Reusable LOX/Kerosene SSTO with drop tanks

GW Johnson · 2017-02-18 15:13:57

Straight conical nozzles are every bit as efficient as curved bells, thrustwise. Their only downside is they are geometrically a little bit longer. But they are far, far easier to fabricate.

Nozzle kinetic energy efficiency correlates very, very well as efficiency = 0.5(1 + cosine(half angle)). For a conical nozzle this is usually very near 15 degrees, for KE efficiency = 0.983.

For a curved bell, there is a bigger half angle just downstream of the throat, and a smaller one at the exit plane lip. You take the arithmetic average of those two, and use it in the very same efficiency equation. That average is just about 15 degrees for almost all practical curved bells, so the nozzle kinetic energy efficiency is still just about 0.983.

When using that efficiency, it applies only to the mV terms, not at all to the PA terms.

Look at NASA SP 8076 on how to do ballistics if you don't believe me. That was the ballistics monograph written by W. Ted Brooks. Ted was a good friend and colleague long ago, and my mentor in solid ballistics and the ballistics of nozzles.

GW

Last edited by GW Johnson (2017-02-18 15:23:53)

RGClark · 2017-02-19 09:21:37

Quaoar wrote:

GW Johnson wrote:
Hi Bob, long time no talk:
Glad you like the idea. If you have a deep flame pit, you can use an even simpler and lighter fixed spike. Elderflower is right: just go with attitude thrusters. KISS is beautiful, ain't it?
GW
Hi GW.
Japanese project Kankou-maru used a fixed spike, like yours:
"Thrust for takeoff is supplied by 12 Mitsubishi LE-9 engines, burning liquid oxygen and liquid hydrogen. 4 of the engines are LE-9B-3 "booster" engines, optimized for low altitude operation. The other 8 engines are LE-9S-3 "Sustainer" engines, optimized for vacuum operation. The vehicle afterbody is designed to use the vehicle exhaust and the atmosphere as an "aerospike" nozzle to increase efficiency at all altitudes.

http://www.spacefuture.com/archive/kank … nual.shtml

This might be similar what SpaceX is doing with their upper stage on the Interplanetary Transport System (ITS):

It uses 3 sea level engines and 6 vacuum engines. In his now famous video introduction of the ITS, Elon discussed that the upper stage could also launch from ground. It is known that high velocity fluid flow acts like a low pressure region, by the Bernoulli principle. Then the exhaust from the three sea level engines could provide a low pressure region for the vacuum engines.

EDIT: I just realized looking at the image that the actual arrangement of the engines is opposite to what I was thinking. The smaller sea level engines are on the interior of the engine set, not the exterior.

Bob Clark

Last edited by RGClark (2017-02-19 09:52:03)

SpaceNut · 2017-02-19 10:29:31

I see that you are looking at the veturi or vortex effect but the arrangement as far as I can see will not create this as sea level air is always in contact to the vacuum compensated engines on the outside.

https://www.festo.com/net/SupportPortal … ciples.pdf

Now reverse the count and location and it would work to create the vortex for the vacuum engines being in the center....

GW Johnson · 2017-02-19 12:05:08

Out in the open, the jet entrainment effect makes very small changes in local pressure. That's why aerodynamic base drag is no larger than it is, all across the spectrum of Mach numbers.

Enclosed inside an appropriately-shaped tube, the jet entrainment effect can cause very large changes in pressure, which is how an air or steam ejector pump works. In making composite propellants, we had to mix under vacuum to avoid air bubbles in the propellant (such a foam is useless), and we used ejectors to suck the mix bowl down to vacuum. Reasonably hard vacuum in under a minute.

GW

RGClark · 2017-03-07 08:54:45

Thanks for that info, GW. Is it possible to create an arrangement to make full use of this Venturi effect? Use full vacuum engines for the majority of the thrust on the booster. Surround the base of the nozzles with a tube as you suggested. But have a separate engine or engines whose exhaust is at sea level pressure and whose exhaust is directed down the sides of the interior of the tube so the inside vacuum engines effectively see a near vacuum.

I think this is doable, but a question is whether you would need the Venturi-creating engines to vary in exhaust pressure as the booster gained altitude and the ambient pressure decreased. But if they could do that you might as well give all the engines on the booster that capability.

But it may not be necessary. For instance sea level engines can work effectively from sea level to vacuum, though not as efficiently.

Bob Clark

GW Johnson · 2017-03-07 17:16:28

Hi Bob:

I really don't think you'll get any significant "venturi effect" around rocket nozzles at the base of a stage. The way a jet pump works is entirely a different picture. The picture in post 78 above is not what works. What works is a convergent section leading into a straight-tube mixer, followed by a subsonic diffusion bell. The primary stream causing the ejection is supersonic, with its exit plane about a mixer tube diameter upstream of the straight mixer section.

The supersonic stream from the primary nozzle expands further until it hits the walls of the straight mixer tube section. That expansion is a sharp pressure dop, which is exactly what induces secondary (ejected) massflow to flow into the convergent section toward the straight mixer tube section. There is turbulent shear mixing between the two streams, but it is very, very inefficient, which is precisely why jet pumps are energetically very inefficient devices.

When the primary-mixed-with-secondary smacks the wall of the straight mixer section, it is very supersonic, and oblique-shocks into a stream paralleling the mixer walls, which is a big piece of the pressure rise. It is also a big piece of the stream deceleration. There is a whole series of oblique shocks in a train that reduce stream speed and raise its pressure a little, followed by Rayleigh flow-type friction choking down to just subsonic speed. The mixer tube must be at least as long, or longer, than this shock-plus-friction deceleration regime requires for practical design.

From there, the other big piece of the pressure rise is subsonic diffusion in the exit bell. Exit velocities are actually very subsonic in most good designs, around 0.2-0.3 Mach. All of these things that increase pressure rise act to reduce velocity (and stream thrust).

Since you are trying to increase thrust of a rocket engine, I think the ejector effect is not what you are looking for, really. It works in fanjets by reducing exit velocity to increase stream massflow. Really good designs get factor 1.4 static (!) thrust increase because the massflow increase outweighs the velocity decrease, but exit streams are subsonic.

The so-called "air-augmented" or "ducted" rockets work by the fanjet mechanism: velocity loss to gain massflow increase, in a speed regime where the massflow effect outweighs the velocity effect. That's no-to-low subsonic flight speed, as it turns out. No way around that. This isn't useful at all for compensating expandedness as a function of altitude backpressure. Doesn't really work that way.

GW

Last edited by GW Johnson (2017-03-07 17:19:49)

RGClark · 2017-03-15 07:36:30

In his presentation last year on the Interplanetary Transport System (ITS), at about the 54 minute mark Musk discusses that the second stage in its tanker form or in its spaceship form will be able to reach orbit when used as a single stage. He states though the tanker will not be able to land, presumably because of insufficient reserve fuel. Then it could be an expendable SSTO.

Making Humans a Multiplanetary Species - YouTube.
http://www.youtube.com/watch?v=H7Uyfqi_ … .be&t=3240

A simulation of the ITS upper stage tanker as an SSTO:

ITS Tanker SSTO - YouTube.
http://www.youtube.com/watch?v=Kzyzwr-5XXY

It suggests it can get a total mass of 190 metric tons to LEO as an expendable. Since the dry mass is 90 metric tons, this means a 100 metric ton payload to orbit.

Bob Clark

Antius · 2017-03-16 15:13:39

RGClark wrote:

In his presentation last year on the Interplanetary Transport System (ITS), at about the 54 minute mark Musk discusses that the second stage in its tanker form or in its spaceship form will be able to reach orbit when used as a single stage. He states though the tanker will not be able to land, presumably because of insufficient reserve fuel. Then it could be an expendable SSTO.
Making Humans a Multiplanetary Species - YouTube.
http://www.youtube.com/watch?v=H7Uyfqi_ … .be&t=3240
A simulation of the ITS upper stage tanker as an SSTO:
ITS Tanker SSTO - YouTube.
http://www.youtube.com/watch?v=Kzyzwr-5XXY
It suggests it can get a total mass of 190 metric tons to LEO as an expendable. Since the dry mass is 90 metric tons, this means a 100 metric ton payload to orbit.
Bob Clark

Interesting. Whilst it isn't much use in the short term, in the longer-term, if asteroid mining and space manufacturing take off, a heat shield would be an item that could be manufactured in space. A sintered titanium oxide shield with a steel skeleton, that can be bolted to the vehicle and simply dropped into the ocean when terminal velocity declines sufficiently. This would make an SSTO technically easier to build.

Last edited by Antius (2017-03-16 15:37:27)

SpaceNut · 2017-03-16 15:55:42

I see launch abort as a problem area for the ITS....

GW Johnson · 2017-03-16 16:28:07

Just FYI, titanium is not a high-temperature material. Its strength has pretty much gone bye-bye at the same ~750 F as plain mild carbon steel. Most of the 300-series stainless steels aren't structural crap until about ~1200 F. The superalloys will go somewhat hotter still, but not really all the way to Earth orbital reentry temperatures.

With a ceramic or metal heat shield, the key is a radiationally-black high-emissivity surface. The idea is to re-radiate all the energy convected to the skin panel, at a temperature that skin panel can survive. You don't want conduction into the interior, that's what you are trying to shield.

What you cannot deal with steady-state, you must heat-sink your way through on a transient. Peak heating rate is usually a pulse only several seconds long. Not quite simultaneous with peak deceleration, but close by in time.

GW

Last edited by GW Johnson (2017-03-16 16:30:39)

kbd512 · 2018-04-18 18:40:26

GW,

For vehicles like ITS that have surface geometries and reentry profiles like the Space Shuttle, if the vehicle had RCC leading edges with "weep holes" in them that could "bleed" a gas, like LN2 or liquid CO2 for example, could that gas provide a cooler boundary layer between the surface of the vehicle and the super heated atmosphere created by the vehicle's friction with the oncoming flow? Would it be possible to very briefly "trap" some of that cooler gas between the surface of the vehicle and the shockwave to carry some of the generated heat away?

I have no clue how practical such a solution would be, I just thought I'd ask because it's something I've always wondered about.

GW Johnson · 2018-04-19 11:37:20

I think what you're talking about goes under the names "transpiration cooling" or "film cooling". It works in wind tunnel tests, but you have a large quantity of expended liquid to deal with in or design. I don't think there is any "trapping" in the flow field, but there is some mass-mixing cooling of the thin film right adjacent to the surface. It was one of the options to be experimentally-investigated on the X-20 Dyna-Soar. That never got done as a flight test, though.

GW

Oldfart1939 · 2018-04-19 12:25:13

This transpiration-type cooling was utilized in the rocket motor of the V-2 (or A4, to be exact!) in order to keep the throat from burning through. In fact, I believe that some of Goddard's last rockets used a film or transpiration cooling for the same purpose. I'll check on that, since I have a hard cover of Goddard's "Rocket Development."

kbd512 · 2018-04-19 18:35:58

GW,

Apparently JAXA asked and answered this question with an experiment. With a mass flow of approximately .2kg/m^2/s, it only lowered the surface temperature by about 400F by injecting Nitrogen through a porous ceramic material. The material suffered no degradation in 5,100 seconds of cumulative testing, but there's a rapidly diminishing return on cooling effect achieved with increased mass flow and I believe it's already impractical at .2kg/m^2/s. The paper was published in AIAA in 2007. At least we can cross that idea off the list.

GW Johnson · 2018-04-20 14:18:24

Well, it's not efficient, but it does work up to a limited point. I think transpiration cooling with water was going to be tested as one of three approaches for X-20 entry. It was a lot of water, though.

GW

Last edited by GW Johnson (2018-04-20 14:18:47)

Oldfart1939 · 2018-04-20 14:35:48

Water would be the best substance to use for several reasons: (1) non-flammable, (2) cheap and available, (3) high heat of vaporization.

RGClark · 2018-07-11 13:49:08

RGClark wrote:

GW, that DARPA proposal I wrote was not successful. It was the one we were discussing about new methods of thermal protection in 2013 here:
Index » Interplanetary transportation » Reusable Rockets to Orbit.
2013-10-10 16:47:00
http://newmars.com/forums/viewtopic.php … 19#p117119
This discussion was from 2013. I finally got around to writing the proposal in 2015.
However, what DARPA really wants now in regards to launch access is low cost flights for small payloads, possibly using a reusable booster. The thermal protection issue was not key to that. But what is still a key question and what doomed the X-33 was the inability to get lightweight conformal tanks.
I took a look again at your blog post:
Sunday, October 6, 2013
Building Conformal Propellant Tanks, Etc.
Done successfully, you have a tank only a few percent heavier than a cylinder of the same volume, but not heavier by factors. It will be at least a little bit heavier, that is inevitable. That’s simply the price you must pay for the shape you want. Update 10-7-13: for the same panel thicknesses and weights as cylindrical construction, a lower-bound estimate of the weight growth factor is the perimeter length ratio, computed from cross-section views.
http://exrocketman.blogspot.com/2013/10 … s-etc.html
I'm still struck by your statement that you can get close to the same weight efficiency for lobed tanks as for cylindrical ones using metal tanks. You mentioned the figure of merit that determines the weight growth is perimeter to length...

GW, I wrote a blog post about getting the noncylindrical tanks of the X-33 getting the weight efficiency of usual cylindrical tanks:

DARPA's Spaceplane: an X-33 version, Page 3.
http://exoscientist.blogspot.com/2018/0 … age-3.html

One method was by using multiple cylindrical tubes to make up the tank. The other was by partitioning the tank into smaller sections. It occurs to me for this second method it has the effect of making the perimeter to length ratio small for each section, which gives confidence in the validity of the method.

Bob Clark

GW Johnson · 2018-07-12 12:09:26

Hi Bob:

Aside from favorite signs, as it turns out I had forgotten I already did the blog post about conformal tanks back in 2013. That one gave the best results I ever had available for the air mattress design. It won't fill an arbitrary 3-D space very efficiently, but some of those wing tank positions might get filled fairly efficiently with it. You can thin down the joint tension membrane if you can get away with only venting from cell to cell in the hemispherical ends.

If the cells are large enough for a man to enter, the aluminum alloys work with wire feed welding, or you might even attempt a composite with glassed-in joints. Do NOT attempt fastener-type joints without doing embedded metal shims to take the bearing and shear loads of the fasteners, and to take the end load in shear from shim to composite. Somewhere, I've got a posting about that issue too. Edit: see "exrocketman.blogspot.com" 6-13-15 "Commentary on Composite-Metal Joints".

GW

Last edited by GW Johnson (2018-07-12 12:40:41)

New Mars Forums

Announcement

#76 2017-02-18 15:13:57

Re: Reusable LOX/Kerosene SSTO with drop tanks

#77 2017-02-19 09:21:37

Re: Reusable LOX/Kerosene SSTO with drop tanks

#78 2017-02-19 10:29:31

Re: Reusable LOX/Kerosene SSTO with drop tanks

#79 2017-02-19 12:05:08

Re: Reusable LOX/Kerosene SSTO with drop tanks

#80 2017-03-07 08:54:45

Re: Reusable LOX/Kerosene SSTO with drop tanks

#81 2017-03-07 17:16:28

Re: Reusable LOX/Kerosene SSTO with drop tanks

#82 2017-03-15 07:36:30

Re: Reusable LOX/Kerosene SSTO with drop tanks

#83 2017-03-16 15:13:39

Re: Reusable LOX/Kerosene SSTO with drop tanks

#84 2017-03-16 15:55:42

Re: Reusable LOX/Kerosene SSTO with drop tanks

#85 2017-03-16 16:28:07

Re: Reusable LOX/Kerosene SSTO with drop tanks

#86 2018-04-18 18:40:26

Re: Reusable LOX/Kerosene SSTO with drop tanks

#87 2018-04-19 11:37:20

Re: Reusable LOX/Kerosene SSTO with drop tanks

#88 2018-04-19 12:25:13

Re: Reusable LOX/Kerosene SSTO with drop tanks

#89 2018-04-19 18:35:58

Re: Reusable LOX/Kerosene SSTO with drop tanks

#90 2018-04-20 14:18:24

Re: Reusable LOX/Kerosene SSTO with drop tanks

#91 2018-04-20 14:35:48

Re: Reusable LOX/Kerosene SSTO with drop tanks

#92 2018-07-11 13:49:08

Re: Reusable LOX/Kerosene SSTO with drop tanks

#93 2018-07-12 12:09:26

Re: Reusable LOX/Kerosene SSTO with drop tanks

Board footer