NASA again looking at Nuclear Rockets

[linuxgurugamer]

The Nerva that we all love (and the Nuclear Lightbulb from Porkjet's old mod Atomic Age) are both based on real research done in the late 50's, 60's and 70's.  The goal was to replace the Saturn 1b upper stage.  It was tabled both because of the change in direction of NASA, overall public fear of nuclear energy and international treaties regarding nuclear energy in space.

Well, they are looking at it again.  See the following, this is a current article:

https://www.nasa.gov/centers/marshall/news/news/releases/2017/nasa-contracts-with-bwxt-nuclear-energy-to-advance-nuclear-thermal-propulsion-technology.html

 

[Benjamin Kerman]

This is a big step int he right direction! I'm glad they are looking at this for Mars missions, but they should really get there first before trying out new tech...

[Bill Phil]

Nerva would have replaced the S-IVB on the Saturn V.

It was cancelled because they stopped building Saturns in 68.

[Dman979]

Moved to Science And Spaceflight, as this is about real science.

[WinkAllKerb'']

energy can be used in many ways ... in/out choose yours 4 ... & keep in mind whine afterward are not allowed

Edited August 29, 2017 by WinkAllKerb''

[Racescort666]

NTRs have some serious advantages. GRC took a serious look at them not too long ago: https://ntrs.nasa.gov/search.jsp?R=20150023036

I've been thinking about (procrastinating) running the numbers to see what a hypothetical NTR powered Ice Giant mission would need. Between GRC and JPL, most of the pertinent information is there. I was thinking something of an A to B comparison of a DCSS or Centaur to a hypothetical ammonia fueled NTR transit/injection stage. 

[WinkAllKerb'']

hum amusing ^^ make me ask are we we really talking about a choice or something that poped out of nowwhere at somepoint and we not really satisfied with ; ) amusing is it ?

ùeanwhile i m gonna agree there a difference assuming it 100% or not and less  ^^ so true ^^

Edited August 30, 2017 by WinkAllKerb''

[Bottle Rocketeer 500]

Yay! We are getting ever closer to a Mars mission! @NASA, a correct step towards Mars.

[regex]

19 hours ago, Racescort666 said:

NTRs have some serious advantages. GRC took a serious look at them not too long ago: https://ntrs.nasa.gov/search.jsp?R=20150023036

I've been thinking about (procrastinating) running the numbers to see what a hypothetical NTR powered Ice Giant mission would need. Between GRC and JPL, most of the pertinent information is there. I was thinking something of an A to B comparison of a DCSS or Centaur to a hypothetical ammonia fueled NTR transit/injection stage. 

Your comment caught my interest so I ran some back-of-the-napkin numbers assuming Centaur hardware mass in three configurations: single RL-10, single Pewee using NH3, and single Pewee using H2. In the Pewee case I subbed the mass of an RL-10 (277kg) for the mass of a Pewee (from Atomic Rockets, 3240kg), assuming no needed changes in thrust structure (would likely need some), plumbing (would definitely need some), and support hardware (very likely as well) keeping the original Centaur total mass.

Item	Mass RL 10	Mass ANTR	Mass HNTR
Hardware	1970	1970	1970
Engine	277	3240	3240
Fuel	20830	17867	17867
Total	23077	23077	23077
Final	2247	5210	5210
exhaust m/s	4565	5101	9200
½-tons	9.8 km/s	7.2 km/s	13.0 km/s
1-tons	9.1 km/s	6.9 km/s	12.5 km/s
2-tons	8.1 km/s	6.4 km/s	11.5 km/s
4-tons	6.7 km/s	5.5 km/s	9.9 km/s

Ammonia loses out in pretty much all cases except for storability if you plan to take the Centaur stage past Earth. A lighter NTR would definitely help but the RL-10 and Pewee have roughly similar thrust. Looking at a subway map of the solar system you're needing somewhere around 8.5km/s to sling something out to Uranus or Neptune which ammonia can't provide without other means (gravity assist). OTOH, you could easily send 6 tons to Uranus using H2, with some of that being a hypergolic braking stage, or probably even more with gravity assists.

E: And looking over these numbers I should also mention that NH3 would produce a much higher thrust than H2, so if we're going for comparable thrust the Pewee is probably too massive for our purposes. However, even reducing the A(mmonia)NTR engine mass to like, say, 1/4 of the value there would place it somewhere in the neighborhood of the RL-10 in terms of performance (as an estimate slightly better, half ton gets 10.1km/s and four ton gets 7.1km/s). At that point cost concerns mean the RL-10 wins out every time if all we want is a pure transfer stage (existing hardware, no nuclear component, etc...)

Edited August 30, 2017 by regex

[Racescort666]

4 hours ago, regex said:

Your comment caught my interest so I ran some back-of-the-napkin numbers assuming Centaur hardware mass in three configurations: single RL-10, single Pewee using NH3, and single Pewee using H2. In the Pewee case I subbed the mass of an RL-10 (277kg) for the mass of a Pewee (from Atomic Rockets, 3240kg), assuming no needed changes in thrust structure (would likely need some), plumbing (would definitely need some), and support hardware (very likely as well) keeping the original Centaur total mass.

Item	Mass RL 10	Mass ANTR	Mass HNTR
Hardware	1970	1970	1970
Engine	277	3240	3240
Fuel	20830	17867	17867
Total	23077	23077	23077
Final	2247	5210	5210
exhaust m/s	4565	5101	9200
½-tons	9.8 km/s	7.2 km/s	13.0 km/s
1-tons	9.1 km/s	6.9 km/s	12.5 km/s
2-tons	8.1 km/s	6.4 km/s	11.5 km/s
4-tons	6.7 km/s	5.5 km/s	9.9 km/s

Ammonia loses out in pretty much all cases except for storability if you plan to take the Centaur stage past Earth. A lighter NTR would definitely help but the RL-10 and Pewee have roughly similar thrust. Looking at a subway map of the solar system you're needing somewhere around 8.5km/s to sling something out to Uranus or Neptune which ammonia can't provide without other means (gravity assist). OTOH, you could easily send 6 tons to Uranus using H2, with some of that being a hypergolic braking stage, or probably even more with gravity assists.

E: And looking over these numbers I should also mention that NH3 would produce a much higher thrust than H2, so if we're going for comparable thrust the Pewee is probably too massive for our purposes. However, even reducing the A(mmonia)NTR engine mass to like, say, 1/4 of the value there would place it somewhere in the neighborhood of the RL-10 in terms of performance (as an estimate slightly better, half ton gets 10.1km/s and four ton gets 7.1km/s). At that point cost concerns mean the RL-10 wins out every time if all we want is a pure transfer stage (existing hardware, no nuclear component, etc...)

Hey look almost all the work already done for me. 

Without having an enormous rocket, there is a practical limit to how much dV something with an ISP of 325 can get you. Unfortunately, as pointed out, NTRs are considerably heavier than their chemical counterparts. In fact, they pretty much always will be because there is a limit to how small they can be made. So even if TWR had no bearing on the decision to use one, the smallest an NTR can be made and still function is 1770 kg per the study I linked.

So, where is the advantage? Storable propellants. While an RL-10 makes a far better upper stage engine than an NTR does, it's usefulness ends during a long duration mission. That is, >5 years even though ULA has proposed using the RL-10 in missions lasting as long as a week. Currently, it's impractical to store LH2 longer than the current time frames. Ammonia, on the other hand, would be relatively easy to store. Yes, you take a performance hit due to the weight of the engine, but having a long duration storage with a high ISP engine could hypothetically make up for it. 

Fuel storage seems to be the problem with an Ice Giant mission. After all, the JPL proposal was going to use a hypergolic bipropellant braking stage and their dV budget was in the neighborhood of 2.5-3 km/s. Several of the proposals were to use a solar electric propulsion stage during the inner solar system gravity assists but looking at the report, it seems to have just complicated the entire mission. They briefly mention radioisotope electric propulsion and it seems like an attractive option as it considerably cuts down on flight time.

Anyway, my goal wasn't to compare directly to DCSS or Centaur, I was just considering structural requirements and tankage weights since they see similar tankage pressures to what you would need in order to keep ammonia liquid at a reasonable temperature. Also, they are in line with what can be reasonably fit within existing payload fairings. But now, on to my numbers:

• Orbiter Mass: 2500 kg
• NTR Mass: 1770 kg
• Tankage mass: 1170 kg (this is taking into account the higher density of NH3 vs LH2 thus the lower dry mass than Centaur)
• Fuel mass: 48790 kg (again, NH3 is around 9 times as dense as LH2 so you don't need nearly the tank volume)
• Launch vehicle payload (reduced to 85%): 54230 kg (this is Falcon Heavy payload and it has been reduced for margin)
• Exhaust velocity: 5101 m/s 
• Delta V once in LEO: 11.7 km/s

It's also worth noting that this would likely fit within the Falcon Heavy payload fairing. The Critically Limited NTR proposed by GRC has a length of 4.9 m and I calculate my tankage to be around 4.7 m in length (4.5 m in dia) leaving around 1.8 m for the probe. Although it would probably have to be mounted upside down.

Things to criticize:

• I averaged tankage densities of Centaur (minus engine) and the space shuttle ET and came up with 15 kg/m3 for an empty tank. Maybe this wasn't entirely realistic, I probably should have gone by surface area but this was the easiest way I felt to do it.
• Falcon Heavy has yet to fly and there has been some discussion as to whether it can actually lift the advertised 63,800 kg into LEO. I will note that the JPL proposal also considers SLS to be a viable launch vehicle for an Ice Giant mission. Regardless, I needed someplace to start for an LEO mass. Falcon Heavy sounded like a great option.
• Actual ISP/Exhaust velocity for an ammonia NTR.  I think it will probably be good, better than any hypergolic available.
• The ability for an NTR to have multiple starts. This is really unknown. The GRC report considered the critically limited engine as a single use demonstrator so who knows what can be built as a multi start capable engine. 
• C3 dV of the launch vehice. I'm going to be honest, this is way beyond my back of napkin calculations. It seems that due to the engine mass minimum of NTRs, it makes some amount of sense to use 1 engine to leave earth orbit and arrive at an Ice Giant. 
• Earth orbit fueling. (not something I was originally proposing, but...) This would be an easy one to criticize but hypothetically you could fuel an ammonia NTR in orbit with 2 or more launches. Putting the dry payload up would be well within most rockets' capabilities and something like Falcon 9 could send up fuel trucks to top off the probe before it leaves. You could even leave the upper stage attached as a kick stage.

All of this was done while trying to think of ways to reduce transit time to Uranus and Neptune. Per the JPL report, basically anything done to try and decrease transit time increases the injection dV so they put a limit on how much injection dV they would allow for mission proposals and the entire report is considering that. I thought, why not keep the probes the same but use a different propulsion method to cut transit time? My first thought was using ClF3 as an oxidizer but you know... running shoes. Then I realized that NTRs can use basically anything for a reaction mass and ammonia seemed like the best candidate.

This is probably way more involved on something entirely hypothetical but I'm glad I did it.

[Rune]

On 29/8/2017 at 9:17 PM, Benjamin Kerman said:

This is a big step int he right direction! I'm glad they are looking at this for Mars missions, but they should really get there first before trying out new tech...

This is a tiny step, at most. And we might differ on it being the right direction slightly, nukes have a lot of hidden quirks that make them troublesome...

Anyhow, 18.8 million dollars is about 1/1000 of the entire NASA budget. That is what they are spending on this. In the same budget, they spend upwards of five times that on education, and that is after massive cuts.

 

Rune. So maybe the new budget is a 5x bigger step into becoming an education agency?

[OrbitalBuzzsaw]

About time.

[NSEP]

Yay?!

So, is NASA actually wanting to use NERVA's on actual spacecraft in the actuall future? Is it just a 'hmmm cool lets revisit this idea and forget it later' kind of thing? or am i dreaming!? Am i too unnecessarily confused? Am i bad at understanding English scientific articles?

Edited August 31, 2017 by NSEP

[magnemoe]

3 hours ago, Rune said:

This is a tiny step, at most. And we might differ on it being the right direction slightly, nukes have a lot of hidden quirks that make them troublesome...

Anyhow, 18.8 million dollars is about 1/1000 of the entire NASA budget. That is what they are spending on this. In the same budget, they spend upwards of five times that on education, and that is after massive cuts.

 

Rune. So maybe the new budget is a 5x bigger step into becoming an education agency?

Yes its an case study, one part was how to run an NTR in an closed loop, you add oxygen to burn the hydrogen into water, add more water to cool it and an heat excanger, then filter it for radioactivity who would be radioactive particles from the reactor so simple to clean. 
Only issue would be to keep low pressure at the nozzle. 
Both the ntr and the test setup would be more expensive than 19M. 

[PB666]

https://www.bloomberg.com/news/articles/2018-02-15/nasa-is-bringing-back-cold-war-era-atomic-rockets-to-get-to-mars

looks like a partial repetition of an old story, but looks like exactly same title so does belong in this necrotic thread.

[DerekL1963]

15 hours ago, PB666 said:

https://www.bloomberg.com/news/articles/2018-02-15/nasa-is-bringing-back-cold-war-era-atomic-rockets-to-get-to-mars

looks like a partial repetition of an old story, but looks like exactly same title so does belong in this necrotic thread.

Either way, this is just a study contract.  NASA generates study contracts the way you or I generate dead skin cells - constantly and without conscious action.  They don't represent policy, they don't represent plans, they don't represent anything NASA "wants" to do.  One in a million goes beyond filing the completed study in the same warehouse the government keeps the Ark of the Covenant.

[DDE]

5 hours ago, DerekL1963 said:

Either way, this is just a study contract.  NASA generates study contracts the way you or I generate dead skin cells - constantly and without conscious action.  They don't represent policy, they don't represent plans, they don't represent anything NASA "wants" to do.  One in a million goes beyond filing the completed study in the same warehouse the government keeps the Ark of the Covenant.

R2-D2, I need to record a message...

[Starman4308]

On 8/30/2017 at 2:47 PM, regex said:

Your comment caught my interest so I ran some back-of-the-napkin numbers assuming Centaur hardware mass in three configurations: single RL-10, single Pewee using NH3, and single Pewee using H2. In the Pewee case I subbed the mass of an RL-10 (277kg) for the mass of a Pewee (from Atomic Rockets, 3240kg), assuming no needed changes in thrust structure (would likely need some), plumbing (would definitely need some), and support hardware (very likely as well) keeping the original Centaur total mass.

Item	Mass RL 10	Mass ANTR	Mass HNTR
Hardware	1970	1970	1970
Engine	277	3240	3240
Fuel	20830	17867	17867
Total	23077	23077	23077
Final	2247	5210	5210
exhaust m/s	4565	5101	9200
½-tons	9.8 km/s	7.2 km/s	13.0 km/s
1-tons	9.1 km/s	6.9 km/s	12.5 km/s
2-tons	8.1 km/s	6.4 km/s	11.5 km/s
4-tons	6.7 km/s	5.5 km/s	9.9 km/s

Ammonia loses out in pretty much all cases except for storability if you plan to take the Centaur stage past Earth. A lighter NTR would definitely help but the RL-10 and Pewee have roughly similar thrust. Looking at a subway map of the solar system you're needing somewhere around 8.5km/s to sling something out to Uranus or Neptune which ammonia can't provide without other means (gravity assist). OTOH, you could easily send 6 tons to Uranus using H2, with some of that being a hypergolic braking stage, or probably even more with gravity assists.

E: And looking over these numbers I should also mention that NH3 would produce a much higher thrust than H2, so if we're going for comparable thrust the Pewee is probably too massive for our purposes. However, even reducing the A(mmonia)NTR engine mass to like, say, 1/4 of the value there would place it somewhere in the neighborhood of the RL-10 in terms of performance (as an estimate slightly better, half ton gets 10.1km/s and four ton gets 7.1km/s). At that point cost concerns mean the RL-10 wins out every time if all we want is a pure transfer stage (existing hardware, no nuclear component, etc...)

Quick question: is that total stage delta-V, or delta-V after reaching LEO parking orbit?

One of the major factors I can think of, though, is the question of how necessary these high delta-V values are. We've gotten quite good at creating some excellent multi-slingshot trajectories.

Similar back-of-the-napkin (well, more like back-of-the-Excel-spreadsheet) math puts a conventional Atlas 551 with short payload fairing as being capable of putting 1.7 tons onto a Jovian slingshot, and if we assume an EEJ slingshot (5.4 km/sec ejection burn from LEO parking orbit), that goes up to about 2.8 tons.

My best guess is that I've lowballed it, since Juno was 3.6 tons at launch, from an Atlas 551.

Probes are getting smaller and lighter recently; a recent flagship-class proposal for the ice giants proposed a 150 kg orbiter plus 50 kg probe as its most ambitious proposal. I'll be conservative and put an orbiter dry mass at 500 kg. Even at that level, with a 310 s-1 hypergolic engine and 2.8 tons ejected on an EEJ trajectory, that still gives you 5.2 km/sec to play with once you hit Jupiter, or presumably whatever you're using Jupiter to slingshot to.

For unmanned missions, then, I would agree with NASA's trend of keeping it conservative, of keeping development costs low by relying on chemical engines supplemented by slingshots.

 

Where these very high-velocity Earth ejections really come in handy is for manned missions, where you can't spend years mucking around getting your slingshots, because your astronauts are not exactly getting any younger. For that, I'd probably assume a custom-built vehicle assembled over multiple HLV/SHLV launches, because I don't think you're sending astronauts to Mars on top of a single Atlas 551, even if you do replace the RL-10s with nuclear thermal rocket engines.

Edited February 17, 2018 by Starman4308

[PB666]

25 minutes ago, Starman4308 said:

Where these very high-velocity Earth ejections really come in handy is for manned missions, where you can't spend years mucking around getting your slingshots, because your astronauts are not exactly getting any younger. For that, I'd probably assume a custom-built vehicle assembled over multiple HLV/SHLV launches, because I don't think you're sending astronauts to Mars on top of a single Atlas 551, even if you do replace the RL-10s with nuclear thermal rocket engines.

UNless ISP is over 1000, the 2.5t engine versus a 277 kg engine, you are going to lose alot of dV burning out the bottom of the fuel tank. Nuclear rockets make alot of sense pushing huge payloads with fuel tanks that drop off. Pushing drop satellites into satellite atmospheres, thats  more of an issue of its easier to land something light and fluffy versus a gold brick. The bigger problem is that a nuclear engine is not socially acceptable inside of Earth's SOI, but if you are not burning from the orbital minimum, you are just throwing away dV.

[KerikBalm]

On 8/30/2017 at 9:47 PM, regex said:

Ammonia loses out in pretty much all cases except for storability if you plan to take the Centaur stage past Earth. A lighter NTR would definitely help but the RL-10 and Pewee have roughly similar thrust. Looking at a subway map of the solar system you're needing somewhere around 8.5km/s to sling something out to Uranus or Neptune which ammonia can't provide without other means (gravity assist). OTOH, you could easily send 6 tons to Uranus using H2, with some of that being a hypergolic braking stage, or probably even more with gravity assists.

It would be great if the same NTR rocket could manage to use two propellents, LH2 for ejection, NH3 for the capture burn-return. Empty fuel tanks would be staged off, but you'd still have an over powered engine for the return/capture.

Methane should get ~3% higher Isp. Its got a low boiling point, but its not nearly as bad as H2.

Boiling/Melting (1 Atmosphere):

H2: 20.3 / 14 K

CH4: 111.7 K / 90.7 K

NH3: 239.8 K / 195.4

With a mass difference of 16 vs 17, its not going to be much better. Atomic rockets says something about problems with carbon deposits in the engine.

I think it would be easier to split the H from C than to split the H from N. So if you've got a nuclear lightbulb that gets hot enough to cause decomposition of the methane but not Nitrogen, then your Isp is going to be significantly better... I'm not sure that temperature is reachable though.

[DDE]

18 minutes ago, KerikBalm said:

It would be great if the same NTR rocket could manage to use two propellents, LH2 for ejection, NH3 for the capture burn-return. Empty fuel tanks would be staged off, but you'd still have an over powered engine for the return/capture.

Not sure if at all doable. It's about neutron physics, not chemistry; the neutron moderation properties of the rapidly flowing propellant require extreme care in operating even a single-propellant design.