Horizontal launch as an option for reusability

22 views
Skip to first unread message

vid.b...@gmail.com

unread,
Feb 8, 2024, 2:03:59 PMFeb 8
to power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu

https://www.space.com/rocket-exhaust-pollution-upper-atmosphere

Thinking about the impact of rocket launch on the upper atmosphere with the potential for a 100 fold increase in launch rates does horizontal launch offer an option?  It appears limits may need to be imposed on launches that degrade the upper atmosphere.

 

Above the cloud layer laser power can be beamed to a launching space craft enabling low emission propulsion options to be considered.  https://en.wikipedia.org/wiki/Beam-powered_propulsion

 

Ukraine had earlier planned to use the AN-225 for horizontal launch.  It was designed to carry the Buran, a larger craft than the space shuttle.  While Russian forces destroyed the Mriya when attacking Kyiv.  https://en.wikipedia.org/wiki/Antonov_An-225_Mriya

https://en.wikipedia.org/wiki/Buran_(spacecraft)

 

Thermal barrier shields could be constructed from lunar basalt fiber and installed in orbit – concept proposed by Mike Turner.  This would further reduce the launch weight of the spacecraft. The Space Review: Space resources: the broader aspect  The shielding would be needed for reentry of spacecraft returning to Earth.

 

Vid Beldavs

Keith Henson

unread,
Feb 8, 2024, 4:44:23 PMFeb 8
to vid.b...@gmail.com, power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu
On Thu, Feb 8, 2024 at 11:04 AM <vid.b...@gmail.com> wrote:
>
> https://www.space.com/rocket-exhaust-pollution-upper-atmosphere
>
> Thinking about the impact of rocket launch on the upper atmosphere with the potential for a 100 fold increase in launch rates does horizontal launch offer an option? It appears limits may need to be imposed on launches that degrade the upper atmosphere.
>
The main problem is the NOx from reentry. It chews up ozone. I
talked NOAA into looking at this problem some years ago in the context
of Skylon.
https://htyp.org/design_to_cost#Skylon_Ozone_damage_.28NOx.29
>
> Above the cloud layer laser power can be beamed to a launching space craft enabling low emission propulsion options to be considered. https://en.wikipedia.org/wiki/Beam-powered_propulsion
>
It takes a really big laser, 4 GW or more.
https://htyp.org/design_to_cost#Space_laser

Keith

> Vid Beldavs
>
> --
> You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/088f01da5ac1%2491152140%24b33f63c0%24%40gmail.com.

Paul Werbos

unread,
Feb 8, 2024, 6:12:37 PMFeb 8
to vid.b...@gmail.com, power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu
In ancient days. when Ramon Chase gave us the best option for horizontal launch (and return), while Jess Sponable probed the design limits for vertical launch...
Ray DID look at the ozone issue in serious detail.

I have the serious technical discussion underneath these debates in my files, which Gary Barnhard MIGHT make available to the world via the NSS web site,
if we all get time and approvals and organization.

In essence, it was no problem. Those of us who worry VERY seriously about the stratospheric ozone layer would focus more on other issues,
like how to get to zero net greenhouse emissions from making electricity worldwide.



--

Keith Henson

unread,
Feb 8, 2024, 11:12:02 PMFeb 8
to Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu
On Thu, Feb 8, 2024 at 3:12 PM Paul Werbos <paul....@gmail.com> wrote:
>
> In ancient days. when Ramon Chase gave us the best option for horizontal launch (and return), while Jess Sponable probed the design limits for vertical launch...
> Ray DID look at the ozone issue in serious detail.

That is most interesting. When I talked NOAA into looking at ozone
damage from rocket flights, we were not aware of any previous work.
Was it published? The NOAA one is available online at

http://onlinelibrary.wiley.com/doi/10.1002/2016EF000399/full

Click on the PDF symbol next to the journal title.

"Serious detail" for NOAA took hundreds of hours of supercomputer time
and a lot of cooperation with Reaction Engines.

> I have the serious technical discussion underneath these debates in my files, which Gary Barnhard MIGHT make available to the world via the NSS web site,
> if we all get time and approvals and organization.
>
> In essence, it was no problem. Those of us who worry VERY seriously about the stratospheric ozone layer would focus more on other issues,
> like how to get to zero net greenhouse emissions from making electricity worldwide.
>
Hmm. The NOAA study showed that enough Skylon flights to support a
substantial power satellite construction program did do some damage to
the ozone, about the same as the chlorofluorocarbon releases.

Keith
>
> On Thu, Feb 8, 2024 at 2:04 PM <vid.b...@gmail.com> wrote:
>>
>> https://www.space.com/rocket-exhaust-pollution-upper-atmosphere
>>
>> Thinking about the impact of rocket launch on the upper atmosphere with the potential for a 100 fold increase in launch rates does horizontal launch offer an option? It appears limits may need to be imposed on launches that degrade the upper atmosphere.
>>
>>
>>
>> Above the cloud layer laser power can be beamed to a launching space craft enabling low emission propulsion options to be considered. https://en.wikipedia.org/wiki/Beam-powered_propulsion
>>
>>
>>
>> Ukraine had earlier planned to use the AN-225 for horizontal launch. It was designed to carry the Buran, a larger craft than the space shuttle. While Russian forces destroyed the Mriya when attacking Kyiv. https://en.wikipedia.org/wiki/Antonov_An-225_Mriya
>>
>> https://en.wikipedia.org/wiki/Buran_(spacecraft)
>>
>>
>>
>> Thermal barrier shields could be constructed from lunar basalt fiber and installed in orbit – concept proposed by Mike Turner. This would further reduce the launch weight of the spacecraft. The Space Review: Space resources: the broader aspect The shielding would be needed for reentry of spacecraft returning to Earth.
>>
>>
>>
>> Vid Beldavs
>>
>> --
>> You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
>> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/088f01da5ac1%2491152140%24b33f63c0%24%40gmail.com.
>
> --
> You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CACLqmgcWD8JumyYAKCXCH%3DuSriK2vCd1p_a_7dZ%3Dis%2BOi2sQBA%40mail.gmail.com.

Keith Henson

unread,
Feb 9, 2024, 3:00:27 PMFeb 9
to Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu
On Thu, Feb 8, 2024 at 8:11 PM Keith Henson <hkeith...@gmail.com> wrote:

snip

> That is most interesting. When I talked NOAA into looking at ozone
> damage from rocket flights, we were not aware of any previous work.
> Was it published? The NOAA one is available online at
>
> http://onlinelibrary.wiley.com/doi/10.1002/2016EF000399/full

Jess Sponable let me know that this URL didn't work in spite of it
having a doi in the URL

Using the title, "Global atmospheric response to emissions from a
proposed reusable space launch system" I found one that works.

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2016EF000399

Fixed https://htyp.org/design_to_cost as well.

Thanks Jess.

Keith

PS I started talking with NOAA about ozone damage from high rocket
traffic 9 years ago.

vid.b...@fotonika-lv.eu

unread,
Feb 10, 2024, 2:27:43 AMFeb 10
to Keith Henson, Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com
The article referenced emissions that contributed to climate change. If there are thousands of launches per annum the problem could become serious. What I raised is the possibility of beam launch for horizontally launch spacecraft that are beamed power via laser when the spacecraft is about the atmosphere layers that could significantly affect laser beamed power. In the lower atmosphere the AN-225 or other suitable carrier draws O2 from the atmosphere reducing the mass of fuel to carry. After the spacecraft disengages from the carrier, which could land in most larger airports. Unlike the Shuttle the mass of thermal barrier shields would not need to be launched to orbit further significantly reducing the launch mass. If thermal barrier shielding would be needed it would be installed in orbit ideally using lunar basalt fiber shielding material further reducing the ecological footprint of the operation on the Earth.

Paul Werbos

unread,
Feb 10, 2024, 8:29:28 AMFeb 10
to vid.b...@fotonika-lv.eu, Keith Henson, vid.b...@gmail.com, power-satell...@googlegroups.com, Geordie Keitt
Beam launch...

Back when I led the joint NSF-NASA research call for technologies to enable SSP (JIETSSP, joint with John Mankins),
we had a proposal from Leik Myrabo, who was the world's leader in exploring/developing that kind of technology.

We DON'T need it to attain the 10 cents per kwh estimate explained in Mankins' book, The Case for SSP, and I always remember the mantra: we must learn to crawl before we can walk, run and fly. Usually it is so. With heavy constraints on funding, I did not advocate funding that, even though the long-term porribilities are worth exploring.

I spoke a lot to Myrabo and his family. The main problem is the difficulty of control of a narrow beam over thousands of miles. New technology began appearing, based on algorithms I developed myself in previous decades and refined a lot since then. **IF** Myravbo had teamed with one of the folks proficient enough in RLADP control technology, I would have funded initial exploratory work to prove they could do it. But that did not happen. 

Ironically, my greatest struggle lately has been to push the development of Quantum RLADP, much more capable for this kind of control challenge, but not having a funding program has gotten very much in the way, despite much more complete details than we had decades ago when we first developed and pushed classical RALSP. I cc Geordie Keitt who is compiling a kind of integrated resource on quantum AGI, where quantum control as quantum RLADP, plays a lead role. He might send a link to those interested. In one hour
he and I go over the final draft of a "deck" describing this area to the investors in India who say they have allocated funds to start implementation, using
semiconductor fabs in India.

a.p.kothari astrox.com

unread,
Feb 10, 2024, 2:59:20 PMFeb 10
to vid.b...@fotonika-lv.eu, Keith Henson, Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com, Gordon’s Mail, john.livingston astrox.com

One of THE MAJOR problems with HTHL (and Jess may confirm this) is that it limits the GTOW to what gears (takeoff speed and hence and runway lengths) can handle. It is approximately 1.5 Mlb GTOW (for 20K lb payload to LEO). There is no room for growth. Then why do this at all, and why not just go with VTVL or VTHL?

SKYLON SSTO is nonsense. It cannot be done. Physics. People NEED to do System Level Analysis to see the impacts of various components on a mission driven system (payload to LEO as e.g.) or use that in their inferences.

 

The attached paper was given at AIAA ISH conference in 2011. Approved for public release by AFRL.

 

And another thing.

The total propellant/fuel used today by rockets is less than 0.04% of what auto vehicles use and thus contribute to climate/CO2 issues.

Even with 3 flights a day by Starship (which I do not see happening for a decade at least or more). It would be less than 1.2%.  

Best way to attack this FIRST is to do what Elon is doing. E-Vehicles. But that electricity has to come from nuclear or it is no go. Here Elon is wrong about putting the entire burden on battery storage.

 And there it has to be Gen IV Thorium Molten Salt Reactors (or something as promising). Most everything else is a band-aid or pure hogwash.  

 

A diagram of different types of aircraft parts

Description automatically generatedA graph of different colored columns

Description automatically generatedA graph of different colored columns

Description automatically generatedA graph of different colored columns

Description automatically generated with medium confidenceA graph of different colored bars

Description automatically generated with medium confidenceA group of different types of aircraft

Description automatically generated

 

-------------------------------------------------------

Dr. Ajay P. Kothari

President

Astrox Corporation

 AIAA Associate Fellow

 

Ph: 301-935-5868

Web:  www.astrox.com

Email: a.p.k...@astrox.com

-------------------------------------------------------

--

You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.

To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

AIAA-2011-2338-500.pdf

Paul Werbos

unread,
Feb 10, 2024, 3:14:44 PMFeb 10
to a.p.kothari astrox.com, vid.b...@fotonika-lv.eu, Keith Henson, vid.b...@gmail.com, power-satell...@googlegroups.com, Gordon’s Mail, john.livingston astrox.com
On Sat, Feb 10, 2024 at 2:59 PM a.p.kothari astrox.com <a.p.k...@astrox.com> wrote:

One of THE MAJOR problems with HTHL (and Jess may confirm this) is that it limits the GTOW to what gears (takeoff speed and hence and runway lengths) can handle. It is approximately 1.5 Mlb GTOW (for 20K lb payload to LEO). There is no room for growth. Then why do this at all, and why not just go with VTVL or VTHL?


Ray Chase was 100% familiar with the constraints imposed by weight on horizontal takeoff. His proposal for a near-term vehicle had something like 10 tons payload,
but with modular SSP design at that is not the barrier many lobbyists tell Congress it is. (They want funds for a bigger vehicle.)

Back when I handled policy for NSS, I heard very serious talks (some with McCulloch of Boeing and AIAA? Lots of files)
proposing what I thought of as "Space Spruce Goose." Yes, there are economies of scale allowing $/kg-LEO and $/kg-GEO even lower than what Chase estimated,
AS Howard Highes knew, using water-based takeoff and landing to allow greater weight. YES, that was a a serious option. Chase was surprised I supported his near-term vehicle idea as hard as I did, after I saw the numbers, but "we must crawl before we walk, run, fly." His near-term vehicle proposal MAINLY requires demonstration and refinement of hot structures needed ANYWAY for the longer-term, lower-cost options.


SKYLON SSTO is nonsense. It cannot be done. Physics. People NEED to do System Level Analysis to see the impacts of various components on a mission driven system (payload to LEO as e.g.) or use that in their inferences.

 

The attached paper was given at AIAA ISH conference in 2011. Approved for public release by AFRL.

 

And another thing.

The total propellant/fuel used today by rockets is less than 0.04% of what auto vehicles use and thus contribute to climate/CO2 issues.

Even with 3 flights a day by Starship (which I do not see happening for a decade at least or more). It would be less than 1.2%.  

 

Best way to attack this FIRST is to do what Elon is doing. E-Vehicles.


At the time of JIETSSP (search www.nsf.gov), john Mankins was VERY excited by maglev type launch, like what ONeill had proposed for the moon.
But Chase's vehicle certainly could have been running missions much sooner than that.

Electricity by BATTERY ... well, the best I saw for an aerospace vehicle was by Lonnie Johnson and Rolls Royce for an earth-based 4000 km 7777 type vehicle.
It was a great and credible serious engineering proposal, described in our IEEE book draft. But present politics is vblocking even the first baby steps proposed in that book, despite clear immediate economic needs. There ARE vested interests getting in the way.

Bryan Zetlen

unread,
Feb 10, 2024, 4:23:16 PMFeb 10
to Ed Tate, Francois Lambert, Keith Henson, Paul Werbos, power-satell...@googlegroups.com, vid.b...@fotonika-lv.eu, vid.b...@gmail.com

A note about rocket launch pollutants. In the mid 1960s, the USAF selected Vandenberg Air Force base, VAFB, an ICBM base, to construct a new launch pad SLC-6 built to launch rockets supporting the Manned Orbital Laboratory. Fast forward to the early 1970’s. As part of the overall STS launch facilities and operations support, USAF modified the original SLC-6 as the West Coast Space Shuttle STS launch site. SLC-6 was never used for MOL, and it was never used for STS. In fact, it remains an unanswered question why NASA and or USAF would even consider an eats-to-west launch option, albeit over water.

 

Eventually, MOL was abandoned, (my former father-in-law was Douglas’s chief MOL engineer). Getting to the point about launch pollution – for STS, one of the principal reasons it was never utilized, either for launch or for back up or emergency landing was political. Vandenberg is a bit over 100 miles northwest of Los Angeles. The STS program was initiated during an era when military and government operations of various kinds including weapons testing, ground battle operations and rocket launches were not closely scrutinized or regulated for environmental impacts., The EPA had just been established in 1970, and for the moist part, states and local jurisdictions exercised environmental supervision loosely if at all. 

 

Adjoining VAFB to the southeast and east are the communities of Santa Barbara and Ojai. Rocket launches, both liquid and solid emit tons fo pollutants, among them raw hydrogen, raw oxygen, O, O2, O3, carbon soot, hydroxyls, hydrochloric acid, aluminum oxide, nitrogen oxide, etc. In the event of an STS launch and given prevailing winds, an STS launch would inundate these communities with tons of corrosive acids, edged granular and crystalline particulates and black soot sufficient to strip paint from vehicles and houses, render the air temporarily toxic, contaminate local drinking water reservoirs and kill or injure pets and gardens.  These communities are second only in republican financial supporters for the then in power Reagan and other republican administrations. STS would never be launched form VAFB SLC-6



--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

a.p.kothari astrox.com

unread,
Feb 10, 2024, 5:57:18 PMFeb 10
to Keith Henson, vid.b...@fotonika-lv.eu, Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com, Gordon’s Mail

No, Keith. You may have spent 100 hrs with Skylon excel but I have actually designed and built Skylon in Astrox code HySIDE which is not a spreadsheet. The code is much more complicated and also ITAR so I cannot discuss it here in detail.  

I sent the results of Skylon trade study previously but apparently you have not looked at them.

 

From this paper:

SIDE/HySIDE SIDE© (System Integrated Design Environment) is a component building software; either from scratch or from existing components. The version for building components related to hypersonic vehicles (both airbreathers and rockets) is HySIDE©. Component parts are created first and vehicles and vehicle systems are assembled from components. Components are stored in directories as part of the HySIDE library. New systems can be built and saved. New directories can be built and can be added to the library. Components and directories can be deleted if no longer used. Tools for analysis, visualization and report creation are included. Variety of geometries can be created and are available for further detailed CFD analysis and CAD modeling Vehicles can be sized until “closed”. SIDE/HySIDE runs on Microsoft Windows based personal computers and workstations.

 

From our JSR paper(attached herewith):

Design Code

All vehicles in this design study have been configured with the

SpaceSIDE code developed by Astrox Corporation.1 The code is a

component-based object-oriented design package within a systems

engineering software environment. SpaceSIDE uses analytical solutions

and tabulated data as available rather than detailed computational

fluid dynamic solutions to be speedy and flexible while still

maintaining a high degree of accuracy. Use of the code’s rapid design

and analysis capabilities allows for the quick systematic comparison

of hundreds of design parameters and input cases.

To design a hypersonic vehicle, the code uses the freestream Mach

number and altitude at a chosen design point and specified bowshock

strength, from which the method of characteristics and streamline

tracing methods2 are used to form the inlet surface. After the trace,

the surface inviscid forces are known as is the inlet exit flow state.

A quasi-one-dimensional combustor model is used to model the

mixing and burning of hydrogen or hydrocarbon, and a combustor

surface is defined. The nozzle flowfield is then also created

using the method of characteristics. An external surface joins the

inlet capture area and nozzle exit. A reference temperature method

161 is then applied to determine the viscous forces, heat transfer, and

boundary-layer displacement thickness on each surface. The aerodynamic

forces are determined by integrating the pressures on each

surface’s gridpoints.3 A rocket vehicle is analyzed with the same

methods, but without the internal flowpath surfaces.

The code has the ability to perform analysis in a completely integrated

fashion (propulsion-airframe-massproperties-aero-gravlossheating-

volumes, etc.). Individual components include either hypersonic

airbreathing or rocket engines integrated into a full vehicle

model; their performance is calculated over the complete mission

trajectory. Vehicle sizing is done in an iterative loop. The vehicle is

scaled until the volume available for the fuel is equal to the fuel volume

needed based on individual component weights and densities.

The code calculates the volumes and areas of all of the components

and from this subtracts the volumes of payload, equipment, thermal

protection system (TPS), etc. The resulting volume is multiplied by

a tank packaging efficiency as a measure of how well the tank shape

is able to use the available volume. The resulting value is the volume

available for propellant and must equal the fuel volume required to

complete the mission trajectory to “close” the vehicle. All of the

components will require resizing, because the vehicle is continuously

scaled to match all of these requirements simultaneously.

The entire code consists of over 200 subroutines and functions

that account for approximately 12,000 executable lines of code. Several

standard codes, such as Missile Datcom for aerodynamics, have

been integrated into the code’s suite of analysis tools. Setup time for

the complete analysis of a new system requires several days, and,

once the included components of the specific vehicle system are

connected, the system calculations for each solution run are done

in about 10 min on a standard desktop PC. The code has the ability

to model 21 different commercially available rocket engines as

well as airbreathing scramjet-based engines and traditional turbine

engines using a variety of inlet geometries. Rocket geometries are

also included, as represented in Fig. 1.

 

 

-------------------------------------------------------

Dr. Ajay P. Kothari

President

Astrox Corporation

 AIAA Associate Fellow

 

Ph: 301-935-5868

Web:  www.astrox.com

Email: a.p.k...@astrox.com

-------------------------------------------------------

 

From: Keith Henson
Sent: Saturday, February 10, 2024 4:06 PM
To: a.p.kothari astrox.com <a.p.k...@astrox.com>
Cc: vid.b...@fotonika-lv.eu; Paul Werbos <paul....@gmail.com>; vid.b...@gmail.com; power-satell...@googlegroups.com; Gordon’s Mail <gm...@earthlink.net>
Subject: Re: Horizontal launch as an option for reusability

 

On Sat, Feb 10, 2024 at 10:37AM a.p.kothari astrox.com <a.p.k...@astrox.com> wrote:

One of THE MAJOR problems with HTHL (and Jess may confirm this) is that it limits the GTOW to what gears (takeoff speed and hence and runway lengths) can handle. It is approximately 1.5 Mlb GTOW (for 20K lb payload to LEO). There is no room for growth. Then why do this at all, and why not just go with VTVL or VTHL?

SKYLON SSTO is nonsense. It cannot be done. Physics.

 

AJ, I think you are speaking from ignorance.  I have spent well over a hundred hours with the Skylon Excel spreadsheets and even contributed to reducing the takeoff mass by finding another way to stop the Skylon in the event of a rejected takeoff.  The payload fraction is not very high, 15 tons out of 325, (4.6%) but at high flight rates the cost goes under $100/kg, which is low enough for power satellites.

 

The innovation for the Skylon engines is the precooler.  That was tested with a J79 on full afterburner simulating Mach 5 intake air. It was a joint project between Reaction Engines and the AFRL and worked as predicted.

 

SpaceX may beat out Skylon on cost, but the ozone damage might not permit the required flight rate (25,000 per year, 2.5 million tons per year) for power satellites..  The NOAA paper indicated that the ozone damage was acceptable even at a million Skylon flights per year.  I suspect this is not the case for LNG/LOX rockets, but I don't know because the study has not been done.

 

Keith

 

People NEED to do System Level Analysis to see the impacts of various components on a mission driven system (payload to LEO as e.g.) or use that in their inferences.

 

The attached paper was given at AIAA ISH conference in 2011. Approved for public release by AFRL.

 

And another thing.

The total propellant/fuel used today by rockets is less than 0.04% of what auto vehicles use and thus contribute to climate/CO2 issues.

Even with 3 flights a day by Starship (which I do not see happening for a decade at least or more). It would be less than 1.2%.  

Best way to attack this FIRST is to do what Elon is doing. E-Vehicles. But that electricity has to come from nuclear or it is no go. Here Elon is wrong about putting the entire burden on battery storage.

 And there it has to be Gen IV Thorium Molten Salt Reactors (or something as promising). Most everything else is a band-aid or pure hogwash.  

  

-------------------------------------------------------

Dr. Ajay P. Kothari

President

Astrox Corporation

 AIAA Associate Fellow

 

Ph: 301-935-5868

Web:  www.astrox.com

Email: a.p.k...@astrox.com

-------------------------------------------------------

 

-----Original Message-----
From: power-satell...@googlegroups.com On Behalf Of vid.b...@fotonika-lv.eu
Sent: Saturday, February 10, 2024 2:28 AM

--

You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.

To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

AIAA-13392-195 HTHL vs VTHL comparision.pdf

Keith Henson

unread,
Feb 10, 2024, 9:01:38 PMFeb 10
to a.p.kothari astrox.com, vid.b...@fotonika-lv.eu, Paul Werbos, vid.b...@gmail.com, power-satell...@googlegroups.com, Gordon’s Mail
 On Sat, Feb 10, 2024 at 2:51 PM a.p.kothari astrox.com <a.p.k...@astrox.com> wrote:

No, Keith. You may have spent 100 hrs with Skylon excel but I have actually designed and built Skylon in Astrox code HySIDE which is not a spreadsheet. The code is much more complicated and also ITAR so I cannot discuss it here in detail.  

I sent the results of Skylon trade study previously but apparently you have not looked at them.


I dug it out of my email and remember looking at it and deciding it didn't make a lot of sense.

1. Skylon does not close even for 20K lbs payload to LEO Easterly to 100 nm orbit

Why you are not using metric is beyond me.  20k pounds is a little over 9 metric tons.  100 nautical miles is 185 km.

Reaction Engines D1 configuration is supposed to deliver 15 metric tons to a 300 km equatorial orbit.  That's the numbers in the Wikipedia page.  If you get different numbers than Reaction Engines did, why?

5. Horizontal takeoff still requires reasonable size wings although we gave them
benefit of 331 nm velocity at  takeoff

What does "nm" mean in this context?

Keith

Arthur Woods

unread,
Feb 11, 2024, 1:07:28 PMFeb 11
to power-satell...@googlegroups.com, Keith Henson, jo...@virtussolis.space

Speaking of using Skylon to launch SPS,

In the Virtus Solis Whitepaper that John Bucknell posted on the list on 02.02.2024

https://www.researchgate.net/publication/377865839_Survey_of_Space_Based_Solar_Power_SBSP

On page 9, "Table 1:  Summary of SBSP Concept Parameters and Performance" references the Frazer-Nash ESA 2022 study using Skylon for launching CASSIOPeiA
with following figures:

CASSIOPeiA
Rated Power to Grid: 1440 MW
Orbital Mass 2,064 MT
Total CAPEX: $4,537 million (€4,210 million)

Spacelift Cost: $2,464 million  (€2,286 million) 
Spacelift: Reaction Engines Skylon
Stated LCoE: $63.64
(€59)  MWh / Nornalized LCoE: $54.33 (€50) MWh

($2,464 million / 2,064,000 kg = $ 1,194/kg)  (€1,108/kg)

From the FNC_014843_53334R_TN3_System Breakdown and Technical Feasibility_ISSUE_1.3.pdf

On page 37 Frazer-Nash writes:

The total spacelift mass that is needed to put a 1.44 GW SBSP system into orbit is 2,491 metric tons. This is equal to the satellite mass of 1,816 metric tons plus the mass of station keeping propellant, assembly robots, and OTV.

Starship, a planned fully reusable super heavy-lift launch vehicle that is being developed by SpaceX, represents the only near-term launch concept which can deliver SBSP’s modular structures to GTO at a reasonable cost and in the right orbit. This system can deliver a total mass of between 21-29 T to GTO, assuming that Starship is refuelled in orbit using propellent that is also delivered to GTO.

Taken together, these two assumptions suggest that between 86 and 119 Starship launches are required to deliver a single 1.44 GW SBSP system into orbit.

In the FN study (page 30), Spacelift costs range from: 454 €/kg  - 1,877 €/kg - 3,301 €/kg

If the spacelift cost can reflect costs observed with reusable rockets in the USA that are launched via the public/private partnership between NASA and SpaceX (~€400/kg), then the LCOE of SBSP is highly competitive compared to other electricity generating assets.

***

From page 60:

We consider two systems to explore the likely range of cost figures for the future space launch market: 

  1. A fully reusable single stage to orbit Reaction Engines SABRE powered horizontal take-off and landing spaceplane.
     
  2. The SpaceX Starship which features fully reusable first and second stages, and with the capability of refuelling in orbit. 

Both systems are designed to require only modest maintenance / refurbishment between flights, rapid turnaround time, high flight rate and high utilisation. 

The life of the Starship is assumed to be up to 100 flights, and that of the spaceplane up to 200 flights.

The SPS payloads would be launched to LEO, where the vehicle would be refuelled to provide an efficient transfer to a geostationary transfer orbit. A dedicated orbit transfer vehicle would deliver the payload to the final orbit. The orbit transfer vehicle would be refuelled from the launch vehicle. Using published performance data for Starship it is estimated that 57.9 tonnes could be delivered to GEO with two launches, one to provide refuelling in LEO. Elon Musk has quoted very ambitious figures for Starship launch costs, which assumes high flight rates and a life of up to 100 flights.  Assuming a cost per launch comparable with the current cost for SpaceX Falcon Heavy, $100M, the cost is $3,453/kg.

As Spacefreight becomes commoditised, as airfreight is today, the costs reduce. Reaction Engines predict  that as the flight rate per year increases from 1,000 to 10,000 per year the cost falls from $400/kg to $100/kg.

From: FNC_TN1_Issue 3 - FINAL_S2C120822.pdf

This FrazerNash report (page 16)  estimates the levelised cost of electricity from space-based solar power systems to be €59 (£50) per MWh at a hurdle rate of 20%. ($ 63.6 /MWh)

The cost is sensitive to hurdle rate and at a 10% rate, the LCOE is €31 (£26) per MWh. ($33.4 /MWh)

The Virtus Solis SPS architecture mentioned in their white paper anticipates a LCoE of $25/MWh (€23.2/MWh).

**

Does anyone know the status of Skylon / SABRE development today?

Regards, 

Arthur Woods

https://astrostrom.ch


----------------------

--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages