+jellied petrol hybrid rocket should be able to push at least 250 seconds at sea level and as high as 300 seconds in vacuum specific impulse. And unlike some other hybrid-fuel oxidizers, HTP is exceedingly dense. Definitely! You could also get corporate sponsorships for high-profile amateur work. First step, I think, is doing some optimization to figure out what staging combination will result in the smallest overall LV size. Gross vehicle size, more than any other single factor, is going to be the greatest determiner of overall project cost.

Yes, it's all-solid. Hobby-class solid rocket motors are small enough that handling is not dangerous, and stuff like combustion instability, burn rate, thrust curve, and propellant grain never really come up. But when you move into larger SRBs, these things all become problematic. Building solid-fueled first and second stages large enough to get a cubesat-sized payload into orbit is just a little beyond what amateurs can really reasonably attempt, even if the upper stages would be more manageable. Parallel staging with a high-impulse-density hybrid fuel allows all the stages to be of a reasonable size but have a stack with potentially enough energy to actually reach orbital speeds. Oh, I would definitely say yes about keeping the solid kick stage attached to the payload. No reason to bother with the added weight of a payload deployment system. There would definitely be a lot of permitting involved, but starting with the right propellant choice helps a little. There are some permitting costs that are unavoidable but the idea is to drive down overall costs. This is just a pipe dream, but it REALLY would be cool for a bunch of video game nerds on a game forum to actually design (and perhaps publish?) plans for an orbital rocket that amateurs could reasonably construct. I certainly don't have the time, space, or funds to actually build something like this, but we could definitely come up with the designs, do the math, simulate it, investigate permitting, and all that stuff.

Air-ignitions for solid rockets aren't as much of a problem. It's just a liquid or hybrid rocket that is difficult. My ideal design is very strongly influenced by the Atlas rocket that put John Glenn in orbit. Parallel/drop staging, MECO just short of orbit, solids for circularization. I mean, technically you could use a tiny hybrid rocket and plumb it from the main stage's HTP tank, but since HTP is a monopropellant it is MUCH easier to just use that. Plus, with Verniers you're still doing differential throttling.

I'm worried less about the actual material itself and more about the actuating mechanism. It would be complex, heavy, and fragile, and the cost reduction of chuting down the otherwise-sturdy spent stages is substantial for an amateur approach.

DIRECT was the brainchild of the NasaSpaceFlight forums, where a bunch of industry professionals got together and proposed a design to repurpose Shuttle-era hardware. This ultimately evolved into the SLS. Unfortunately, the SLS has evolved since then, to the point that it's no longer really useful for much of anything.

Vernier thrusters for a hybrid rocket would need to be their own whole hybrid rocket engines, which seems like a large complication when differential throttling on the parallel cores can do the trick. Though I would imagine angling the integrated RCS so every puff pushes at least a little bit prograde. As far as payload, I'm thinking just getting a guidance and comm system into orbit would suffice to win the "first amateurs to orbit" trophy.

I don't think the individual cores would be larger than a serial first stage. Falcon Heavy could still take a reasonable amount of payload to orbit even without its upper stage, and its cores are the same size as the F9 first stage. And here we'd be dealing with three boosters instead of two. That being said, a kick stage (maybe made from a cluster of separation motors) for the final push into orbit is still probably a good idea. If we were dealing with low-energy propellants, fins might work...but with medium-energy propellants like HTP and jellied petrol, specific impulse is high enough that you'd run out of aerodynamic control authority before the boosters ran out of propellant. And the vanes are a REALLY complex metallurgical problem. RCS allows replication of the HTP injection system and allows attitude control even while the engine is off. LOX is definitely higher-isp, but it is harshly cryogenic and eats anything it touches. It's also not self-pressurizing since it isn't a monopropellant like HTP. Finally, since the impulse density of jellied petrol is lower than most hybrid rockets, the higher impulse density of HTP helps. You can also synthesize HTP yourself...can't do that with LOX, not in any significant quantities.

A Z-pinch's capacitor bank can charge on solar power if the pulse rate is lower, and still outperform NERVA. But a Rankine-cycle small nuclear reactor would likely be a good idea anyway for the ship's power systems.

Fair point about turbopumping the oxidizer, but I was more thinking about tank pressure. Hybrids and solids, like pressure-fed liquid rockets, have to have a body that can contain pressures equal to or greater than combustion pressures, since the combustion takes place in the "tank". This makes mass fraction higher, but it also means going asparagus certainly wouldn't be a strength problem.

Well, by "should" I meant from an engineering and rocket science perspective.

I still think they should replace the square-arranged 4-engine stack on the EUS with a 1+3 arrangement that allows three engines to be dropped mid-burn, Atlas-style. It's not a ton of added complexity but it significantly increases payload capability.