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.

#26 2017-03-07 19:49:49

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

Re: Kbd512's human mission design for Mars

Ya once it all sinks in it will...

Offline

#27 2017-03-07 20:23:08

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

Re: Kbd512's human mission design for Mars

GW,

Can you explain how pressurization in pressure fed systems relates to engine cycle losses or other issues affecting accurate determination of appropriate expansion ratios and Isp?

Offline

#28 2017-03-07 21:11:36

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

Re: Kbd512's human mission design for Mars

The chamber pressure I have referred to is that in the chamber that actually feeds the nozzle.  That's the downstream-most in a staged combustion scheme.  The L* is that chamber volume divided by throat area.  That parameter includes both chamber length and nozzle contraction ratio toward the throat. 

High c* efficiency correlates with long chamber length (L/D > 1+) and larger contraction area ratios (Ach/At > 4),  both of which raise the L* parameter.  Once you get to an acceptable L*,  c*-efficiency should be above 90%.  In a good design with well-designed injection spray patterns and with long chamber and large contraction ratio,  c* efficiency is above 98%. 

Isp needs to figure on delivered thrust divided by the actual propellant flow rate.  That propellant flow rate is the ideal value in the final chamber feeding the nozzle,  plus whatever is bled off upstream to feed the turbopumps. 

For a pressure-fed design,  there is only one chamber and it has a chamber pressure.  All the flow rate is ideal,  so wdot = Pc At g / c* is all the massflow that there is from the tankage,  while thrust F = CF Pc At.  In that case (and only that case)  Isp = CF c* / g.

Even for pressure-fed designs,  you still have to worry about selecting chamber length and nozzle contraction,  and getting good spray patterns,  to achieve high c* efficiencies.  I would use L/D around 1.5 and Ach/At exceeding 4 to get efficiencies at 99+%,  given good spray patterns.  Most violate this,  and so most have c* efficiency losses.  It's a tradeoff of achievable performance versus the weight of the components you design. 

In pressure-fed designs,  the old 1950's - 1960's rule of thumb was you need a feed pressure / chamber pressure ratio near 2 to achieve combustion stability.  They tell me that ratio is closer to 1.2 today.  There's a similar ratio across whatever kinds of valves control the propellant feed rate for a given pressure ratio.  My best guess is a tank pressure around 1.4-1.5 times whatever chamber pressure you end up designing to.  That's just a rough figure. 

There's just no way to make pressure-fed propellant tankage really lightweight.  For a 2000 psig rocket chamber,  you need something like a 3000 psig tank.  The modified Barlow's formula (using vessel ID) relates hoop stress to vessel pressure,  which in turn sets a lower bound on your tank weight:  stress x 2 x twall = Pressure x ID. 

You simply cannot use material ultimate stress capability in a practical design.  Even a 1-shot design should stay under material yield stress.  The hotter you let the materials get,  the lower those allowable-stress numbers get.   Reused designs need to set an allowable stress about 10 times less than the S-N curve levels indicate (and that's before you knock down for temperature),  for the appropriate load case in Mil Handbook 5,  and an N defined by the expected life.  If the structure carries other loads as well (most do),  then those may (or may not) actually set your design.  Every situation is different. 

It does get quite complicated.  Sorry,  that's just life.

Titanium won't save you.  It's hard to fabricate with,  and it's no better hot than plain carbon steel (pretty much crap above 750 F).  Its only advantage is that it's just about half the weight of steel.  Austenitic and martensitic stainless steels will go a lot hotter,  most near 1200 F.  Don't even think about aluminum.  It's crap at 300 F. 

There are a couple of austenitic stainless alloys usable even hotter (near 1700 F),  but they are quite a bit weaker there than at 1200 F,  where they are just about the same strength as the others.  The only real difference is freedom-from-scaling to higher temperatures,  not strength.  316 and 310 are the ones.  Softened-butter weak,  but free of scale. 

Austenitics (300-series) must be used at low annealed strength.  They do work-harden,  biut modest heating anneals them again.  Martensitics (D6ac and the 4000-series,  and T-250 and similar) can be heat-treated to high strength,  but there are severe exposure limits above which they revert to low annealed strength (usually under 1200 F).  Scaling and corrosion get very bad beyond 1200 F with nearly all of these. 

There are some super alloys (that really aren't steels),  that will go hotter still,  but they are super-expensive,  and very difficult to fabricate.  You find them in combustor cans,  turbine blades,  and afterburner components. 

GW

Last edited by GW Johnson (2017-03-07 21:35:46)


GW Johnson
McGregor,  Texas

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

Offline

#29 2017-03-07 21:30:48

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

Re: Kbd512's human mission design for Mars

https://en.wikipedia.org/wiki/Pressure-fed_engine
250px-Pressure_fed_rocket_cycle.png

It's easy enough to have sufficient pressure when the tank is full. But you need to pump in enough gas to keep pressure up until the tank is empty. Say you'll use the tank until 5% of the fuel is left. In that case, your pressurant gas has expanded by a factor of 20, and has dropped pressure by a factor of 20. If the pressure at that point is 1 bar, you need to start at 20 bar. That means you need a very heavy tank: 20 times heavier than a tank that's good for 1 bar. Its also why its heated to make it expand once out of the tank that its stored in.

http://space.stackexchange.com/question … tabilities

To be theoretical the fuel residence time in a combustion chamber is given by the Characteristic length(usually denoted by L*)(minimum length that the fuel will remain in the combustion chamber and nozzle for complete combustion to take place)

L∗=q∗V∗ts/A

q is the propellant mass flow rate, V is the average specific volume, and ts
is the propellant stay-time A is the sonic throat area

Offline

#30 2017-03-07 21:44:52

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

Re: Kbd512's human mission design for Mars

Spacenut:

What governs propellant flow rate from the tank in a pressure-fed system is pressure drop across the tank outlet fitting.  The outlet side of that is set by engine chamber pressure. That means you maintain a high CONSTANT pressure in the propellant tank for a constant flow rate,  constant thrust burn. 

To do that means you have a high-pressure gas reservoir fed to the propellant tank through a pressure regulator.  For a 2000 psig engine,  the tank might be 3000 psig.  The regulator must keep that 3000 psig on the tank as the gas reservoir pressure drops.  It takes a positive pressure drop across that regulator to flow gas through it at all. 

Just hazarding realistic guesses,  say your reservoir operates between 8000 and 4000 psig.  The reservoir volume must be such that gas mass expelled between those pressures equals the mass that fills the propellant tank as it empties at a constant 3000 psig.  It's not a trivial gas bottle volume in relation to the propellant tank volume.  You minimize gas bottle volume by going to very high initial gas pressures indeed,  which does make that bottle heavier. 

L* = chamber volume/nozzle throat area = 100+ inches for most propellants.  Determined initially by von Braun at Penemunde for LOX-ethanol in his A-4/V-2 work.  Varies by propellant combination,  but not all that much. 

GW

Last edited by GW Johnson (2017-03-07 21:48:59)


GW Johnson
McGregor,  Texas

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

Offline

#31 2017-03-08 11:25:54

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

Re: Kbd512's human mission design for Mars

SpaceNut,

Obviously this is very high level, but the following document explains a bit about NTO/MMH tank fabrication and testing:

Design and Manufacture of a Propellant Tank Assembly

This explains the method ESA developed to decontaminate Rosetta's MMH tank assemblies without damaging them:

DECONTAMINATION OF MMH- AND NTO/MON -PROPELLANT TANKS

Offline

#32 2017-03-08 17:39:35

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

Re: Kbd512's human mission design for Mars

GW thanks for why pressure needs to be much higher than what I had put in my post. I actually have worked on a breathing system that used composite tanks pressurized to 5,000 psi and was regulated down by manifold orifices and regulator diaphram unit.

Kbd512 thanks for the links

Offline

#33 2017-03-09 11:02:26

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

Re: Kbd512's human mission design for Mars

The lower tank pressure goes with a lower engine chamber pressure.  For deep space thruster propulsion,  this is OK,  because you cannot expand to zero backpressure anyway.  The only downside is that c* is a little lower at lower pressure.  For most propellant combinations,  it's a power function of chamber pressure,  with an exponent on the order of 0.01. 

My ancient Pratt&Whitney handbook lists c* for NTO-hydrazine as 5860 ft/sec at 1000 psia versus 5740 ft/sec at 100 psia.  Small systems are almost always pressure-fed.  So if you figure a vacuum thrust coefficient for whatever expansion you can achieve,  then Isp = CF c* / gc.  Pick gc to make units consistent.  If they already are,  then gc = 1. 

Same ancient handbook shows a sea level Isp = 292 sec at 1000 psia chamber,  and a vacuum Isp = 342 sec at 100 psia.  I don't think those are properly corrected for nozzle KE efficiency,  though.  The vac numbers are for expansion area ratio 40.  The sea level numbers are for expanded pressure equal to sea level backpressure (perfect expansion). 

Nozzle KE efficiency is typically very close to 0.983,  but it only applies to the mdot V term,  not the P A terms.  If you have an off-angle nozzle orientation,  you must use its cosine to knock down thrust and Isp.  That applies to all the terms. 

GW

Last edited by GW Johnson (2017-03-09 11:04:36)


GW Johnson
McGregor,  Texas

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

Offline

#34 2023-08-17 10:11:06

Mars_B4_Moon
Member
Registered: 2006-03-23
Posts: 8,893

Re: Kbd512's human mission design for Mars

Some news on possible NASA NEP, new designs for Reactors in space or Space-Tugs

Space mining company developing nuclear reactor and more for moon projects
https://www.space.com/space-mining-comp … ew-feustel

Atomic rockets are back
https://www.aerosociety.com/news/atomic … -are-back/

NASA’s Building a Nuclear Rocket That Would Get Us to Mars in Just 6 Weeks
https://singularityhub.com/2023/08/07/n … t-6-weeks/

Offline

Board footer

Powered by FluxBB