You are not logged in.
I (Be aware that free-expansion designs operating far off design will have far lower nozzle kinetic efficiencies, too many folks assume high efficiency from such designs, and that is simply not correct.)
GW
Please, forgot my ignorance: what is exactly a free expansion design?
GW Johnson wrote:To answer Quaoar in post 7, yes, I think your Isp simulator software is doing a decent job, as far as it goes. I think you might find a nozzle with an area ratio = 1000 is not something easily built, and likely rather heavy.
...
GWQuaoar, did you find the Cearun simulator did a good job for the sea level ISP? I’ve been using the free version of the Rocket Propusion Analysis program, http://propulsion-analysis.com/index.htm.
By comparing to known engines, I found it does a good job for the vacuum ISP, but poorly for the sea level value. I’d use Cearun eventhough it doesn’t have a good GUI, it it did do a good job simulating sea level Isp’s.
Bob Clark
Hi, Bob
I've also tried to simulate the glorious F-1: the vacuum and sea level Isp of the real rocket were about 96% of the virtual one
OF:
My ancient reference is too old to reflect the modern views on toxicity. Typical at NASA they use full suits with self-contained breathing gear to handle the NTO-hydrazine systems. Actually getting some NTO in your lungs was usually an immediate death sentence. It was a worse threat than the hydrazines, which were bad enough. [...]
Soldiers in the field without such gear were able to handle IRFNA-kerosene without much problem. That's a lower-performing system, but it works, and with hypergolic ignition. Such missiles were common in the 1950's, but pretty much got replaced by the "wooden-round" solid propellant systems by the 1970's. Far easier to handle. The still-lower performance was not objectionable for those applications.
I've been told that H2O2-kerosene is not really hypergolic, except maybe at H2O2 purity near 100%. Some claim stability for high test H2O2, but its reputation was poor in prior decades. Lots of stuff blew up, including one USN submarine. Even today, no one is claiming storage times longer than a few weeks for high-test H2O2. Most only claim days.
IRFNA isn't "just" nitric acid, there is some NTO in it. But it's mostly just a strong corrosive that must also be isolated from cellulose to prevent nitration into something explosive. It's probably easier to handle in the field than LOX. Certainly far less complex to handle.
GW
Hi, GW
I used NASA software to simulate a real rocket like the AMBR (NTO-MMH, O/F 1.2, chamber pressure 18.96 bar, expansion ratio 1:400)
with these parameters it gives a vacuum exhaust velocity of 3340.7 m/s versus 3283 m/s of the real rocket. So the real rocket is about 98.3 % of the ideal simulated rocket.
Changing the O/F ratio to 2.4 and enhancing the nozzle expansion ratio to 1000, the simulator gives an exhaust velocity of 3722.2 m/s: taking the 98.3% of this value, a real rocket with these parameters should have an exhaust velocity of 3658.9 m/s: do you think we can thrust the simulator?
Quaoar,
I presume the stability of this type of combustion process is well-characterized, since this looks like an application of combustion instability to prevent the effects of combustion instability from destroying the engine. Would that be correct, or is this technology still in development?
It looks very interesting, though. I'm definitely going to read more about it.
Orbital Technology Corporation (Orbitec) still produces commercial vortex cooled rocket
https://www.sncorp.com/press-releases/s … ex-rocket/
This is the takeoff of a vortex cooled rocket
Brian Palaszweski has investigated the possibility to freeze 15% weight of single H hydrogen atoms in hydrogen slush, making a very high energy density propellant.
https://ntrs.nasa.gov/archive/nasa/casi … 009862.pdf
https://ntrs.nasa.gov/archive/nasa/casi … 011425.pdf
Using the NASA rocket simulator, I tried this mixture and found an an exhaust velocity more than 7500 m/s with an expansion ratio of 500 and a chamber temperature of 2900°K
https://cearun.grc.nasa.gov/OFILES/quao6492.html
adding oxygen at an O/F ratio of 1.5, the exhaust velocity is lowered to about 6700 m/s, but thrust is increased: we can add it in the nozzle as an afterburner like in the LANTR.
Hypothetically, if we succeed in storing 30% weight of single H in hydrogen slush, we can reach exhaust velocity of about 9800 m/s like an advanced solid core nuclear rocket, but with chemical rocket like T/W ratio
Playing with NASA rocket simulator, I "discovered" that NTO-MMH rockets have not the optimal O/F ratio: it is kept near 1-1.2 to have an acceptable chamber temperature near 2500 °K, but the optimal ratio, which gives the best Isp, is about 2.4. Using this O/F ratio the vacuum exhaust velocity rises over 3600 m/s (with an expansion ratio of about 500) but the drawback is that even the chamber temperature rises over 3200°K.
Anyway, in Orbitec vortex-cooled rockets, the oxidizer is injected tangentially to the chamber, creating a vortex that keeps the combustion in the middle without touching the wall.
http://www.celestialmechanics.co.uk/vortex.html
Combining these two items, we can make high Isp storable propellant rockets, ideal for a Mars manned mission, because NTO-MMH tanks can be easily sent to Mars orbit as return propellant with electric propulsion, without boil-off trouble.
SpaceNut,
Thanks to the hard work done by Ascent Solar, the claim that solar power could not possibly provide the type of energy required for VASIMR to get a craft to Mars in 39 days is simply not true. Dr. Zubrin made the assertion that a 1kW/kg power source simply didn't exist and at the time he made that assertion he was correct. Ascent Solar has since had 2kW/kg (about 2.45kW/kg, actually) thin film photovoltaic arrays in testing as I write this and their 1.25kW/kg photovoltaic arrays are current production technology, slated to power Japan's Jupiter-bound solar sail-powered probe, and actually powering long duration lighter-than-air drones made by Airbus Defense. Even after you factor PMAD and propulsion hardware into the equation, you arrive at a completely realistic power-to-weight ratio (PWR) for the complete system. I don't think VASIMR is the optional propulsion solution, given recent developments with solid-fueled (Iodine crystal) plasma thrusters with better thrust-to-weight ratios (TWR), but that takes nothing away from the fact that VASIMR would actually perform as advertised. At the time that Dr. Chang-Diaz made his assertion about his propulsion system, it was utterly fantastic because no such solar or nuclear propulsion system with the requisite PWR existed. Such is no longer the case. Time marched on and technology advanced.
I've seen their site:
http://www.ascentsolar.com/custom-solutions.html#wide
They claim 1100 W/kg for bare modules (without the supporting structure), 200 W/kg for the superlight solar blanket and 30 W/kg for the extreme blanket, but I couldn't find anything about 2450 W/kg
The X3 thruster weighs 227kg and produced 5.4N of thrust at that steady state power level. It's full input capability is 200kW, but that's for "low gear", like the gears in a transmission. The 200kW run is expected to produce a bit more than double that figure. Let's just call that 54N/MW. Aerojet-Rocketdyne has a design for a version of the X3 technology with more channels (gears) that weighs 320kg and accepts 1.2MW to 1.4MW of input power and projected produce 36N to 216N of thrust (output does not scale linearly with input power in these electric thrusters and also increases as input voltage increases). X3 and all variations of X3 are NHT's or Nested Hall Thrusters with concentric ring channels. No major design hurdles had to be cleared to get X3 to work.
Edit: Isp ranges between 1200s for low gear and 3800s for high gear
The internet is littered with research papers about MW-class MPD thrusters that were run by universities or NASA or other space exploration agencies. Google "MW class MPD thruster" and start reading.
MW-class MPD thrusters from:
NASA Glenn Research Center
Georgia Tech (variant closely associated with the one used at NASA GRC)
various universities in Japan, Europe, and Russia
There are some issues whit electric propulsion: a 5 MW solar panel array, considering a very optimistic triple junction panel with 40% of efficiency and a mean solar irradiance of 1000 W/m2, must have a surface of 12500 square meters, and that huge surface has to continuously face the sun during propulsion. The best way to do it, avoiding shadowing between panels, is to split the surface in two 80-meter-sided squares and mount them on a rotating platform, that can be oriented to the sun regardless of the thrust direction. This is not impossible to do, but is not very simple - given that now we have no more Space Shuttle and Canadarm - and even if the panels are light, the whole structure - panel arrays plus support and motors - can be heavier than 1400W/kg.
Another issue is the artificial gravity: a chemical rocket or a NTR-propelled space ship can be made as a GW's rigid baton and can spin while coasting, providing artificial gravity to the astronauts. A solar electric spaceship has to use the ion engines for the whole part of her trip, so if you want artificial gravity you are forced to build a ship with a separate spinning section. Event this is not impossible but is difficult, massive and expensive.
For those reasons I think a chemical or NTR spaceship is best suited for the crew, while a solar electric one may be useful as a cargo-tug.
Quaoar,
We've tested MPD thrusters up to 1.5MWe at universities and NASA has tested MPD thrusters up to 1MWe. Perhaps we'd need multiple engines, but the point is that it can be done. The Ascent Solar folks have tested thin film arrays of up to 2.25kW/kg in their labs. The tonnage involved with a multi-MW-class thin film solar array is clearly not a major problem. It's an engineering problem to be sure, but the basic 1.4kW/kg thin film solar technology has already flown in LEO for more than 6 months in preparation for use on JAXA's Jupiter solar sail probe.
If the multi-MW thrusters were problematic, although clearly it's just a matter of feeding in enough power, then an array of 25 of the 200kW X3 thrusters would still work. The X3's would weigh about 5.6t, whereas 5 of the MW-class MPD thrusters would weigh than 1t, but it's still feasible.
The propellant and launch costs savings are considerable, to say the least.
Interesting. Which is the power/weight ratio of the ion engines?
Quaoar,
For whatever it's worth, ion engines have already operated continuously for years without maintenance of any kind. There is no chemical rocket engine that comes close to touching that reliability figure. By their very nature, absent considerable maintenance, all combustion engines are limited lifetime propulsion systems. The greater the power output level, the shorter the lifetime and the more involved the maintenance operations become. I'm not sure there's any way around that.
With 600kWe to 700kWe class ion engines, the number of days it takes to get to Mars, including spiral out / spiral in time, is no more than the number of days BFS requires. Argon may require a little more input power than Xenon, but it's significantly more affordable and no new development is required. The thin film arrays from Ascent Solar have already demonstrated 1400W/kg in LEO. If a ship required 1MWe of input electrical power, then that's about 1.5t worth of hardware using multi-stranded array tethering materials and gyroscopic deployment and stabilization of a circular array.
It depends on the ion engine exhaust velocity and the mass of the spaceship. For example, an hypothetical NSTAR-like ion engine, with an exhaust velocity of about 30 km/s, with 1 MWe of power has a propellant flux of about 0.0022 kg/s (2E/V^2) and a thrust of about 66 N. The non impulsive delta-V transfer from Earth to Mars is about 10 km/s, so a round trip has a delta-V of about 20 km/s, so a ship with the aforementioned ion engine needs a Mass ratio of about 2.
Let's take a BFR-like spaceship with a inert mass of 85 ton and 150 ton of payload (lander + supplies): she has a dry mass of 235 ton, 235 tons of propellant and a wet mass of 470 tons.
A 470 ton spaceship with a 66 N ion-engine has an acceleration of 0.00014 m/s2, so to gain 10000 m/s of delta-V, she needs 71428571 seconds or 826 days. 826 days are more than 2 years: a much longer time than the almost 8.5 months of a Hohmann impulsive transfer.
To have a BFR-like transit time you needs at least a 3-5 MWe ion engine but I think there are no engine of this class ready to use by now.
Hi Quaoar:
I honestly don't know the answer to your question. When expanding drive gas to do work in a heat engine, there is some drop in temperature as the pressure drops. Whether that is enough to reliquify in this particular design, I don't know. If not, you either need a heat exchange radiator to cool it to reliquifaction, or else you need to compress gas instead of liquid back up to the engine-cooling pressure, which requires a lot more power and involves heavier equipment.
GW
So, to recap if you have to project an orbit-to-orbit spaceship to Mars, which has to be reused many times, which kind of rocket would you use: expander cycle, gas generator, pressure fed, pistonless pressure pump or piston pump?
After driving the piston engine, this fluid re-liquifies, and is pumped back up to pressure for recirculating into the coolant passages in the rocket engine. GW
To be re-liquified, the fluid has to be cooled, I suppose. Is the fluid cooled regeneratively in some kind of heat-exchanger by the incoming propellant?
Quaoar:
I had not seen the electric-pumped designs before, been out of the industry for a long time. Looks good except for the weight of the electrical gear. For an attitude/translation thruster system, increased weight of a small-percentage system is no big deal, and such, if present, can serve the ullage function, too. For a main engine system which is a bigger percentage of the weight, any weight addition is a bigger deal.
I think any of these ideas could easily serve multiple functions as thrusters and ullage for the main engines. The higher the thruster Isp, the smaller its tank can be for a given delta-vee budget. And then there's Kbd512's idea of using propellant boil-off gases as the cold-gas ullage thruster without adding any other thruster system. Same thing could serve both ullage and attitude/translational thruster functions.
GW
Thanks GW,
How does exactly work the piston rocket?
A working fluid in a closed cycle cools the rocket becoming hot, then it moves the pistons of the pump, it is cooled by the incoming propellant via an heat-exchanger, and it comes back again to cool the rocket?
To answer Quaoar:
Sure, any storable-propellant system that is expelled by gas pressure from a bladder inside the tank can be used as an ullage thruster. Most hydrazine monopropellant or hydrazine-NTO systems are easily capable of this function. And have been used in this way.
You can expel from the bladder at low pressures, and pump it up to high pressures for the thruster's chamber, or you can expel at the injection pressure at the cost of a heavier, high-pressure tank. The high pressure tank approach gets you the instant response you want in an attitude, etc., thruster.
If it's a one-time deal (like the Saturn stages), the solid cartridge is lighter and cheaper. But it's a one-shot device. You would have to have a set of them dedicated to each relight that you want. More than just one or two relights, and the bladder-expelled storables are the better deal.
GW
Thanks GW,
I also discovered the existence of electric-feed rockets
https://en.wikipedia.org/wiki/Electric-pump-fed_engine
https://en.wikipedia.org/wiki/Rutherfor … et_engine)
May they be useful for a Mars mission, given that a manned spaceship with life support needs battery and solar panels anyway?
I am sure that a hybrid could be used as an ullage motor. Be aware that the state of hybrid technology is not yet as advanced as solid or liquid rocket technology. But it could be, given the effort to do so. I'd recommend a storable oxidizer like a strong acid, to get hypergolic ignition of a fuel grain that is a simple particulate-loaded rubber. Such storables can be pressure-expelled from bladders in tanks, eliminating their ullage problem, something we still cannot yet do with cryogens, even "soft" cryogens like liquid oxygen.
GW
So can we also use a classic pressure-feed liquid propellant NTO-MMH with bladder-tank as an ullage motor?
Second, the longer you intend to burn, the bigger your supply of high pressure gas must be. The XCOR piston-pump design I mentioned avoids this issue by recirculating its heat engine drive fluid (which is the rocket engine cooling fluid), taking advantage of phase changes liquid vs gas. I don’t know the details, but I presume there is an electrically-driven pump that starts the recirculating process, until the heat engine can take over that duty.
Do you think it's possible to build a bigger version of the piston rocket, something like 6-8 MN of thrust and 70-100 bar of chamber pressure?
It might be possible to replace the high-pressure gas source with a much smaller high-pressure gas tank fed by a solid gas generator, or a whole series of them. These would be propellants designed to burn cleanly, and at modest chamber temperatures. Such things already exist, such as the solid starter cartridges for starting aircraft engines with no battery, since WW2. Automotive air bag cartridges are another example.
What kind compound would you use for this solid gas generator?
The tank ullage problem has been solved since at least the Saturns of the 1960’s by solid propellant ullage motors. These are small solid motors that provide just enough thrust to settle the propellant liquids into a pool, for just long enough to have that settling occur, plus get a liquid engine start. On the Saturn 5 second stage, there were 3 of these, each a small pancake motor about 6-7 inches diameter and 2-2.5 inches thick. These were made at the plant where I once worked, where Spacex tests its rockets now.
It is possible to use a miniaturized version of the Space Ship One hybrid rocket as a multiple-restart ullage rocket?
I beg to differ with Kbd512, but only a little bit.
Production pressure-fed engines have not existed in a long time, other than small thrusters. But with no moving parts at all, they can be very reliable, at the cost of extremely heavy propellant tankage.
Now-bankrupt XCOR Aerospace developed and flew in manned airplanes a unique piston-pumped rocket engine technology, including rating it for harsh cryogens to include liquid hydrogen. What they found was that this technology worked in smaller engines only, not seemingly being scalable to very large thrusts per engine. The biggest they looked at was the RL-10 of the Centaur upper stage.
They never did demonstrate anything but 10 times smaller thrust than an RL-10, though. Uniquely, in that smaller scale, they did finally demonstrate driving that piston pump assembly off the waste heat from the engine, via a third cooling fluid through some sort of heat engine. This is a technology I'd really like to see someone take on, and take to the next level. It has great promise, but no backing that I am aware of.
Thank, GW
So, if we want to build our orbit-to-orbit spaceship (something like your modular Johnson-express) in the next 10-15 years, which kind of feed-system would you use for the rocket?
P.S.
What about pistonless pump rocket?
http://www.flometrics.com/wp-content/up … PC2003.pdf
Quaoar,
The most practical orbit-to-orbit interplanetary transportation system is the SEP-enabled ITV that I've described elsewhere. This ITV concept is modular. The engineering and propulsion module (ITV-E) would have BFS (ITV-C configuration) or a rotating wheel artificial gravity habitat module (ITV-P configuration) strapped to the front of the ITV-E. By virtue of the ITV-E's megawatt class solar power array and efficient ion propulsion, transit duration to Mars from LLO is identical to any practical form of chemical propulsion.
It may even be more cheap using iodine instead of xenon. I've only some concern about spiraling inside the Van Allen Belt, but the astronauts can reach the ship with a capsule when she is out, and however she may also be very useful to send landers and other payload one-way to low Mars orbit. Please can you post some link to your spaceship, so I can make some comparison?
SpaceNut,
On that note, RL-10, J-2, and RS-25 are all real man-rated rocket engines. If I had to design a man-rated upper stage for a vehicle, then I'd pick an engine technology that's already exceptionally well proven. That means using one of those engine types or some combination of them.
RL-10 uses an expander cycle, while RS-25 uses a stage-combustion cycle. RS-25 needed a lot of maintenance and were disassembled after each Space Shuttle fly, so it might be not the best choice for an orbit-to-orbit spaceship that uses it many times during her trip. I don't know if pressure-feed rockets or expander-cycle rockets are more reliable or not.
I would like to imagine a modular, completely reusable orbit-to-orbit spaceship, also able to spin for artificial gravity.
She does all the maneuvers propulsively, so the delta-V budget is almost 5.7 km/s for the outward trip and 5.7 km/s for the inward one. Let's take 12 km/s, considering plane changes and course corrections.
The lander must be reusable, must be also used as an habitat, and can be send one-way with propellant tanks: an orbit-surface-orbit two-way trip costs almost 5 km/s of delta-V (please correct me if I'm wrong)
1) NTO-MMH
pros: self-igniting and storable
cons: expensive, highly toxic, the exhaust velocity is only about 3.3 km/s.
Mass ratio for 12 km/s of delta-V: 38
Propellant needed for a 100 ton spaceship: 3700 tons
Mass ratio for a two-way Mars lander: 4.6
Propellant needed for a 100 ton lander: 360 tons
2) LOX-RP1
pros: better exhaust velocity than NTO-MMH, cheap, non-toxic, RP1 is storable.
cons: not self-igniting, cryocooler needed for oxidizer, limited exhaust velocity of about 3.4 km/s.
Mass ratio for 12 km/s of delta-V: 34.1
Propellant needed for a 100 ton spaceship: 3310 tons
Mass ratio for a two-way Mars lander: 4.4
Propellant needed for a 100 ton lander: 340 tons
3) LOX-CH4
pros: better exhaust velocity than NTO-MMH, cheep, non toxic, possibility of production in situ.
cons: not self-igniting, cryocooler needed for both oxidizer and propellant, limited exhaust velocity of about 3.65 km/s.
Mass ratio for 12 km/s of delta-V: 26.8
Propellant needed for a 100 ton spaceship: 2580 tons
Mass ratio for a two-way Mars lander: 3.94
Propellant needed for a 100 ton lander: 294 tons
3) LOX-LH2
pros: it has the best exhaust velocity obtainable from a realistic chemical propellant (4.6 km/s), possibility of production in situ from a buried glacier or from the ice of Phobos (if there is any)
cons: not self-igniting, cryocooler needed for both oxidizer and propellant. LH2 has a lower temperature than other cyogenic propellants and is more difficult to maintain. LH2 has a very low density and needs a bigger tank that weights almost 10% of the propellant.
Mass ratio for 12 km/s of delta-V: 13.6
Propellant needed for a 100 ton spaceship: 1260 tons
Mass ratio for a two-way Mars lander: 3
Propellant needed for a 100 ton lander: 200 tons
I don't want to talk about nukes in this post, but just out of curiosity, I'd like to report the data for a solid core NTR
4) LH2 NTR
pros: exhaust velocity of about 9.2 km/s, it may be also used as a power generator in a bimodal rocket
cons: it would be surely boycotted by public opinion, cyocooler and huge tanks needed for LH2, further R&D needed to resume the studies of 60 years ago.
A nuclear lander is unlikely the best option: a lander must be compact so it might be difficult to shield the astronauts. But just out of curiosity:
Mass ratio for a two-way Mars lander: 1.73
Propellant needed for a 100 ton lander: 73 tons
Then I would like to talk about the rocket type: which kind of cycle? Gas-generator, expander-cycle, staged-combustion, or pressure-feed?
Which one may be the best suited for this kind of spaceship?
Or we could finish the Mars Gravity Biosatellite. That should fit in the trunk?
Mice would have been a good model for starting, and we would surely have gained more knowledge on the issue. But it was cancelled, like almost all that might have been of some interest for a real Mars mission. I think that NASA guys have really no intention to go everywhere but nowhere.
That centrifuge module would have come in real handy... any chance we can fly one up on a Falcon rocket?
It must be reprojected as inflatable to be stored in the trunk of a Dragon capsule.
I think it can be done in three pieces and three missions: one for the central hub with the electric engine and the docking port and the other two for the arms. Two inflatable arms of about 22 meters, rotating at 4 RMP can produce an artificial gravity of 0.4 gee, simulating Mars surface gravity.
There's no such thing as a solar array made of the lightest materials available located 1km away from a launch vehicle with a rocket that has 4 engines with RS-25 thrust levels. The rocket will literally blow the array away at that distance. Landing another BFS 1km away from the first is probably out of the question as well. People are just going to have to get used to the fact that everything is going to be a little heavier than they'd like it to be and separation distances between rockets will also be greater than they'd like. PMAD will be substantially heavier than the array, for example.
How many km is the safe BFS-landing range on Mars?
"Exercise" is a bit of a misnomer. The body needs the workload of carrying the torso. It will get that when on Mars in a weighted suit. There is no evidence that a gravity-mitochondrial system is crucial in bone/muscle loss or gain. As I said previously, if it were then we could expect people to maintain bone and muscle during bed rest in a 1G environment.
Gravity is not enough without workload and workload is not enough without gravity. Bodies need both. In fact ISS astronauts lose bone mass despite workload, less than what they would lose without it, but still they have a loss.
The study on microgravity alteration on a cellular level are just at the beginning: we know very little about this issue, it's too early to say which role it might play, and while we are posting, many fine scientist are researching to give an answer to this question. But reasoning about human physiology from a mechanical-only point of view seems like a watchmaker who pretend to apply his knowledge to a silicon CPU.
Gravitational effects at a molecular/cellular level are a red herring here I think because the body reacts to workload not gravitation. If that were not the case, then there would be no bone-muscle loss when you have constant bed rest - you're getting the full 1G gravitational load, your mitochondria are in tip top condition, but you still lose bone and muscle.
Bodies need both: gravity and exercise. It's even a fact that despite 2 h/day of intense exercise ISS astronaut lose muscular and bone mass. And that's the reason for why the aforementioned studies, performed by highly qualified scientists, that you call "red herrings", are instead a very interesting and promising line of research.