New Mars Forums

Quaoar

Member

Registered: 2013-12-13

Posts: 665

Steam nuclear rocket

As a steam punk passionate, I love the idea of a steam Spaceship, but there are also other reasons to love her: unlike hydrogen, water is very easy to store and can be found everywhere in the Solar System. A water propelled Spaceship can find the return propellant from Mercury poles to Jovian moons. Water is also a very good cosmic ray shielding material: we can imagine a multiple shells propellant tank that surrounds the habitat, solving another nasty issue. Using the same substance for propulsion and for life support is also very safe: imagine a Ship returning to Earth that fails orbital insertion burn for rockets failure: the unused water can keep astronauts alive for years, giving time to set-up a rescue mission.

A water propelled Spaceship may be strategic for future human space exploration, but surfing on Internet I found no serious work about steam NTR: the web is full of  very interesting LH2 NTR study, but completely empty about water NTR.

The only study I found is this ( http://www.permanent.com/space-transpor … ckets.html ) but it propose a ridiculous specific impulse of 190 s: it would be wiser using nuclear energy to split  water in LOX and LH2 and burn them in a good chemical rocket (adding a small percentage of RP1 or LCH4 to fix oxidizer/fuel ratio; but this is another story: I’m here to talk about steam NTR).

Even if water has an higher molecular weight than hydrogen, I found hard to believe that a NTR can do anything better than 190 miserable seconds. But there are no works on the issue. So I wanna try to imagine how a steam NTR can be and what kind of performance can reach (it’s to note that an experienced rocket man like GW Johnson talks about 600 s or more for water NTR in his blog).

I’m not an expert, only an amateur, so I beg the engineers of this forum to correct my (many) mistakes.

Let’s start from a basic design: a classical Pratt & Whitney with a cermet W/UO2 core, derived from the Rover Pewee.
Core temperature 3000 °K
chamber pressure 136 atm
Isp 940 s with LH2

First of all, we have to note that the cermet core is conceived to run with LH2, so if we want to adapt it to run with an oxidizer propellant like water, we have to protect it with some kind of  coating: I guess Thorium dioxide, because it’s the oxide with the highest melting point (3660 °K, very near to tungsten). Let’s imagine it works and go on.

Now let’s calculate the specific impulse, that is proportional to the square root of the temperature/propellant molecular weight ratio. Hydrogen has a molecular weight of 2, water of 18, so 940 s with LH2 will become almost 313 s using water: more than 190s, just enough for a MAV, but to low for an orbit to orbit Spaceship.

If we want more, we have to rise the temperature: just bring it to 3500 °K, like the Russian Superraket Block B ammonia NTR ( http://www.astronautix.com/stages/suplockb.htm ).

Working at 3500 °K, our steam NTR reach 338 s of specific impulse, quite better, but not very useful for interplanetary travels.

At this point, we are just 195 °K below tungsten melting point, very near the temperature limit of a solid core NTR. If we want to enhance Isp, we can only lower propellant molecular weight. At 3500 °K and 136 atm, water is almost integer and to dissociate needs temperature higher than reactor melting point.
So, if we think 338 s are not enough, we have to build a different kind of rocket, working at very low chamber pressure (1 atm or even below), for achieving water dissociation at sustainable temperature.

A very promising high temperature-low pressure NTR is the pressure feed MITEE monoatomic H, designed by the same guys the Timberwind. It may be a good candidate to be adapted for water.

http://web.archive.org/web/200503071638 … mitee.html

http://web.archive.org/web/200503171455 … /PUR-8.PDF

Let’s start with this MITEE:
core temperature 3000 °K
chamber pressure 1 atm
Isp 1270 s with LH2

At 3000 °K and 1 atm, water is partially dissociated in a mix of: H2O, H2, O2, HO, H, O. To
calculate the possible specific impulse of a hypothetical water version of MITEE, we need to know the fraction of every component of the cocktail. I used this curve, that show the dissociation of water at 1 atmosphere of pressure:

According to the graphic, at 3000°K and 1 atm the mixture is composed of: 40% H2O, 18% H2; 16% H, 14% HO, 6% O, 6% O2 and the mean molecular weight is 12.98. So the specific impulse will be 429 s: slightly less than a good LOX-LH2 rocket, but quite good.

Now let’s rise core temperature at 3400 °K: we have 34% H, 16% H2, 16% O, 15% H2O, 14% HO, 5% O2, with a mean molecular weight of 9.9. Now the specific impulse rises to 530 s: better than any chemical rocket, and near GW’s 600 s target.

Rising core temperature at 3500°K, we have 40% H, 20% O, 14% H2 13% HO, 9% H2O, 4% O2, with a mean molecular weight of 8.99. The specific impulse is now 565 s.

If we want to obtain more, now we can only lower the pressure: let’s put it to 0.4 atmosphere. Unfortunately, I have not a curve for 0.4 atmosphere, so I move slightly to the right side and guess that dissociation at 3500 °K and 0.4 atm would be almost like 3750°K and 1 atm. It’s the best I can do, and I beg you to correct me if you have better data.

At 3500°K and 0.4 atm, I guess: 51% H, 26% O, 10% H2, 8% HO, 3% O2, 2% H2O. With a mean molecular weight of  7.55, the specific impulse would be now 616 s (GW was right!), more than 6 km/s of exaust velocity!

With this rocket we can build a completely reusable and versatile spaceship, that can be used from Mercury to Jovian moons, without all the nasy trouble of storing cryogenics. Isn't she lovely?

Last edited by Quaoar (2014-04-24 07:43:30)

GW Johnson

Member

From: McGregor, Texas USA

Registered: 2011-12-04

Posts: 5,801

Website

Re: Steam nuclear rocket

Thanks Quaoar.  Nice to know my hunch was about right. 

I think the steam NTR would be even more attractive,  and practical,  done as an open-cycle gas core design:  somewhere between 1500 and 4000+ s Isp,  depending on achieved T/W.  Too bad nobody ever tested such a thing,  other than a couple of academic bench tests of a couple of principles,  about half a century ago. 

GW

---

GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Quaoar

Member

Registered: 2013-12-13

Posts: 665

Re: Steam nuclear rocket

Thanks Quaoar.  Nice to know my hunch was about right. 

I think the steam NTR would be even more attractive,  and practical,  done as an open-cycle gas core design:  somewhere between 1500 and 4000+ s Isp,  depending on achieved T/W.  Too bad nobody ever tested such a thing,  other than a couple of academic bench tests of a couple of principles,  about half a century ago. 

GW

With a solid core NTR steam rocket we can add an arc-jet afterburner, like the hybrid electro*thermal MITEE, that super heat propellant at 4000 °K, reaching an Isp of 800-850 s, but I guess we are reached the limit.
Gas-core is very interesting: I read that the most promising type is a toroidal counterflow vortex in a spherical chamber. With core temperature of 15-20k °K, hydrogen is almost transparent to radiation, I read, and it needs to be seaded to have a good heat transfer. So water may be probably a better propellant, reaching an Isp of 1200-1500 s.
UF2 gas is stored at low pressure in big tanks, to not be critical. Before start-up it is pressurized and injected in the vortex inside the chanber. When the burn is finished, the chamber is vented and all the trouble are finished: very safe.
Even if may be dangerous to test it on Earth, why not to study all the issues of core confinement with some heavy not radioactive gas at high temperature?

RobertDyck

Moderator

From: Winnipeg, Canada

Registered: 2002-08-20

Posts: 7,934

Website

Re: Steam nuclear rocket

Hydrogen has a molecular weight of 2, water of 18

You could be more precise. Hydrogen has atomic weight of 1.00794, and oxygen 15.9994, so H2 has molecular weight of 2.01588, and H2O is 18.01528.
http://www.webelements.com/

GW Johnson

Member

From: McGregor, Texas USA

Registered: 2011-12-04

Posts: 5,801

Website

Re: Steam nuclear rocket

Precision:  hi RobertDyck.  Awww,  don't quibble.  We're talking something still sci-fi here.  (ha ha). 

Testing:  why not test such things on the moon?  No need to build an exotic and expensive facility with plume capture capability.  That should almost pay for building the base right there.  Put the thrust stand down in a crater,  to contain the debris from a failed test (and there will be failures,  that's inevitable in rocket development testing).  No air and water to pollute,  no neighbors to bother.  It's perfect.

GW

---

GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

RobertDyck

Moderator

From: Winnipeg, Canada

Registered: 2002-08-20

Posts: 7,934

Website

Re: Steam nuclear rocket

Spray plastic sealant on the since wall of a cavern? A concrete pad beneath the test stand to deal with catastrophic engine failure? A natural cavern would have a lot of detail, stalactites and stalagmites; would a salt cave work better?

GW Johnson

Member

From: McGregor, Texas USA

Registered: 2011-12-04

Posts: 5,801

Website

Re: Steam nuclear rocket

RobertDyck: 

I'm guessing you are talking about the plume capture facility to test such things on Earth.  Your cavern idea is just part of such a thing.  Yes,  you must have a gigantic volume into which you store the entire accumulated plume from the test,  but without upsetting the backpressure on the nozzle,  I might add.  So,  you don't just fire the engine inside the space.  You must capture the plume real-time at ambient pressure,  and pump it real-time into your storage space.  That sort of thing is horribly complicated and horribly expensive.  Then you have to decontaminate all that gas of radioactivity,  before you can let it go.  That ain't cheap,  either.

So for Earthly testing,  you have a huge,  horribly expensive facility,  and every test is a bank-breaker.  Plus all the neighbors will fear you and complain,  whether justifiable or not.  Eventually they'll shut you down,  the way our politics-of-money has been working the last 40-some years. 

On the moon,  it's just an open thrust stand (steel on concrete,  with piping and cables out of the crater over to the base proper,  where all the tanks are).  Very expensive to build a simple base and even simpler facility there,  because of launch costs to the moon.  But every test is essentially dirt-cheap in comparison,  and there are no neighbors to complain.  In the long run,  almost regardless of the lifetime you expect out the base,  this is cheaper,  and politically far easier to maintain.

GW

---

GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

RobertDyck

Moderator

From: Winnipeg, Canada

Registered: 2002-08-20

Posts: 7,934

Website

Re: Steam nuclear rocket

Just seal a large cavern, set up the test stand in the cavern, let exhaust go. Keep it sealed closed. Why do anything more?

Pressure increase would be very slight due to volume of the cavern. Really fine filters. The cavern itself will settle out particulates. Let pressure push air through filters, no pumps.

GW Johnson

Member

From: McGregor, Texas USA

Registered: 2011-12-04

Posts: 5,801

Website

Re: Steam nuclear rocket

F = m*Vex + (P-Pback)*Aex  where m is the rocket massflow rate.  Pback is the pressure surrounding the exit area.  As this changes,  thrust is directly affected. 

Pback is now a rising function as you add both mass and heat to the cavern volume,  when you fire your engine inside a sealed cavern.  How many tons do you add in a 1 minute burn,  at around 2000 K for a solid core NTR of several tons thrust?  How big is your cavern?

m = (dP/dt)*V/R*T  where T is the mass fraction mixed temperature,  actually a transient here as well,  but assumed constant in the formula given.  dP/dt refers to the time dependence of Pback. 

There's a 1st order model.  V will have to be truly enormous (enormous = huge facility expense) for there to be a negligible change in Pback.  Otherwise,  you must correct test thrust over time for variable backpressure effects,  a result one further step removed from what you are trying to measure with high confidence. 

Once the backpressure rises above about 30% higher than expanded exit plane pressure,  the nozzle flow separates "for sure",  and all bets are completely off about what your thrust data mean.  That 30% figure is a only rough-and-ready approximation.  There is no real reliability to predicting nozzle flow separation. 

Compare that to simply firing out in the open vacuum every single time on the moon,  and not having to do anything to clean up the radioactive residues afterward,  since Vex exceeds lunar escape. 

GW

---

GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Quaoar

Member

Registered: 2013-12-13

Posts: 665

Re: Steam nuclear rocket

Compare that to simply firing out in the open vacuum every single time on the moon,  and not having to do anything to clean up the radioactive residues afterward,  since Vex exceeds lunar escape. 

GW

I think you are right, but unfortunatly, by the moment, we have not a lunar test facility. I guessed if it may be possible to simulate a gas core NTR on Earth, with a very safe and cheep argon vortex core, heated with some kind of microwave device (something like VASIMR ion-cyclotron resonance heater) at 10-20k °K and use it to study all the issues of core-propellant heat transfer, core confinement, chamber cooling, plume contamination and multi-propellant use.
When all the issues are solved, we can go to the Moon to perform the final tests with a real gas core prototype, saving a lot of money.

A solid core NTR can be designed for a reducing propellant like hydrogen or an oxydizing propellant like water, but not for both. A gas core has not these problems and can work with any kind of propellant. It can do the departure burn with hydrogen at very high specific impulse of 2500 s, then switch for easy storable water and use it for arrive burn at a lower specific impulse of 1000 s.

I imagine a very verastile ship with a small habitat surrounded by a multiple shells water tank and a bigger "habitank" holding the departure burn LH2, that is vented after departure and used as an habitat extention douring coasting.
After the burn, instead to be vented out, the gas core can be aspired in a bigger tank, where it adiabatically cools, and used as a gas core reactor for generating electrical power.

Last edited by Quaoar (2014-04-26 09:31:49)

RobertDyck

Moderator

From: Winnipeg, Canada

Registered: 2002-08-20

Posts: 7,934

Website

Re: Steam nuclear rocket

Would a salt cavern have to be sealed at all? Would the salt walls contain nuclear waste? Salt caverns are currently used for natural gas storage, which implies they're pressure tight. So just use a salt cavern.

Mars_B4_Moon

Member

Registered: 2006-03-23

Posts: 9,776

Re: Steam nuclear rocket

Nuclear power on the moon: NASA wraps up 1st phase of ambitious reactor project
https://www.space.com/nasa-moon-nuclear … e-complete

2027 DARPA NASA DRACO Nuclear Thermal Rocket That is Up To Three Times Better Than Chemical Rockets
https://www.nextbigfuture.com/2023/07/d … ear-t.html