Closed cycle gas core NERVA Scram jet?

[drakesdoom]

I have this crazy idea in my head. Use a Nuclear lightbulb type design engine to preheat hydrogen for injection into a scram jet nozzle for burning for extremely high thrust and isp in atmosphere with no fallout (assuming you don't crash). Another advantage would be once in vacuum you still have a fully functional NERVA engine to thrust on. Basically this would result in a Nuclear Sabre Engine.

Now tell me what I didn't think of that would make this impossible.

[Pyotor Gagarin]

Who needs closed cycle? Just look at Project Pluto

Edited May 2, 2013 by Pyotor Gagarin
Seriously though, Project Pluto is horrifying.

[Stochasty]

Now tell me what I didn't think of that would make this impossible.

Impossible in principle? Very little, I suspect. The energy density is there, it's just a technological challenge.

Impossible today? The are a large number of technical issues with NERVAs which need to be resolved before they see the light of day; foremost of these is developing a NERVA with sufficient thrust to weight ratio to get off the ground. However, even before these technological issues are dealt with, the political issue much be overcome - nukes are "scary." You'll never get a chance to improve the technology if protesters picket you every time you try to find funding.

[architeuthis]

The Reaction Engines Sabre, isn't actually a scramjet, rather it is a hydrogen fueled hybrid turbojet/rocket engine. One issue with scramjets is that they don’t start working until they are moving at over Mach 4, so you need something else to taxi and takeoff with. The funny thing about Project Pluto, if such a ghastly artifact could ever truly be said to be funny, is that it didn't need propellant. The drone was powered by expanding the atmospheric air flowing through the ram intake using the heat of the nuclear reactions. This was in a way the fulfillment of a dream dating back to the 50's; using air-breathing nuclear jets for nearly infinite range (in realty not so easy, supersonic flow ablates away the reactor elements; still Project Pluto's nightmare engine gave it a 180,000 km range, which is colossal for something flying at Mach 4 at essentially tree top altitude). With scramjets though the propellant here is your ally: you can use hydrogen's high specific heat to good effect for regenerative cooling of the engine during flight.

Impossible in principle? Very little, I suspect. The energy density is there, it's just a technological challenge.

Impossible today? The are a large number of technical issues with NERVAs which need to be resolved before they see the light of day; foremost of these is developing a NERVA with sufficient thrust to weight ratio to get off the ground. However, even before these technological issues are dealt with, the political issue much be overcome - nukes are "scary." You'll never get a chance to improve the technology if protesters picket you every time you try to find funding.

I agree that the main problem here is social and political, not technical. Project Rover did build NTRs with TWR>1, and generally the program substantially validated the concept, the performance and safety of NTR. The main technical hurdles were in moving past graphite to carbon-carbide composite reactor elements which would have allowed higher operating termperatures (and thus higher isp) as well as longer operating times and more operating cycles. Gas core is really where the demons live in terms of technical unknowns.

[drakesdoom]

I don't expect it would work as a first stage alone but my thoughts were that once it was moving through atmosphere Thrust could be increased by burning the already heated hydrogen leaving the reactor as reaction mass with incoming scooped air for extra exhaust temperature and reaction mass. Does that particular aspect sound possible?

[Kryten]

(assuming you don't crash)

Found the main problem with this.

[magnemoe]

The Reaction Engines Sabre, isn't actually a scramjet, rather it is a hydrogen fueled hybrid turbojet/rocket engine. One issue with scramjets is that they don’t start working until they are moving at over Mach 4, so you need something else to taxi and takeoff with. The funny thing about Project Pluto, if such a ghastly artifact could ever truly be said to be funny, is that it didn't need propellant. The drone was powered by expanding the atmospheric air flowing through the ram intake using the heat of the nuclear reactions. This was in a way the fulfillment of a dream dating back to the 50's; using air-breathing nuclear jets for nearly infinite range (in realty not so easy, supersonic flow ablates away the reactor elements; still Project Pluto's nightmare engine gave it a 180,000 km range, which is colossal for something flying at Mach 4 at essentially tree top altitude). With scramjets though the propellant here is your ally: you can use hydrogen's high specific heat to good effect for regenerative cooling of the engine during flight.

I agree that the main problem here is social and political, not technical. Project Rover did build NTRs with TWR>1, and generally the program substantially validated the concept, the performance and safety of NTR. The main technical hurdles were in moving past graphite to carbon-carbide composite reactor elements which would have allowed higher operating termperatures (and thus higher isp) as well as longer operating times and more operating cycles. Gas core is really where the demons live in terms of technical unknowns.

Main problem with nerva today is the lack of missions, current deep space missions involving sending probes of a couple of tons to Mars or less than a ton outside Mars. This can be done with chemical engines, using an nerva would increase cost without reducing weight much.

For manned missions past the moon or something like asteroid mining we need something like nerva.

In ksp something like pluto would be perfect for an Eve or Jool accent. It should be possible to modify an pluto so it switched to stored fuel at some height making it work like an nerva and you have an SSTO Eve plane.

[Person012345]

Found the main problem with this.

Mounting nuclear fuel atop a giant stack of explosives then setting it on fire?

What could possibly go wrong?

[magnemoe]

Mounting nuclear fuel atop a giant stack of explosives then setting it on fire?

What could possibly go wrong?

Well it worked well for the curiosity rover and all the deep space probes.

[Person012345]

Well it worked well for the curiosity rover and all the deep space probes.

I was kidding.

I don't really know how much nuclear material nervas take anyway.

[nyrath]

Main problem with nerva today is the lack of missions, current deep space missions involving sending probes of a couple of tons to Mars or less than a ton outside Mars. This can be done with chemical engines, using an nerva would increase cost without reducing weight much.

For manned missions past the moon or something like asteroid mining we need something like nerva.

The way I understand it, in this situation the (minimal) advantage of the NTR is increased thrust as well as increased Isp. The problem is that a manned mission to Mars using a Hohmann orbit takes long enough that astronaut exposure to GCR and solar protons storms becomes prohibitive.

I agree that for unmanned probes there is no need for NTR.

[drakesdoom]

TWR has been brought up as a downside, I would like to mention in reality it is not as bad as in ksp. Test fired nervas at 900isp had a 30 TWR compared to 70 for chemical rockets. Closed gas cycle is expected to hit 1500 isp so if weight and propellant flow stayed constant that would be 60 TWR right there. Now a gas cycle may weigh more or be able to pass less propellant but I think that could be offset if a second stage was used mixing and burning the propellant with air, also from what I can find of space vehicle launch speeds a ram intake may suffice as by the time they are above a ram jets operating speeds they are high enough to not need the extra TWR. Since Ram intakes involve no moving parts making the intake a disposable piece to be dropped when it's maximum effective speed was reached may further improve TWR.

After reading this huge enginelist I think it would be easier to upgrade a LANTR to use a nuclear lightbulb first stage instead of a solid core nerva.

Edited May 3, 2013 by drakesdoom
adding second paragraph

[architeuthis]

Main problem with nerva today is the lack of missions, current deep space missions involving sending probes of a couple of tons to Mars or less than a ton outside Mars. This can be done with chemical engines, using an nerva would increase cost without reducing weight much.

For manned missions past the moon or something like asteroid mining we need something like nerva.

In ksp something like pluto would be perfect for an Eve or Jool accent. It should be possible to modify an pluto so it switched to stored fuel at some height making it work like an nerva and you have an SSTO Eve plane.

Sure, NERVA is overkill for (most) robotic planetary science missions, but if we are ever serious about visiting Mars or for that matter, the moon in a fashion more permanant than the flags and footprints of the past, then NTR will be needed. Chemical rockets are simply too inefficient for moving large payloads beyond LEO. Part of the problem according to David Portree over at Beyond Apollo is that we have been conditioned to 'think small' about space, so in that sense the mission does not exist.

[Also I know people constantly debate the merits of nuclear electric vs solar electric vs NTR; my understanding 'high-thrust' electric drives (relative term, a huge 200 kW VASIMR provides 5N of thrust) will need on the order of 200 MW to be used for fast human piloted Mars missions. Unfortunately no space nuclear reactor has yet been flown with power greater than 10kW, so an improvement of 20,000X is required, alas an improvement for which development if currently unfunded. In the case of solar IRC the ISS 8 arrays provide less than 100 kW, and they are not light weight. NERVA on the other hand is an essentially proven design. The trick is remembering how to build one:)]

Also it looks like the last Waverider worked as designed! Yay for scramjets!

[bs1110101]

In ksp something like pluto would be perfect for an Eve or Jool accent. It should be possible to modify an pluto so it switched to stored fuel at some height making it work like an nerva and you have an SSTO Eve plane.

I made a (rather crappy) project pluto mod for just that

http://kerbalspaceprogram.com/project-pluto-atomic-ramjet/

[Rune]

TWR has been brought up as a downside, I would like to mention in reality it is not as bad as in ksp. Test fired nervas at 900isp had a 30 TWR compared to 70 for chemical rockets. Closed gas cycle is expected to hit 1500 isp so if weight and propellant flow stayed constant that would be 60 TWR right there. Now a gas cycle may weigh more or be able to pass less propellant but I think that could be offset if a second stage was used mixing and burning the propellant with air, also from what I can find of space vehicle launch speeds a ram intake may suffice as by the time they are above a ram jets operating speeds they are high enough to not need the extra TWR. Since Ram intakes involve no moving parts making the intake a disposable piece to be dropped when it's maximum effective speed was reached may further improve TWR.

After reading this huge enginelist I think it would be easier to upgrade a LANTR to use a nuclear lightbulb first stage instead of a solid core nerva.

Actually, it works the other way around, doubling isp means half the thrust. For the same jet power, of course. Look up the entry for fission fragment rocket in that list: 2.6 GWs provide 344 Newtons of thrust, an extreme example.

Rune. Also, making a nuclear lightbulb not melt itself is trickier than it seems.

[nyrath]

Also I know people constantly debate the merits of nuclear electric vs solar electric vs NTR; my understanding 'high-thrust' electric drives (relative term, a huge 200 kW VASIMR provides 5N of thrust) will need on the order of 200 MW to be used for fast human piloted Mars missions. Unfortunately no space nuclear reactor has yet been flown with power greater than 10kW, so an improvement of 20,000X is required, alas an improvement for which development if currently unfunded. In the case of solar IRC the ISS 8 arrays provide less than 100 kW, and they are not light weight

Not to mention the fact that solar electric arrays suffer from that pesky inverse square law. Here in our solar system, a given size of solar electric array that produces 100% power in Earth orbit will only produce 42% power at Mars orbit. And the power drops off rapidly the further away from the sun you travel.

EnergyPercent = 1 / Dist^2

where

Dist = distance the spacecraft is from Kerbal, measured in KAU, or units where 1 = distance between Kerbal and Kerbin

EnergyPercent = percentage of power the array will produce as compared to what it produces at 1 KAU (1.0 = 100%). Of course the array will produce more than 100% of the power if it is closer to Kerbal than 1 KAU.