New Mars Forums

Official discussion forum of The Mars Society and MarsNews.com

You are not logged in.

Announcement

Announcement: As a reader of NewMars forum, we have opportunities for you to assist with technical discussions in several initiatives underway. NewMars needs volunteers with appropriate education, skills, talent, motivation and generosity of spirit as a highly valued member. Write to newmarsmember * gmail.com to tell us about your ability's to help contribute to NewMars and become a registered member.

#1 2018-04-29 09:34:47

RGClark
Member
From: Philadelphia, PA
Registered: 2006-07-05
Posts: 696
Website

Towards highly reusable rocket engines.

I thought I discussed this before on this forum but I couldn't find it when I did a site search so here it is again:

I'm investigating methods to get high reusability of rocket engines like for aircraft engines. I'm operating under the theory jet engines are able to operate for thousands of hours because they have to run under much reduced temperature requirements than for rocket engines. A rocket engine may run at 3,000 C and above, while jet engines may run at only ca. 1,200 C.

There a couple of ways of getting the reduced temperatures. One way is to use liquid air, including the nitrogen, for the oxidizer. The included nitrogen which does not take part in the combustion would serve to cool the combustion. A problem here though is a rocket engine sim only gave the vacuum Isp in this case as only ca. 220 s for kerosene-air.

Another possibility is to use highly fuel-rich mixture ratio. I'm aware that typically rocket engines run at a mixture ratio below stoich anyway, but I'm talking even well below this. For instance while a kerolox engine may typically run at a mixture ratio of ca. 2.3 to 1, I'm considering running at a mixture ratio below even 1.

GW, would you consider tests at such such low mixture ratios?

The performance, i.e., Isp would certainly be reduced but I'm thinking you could still get an orbital launch vehicle by radically reducing stage dry weight. For instance, reducing the combustion temperature, the pressure would also be reduced, say by a factor of three requiring reduced chamber wall thickness.

More importantly you could use lower weight materials for the engine construction. Instead of using a heavy thick wall combustion chamber, and with regenerative cooling channels cut thoughout, just use a thin layer of high temperature metal that faces the combustion flame. Then surround this with highly effective but low weight insulation, like the shuttle tile aerogels. Then finally use carbon fiber for strength for the outer layer of the engine.

With these changes we might be able to get the thrust/weight ratio from the typical 100 to 1 for kerolox engines to perhaps 1,000 to 1(!)

  That's for the engines. I'm still working on some possibilities to increase the tank propellant weight to empty weight ratio from the typical 100 to 1 for kerolox to ca. 1,000 to 1.

  Bob Clark


Old Space rule of acquisition (with a nod to Star Trek - the Next Generation):

      “Anything worth doing is worth doing for a billion dollars.”

Offline

#2 2018-04-29 11:20:16

Terraformer
Member
From: Ceres
Registered: 2007-08-27
Posts: 3,800
Website

Re: Towards highly reusable rocket engines.

Hmmm. That sounds like some sort of chemical-thermal engine, where you're using chemical fuels to heat a hydrogen propellent? Wouldn't that reduce the thrust, though?


"I'm gonna die surrounded by the biggest idiots in the galaxy." - If this forum was a Mars Colony

Online

#3 2018-10-14 09:34:01

RGClark
Member
From: Philadelphia, PA
Registered: 2006-07-05
Posts: 696
Website

Re: Towards highly reusable rocket engines.

To GW Johnson, I seem to remember on your blog of your writing of amateurs you knew doing experiments with their home built liquid rocket engines. I'm interested in doing a test of this theory of mine. Do you know of people I could contact to collaborate with?

Anyone else on the forum who knows of amateurs doing ground tests with liquid fueled rockets I'd like to see links to their pages on the net.

  Bob Clark


Old Space rule of acquisition (with a nod to Star Trek - the Next Generation):

      “Anything worth doing is worth doing for a billion dollars.”

Offline

#4 2018-10-14 13:56:29

JoshNH4H
Member
From: Pullman, WA
Registered: 2007-07-15
Posts: 2,538
Website

Re: Towards highly reusable rocket engines.

This is a worthwhile goal, but the way you've suggested going about it seems to be in most respects the polar opposite of how I would try to do it.  It seems to me that the sacrifice you're trying to make (much lower engine mass in exchange for much lower Isp) won't really pay off.

To get rough numbers, I used the equation here.  It's not perfect, but does give a good approximation of the exhaust velocity for most fuels.  According to this page, kerolox at a mixture ratio of 2.3 has a flame temperature around 3550 K, a mean exhaust molecular weight of 21.7, and a gamma (Ratio of specific heats) of 1.22.

Assuming that the nozzle is perfectly expanded to vacuum, this gives a vacuum exhaust velocity of 3,883 m/s.  Real high performance kerosene engines can get as high as 3600, and 3500 is reasonable for a well designed engine.  I will therefore introduce a "fudge factor" of 0.82 within the square root to account for the finite efficiency of the engine.

The stoichiometric mixture ratio for kerolox is 3.4.  At 2.3 my expectation would be that the exhaust would be a mixture of CO2, CO, and H2O.  It happens to be the case that at a ratio of 2.3 the exhaust will be entirely CO with no CO2 admixed at all.  At mixture ratios below 2.3 the exhaust will contain either Carbon or Hydrogen.  In order to move on, I wrote down the following chemical equation:

CH2 + nO2 → aCO2 + bH2O + cCO + dC + eH2

The equation is valid for values of n between roughly 0.1 (hot enough to dissociate the kerosene into its constituents) and 1.5 (stoichiometric combustion, above which excess molecular oxygen will exist in the exhaust).  Here are the results for three selected values of n:

d14yz2F.png

R, in this case, is the mixture ratio in terms of mass that we use when discussing rocket fuels.

Below n=1.0, combustion will not reach a level near what could be considered "complete" and there will either be uncombusted carbon or hydrogen in the exhaust.  In terms of enthalpy, it is favored to produce water and carbon; in terms of entropy it is favored to produce hydrogen and carbon monoxide.  I'm not sure where the equilibrium lies and it's entirely believable that rocket exhaust is not at equilibrium.  I will therefore consider three possible cases: In the first case, H2O predominates with a small amount of CO and H2.  In the second case, H2O and CO are produced in roughly equal proportion.  In the third case, CO predominates with a small amount of H2O and C.  The numbers I will use are as follows:

zeWnFBW.png

To calculate the combustion temperature, I will assume that the unreacted elements absorb heat in accordance with their heat capacities and proportion.  I will deal with solids by ignoring them (except for the heat they absorb) and reducing exhaust velocity in proportion to their mass.  In other words, I will assume that if the exhaust is 75% gases moving at 4000 m/s and 25% Carbon the effective exhaust velocity will be 3000 m/s.  I will assume the ratio of specific heats remains constant.

Using the equation above with the modifications I have described, here are my results:

dyN4efh.png

Naturally these results are approximate, but they suggest that the actual exhaust velocity is not strongly sensitive to the location of the chemical equilibrium, and furthermore that the vacuum exhaust velocity should probably be in the range of 2500 m/s.  It's also worth noting that at low mixture ratios there's potential to produce really substantial amounts of Carbon in the exhaust.  It' hard for me to believe that an engine will continue to work well for long periods of time when 20% of the exhaust produced is a solid.

You also said this:

RGClark wrote:

For instance, reducing the combustion temperature, the pressure would also be reduced, say by a factor of three requiring reduced chamber wall thickness.

This is not correct.  Reducing the combustion temperature does not reduce the chamber pressure.  The combustion temperature is a function of what is being combusted.  The chamber pressure is a function of the inlet pressure generated by the propellant feed system.  It is possible to lower the chamber pressure in conjunction with other changes, but you will suffer further reduced Isp for doing so if you are operating in the atmosphere.

You certainly could see some mass savings from reduced temperature as you can have a smaller cooling system, perhaps even a passive one, but I don't think they're very substantial.  I don't see why your tankage or structural mass should change at all, for example.

Here's some numbers: Let's say you have a two-stage rocket where each stage has a delta-V of 4700 m/s.  With ideal kerolox at 3500 m/s Vex your dry mass fraction will be 26%.  With the mixture ratio you have suggested (Vex=2500 m/s) the dry mass fraction is just 15%. 

I would choose to go at it the opposite way by allowing for a moderately higher engine mass while keeping performance close to constant.  The biggest change the pumps from turbopumps to something more traditional such as a reciprocating pump.  I suppose you could also build a combustion-based pump where the detonation of a small amount of fuel serves to pressurize a much larger amount of fuel and push it into the combustion chamber.  Turbopumps have a higher efficiency and lower mass but they are also more complicated and last less time (alternatively, you could build a turbopump and run it to failure and see what breaks, make that part stronger and do it again)


-Josh

Offline

#5 2018-10-15 00:32:24

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,362

Re: Towards highly reusable rocket engines.

RGClark,

Most people seem to have a difficult time wrapping their minds around the mechanical forces at work in a rocket engine.  Simply put, jet engines last longer than rocket engines for two principle reasons, the jet engine is a heavier pump design and it typically spins slower.  The heat of combustion has a substantial effect on the nozzle in both applications, but not much else.

Each RS-25 weighs about the same as the pair of F100's in our F-15, yet produces more than 10 times as much thrust.  The RS-25's high-pressure LH2 pump spins at 37,000rpm and the blades are not much bigger or thicker than a quarter.  The F119's that enable our F-22 to accelerate in a vertical climb, much like a rocket, have a HPT shaft speed of around 15,000rpm.  For each pound of weight associated with the HPFTP's total mass, it's making 100hp.  In practical terms, each quarter-sized blade inside that pump housing is making more than 750hp.

Try to let that sink in for a moment.  What object the size of a quarter do you know of that would withstand 750hp applied to it for any substantial length of time?  The 750hp PT6A turboprops I've seen weigh about twice what I do and are rated to produce that power continuously for around 6000 hours.  A variety of components on the engines don't last quite that long, but are you starting to understand what kind of forces are at work in a rocket engine?

The Continental O-200 piston aircraft engine I know and love weighs a little over 200 pounds and makes 100hp reliably for anywhere between 2000 and 3000 hours before wear causes my favorite little four-banger to literally tear itself apart.  There are 2 pounds of mass allocated for every mechanical horsepower that O-200 makes.  The RS25 has a fuel pump bolted to it with a power-to-weight ratio more than 200 times greater.  Consequently, the blades in the RS-25's LH2 pump are trash after a few minutes of operation at full rated output.  The fact that those little things even last for a few minutes is an engineering miracle of epic proportions.

The air moving through a jet engine also removes heat from the engine and the hot gases are expelled through the exhaust nozzle.  The components inside the hot section don't melt because the flame never touches them in normal operation and the boundary layer air is sufficient to prevent most of the heat of combustion from reaching them.  If the flame from the combustor actually touches components in the hot section, then they melt like butter in seconds.  There's no air in space to cool a rocket nozzle, so the oxidizer or fuel has to cool the exhaust nozzle.  To prevent severe oxidation associated with running an oxidizer through a very hot rocket nozzle, the fuel is normally used for that purpose.

When you demand performance levels from components that are up, up, and away, you get component lifetimes measured in minutes.  If you have material that can withstand those environments for a considerable length of time, then you have something I call Unobtanium.  Anyhow, that little marvel of engineering we call the RS-25 is literally guzzling a fuel cold enough to freeze Oxygen solid to produce exhaust hot enough to melt Tungsten.  It's amazing that it works at all and it's positively mind-blowing that we've never grenaded one in flight.

Incidentally, they've been trying to do what you suggested doing to jet engines for decades.  It's been extremely challenging, to put it mildly.  Some of those annoying physics and materials science problems have apparently gotten in the way.  We have no choice, General Clarkrissian.  Our little turbo-blades can't maintain horsepower of that magnitude!

Offline

#6 2018-10-15 11:33:36

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

What you will find trying to go to very off-design mixture ratio is drastically-reduced performance,  if you can get it to burn at all!  Most devices that mix and burn things have mixture limits outside of which you could not light the reaction with an atom bomb.  Ugly little fact of life,  but there it is.

Ignoring dumped drive gas massflows,  liquid rocket Isp = CF c*/gc,  where gc is the gravity constant that makes inconsistent units work out.  Also true for solids and hybrids.  Thrust coefficient CF depends principally upon expansion pressure ratio and physical expansion area ratio.  Low chamber total pressure is low CF,  plain and simple.  It’s quite the significant effect,  too.

Characteristic velocity c* is a book-keeping device containing gas properties,  but principally the total temperature in the chamber.  Low temperature is low c*,  and it,  too,  is quite the significant effect. 

It also governs the nozzle size for a given chamber pressure from the massflow balance.  Thrust is F = CF Pc At.  Nozzle throat flow rate w = Pc At gc/c*.  And so Isp = F/w = CF c*/gc,  as long as nozzle flow rate is the ONLY flow rate drawn from storage. 

And there is thermochemistry:  low pressure is lower calculated theoretical temperature.  That’s a smaller effect,  but still fairly significant.  Efficiencies of mixing and combustion also decrease significantly at lower pressure,  in actual test devices,  and this effect is larger than the thermochemical effect,  but entirely empirical.  (In other words,  your traditional constants are really variables,  too!)

Practical devices nearly always optimize just slightly (not greatly) fuel-rich of the stoichiometric point that is the theoretical optimum point.  This is as true of airbreathing devices as rockets.  If you try to operate far from that optimum,  your performance quickly approaches zero,  if you can light the thing at all!!  THAT is what usually fails first:  stability of flame.  Another entirely empirical thing.

The upshot is,  you want the maximum pressures and temperatures you can possibly stand,  to maximize performance.  The vulnerable hardware is not the chamber and nozzle,  it is the propellant pump(s) assembly,  which is almost invariably a turbine-driven impeller-pump,  driven by (really) hot gas tapped off from somewhere in the combustion chamber,  typically staged these days,  so as not to dump all the drive gas anymore. 

Any sort of gas turbine is always vulnerable,  and relatively short-lived,  because it is so highly-stressed.  Hot materials are very weak.  The impeller is not quite so vulnerable,  but subject to cryogenic stresses due in large part to cold embrittlement,  plus hydrogen embrittlement effects for that fuel. 

The point is,  hot-gas-driven turbopumps for cryogens will always be relatively short-life items.  They WILL NEVER approach 1000+ hour service lifetimes between overhaul.

I think one would be better off pursuing a different pumping means,  since that is the vulnerable part of the whole engine.  Historically,  that alternative has been the pressure-fed approach,  which is inherently very heavy.  But those two are not the only ways to get from low pressure propellant storage to high-pressure injection into the engine. 

You need a different drive mechanism than a hot gas turbine.  And an impeller is not the only way to pump a liquid to high pressure at high flow rates.  Positive displacement (pistons and rotors) are inherently less fragile than impeller blading.

This is something that now-bankrupt/dead-and-gone XCOR Aerospace was pursuing.  They went bankrupt because they got stiffed contractually on their alternate engine for the Centaur stage,  not because of piston-pumping LH2 and LOX into a demonstrator for that Centaur engine. 

They flew two rocket-powered airplanes very successfully with piston-pumped storable propellants for several years.  We’re talking rocket systems providing thousands of hours between overhaul,  and outstanding safety characteristics.  I don’t know how those piston pumps were driven.

For the Centaur engine contracts,  they fired 1/10 thrust scale demonstrators (2500 to 5000 lb thrust),  piston-pumped with the pump drive being a piston heat engine.  That heat engine ran off a third-fluid rocket cooling system operating in a closed loop.  I don’t know details beyond those.

I saw their heat engine/piston pump hardware:  it was similar to a case-and-jug type of multi-cylinder lawnmower engine.  It fed the demonstrator rocket at about 1/10 its delivery capacity.  So,  this kind of technology,  just as it was,  might handle 25,000-to-50,000 lb thrust rocket engines (Centaur engines 25,000 lb with factor-2 growth).  No one ever knew whether it could be scaled up further. 

XCOR was claiming up to a 4000-hour life between overhauls for this technology.  If that’s not a breakthrough in rocket technology,  then I don’t know what a breakthrough really is! 

XCOR is now gone.  Jeff Greason is off doing something else.  I don’t know what happened to the rest of XCOR’s talent.  But I think there is great promise if somebody would pick up this piston-pump technology and run with it.

I see a lot more promise in that,   than in monkeying around with mixture ratios. 

GW


GW Johnson
McGregor,  Texas

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

Offline

#7 2018-10-16 14:04:47

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,362

Re: Towards highly reusable rocket engines.

GW,

Is there any reason why the Rampressor from RamGen can't be considered as a more durable blisk design?

The blades, which cost beaucoup bucks since they're so difficult to fabricate, can be entirely 3D printed to form a single and substantially more durable part that can handle supersonic fluid velocities without significant vibration and wear.  Instead of fighting against the effects of compressibility, it uses compressibility to achieve higher single-stage compression ratios.

Could the blisk be incorporated into the ring of an annular aerospace to eliminate some of the plumbing?

Offline

#8 2018-10-17 19:30:36

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

Kbd512:

Can't even guess an answer.  "blisk" is not a word I have ever heard of.

GW


GW Johnson
McGregor,  Texas

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

Offline

#9 2018-10-17 20:08:38

Void
Member
Registered: 2011-12-29
Posts: 6,976

Re: Towards highly reusable rocket engines.

I am stupid about any of this stuff.  However I had to query for "Blisk".  Was surprised to see it.

Not looking for trouble, have nothing more to add.

https://en.wikipedia.org/wiki/Blisk

Hope you don't mind.


Done.

Offline

#10 2018-10-17 20:19:18

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 28,750

Re: Towards highly reusable rocket engines.

I have seen the unique fan blade design but never knew what it was called...

Offline

#11 2018-10-18 07:22:30

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,362

Re: Towards highly reusable rocket engines.

GW,

A "blisk" is a one-piece bladed disk and a turbine compressor stage, for all intents and purposes.  They can be, and are, machined using multi-axis CNC machines.  However, that process is very time consuming and intricate, thus expensive.  The RamGen design is much simpler from a machining perspective.  It requires a supercomputer to determine how flow is affected across all Mach numbers, so the design is substantially more complicated than a subsonic compressor.  However, in fabrication and operation, the "Rampressor" version of the "blisk" is simpler.

Offline

#12 2018-10-18 10:36:19

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

I don’t know how any of these ideas being discussed apply to a rocket engine pump assembly.  The rampressor might apply to an air breather.  I do appreciate finding out what a “blisk” was,  thanks!  Jargon seems to be everywhere. 

The advantage of an impeller pump is a fairly smooth outlet pressure vs time trace.  The disadvantages are structural fragility of the blading,  and encountering blade stall if you get too far outside its optimum operating conditions (sometime quite narrow!!).  Outlet flow rate responds drastically to demanded outlet pressure,  easily driving the pump to blade stall.  It’s a dynamic machine,  just like any other turbine-like device.

A piston pump delivers definite pulses,  thus requiring some surge chamber volume at the outlet to damp the pulses into a smoother outlet pressure trace.  As long as foreign objects don’t get into the pumped fluid,  the pistons and cylinder heads are not hardly structurally vulnerable at all.  Your maintenance items are valves and bearings,  and these can last a very long time,  as long as you stay away from fatigue-vulnerable reed valves. 

Gear and gerotor pumps are not reciprocating devices,  and have little need for a surge damper volume on the outlet.  These are simple rotation devices,  but they are positive displacement machines,  as is the Archimedean screw.  For these pumps,  keeping foreign objects out of the fluid is even more important,  because of the tight clearances.  Auto engine oil pumps are often gear or gerotor devices.  This is a very well-established mature technology.

The positive displacement devices do not respond to demanded outlet pressure,  it works the other way around:  they determine it.  It’s delivered massflow against the outlet restriction that determines delivered pressure.  There is no upper limit,  and there is no narrow band of stable operation the way there is with dynamic machines.  You simply must have a pressure relief valve diverting back to the inlet side,  in order to control outlet pressure. 

If you make that relief valve a controllable item,  then controlling to the desired injection pressure schedule across a fixed-geometry injector plate is the way to throttle your rocket engine.  Simple.  And adaptable a very wide range of operating conditions,  unlike with a dynamic machine as your pump.

All you need is a rotating source of torque to drive a gear or gerotor pump.  Depending upon the time required,  and whether restarts are required,  this could be any of several power sources.  One-shot devices might be high-pressure bottled gas feeding a pneumatic motor of some kind,  through a pressure-control regulator.   Reusable/multi-start devices might be as simple as a rechargeable battery feeding an electric motor,  with some solar cells to recharge the battery between burns.  XCOR used a heat engine operating off the third-fluid rocket engine cooling system. 

When this kind of simplicity and durability is available,  I have to wonder why turbopumps are still in vogue,  after over half a century.  We could have had reusable rocket engines and stages long ago. 

Which is exactly why I suggested positive-displacement pumping as the better path forward to a long-life rocket engine.

GW

Last edited by GW Johnson (2018-10-18 10:36:46)


GW Johnson
McGregor,  Texas

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

Offline

#13 2018-10-18 13:59:16

JoshNH4H
Member
From: Pullman, WA
Registered: 2007-07-15
Posts: 2,538
Website

Re: Towards highly reusable rocket engines.

GW-

Do you have an estimate of how switching from a turbopump to a positive displacement pump would affect the mass of the engine?


-Josh

Offline

#14 2018-10-19 09:55:22

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

Hi Josh:

Nope,  no real weight estimate for a positive displacement pumping system.  A hunch says it's heavier than turbopump machinery,  by a factor near 3-ish (or more), but it lasts in service a hell of a lot longer. 

The pumps are a smaller piece of the engine weight,  in turn a smaller piece of the stage weight.  So it shouldn't be a show-stopper for anyone.  But it is a price to pay,  for long life. 

You can't have everything,  usually.  But the conceptual thinking change appears to be a bigger stumbling block than any weight penalty.  At least,  that's the way it looks to me. 

GW


GW Johnson
McGregor,  Texas

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

Offline

#15 2018-10-21 03:50:20

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,362

Re: Towards highly reusable rocket engines.

GW,

The RamPressor blisks are not fragile.  There are no small blades for cavitation to damage.  It's a solid chunk of metal shaped like a gerotor, except more circular than a prototypical generated-rotor.  There's a ramp with special geometry that uses shock waves generated in the compressed fluid to assist with rotation of the blisk in the fluid being compressed.  The rotational speed is not limited by the speed of sound in the fluid being compressed, either.  It's designed for supersonic operation.

Mach 1 in various fluids:
LO2 - 1,056m/s
LH2 - 1,246m/s
RP1 - 1,324m/s
LCH4 - 1,420m/s

The tip speed of the turbines in the RS-25's HPFTP is just over 500m/s at 37,000rpm and outlet velocity ranges from roughly 600/ms to 800m/s.  The blisk diameter is 25.88cm, or at least it was back in the early 1990's.  Now the geometry and setup is a big secret.  Presumably, we could spin the crap out of a significantly smaller RamPressor blisk and generate the same outlet pressure with a single stage thanks to the low dynamic viscosity of LH2 and achieve the same or higher outlet pressure using a smaller, lighter, cheaper component that's easier to replace.

The longest serving HPFTP lasted 22 missions before replacement of the blisks was required.  Much of the plumbing and other moving components were rated for at least a couple hundred operational cycles and more than a thousand non-operational cycles.  The new design life of the Block II SSME is 55 operational cycles.  In short, the life of the engine components is not a significant factor.

My personal opinion about this is that we should take our most thoroughly flight proven and tested engine, the RS-25, and continue to engineer the crap out of it until we have something that lasts for 100 cycles or more.  It drastically reduces the propellant mass required to go from the surface of Mars to orbit and only requires water electrolysis to work.

Success Legacy of the Shuttle Program - Space Shuttle Main Engine - Relentless Pursuit of Improvement - Katherine Van Hooser - SSME Chief Engineer - NASA Marshall Space Flight Center

These are the things that cost real money:

* tossing any orbital launch vehicle in the ocean after a single flight
* developing an entirely new launch vehicle from scratch
* developing an entirely new LOX/LCH4 engine for both atmospheric and vacuum flight regimes
* developing a fundamentally new portable LOX/LCH4 propellant plant for use on Mars, since a portable LOX/LH2 plant already exists
* developing an extremely efficient dust separator for a vacuum pump to obtain the LOX from CO2
* obtaining 1,100t of propellant from Earth or Mars to make up for the substantially lower Isp of LOX/LCH4

Any existing engine capable of 50+ flights prior to refurbishment is NOT a major cost factor.  Even if any given rocket engine never achieves commercial jet engine service life levels, so what?  It's not like the engine is operated for hours at a time.  Burn times are measured in minutes before the fuel is consumed.  If we can keep any rocket engine alive for 50+ flights without major overhaul, we're good to go.

Offline

#16 2018-10-21 09:27:42

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

Kbd512:

I thought the rampressor was for compressing gases.  The compressible fluid mechanics that makes it work doesn't apply to liquid media,  because the ideal gas equation of state P = rho R T simply does not apply to fluids of such high bulk modulus.  How in the world would a rampressor compress a liquid?  I don't understand.

Lessee,  I'm not sure I remember the names of the two piston-pumped rocket airplanes that XCOR flew.  Those used storable liquids for propellants.  Alcohol and some oxidizer,  I believe that wasn't LOX. 

There was really nothing about the chamber,  injector plate or nozzle that could wear out,  and they had containment in case they cracked.   The pump assembly (and I don't know how it was powered),  seemed to have a life-between-overhauls exceeding 1000 hours. 

What that really means is thousands and thousands of burns,  each a tiny handful of minutes long.  Pretty much about like conventional aircraft engines of most any type,  except for the short flights.

I kinda liked the sound of that long service life.  But it's not a 1:1 substitution for a turbopump assembly on an existing engine.  Doesn't use chamber bleed gas as the drive.  That hot gas bleed is what makes pumps short life.  That's what you want to avoid.

GW

Last edited by GW Johnson (2018-10-21 09:29:52)


GW Johnson
McGregor,  Texas

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

Offline

#17 2018-10-21 12:01:50

Belter
Member
Registered: 2018-09-13
Posts: 184

Re: Towards highly reusable rocket engines.

Aliens do it with their superior materials technology.

Offline

#18 2018-10-21 12:53:49

kbd512
Administrator
Registered: 2015-01-02
Posts: 7,362

Re: Towards highly reusable rocket engines.

GW,

My understanding, perhaps incorrect, is that LH2 is compressible, which would explain why accurately modeling cavitation took so long.  LH2 was treated as an incompressible fluid, like water, but that didn't reflect reality.  I could be completely wrong or maybe I misunderstood what I was reading.  I've seen the bulk modulus of LH2 cited as 200MPa at 4K.

Offline

#19 2018-10-22 09:33:23

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

I dunno much about the properties of LH2.  But if the bulk modulus is low enough,  then it starts behaving compressibly.  Hydrogen always has been the weird one. 

The compressibility of water and things like it is near 300,000 psi ~ 2000 MPa,  where a factor of 2 lower is considered to be getting slightly compressible (most brake fluids are limited by such incipient compressibility). Not enough to be gas-like,  but enough to cause brake fade problems without friction surfaces getting hot.

If your 200 MPa figure is correct,  there's quite a bit of effective compressibility to LH2. 

I rather doubt that LOX or LCH4 would behave with any significant compressibility.  But I really don't know anything about their compressibility properties. 

GW

Last edited by GW Johnson (2018-10-22 09:39:01)


GW Johnson
McGregor,  Texas

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

Offline

#20 2018-10-23 05:20:21

elderflower
Member
Registered: 2016-06-19
Posts: 1,262

Re: Towards highly reusable rocket engines.

Liquids can have their pressures increased in a diffuser, as described by Bernoulli's equation.

Offline

#21 2018-10-23 22:55:46

RGClark
Member
From: Philadelphia, PA
Registered: 2006-07-05
Posts: 696
Website

Re: Towards highly reusable rocket engines.

JoshNH4H wrote:

This is a worthwhile goal, but the way you've suggested going about it seems to be in most respects the polar opposite of how I would try to do it.  It seems to me that the sacrifice you're trying to make (much lower engine mass in exchange for much lower Isp) won't really pay off.

To get rough numbers, I used the equation here.  It's not perfect, but does give a good approximation of the exhaust velocity for most fuels.  According to this page, kerolox at a mixture ratio of 2.3 has a flame temperature around 3550 K, a mean exhaust molecular weight of 21.7, and a gamma (Ratio of specific heats) of 1.22.

Assuming that the nozzle is perfectly expanded to vacuum, this gives a vacuum exhaust velocity of 3,883 m/s.  Real high performance kerosene engines can get as high as 3600, and 3500 is reasonable for a well designed engine.  I will therefore introduce a "fudge factor" of 0.82 within the square root to account for the finite efficiency of the engine.
...
Naturally these results are approximate, but they suggest that the actual exhaust velocity is not strongly sensitive to the location of the chemical equilibrium, and furthermore that the vacuum exhaust velocity should probably be in the range of 2500 m/s.
...

I used the Rocket Performance Analysis program to estimate performance. Here are the specs for the engine:

RPA-40bar-1-0-MR.png

It uses a 40 bar combustion chamber pressure at a mixture ratio of 1. Note though I'm assuming a very high expansion ratio of 750 to 1 to get the vacuum Isp in the range of 300 s. The highest expansion ratio of any engine currently is 250 to 1 by the RL-10B2. So I'm assuming very lightweight materials for the nozzle or using an aerospike to get such a high vacuum Isp.

Here is listed the thermodynamics data:

RPA-40bar-1-0-MR-thermodynamics.png

The combustion temperature is 1,528.6° K or 1,254.6° C.

My statement about the lowered temperature corresponding to a lower chamber pressure assuming the same bipropellant combination is just from using the Ideal Gas Law.

  Bob Clark

Last edited by RGClark (2018-10-23 23:02:18)


Old Space rule of acquisition (with a nod to Star Trek - the Next Generation):

      “Anything worth doing is worth doing for a billion dollars.”

Offline

#22 2018-10-23 23:50:10

RGClark
Member
From: Philadelphia, PA
Registered: 2006-07-05
Posts: 696
Website

Re: Towards highly reusable rocket engines.

GW Johnson wrote:

Kbd512:
I thought the rampressor was for compressing gases.  The compressible fluid mechanics that makes it work doesn't apply to liquid media,  because the ideal gas equation of state P = rho R T simply does not apply to fluids of such high bulk modulus.  How in the world would a rampressor compress a liquid?  I don't understand.
Lessee,  I'm not sure I remember the names of the two piston-pumped rocket airplanes that XCOR flew.  Those used storable liquids for propellants.  Alcohol and some oxidizer,  I believe that wasn't LOX. 
There was really nothing about the chamber,  injector plate or nozzle that could wear out,  and they had containment in case they cracked.   The pump assembly (and I don't know how it was powered),  seemed to have a life-between-overhauls exceeding 1000 hours. 
What that really means is thousands and thousands of burns,  each a tiny handful of minutes long.  Pretty much about like conventional aircraft engines of most any type,  except for the short flights.
I kinda liked the sound of that long service life.  But it's not a 1:1 substitution for a turbopump assembly on an existing engine.  Doesn't use chamber bleed gas as the drive.  That hot gas bleed is what makes pumps short life.  That's what you want to avoid.
GW

I believe those thousands of burns for an XCOR engine were for the pressure-fed engines used on the rocketplanes of the Rocket Racing League:

https://en.wikipedia.org/wiki/XCOR_EZ-Rocket

These were low performance engines as suggested by the fact the combustion chamber pressure was only 350 psi (24 bar). I'd bet dollars to donuts the combustion chamber temperature also was low.

Note that this idea can actually be more easily tested using pressure-fed engines, by either amateurs or professional companies. This is because the pump-fed engines usually have their mixture ratios fixed by fixed gear ratios.

   Bob Clark

Last edited by RGClark (2018-10-23 23:54:28)


Old Space rule of acquisition (with a nod to Star Trek - the Next Generation):

      “Anything worth doing is worth doing for a billion dollars.”

Offline

#23 2018-10-24 09:59:29

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 5,423
Website

Re: Towards highly reusable rocket engines.

I'd have to do some looking around,  but I thought one of those two airplanes had a piston-pumped rocket.  XCOR isn't there anymore to ask.  That kind of stuff used to be on their website.  I visited them once as a possible consultant for an orbital follow-on to Lynx,  and got to see a lot of their stuff.  If memory serves,  I was told about a piston-pumped rocket on one of those planes during that visit.  It would be easy enough to drive the piston pump with a high-pressure gas bottle and a pneumatic motor,  or by a battery and an electric motor. 

My guess is that you could check out your off-mixture idea easiest with a simple pressure-fed design down on the ground.  But,  I'd bet real money you will have to stage the combustion chamber:  burn to reaction completion at near-proper ratio,  then in a second stage,  add the excess propellant as simple dilution mixing.  Otherwise,  you're going to run into unignitable mixture limitations.  If you don't choke the connection between the two chambers,  they will "talk" acoustically. There's all sorts of instability potential there. 

GW


GW Johnson
McGregor,  Texas

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

Offline

#24 2018-10-25 04:39:10

elderflower
Member
Registered: 2016-06-19
Posts: 1,262

Re: Towards highly reusable rocket engines.

If you use an excess of carbon containing propellants, you will likely make a highly luminous, smoky flame. You may also coke up your combustion chamber. I would look at excess oxidiser to avoid these issues.

Offline

#25 2018-10-25 05:36:58

louis
Member
From: UK
Registered: 2008-03-24
Posts: 7,208

Re: Towards highly reusable rocket engines.

https://www.youtube.com/watch?v=R7pj_W5vMBE

Interesting Q&A at the end of this video - presentation by Hans Koenigsman of Space X.


Let's Go to Mars...Google on: Fast Track to Mars blogspot.com

Offline

Board footer

Powered by FluxBB