Jump to content

HASDA - my virtual, (mostly) Japan-inspired space program (non-Kerbal)


Pipcard

Recommended Posts

  • 7 months later...

First and second stages are both 5.39 m in diameter, and contain methane/LOX propellant. Total mass is over 440000 kg.

First stage: seven engines, total thrust of 5839 kN (5200 kN at sea level), Isp = 363.9 s (323.9 s at sea level)

Second stage: one vacuum-optimized version of the first stage engine, thrust of 833.9 kN, Isp = 380 s

It is based on the SpaceX philosophy of common propellants and mostly-common engines for both stages of the rocket to reduce manufacturing/operations costs.

If you haven't noticed yet, it also has grid fins. So the first stage will be reusable, SpaceX style.

95oZGTU.png

Payload:

21400 kg to LEO, 8200 kg to GTO (expendable mode)

14100 kg to LEO, 4200 kg to GTO (reusable mode, first stage return to launch site)

I have already tested the rocket in Orbiter (but with a placeholder mesh), and it uses modified code from the Falcon 9R add-on by BrianJ.

Edited by Pipcard
Link to comment
Share on other sites

On 10/19/2014 at 1:45 PM, Pipcard said:

 

 

IylNwYy.png

HASDA (Hatsunia Aerospace Science & Development Agency*) is a fictional alternate-reality version of JAXA/NASDA (Hatsunia being an alternate-reality counterpart to Japan - an entire country of Hatsune Miku fans).

It is also known as the "Hatsunese Space Program."

Instead of using KSP, I am making add-ons for Orbiter (space flight simulator).

Bi8lqhe.png

Launches are conducted from Negishima Space Center, a (completely fictional) island spaceport located near the equator in the western Pacific Ocean

s5cJRR9.png

The first orbital launch occurred on August 31, 1967 with the Negi-1 launch vehicle carrying HATSUNE (High Altitude Test Satellite Using Numerous Experiments) (based on the Lambda-4S and Ã…Å’sumi, respectively)

Qj8hmlx.png

Today, HASDA uses the M-II and Negi-5 launch vehicles (based on H-IIA and Mu-V, respectively). Commercial launch services are actually provided by a company known as Hatsunespace, and the vehicle components are manufactured by Mikubishi Heavy Industries, Crypton Future Aerospace, and Yamaha Heavy Industries. The M-II can optionally have 2-4 solid rocket boosters or 2 liquid rocket boosters.

nj2uXCG.png

oRrT9zt.png

HASDA's manned spacecraft system is the Reusable Crew Vehicle: a lifting body which mostly takes its inspiration from an experimental JAXA aircraft (LIFLEX), but is also inspired by Kliper and Dream Chaser. The first RCV is also known as "Hatsune" (RCV01). In-universe, the first launch took place in 2007, but I started making it several weeks ago.

HASDA also has a space station and is planning a manned lunar landing, but I haven't gotten around to making them yet.

 

 

*(retconned from Hatsunia AeroSpace Development Agency)

What's with you and Hatsune? Either way, you probably want to make an Epsilion rocket too now that that's a thing:P

And sounding rockets http://global.jaxa.jp/projects/rockets/s_rockets/

Also, do you have a thread on this on the Orbiter Forums?

On 11/11/2014 at 8:37 PM, Pipcard said:

RCV01 on M-II

pZZEauO.png

Yeah, is the core H2/Lox? If so, you will likely use boosters of sorts on each version, simply due to the fact the TWR of H2 lox engines genreally suck due to large engines being very expensive. Other fuels wouldn't be reminiscent of the H- line of rockets.

But either way, the top looks very unstable. You probably want a thicker adapter covering the entire bottom of the RCV to the rocket, like done on dream Chaser, as that offers more stability.

Also, is there a cargo version of the rocket/RCV?

On 11/12/2014 at 5:19 PM, Pipcard said:

If you're asking about who, she is basically a virtual character that represents a singing synthesizer software. If you're asking what the name means, it means "first sound" (Miku means "future")

Think of Hatsunia as a country with 139 million Miku fans.

Some people have associated her with space flight.

I'm not ready to import it into KSP as of this moment.

(the Pip stands for Piplup, by the way)

It *should* be easy to make this in KSP RSS + Procedural parts. I might do that eventually.

On 12/21/2014 at 10:55 PM, Special Agent Sigvan said:

Even in fictional video games set in alternate realities, Japanese stuff looks better than American stuff.

I am not surprised.

I would argue not. The Delta II and Atlas V are very sleek and unique-looking rockets.

On 1/24/2015 at 10:11 PM, Pipcard said:

Thank you.

3D models are made in Wings3D and converted to Orbiter's .msh format through Hielor's converter.

Textures are made with GIMP and Inkscape.

- - - Updated - - -

zVnyYCS.png

The Reusable Crew Vehicle requires an expendable APAS docking adapter to dock with space stations.

(pictured above is the ISS that comes with Orbiter -which includes all the cancelled modules-, but I plan to make a Hatsunese Space Station after I'm done with the RCV)

Where's the windows? I know that increases complexity, but darn, it, I want to see outside!!

(It's also good incase the auto-landing system fails)

On 3/17/2015 at 11:25 PM, Pipcard said:

jsXzrMt.jpg

Note that this is a Solid Edge render (with post-processing in GIMP) and is not the final version that will appear in Orbiter.

What's up with the diagonal panels? No station construction Arm? What is this madness?:0.0:

Also, that docking port on the front isn't too safe to dock on due to being too close to the other modules. I would add a node area on the nadir if you needed more places to dock (and a copula can't hurt) :)

On 3/21/2015 at 8:13 AM, Pipcard said:

Thanks, but that's not how they would deploy. They just fold out like most solar panels.

Also, hexagons are cool and futuristic-looking.

Why tho? That just increases complexity of the panels :P

On 8/30/2015 at 8:56 AM, Pipcard said:

 

It seems weird to put panels on the truss, especially since you aren't going to get much more power out of it. Just saying :)

Link to comment
Share on other sites

On 4/6/2016 at 11:37 PM, Pipcard said:

First and second stages are both 5.39 m in diameter, and contain methane/LOX propellant. Total mass is over 440000 kg.

First stage: seven engines, total thrust of 5839 kN (5200 kN at sea level), Isp = 363.9 s (323.9 s at sea level)

Second stage: one vacuum-optimized version of the first stage engine, thrust of 833.9 kN, Isp = 380 s

It is based on the SpaceX philosophy of common propellants and mostly-common engines for both stages of the rocket to reduce manufacturing/operations costs.

If you haven't noticed yet, it also has grid fins. So the first stage will be reusable, SpaceX style.

95oZGTU.png

Payload:

21400 kg to LEO, 8200 kg to GTO (expendable mode)

14100 kg to LEO, 4200 kg to GTO (reusable mode, first stage return to launch site)

I have already tested the rocket in Orbiter (but with a placeholder mesh), and it uses modified code from the Falcon 9R add-on by BrianJ.

Hmm, you definitely want to supercool if you're using Ch4. I don't like Methane, since it offers very little advantages from an ISP point of view, and since you can make a overall more efficient rocket and smaller rocket by using H2/Lox + Rp-1 on the 1st stage. (and possibly a solid or hypergolic 3rd stage) but I digress. Common propellants aren't all that great, it only really saves on pad costs.

Ever thought of using the H-II diameter for the rocket, then using the "heavy" configuration (with booster reuse) to get the same payload with less R+D costs? Just a thought.

Is this a modular rocket, or is it just a singular rocket? The IRL H-III is modular, so I was just asking.

BTW, 21T to LEO is actually too small if you want a HLV. Boosters are a good idea here. 4 RP-1 or Ch4 boosters could land on a barge, along with the core  (allowing for a smaller rocket, saving costs on the larger tanks and larger engines).

It would be: 28T to LEO (4 boosters, expendable), 24T to LEO (4 boosters, Barge landing), 20T to LEO (3 Boosters, Barge landing), 17 T to LEO (2 Boosters, Barge Landing, 13 T to LEO (1 Booster, Barge landing), and 9T to LEO (core only, Barge landing). Thus, it would cover a large line of rocket types.

A manned moon launcher would be made by attaching 4 cores to the core, and attaching an upgraded 2-engine 2nd stage. A possible Delta II-class launcher could be made with a solid upper stage on a H-III core.

I don't know what LV capacity is needed for the RCV, so :P.

Also, I'm a fan of modular rockets, if you couldn't tell :P

BTW, I noticed you don't have any insulation on your liquid rockets. Or is it just painted over? Insulation is pretty rough, so the intricate designs on the rocket are going to be a royal pain in the butt on a CH4 rocket.

Link to comment
Share on other sites

Okay, this is a lot to reply to, but...

10 hours ago, YumonStudios said:

[1.] What's with you and Hatsune? Either way, you probably want to make an Epsilion rocket too now that that's a thing:P

[2.] And sounding rockets http://global.jaxa.jp/projects/rockets/s_rockets/

[3.] Also, do you have a thread on this on the Orbiter Forums?

[4.] Yeah, is the core H2/Lox? If so, you will likely use boosters of sorts on each version, simply due to the fact the TWR of H2 lox engines genreally suck due to large engines being very expensive. Other fuels wouldn't be reminiscent of the H- line of rockets.

[5.] But either way, the top looks very unstable. You probably want a thicker adapter covering the entire bottom of the RCV to the rocket, like done on dream Chaser, as that offers more stability.

[6.] Also, is there a cargo version of the rocket/RCV?

It *should* be easy to make this in KSP RSS + Procedural parts. I might do that eventually.

I would argue not. The Delta II and Atlas V are very sleek and unique-looking rockets.

[7.] Where's the windows? I know that increases complexity, but darn, it, I want to see outside!!

(It's also good incase the auto-landing system fails)

[8.] What's up with the diagonal panels? No station construction Arm? What is this madness?:0.0:

[9.] Also, that docking port on the front isn't too safe to dock on due to being too close to the other modules. I would add a node area on the nadir if you needed more places to dock (and a copula can't hurt) :)

[10.] Why tho? That just increases complexity of the panels :P

[11.] It seems weird to put panels on the truss, especially since you aren't going to get much more power out of it. Just saying :)

1. I like Miku and space. And there is a precedent for associating Miku with space and rocketry. The Negi-5 is the Epsilon-equivalent.

2. At one point I was planning to make sounding rockets, but I decided to focus on things such as crew vehicles and space stations instead.

3. It's several threads in the "add-on development" section of the forum, not a single thread.

4. The M-II core is kerosene (RP-1) /LOX.

5. See this as an example. I wanted to have a docking adapter built in (so it wouldn't need to dock to a PMA-style adapter that stuck out of a space station), but couldn't do a Dream Chaser-style adapter because the docking adapter would get in the way when jettisoning. I also wanted to save (virtual, guesstimated) mass.

6. No cargo version, however, there is an HTV-equivalent called the UTV (Unmanned Transfer Vehicle) "Hikyaku"

7. The windows are the circular things, and are there for the astronauts to look out when in orbit. It doesn't have front-facing windows, Kliper wasn't going to have front-facing windows either.

8. It's really for aesthetic reasons, I wanted the shape to be reminiscent of 未来 (Mirai). And if you look really closely, there is a station construction arm.

9. Which is why I moved it up in a later design. That was only preliminary. There was also a cupola at the bottom.

10. Not really that complex. They "look more complicated despite being more efficient to build with (a hexagon has a greater area for the length of its sides than a square, hence why bees use hexagonal honeycombs for storage)" (TVTropes link)

11. The trusses need to rendezvous with the station by themselves.

9 hours ago, YumonStudios said:

[12.] Hmm, you definitely want to supercool if you're using Ch4. I don't like Methane, since it offers very little advantages from an ISP point of view, and since you can make a overall more efficient rocket and smaller rocket by using H2/Lox + Rp-1 on the 1st stage. (and possibly a solid or hypergolic 3rd stage) but I digress. Common propellants aren't all that great, it only really saves on pad costs.

[13.] Ever thought of using the H-II diameter for the rocket, then using the "heavy" configuration (with booster reuse) to get the same payload with less R+D costs? Just a thought.

[14.] Is this a modular rocket, or is it just a singular rocket? The IRL H-III is modular, so I was just asking.

BTW, 21T to LEO is actually too small if you want a HLV. Boosters are a good idea here. 4 RP-1 or Ch4 boosters could land on a barge, along with the core  (allowing for a smaller rocket, saving costs on the larger tanks and larger engines).

It would be: 28T to LEO (4 boosters, expendable), 24T to LEO (4 boosters, Barge landing), 20T to LEO (3 Boosters, Barge landing), 17 T to LEO (2 Boosters, Barge Landing, 13 T to LEO (1 Booster, Barge landing), and 9T to LEO (core only, Barge landing). Thus, it would cover a large line of rocket types.

[15.] A manned moon launcher would be made by attaching 4 cores to the core, and attaching an upgraded 2-engine 2nd stage. A possible Delta II-class launcher could be made with a solid upper stage on a H-III core.

[16.] I don't know what LV capacity is needed for the RCV, so :P.

Also, I'm a fan of modular rockets, if you couldn't tell :P

[17.] BTW, I noticed you don't have any insulation on your liquid rockets. Or is it just painted over? Insulation is pretty rough, so the intricate designs on the rocket are going to be a royal pain in the butt on a CH4 rocket.

12. Raptor (first stage) - 363 s Isp in vacuum (321 at sea level). Merlin (first stage) - 311 s in vacuum (282 s at sea level). Elon Musk made the compelling argument that "Using three kinds of rockets in the same vehicle may optimize its performance, but at a price: 'To a first-order approximation, you’ve just tripled your factory costs and all your operational costs'." Hydrogen/LOX is not a good common propellant for both stages for the reasons you mentioned in [4.]. Switching to an all-methane rocket will "optimize for cost" (per kg) instead of "optimizing for pure performance." Yes, there are R&D costs, but SpaceX is doing it anyway (It is good for reusability because there is less residue buildup, a.k.a. coking). The M-II was designed in the late 1980s (in-universe), decades before SpaceX disrupted the launch industry.

13. I need the larger diameter for really heavy payloads. (There will be a three-core version later)

14. It won't have SRBs, as it is planned to have scalability through reuse modes, just like Falcon 9. It will have a three-core variant sometime later. Landing more than 3 boosters (including central core) is complex operationally, which is why Falcon Heavy will only have 3 cores, and the future BFR will only have one. Small boosters (like what you seem to be proposing) will result in extra production lines and more manufacturing costs.

15. The plan is to use multiple launches of the 3-core M-II Heavy. And it does not need a solid upper stage when it can scale through reusability.

16. About 10-11 tonnes.

17. It's painted. See Vulcan with it's 'Murican decals on the LNG (liquid natural gas, mostly methane) first stage

Edited by Pipcard
Link to comment
Share on other sites

16 hours ago, Pipcard said:

The Negi-5 is the Epsilon-equivalent.

No, it isn't:

This is from the OP:

Quote

Today, HASDA uses the M-II and Negi-5 launch vehicles (based on H-IIA and Mu-V, respectively).

 

16 hours ago, Pipcard said:

2. At one point I was planning to make sounding rockets, but I decided to focus on things such as crew vehicles and space stations instead.

Can I do it then :3 (in KSP) ?

 

16 hours ago, Pipcard said:

It's several threads in the "add-on development" section of the forum, not a single thread.

Can you show me to them please?

16 hours ago, Pipcard said:

4. The M-II core is kerosene (RP-1) /LOX.

Can I have the stats for the M-II in general? Thanks.

16 hours ago, Pipcard said:

5. See this as an example. I wanted to have a docking adapter built in (so it wouldn't need to dock to a PMA-style adapter that stuck out of a space station), but couldn't do a Dream Chaser-style adapter because the docking adapter would get in the way when jettisoning. I also wanted to save (virtual, guesstimated) mass.

? I don't see the point. Just keep it on the space station, the docking adapter can be reused then. Also, even the HL-20 attachment is wder than the one for the RCV.

16 hours ago, Pipcard said:

6. No cargo version, however, there is an HTV-equivalent called the UTV (Unmanned Transfer Vehicle) "Hikyaku"

Then you probably want to make the crew vehicle carry quite a bit of downmass. How much does it have?

16 hours ago, Pipcard said:

8. It's really for aesthetic reasons, I wanted the shape to be reminiscent of 未来 (Mirai). And if you look really closely, there is a station construction arm.

But those diagonal panels are pointless. They're no good as radiators, nor as solar panels (they can't move to face the Sun)

And I could only see the scientific arm for the unpressurized exposure experiments. And speaking of experiements, is this a scientific station? I would think so.

16 hours ago, Pipcard said:

 

9. Which is why I moved it up in a later design. That was only preliminary. There was also a cupola at the bottom.

Ah, didn't notice it. Sorry. The copula looks like it is in a bad position, as it isn't on the Nadir, allowing for better Earth Observations, though.

And it's still kind of a bad position to put a docking port, even as a backup.

On 3/2/2015 at 6:40 PM, Pipcard said:

 

Thanks.

Current WIP: the Hatsunese Space Station

2T6mI9x.png

 

Assuming the positions of the docking ports haven't changed, you can just use the 2 on the front side, which have no obstructions, and get rid of the back one. You only ever need two, (especially since this looks like a small station).

Speaking of docking ports (and assuming you use the ISS US berthing system), it would be ideal to have a 2nd docking port on the nadir. I Know one's on the bottommost module, but a good 2nd one could be located on the bottom of the left module on the back. Berthing is done from the Nadir.

Sorry for all this questioning. I hope you don't mind me criticizing your designs.

Link to comment
Share on other sites

35 minutes ago, YumonStudios said:

[1] No, it isn't:

This is from the OP:

 

[2] Can I do it then :3 (in KSP) ?

 

[3] Can you show me to them please?

[4] Can I have the stats for the M-II in general? Thanks.

[5] ? I don't see the point. Just keep it on the space station, the docking adapter can be reused then. Also, even the HL-20 attachment is wder than the one for the RCV.

[6] Then you probably want to make the crew vehicle carry quite a bit of downmass. How much does it have?

[7] But those diagonal panels are pointless. They're no good as radiators, nor as solar panels (they can't move to face the Sun)

[8] And I could only see the scientific arm for the unpressurized exposure experiments. And speaking of experiements, is this a scientific station? I would think so.

[9] Ah, didn't notice it. Sorry. The copula looks like it is in a bad position, as it isn't on the Nadir, allowing for better Earth Observations, though.

And it's still kind of a bad position to put a docking port, even as a backup.

[10] Assuming the positions of the docking ports haven't changed, you can just use the 2 on the front side, which have no obstructions, and get rid of the back one. You only ever need two, (especially since this looks like a small station).

[11] Speaking of docking ports (and assuming you use the ISS US berthing system), it would be ideal to have a 2nd docking port on the nadir. I Know one's on the bottommost module, but a good 2nd one could be located on the bottom of the left module on the back. Berthing is done from the Nadir.

[12] Sorry for all this questioning. I hope you don't mind me criticizing your designs.

[1] I meant Epsilon seeing as Epsilon is basically a Mu-V but with the H-IIA SRB as its first stage. But in-universe, Negi-5 was introduced in the 90s, like Mu-V.

[2] Whatever you want to do.

[3] Negishima | M-II/Negi-5 | Reusable Crew Vehicle | Space Station Mirai | M-III Launch Vehicle

(a lot of the older posts have defunct images because of imageshack, ugh)

[4] Stage masses (a lot of that was guesstimated based on real stage propellant fractions) and payload capacities

[5] The adapter needs to be brought up there in the first place. The interface between the docking adapter and the RCV doesn't have proper docking mechanisms, similar to how the Soyuz descent module can't re-dock with the orbital module. Making the adapter wider at the top would have made it too tall for the back of the RCV.

The issue with a Dream Chaser style adapter for RCV is that the docking port wouldn't fit through the front end.

[6] Downmass is a few hundred kilograms.

[7] They are supposed to be radiators, and that was not the final design. I wanted to combine form and function in a way that seemed somewhat plausible.

[8] Yes it is.

[9] I moved the cupola so that there could be a free berthing port for a UTV at the nadir.

[10] That was not the final design. One of the docking ports is supposed to be used for temporary visits or crew rotations.

[11] Eh, it's okay.

[12] It is getting a bit excessive.

Edited by Pipcard
Link to comment
Share on other sites

17 hours ago, Pipcard said:

 

17 hours ago, Pipcard said:

11. The trusses need to rendezvous with the station by themselves.

 

17 hours ago, Pipcard said:

15. The plan is to use multiple launches of the 3-core M-II Heavy. And it does not need a solid upper stage when it can scale through reusability.

If my assumption that the M-II is replaced by the M-III, and the M-II is smaller than the M-III in payload, then the M-II heavy is a bad decision to use on a Moon mission. You want the biggest rocket you can make and launch in reasonable amounts (reducing complexity of the spacecraft), and a 4-core expendable supercooled M-III is a great way to make that happen.

The solid upper stage was just an idea, I knew you'd reject it anyways. It wouldn't fall under a commonly used payload class.

17 hours ago, Pipcard said:

 

16. About 10-11 tonnes.

Thanks. But that means the M-III is OP for the RCV.

17 hours ago, Pipcard said:

 

17. It's painted. See Vulcan with it's 'Murican decals on the LNG (liquid natural gas, mostly methane) first stage

Those were just for marketing and aesthetic purposes, like the original paint job on the SLS. Noone in their right mind would actually paint over insulation, it adds a lot of weight for not reason. The newer Vulcan images show an Orange core, which means a non-painted core:

Atlas_Evolution-879x485.jpg

17 hours ago, Pipcard said:

12. Raptor (first stage) - 363 s Isp in vacuum (321 at sea level). Merlin (first stage) - 311 s in vacuum (282 s at sea level). Elon Musk made the compelling argument that "Using three kinds of rockets in the same vehicle may optimize its performance, but at a price: 'To a first-order approximation, you’ve just tripled your factory costs and all your operational costs'." Hydrogen/LOX is not a good common propellant for both stages for the reasons you mentioned in [4.]. Switching to an all-methane rocket will "optimize for cost" (per kg) instead of "optimizing for pure performance." Yes, there are R&D costs, but SpaceX is doing it anyway (It is good for reusability because there is less residue buildup, a.k.a. coking). The M-II was designed in the late 1980s (in-universe), decades before SpaceX disrupted the launch industry.

That's because while Merlin uses a gas-generator cycle, Raptor uses a more effcient staged combustion cycle. It's apples to oranges.

http://www.braeunig.us/space/propel.htm

This link shows the max. conventional engine (no altitude compensation or air-breathing) isp and density (minus supercooling)

RP-1 Lox: 289s ISP sea level

Ch4 Lox: 299s ISP sea level

A 10 s isp difference is not worth it. SpaceX is pursuing it because they want to eventually reuse the 2nd stage, and H2 sucks when it comes to reuse due to hydrogen embrittlement. However, it's not impossible to solve and account for, otherwise DC-X, X-33, the Space Shuttle (original, fully reusable design, before it went to sh*t), and the New Shepard all use(d) H2 propellants. And you want to reuse only for 10 flights max anyways.

Not worth it. However, though Methane can self pressurize, the F9R has shown it's not necessary to have, and helium does the job just fine.

IVF can be used if you really need infinite burns, for H2 only though. It offers a nice ~1T payload capacity boost though. I'm not saying you shoulf use it it was just a suggestion.

Also,

Quote

"Using three kinds of rockets in the same vehicle may optimize its performance, but at a price: 'To a first-order approximation, you’ve just tripled your factory costs and all your operational costs'."

Implies that each stage added increases factory costs by 2x.

A 2-stage vehicle uses 2 rockets.

I'm only adding another rocket. And ULA doesn't seem to think it's a huge deal, they need to reduce prices as fast as they can to compete vs SpaceX on military launches (bribes, and relations to Lockheed and Boeing will only get you so far), and they're still cool about adding boosters to optimize payload.

H-III wants to reduce costs to go into the commerical market, but is still cool about using boosters to optimize payload. https://en.wikipedia.org/wiki/H3_Launch_Vehicle

Hell, Ariane 6 is expected to cost as much as today's F9 per kg to GTO (and possibly better than FH, depending on how well reuse goes), and it still uses boosters to optimize payload. http://spacenews.com/ariane-6-rocket-designers-say-theyll-match-or-beat-todays-spacex-prices-on-per-kilogram-basis/

Even SpaceX wanted to do it with LRBs when F5 was still a thing. http://www.spacelaunchreport.com/falcon9.html

Adding all that together, it can't cost that much more. Especially when you use the same engines and tankage diameter tooling as the core.

And in the launch scales where reuse is better than mass production, you will want to optimize payload, at least a bit.

Quote

Switching to an all-methane rocket will "optimize for cost" (per kg) instead of "optimizing for pure performance. Yes, there are R&D costs, but SpaceX is doing it anyway (It is good for reusability because there is less residue buildup, a.k.a. coking).

It's better to lower R+D costs to build it based off H-II tech, but with modifications to the engine and processes (like Horizontal vs vertical integration) to save costs, then chase after a 10s isp increase and increase R+D costs (and thus cost per launch) (and coking isn't an enormous problem for the Merlin anyways apparently due to using an O2 rich cycle, which also is the most efficient rocket cycle.)

I guess we have different viewpoints.

Link to comment
Share on other sites

56 minutes ago, Pipcard said:

[12] It is getting a bit excessive.

Sorry. :P

I can't help but do it. You're not the only one to have this problem with me.

56 minutes ago, Pipcard said:

 

[10] That was not the final design. One of the docking ports is supposed to be used for temporary visits or crew rotations.

Ok, but isn't that what a backup docking port is for? The only reason I can see the 3rd one being needed is in case to abandon ship in an emergency, but in that case, it's definitely not ideal, but still excusable.

56 minutes ago, Pipcard said:

 

[9] I moved the cupola so that there could be a free berthing port for a UTV at the nadir.

Ah. It's not a huge loss then, the orginal ISS design had the coupula like that too :)

56 minutes ago, Pipcard said:

 

[7] They are supposed to be radiators, and that was not the final design. I wanted to combine form and function in a way that seemed somewhat plausible.

It's still kind of stupid to put it there, but I guess since you already made the station, there's not much you can do about it now. I think.

56 minutes ago, Pipcard said:

[6] Downmass is a few hundred kilograms.

Well, that's better than what Mir had to work with. :)

56 minutes ago, Pipcard said:

[5] The adapter needs to be brought up in the first place. Making the adapter wider at the top would have made it too tall for the back of the RCV.

...And you can't bring it up with the module?

Either way, the modules also lack a propulsion system in the video. How did they get up there in the first place? :confused:

56 minutes ago, Pipcard said:

[5] The adapter needs to be brought up in the first place. Making the adapter wider at the top would have made it too tall for the back of the RCV.

Looking at the videos above again, that hole between the adaptor and the RCV is pretty darn skinny. Dang, I guess this station has a pretty strict waist diameter and cup size limit :) (seriously though, it couldn't be designed wider? What if someone gets stuck?)

56 minutes ago, Pipcard said:

 

[4] Stage masses (a lot of that was guesstimated based on real stage propellant fractions) and payload capacities

Thanks :)

And this http://www.silverbirdastronautics.com/LVperform.html

is a good tool to use for your rockets.

56 minutes ago, Pipcard said:

Thanks.

56 minutes ago, Pipcard said:

[1] I meant Epsilon seeing as Epsilon is basically a Mu-V but with the H-IIA SRB as its first stage. But in-universe, Negi-5 was introduced in the 90s, like Mu-V.

Oh. The IRL Mu-V stopped being used due to high costs, but I guess since it could be used as an ICBM, it could benefit from Mass production and make the Epsilon unnecessary...

Ok, I'll stop.

 

I'm actually a fan of this project, if I didn't care, I wouldn't be bothering to reply to this thread. :P

It's pretty impressive.

Edited by YumonStudios
Link to comment
Share on other sites

Meanwhile in KSP:

"Alert! We've got a competitor who launched rocket like us! The name is... HSP (Hatsunese Space Program)!"

Later, both space program started and battled in the Space Race!

 

Edited by Anbang11
Link to comment
Share on other sites

6 hours ago, YumonStudios said:

[1] If my assumption that the M-II is replaced by the M-III, and the M-II is smaller than the M-III in payload, then the M-II heavy is a bad decision to use on a Moon mission. You want the biggest rocket you can make and launch in reasonable amounts (reducing complexity of the spacecraft), and a 4-core expendable supercooled M-III is a great way to make that happen.

The solid upper stage was just an idea, I knew you'd reject it anyways. It wouldn't fall under a commonly used payload class.

[2] Thanks. But that means the M-III is OP for the RCV.

[3] Those were just for marketing and aesthetic purposes, like the original paint job on the SLS. Noone in their right mind would actually paint over insulation, it adds a lot of weight for not reason. The newer Vulcan images show an Orange core, which means a non-painted core:

[4] That's because while Merlin uses a gas-generator cycle, Raptor uses a more effcient staged combustion cycle. It's apples to oranges.

http://www.braeunig.us/space/propel.htm

This link shows the max. conventional engine (no altitude compensation or air-breathing) isp and density (minus supercooling)

RP-1 Lox: 289s ISP sea level

Ch4 Lox: 299s ISP sea level

A 10 s isp difference is not worth it. [5] SpaceX is pursuing it because they want to eventually reuse the 2nd stage, and H2 sucks when it comes to reuse due to hydrogen embrittlement. However, it's not impossible to solve and account for, otherwise DC-X, X-33, the Space Shuttle (original, fully reusable design, before it went to sh*t), and the New Shepard all use(d) H2 propellants. And you want to reuse only for 10 flights max anyways.

[6] Not worth it. However, though Methane can self pressurize, the F9R has shown it's not necessary to have, and helium does the job just fine.

IVF can be used if you really need infinite burns, for H2 only though. It offers a nice ~1T payload capacity boost though. I'm not saying you shoulf use it it was just a suggestion.

Also,

[7] Implies that each stage added increases factory costs by 2x.

A 2-stage vehicle uses 2 rockets.

I'm only adding another rocket. And ULA doesn't seem to think it's a huge deal, they need to reduce prices as fast as they can to compete vs SpaceX on military launches (bribes, and relations to Lockheed and Boeing will only get you so far), and they're still cool about adding boosters to optimize payload.

H-III wants to reduce costs to go into the commerical market, but is still cool about using boosters to optimize payload. https://en.wikipedia.org/wiki/H3_Launch_Vehicle

Hell, Ariane 6 is expected to cost as much as today's F9 per kg to GTO (and possibly better than FH, depending on how well reuse goes), and it still uses boosters to optimize payload. http://spacenews.com/ariane-6-rocket-designers-say-theyll-match-or-beat-todays-spacex-prices-on-per-kilogram-basis/

[8] Even SpaceX wanted to do it with LRBs when F5 was still a thing. http://www.spacelaunchreport.com/falcon9.html

Adding all that together, it can't cost that much more. Especially when you use the same engines and tankage diameter tooling as the core.

And in the launch scales where reuse is better than mass production, you will want to optimize payload, at least a bit.

[9] It's better to lower R+D costs to build it based off H-II tech, but with modifications to the engine and processes (like Horizontal vs vertical integration) to save costs, then chase after a 10s isp increase and increase R+D costs (and thus cost per launch) (and coking isn't an enormous problem for the Merlin anyways apparently due to using an O2 rich cycle, which also is the most efficient rocket cycle.)

I guess we have different viewpoints.

[1] M-II Heavy isn't going to be used for a Moon mission, multiple launches of M-III Heavy (~60 tonnes to LEO expendable) will. There are a few reasons why Falcon 9 won't have a four core variant: it's more complex to reuse, and the second stage would be underpowered for launching heavy payloads anyway (diminishing returns). They will not create an extra variants of the second stage (like Angara), and neither will I.

[2] Falcon 9 is thought to have a capacity of a few tonnes greater than the advertised capacity of 13 tonnes to LEO (something like 16-17 tonnes). Dragon has a mass of 4 tonnes, and Dragon V2 might be the same. If that is just the mass of the capsule, the trunk can't weigh much more than that.

[3] I don't know, Delta II and Falcon are both painted (I know, they don't have liquid methane). Let's see if BFR is painted.

[4] Well, M-III will have the same kind of engine as Raptor, so it will have a better increase in efficiency.

[5] Exactly why the rocket is all-methane.

[6] And yet they're switching to all-methane anyway. Helium is an extra fluid to deal with.

[7] At most, they might only be competitive with F9 as it is today.

[8] and they abandoned that idea.

[9] It won't be a 10 s Isp increase. SpaceX is developing Raptor anyway, despite the R&D costs.

6 hours ago, YumonStudios said:

Sorry. :P

I can't help but do it. You're not the only one to have this problem with me.

Ok, but isn't that what a backup docking port is for? The only reason I can see the 3rd one being needed is in case to abandon ship in an emergency, but in that case, it's definitely not ideal, but still excusable.

Ah. It's not a huge loss then, the orginal ISS design had the coupula like that too :)

It's still kind of stupid to put it there, but I guess since you already made the station, there's not much you can do about it now. I think.

Well, that's better than what Mir had to work with. :)

[10] ...And you can't bring it up with the module?

[11] Either way, the modules also lack a propulsion system in the video. How did they get up there in the first place? :confused:

[12] Looking at the videos above again, that hole between the adaptor and the RCV is pretty darn skinny. Dang, I guess this station has a pretty strict waist diameter and cup size limit :) (seriously though, it couldn't be designed wider? What if someone gets stuck?)

Thanks :)

[13] And this http://www.silverbirdastronautics.com/LVperform.html

is a good tool to use for your rockets.

Thanks.

[14] Oh. The IRL Mu-V stopped being used due to high costs, but I guess since it could be used as an ICBM, it could benefit from Mass production and make the Epsilon unnecessary...

Ok, I'll stop.

 

[15] I'm actually a fan of this project, if I didn't care, I wouldn't be bothering to reply to this thread. :P

It's pretty impressive.

[10] If the adapters were launched with the core module, it wouldn't fit in the fairing.

[11] Like this. And if you're going to say something about the detachment process being too risky, I just really needed to have RCS thrusters at the front in order for it to not generate too much torque when making translational maneuvers.

[12] About the same size as hatches on Soyuz.

[13] I already know about that.

[14] But is Japan really willing to make a nuclear weapon? Even it they did, it would probably be an unpopular decision with the public.

[15] Thank you.

 

Edited by Pipcard
Link to comment
Share on other sites

On 4/10/2016 at 8:03 AM, Pipcard said:

[10] If the adapters were launched with the core module, it wouldn't fit in the fairing.

Why not? If you have a robotic arm, you can just place it on the top of the module  (then redock it to the proper docking port) or do something like what Skylab did to deploy the ATM.

 

On 4/10/2016 at 8:03 AM, Pipcard said:

 

[11] Like this. And if you're going to say something about the detachment process being too risky, I just really needed to have RCS thrusters at the front in order for it to not generate too much torque when making translational maneuvers.

Why not just dock two together, on both sides? :P

But that's actually a decent idea- unfortunately, it can't be extended to normal space tug use (LEO to GEO) without significant modifications. (or can you add a Xenon motor and tank without significant mods?

After the station construction missions, the space repair tug would be used in the same way as OrbitalATK's Vivisat project http://www.vivisat.com/

and be able to offer commercial services.

I'm assuming the space tug is reusable and refuelable.

 

The space "tug" is honestly too good to waste. We need to get more uses for it and not let it die! :D

Another idea is a reatachable mini space station (like the https://en.wikipedia.org/wiki/Columbus_Man-Tended_Free_Flyer )

which would periodically dock (not berth) so that it can do ultra-sensitive experiments.

 

On 4/10/2016 at 8:03 AM, Pipcard said:

[12] About the same size as hatches on Soyuz.

Didn't know that, it just looks so tiny :P

I guess the spaceplane is just bigger then.

On 4/10/2016 at 8:03 AM, Pipcard said:

 

[14] But is Japan really willing to make a nuclear weapon? Even it they did, it would probably be an unpopular decision with the public.

This is not Japan, it's Hatsunia! Anything is possible, you said it yourself! :)

So it's your choice to decide.

On 4/10/2016 at 8:03 AM, Pipcard said:

 

[9] It won't be a 10 s Isp increase. SpaceX is developing Raptor anyway, despite the R&D costs.

STOP COMPARING APPLES TO ORANGES!

Did I get you attention? Good.

 

Let me simplify this:

Merlin uses Gas-generator cycles to power its engine.

https://en.wikipedia.org/wiki/Gas-generator_cycle

Quote

The main disadvantage is lost efficiency due to discarded propellant. Gas-generator cycles tend to have lower specific impulse than staged combustion cycles.

On the other hand, Raptor is a Staged Combustion engine, which is far more efficient.

https://en.wikipedia.org/wiki/Staged_combustion_cycle#Raptor

Quote

The Raptor engine currently under development by SpaceX is a full-flow staged combustion engine that will be powered by liquid methane and liquid oxygen,[5][6] a departure from the 'open cycle' gas generator system and LOX/kerosene propellants used by SpaceX's current Merlin engines.[6]

 

 

I am using an Apples to Apples comparison when I compared Max. ISP, which shows only a ~10s increase. Keep in mind the larger tanks eats away a lot of even that performance increase.

 

Even comparing Russian Staged combustion engines like https://en.wikipedia.org/wiki/RD-120

and https://en.wikipedia.org/wiki/RD-191

you get a ISP increase of 20-26s. Not a huge deal, you might say.

 

Until you consider that the RD-170, 180, 190, 120 line are old rocket engines from the 80s, while Raptor is a modern engine.

 

Worse, the performance of a efficient H2Lox engine (465s isp) like https://en.wikipedia.org/wiki/RL60

is leaps and bounds above the 380s isp number for a vaccum-optimized Raptor.

 

Since a reusable rocket like this will (if anything like SpaceX) be using the upper stage to get the majority of the Delta V... a super-cooled H2 engine is FAR better. If you don't plan to reuse the 2nd stage, which you don't. (but even then, it's possible to bypass the problems with H2 reuse)

 

On 4/10/2016 at 8:03 AM, Pipcard said:

[8] and they abandoned that idea.

Because they realized the initial payload offering for the F5 was not big enough of a market.

On 4/10/2016 at 8:03 AM, Pipcard said:

[3] I don't know, Delta II and Falcon are both painted (I know, they don't have liquid methane). Let's see if BFR is painted.

:mad:

Really?

RP-1 rockets don't need insulation due to being only semi-cryogenic (unless the Lox is supercooled, which needs insulation if you don't want to miss launch windows constantly due to it making every launch window a instantaneous launch window (the heating in the Lox tank is a huge problem otherwise, especially on high-performance missions).

That kind of thing can delay missions for weeks, and since time is money, I think Elon will eventually insulate the F9 core, even if it makes it look generic.

 

He may not though, public relations is a big part of his company and why it is successful, and a generic-looking rocket kind of kills that.

 

Methane rockets need insulation, though, you can't get away with that. It's even colder than slush-supercooled Lox https://en.wikipedia.org/wiki/Methane

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

 

I don't know if there is turqouise rocket insulation, all of it is some shade of orange....

On 4/10/2016 at 8:03 AM, Pipcard said:

[4] Well, M-III will have the same kind of engine as Raptor, so it will have a better increase in efficiency.

And if the M-II uses stage combustion Rp-1, that performance increase is only ~20s isp. Worse once you account for the much larger tanks needed, as that means a higher tank mass.

http://www.answers.com/mobile/Q/What_is_a_density_of_solid_methane

http://encyclopedia2.thefreedictionary.com/Solid+oxygen

On 4/10/2016 at 8:03 AM, Pipcard said:

[5] Exactly why the rocket is all-methane.

The M-III is planning on 2nd stage reuse? If so, brace yourself for even WORSE payload fraction. F9 is around 4-5T less payload for the 2nd stage reuse, and that's for LEO missions only. GTO is even worse, and don't even think about GEO. (yes, you can aerobrake, but all that fuel needs to be carried for longer in terms of Delta V)

I'd only bother with reusing the 2nd stage engines via SMART reuse, until the reuse turns out to be cheap enough to justify making an even bigger rocket to reuse the 2nd stage.

And even then, reusing H2 Lox is not impossible. In fact, we have arguably, the most experience with it. Remember the DC-X? New Shepard? The reusable SSMEs?

Hell, the former 2 achieved fast reuse. Yes, a 2nd stage is under worse conditions, but no reason to believe that engine reuse on the 2nd stage won't be economical.

On 4/10/2016 at 8:03 AM, Pipcard said:

[6] And yet they're switching to all-methane anyway. Helium is an extra fluid to deal with.

No, they aren't. I would like a Source. The only one we now of that they are use Ch4 on is MCT, and maybe the upper stage of F9 (which is probably only because they have so much research into Ch4 and none into H2. )

Also, I had a brainfart, you can't self-pressurise and have infinite burns. You can with the CH4, but the Lox will not self-pressurise, so no infinite burns :( .

And don't even think about putting the CH4 in the Lox tank to pressurize it. You know what will happen, hopefully. :)

On 4/10/2016 at 8:03 AM, Pipcard said:

[7] At most, they might only be competitive with F9 as it is today.

"At most".

Considering ArianeSpace has made huge design changes to hit current SpaceX prices (and that they will certainly have a mid-life evolution, like Ariane V...

I wouldn't cast them out yet. Also, I calculated elsewhere that the cost savings from reuse is only around 15%, if SpaceX hits their stated cost goals, accounting for the added cost of a bigger rocket. I can give you may calculations, if you really want to, but then I'll have to dig very deep...

 

On 4/10/2016 at 8:03 AM, Pipcard said:

[1] M-II Heavy isn't going to be used for a Moon mission, multiple launches of M-III Heavy (~60 tonnes to LEO expendable) will. There are a few reasons why Falcon 9 won't have a four core variant: it's more complex to reuse, and the second stage would be underpowered for launching heavy payloads anyway (diminishing returns). They will not create an extra variants of the second stage (like Angara), and neither will I.

The Angara solution (minus the numerous upper stage) is honestly pretty smart.

Ok, I'll make an adjustment.

The basic core + upper stage takes 8T to LEO under propulsive Barge landings. 5T or so under RTLS.

Adding one core (asymmetric) increases payload to ~15T to LEO under barge landings. 13T or so under RTLS.

Adding another core (symmetric) increases payload to ~22T to LEO under barge landings, and 20T or so with RTLS. ~30T to LEO or so in expendable mode, which is plenty for space station modules.

 

Only the core for 1st version, and boosters for the others, are reused, at first. If doing booster + core reuse + high-speed barge landgings for version 1 and 2, use the RLTS payload numbers.

For all three, a STAR motor is a option to do direct to GEO or MEO transfers (like what was done on the Delta II)

 

Also, a underpowered 2nd stage is not a problem for BLEO missions. Hell, SLS has a underpowered 2nd stage powered by puny RL-10s.

https://en.wikipedia.org/wiki/Merlin_%28rocket_engine_family%29#Merlin_1D

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

Quote

In June 2015 Tom Mueller answered a question about the Merlin 1D thrust/weight ratios on Quora. He specified that the Merlin 1D has a weight of 1,030 lb (470 kg) including thrust actuators, a current vacuum thrust of 162,500 pounds-force (723 kN), and an uprated vacuum thrust of 185,500 pounds-force (825 kN), which still weighs the same.

Quote

The Exploration Upper Stage (EUS) is being developed as a large second stage for Block 1B and Block 2 of the Space Launch System (SLS), replacing Block 1's Interim Cryogenic Propulsion Stage. It will be powered by four RL10 engines burning LOX/LH2 to produce a total of 440 kN (99,000 lbf) thrust.

Note that the F9 upper stage only carries 88T of propellant.

The EUS carries 129T of propellant.

In other words, the F9 2nd stage uses a engine with a thrust of almost 2x of the EUS, despite the EUS being ~1.5X bigger.

As a result, a 2-launch lunar mission for SLS is planned to use LOR instead of EOR to allow for the use of such a low TWR stage.

 

The lack of a 4-core F9 is more due to sheer lack of demand for an even bigger rocket than F9FT.

 

However, a Lunar version of my proposed M-III with 4 cores as boosters could carry 55T to LEO, according to delta_iv_heavy_lift.jpg

an expendable version would probably be more like ~65T to LEO, and a potential 6-booster core version could carry even more, if needed.

The RL-60 (aka MB-60 has 266kN thrust. 3 are needed to meet F9 2nd stage thrust). However, structural mods are needed for 6-booster configs, and should only be done when necessary.

https://www.nasaspaceflight.com/2013/08/dual-sls-required-nasas-lunar-landing-option/

3 launches of my M-III would be required, assuming an SLS-style lunar mission . Less than optimal, but not bad. Considering you could reduce mass even more if you used a Soyuz-style config with separate orbital and descent modules (this shaves off lots of CM mass), and also use a smaller SM to compensate for the smaller CM, plus a H2lox lunar lander with IVF, you might even get away with 2 launches, plus another launch of a smaller LV to carry only the crew to LEO to dock with the lunar transfer vehicle.

 

 

Link to comment
Share on other sites

16 hours ago, fredinno said:

[1] Why not? If you have a robotic arm, you can just place it on the top of the module  (then redock it to the proper docking port) or do something like what Skylab did to deploy the AT

[2] Why not just dock two together, on both sides? :P

[3] The space "tug" is honestly too good to waste. We need to get more uses for it and not let it die! :D

[4] This is not Japan, it's Hatsunia! Anything is possible, you said it yourself! :)

[5] STOP COMPARING APPLES TO ORANGES!

Did I get you attention? Good.

Let me simplify this:

Merlin uses Gas-generator cycles to power its engine.

https://en.wikipedia.org/wiki/Gas-generator_cycle

On the other hand, Raptor is a Staged Combustion engine, which is far more efficient.

https://en.wikipedia.org/wiki/Staged_combustion_cycle#Raptor

I am using an Apples to Apples comparison when I compared Max. ISP, which shows only a ~10s increase. [5a] Keep in mind the larger tanks eats away a lot of even that performance increase.

[6] Worse, the performance of a efficient H2Lox engine (465s isp) like https://en.wikipedia.org/wiki/RL60

is leaps and bounds above the 380s isp number for a vaccum-optimized Raptor.

Since a reusable rocket like this will (if anything like SpaceX) be using the upper stage to get the majority of the Delta V... a super-cooled H2 engine is FAR better. If you don't plan to reuse the 2nd stage, which you don't. (but even then, it's possible to bypass the problems with H2 reuse)

[7]

:mad:

Really?

RP-1 rockets don't need insulation due to being only semi-cryogenic (unless the Lox is supercooled, which needs insulation if you don't want to miss launch windows constantly due to it making every launch window a instantaneous launch window (the heating in the Lox tank is a huge problem otherwise, especially on high-performance missions).

That kind of thing can delay missions for weeks, and since time is money, I think Elon will eventually insulate the F9 core, even if it makes it look generic.

He may not though, public relations is a big part of his company and why it is successful, and a generic-looking rocket kind of kills that.

Methane rockets need insulation, though, you can't get away with that. It's even colder than slush-supercooled Lox https://en.wikipedia.org/wiki/Methane

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

I don't know if there is turqouise rocket insulation, all of it is some shade of orange....

And if the M-II uses stage combustion Rp-1, that performance increase is only ~20s isp. Worse once you account for the much larger tanks needed, as that means a higher tank mass.

http://www.answers.com/mobile/Q/What_is_a_density_of_solid_methane

http://encyclopedia2.thefreedictionary.com/Solid+oxygen

[8] The M-III is planning on 2nd stage reuse? If so, brace yourself for even WORSE payload fraction. F9 is around 4-5T less payload for the 2nd stage reuse, and that's for LEO missions only. GTO is even worse, and don't even think about GEO. (yes, you can aerobrake, but all that fuel needs to be carried for longer in terms of Delta V)

I'd only bother with reusing the 2nd stage engines via SMART reuse, until the reuse turns out to be cheap enough to justify making an even bigger rocket to reuse the 2nd stage.

And even then, reusing H2 Lox is not impossible. In fact, we have arguably, the most experience with it. Remember the DC-X? New Shepard? The reusable SSMEs?

Hell, the former 2 achieved fast reuse. Yes, a 2nd stage is under worse conditions, but no reason to believe that engine reuse on the 2nd stage won't be economical.

[9] No, they aren't. I would like a Source. The only one we now of that they are use Ch4 on is MCT, and maybe the upper stage of F9 (which is probably only because they have so much research into Ch4 and none into H2. )

Also, I had a brainfart, you can't self-pressurise and have infinite burns. You can with the CH4, but the Lox will not self-pressurise, so no infinite burns :( .

And don't even think about putting the CH4 in the Lox tank to pressurize it. You know what will happen, hopefully. :)

[10] "At most".

Considering ArianeSpace has made huge design changes to hit current SpaceX prices (and that they will certainly have a mid-life evolution, like Ariane V...

[11] I wouldn't cast them out yet. Also, I calculated elsewhere that the cost savings from reuse is only around 15%, if SpaceX hits their stated cost goals, accounting for the added cost of a bigger rocket. I can give you may calculations, if you really want to, but then I'll have to dig very deep...

[12] The lack of a 4-core F9 is more due to sheer lack of demand for an even bigger rocket than F9FT.

(Okay this is annoying, pressing ctrl+z here reverts everything, and I can't redo)

[1] During the start of Mirai's construction, there were no Canadarm-style robotic arms, and they couldn't have been launched with the module as they would not fit in the fairing. A Skylab-like mechanism is more mechanical complexity.

[2] Because that would introduce extra complexity in avionics and detachment operations.

[3] It can't be a reusable space tug because it is designed to carry modules with berthing ports and can't redock or refuel.

[4] But it is based on post-WWII Japanese society.

[5] Okay, so M-III's engines will be like those figurative "oranges." A significant improvement.

[5a] and you need even larger tanks for hydrogen

[6] If you read this thread on NASASpaceflight starting from here, you'd see how methalox or kerolox are better than hydrolox for "cost optimization," despite their lower Isp.

Quote

"Rockets with hydrogen upper stages are known for being expensive (Atlas, Delta, H-II, Ariane).  The low cost rockets (Falcon, Soyuz, Proton) do not use hydrogen in the upper stages." - LouScheffer

_________

"Higher efficiency means nothing if the entire system is not efficient.  For instance, the Shuttle [hydrolox] SSME's were very efficient, but on a $/kg to orbit basis the Shuttle system was pretty expensive.  And cost is the most important factor limiting us from doing more in space." - Coastal Ron

_________

" Look, it's perfectly clear hydrogen has more energy per kg than kerolox, and hence allows a lighter first stage for the same performance.   That's simple physics and not in dispute.  But hydrogen has drawbacks as well, and hence may not be the most economical choice.  It's not a good first stage fuel (not dense enough).  So now you need a two-fuel system.  This implies different engines for the different stages, more specialists on your launch team, and now your second stage engine is produced in low volume.  All of these can be solved, but it costs money.  On the whole, is the hydrolox upper stage cheaper?  Like all engineering, it's a question of tradeoffs...

Overall, there can be no credible claim for a hydrolox upper stage reducing cost without running the numbers.  And the empirical evidence runs the other way - the hydrolox upper stages belong to the high cost vendors.  Why do you suppose that is, if a hydrolox upper stage should lead to a low system cost?" - LouScheffer, again

_________

" Now sure, if development cost doesn't matter (so you build multiple, totally different first stage engine(s) from upper stage engine(s)) and you ignore operations cost, it's clearly better to do both. But if you include development costs and operations cost, it's not at all clear that doing both is optimal...


From an economics perspective where money is actually limited, I'd say a full-kerolox rocket is far superior to a full-hydrolox rocket. Maybe methane/LOx (or another simple high-Isp and high-density hydrocarbon like propylene or propane) would be even better, I'm not sure." - Robotbeat

_________

" Not sure why you think having multiple engine designs is a good thing.  Sure it may wring out the last percentage of "efficiency", but the #1 goal should be cost, which is the efficiency of the entire system.  And multiple engine designs, while maybe individually more efficient, are a drag on overall costs compared to a single engine type system like Falcon Heavy. " - Coastal Ron, again

_________

" Falcon 9 and Heavy are basically just variations of one rocket engine using a single propellant combo and a single stage type. That incredibly streamlines manufacturing, testing, and ground support equipment.

Add another propellant combination, especially hydrolox, and you need a new type of rocket stage with different manufacturing considerations (significantly different temperatures changes what the optimal materials are, hydrogen embrittlement becomes a concern, insulation becomes very important whether foam or MLI, etc), a totally new engine that needs to be tested from scratch, new ground support equipment, different training, for hydrogen you have to be really careful about leaks and even condensing out oxygen from the air onto your pipes and stuff, etc.

Basically, you have double as much equipment. Maybe you can get double the payload to GTO for the same lift-off mass, but you might be just better off with another stage of the same propellant combo and same engine and stage type, etc... Basically, Falcon Heavy. Which also has the bonus of getting MUCH more payload to LEO. " - Robotbeat, again

[7] Let us wait and see. If SpaceX unveils the BFR with unpainted insulation, I will change it. A speculative render of the BFR made by a NASASpaceflight member seems to show it painted. Methane and LOX also have similar boiling points (110 K and 90 K, respectively, while hydrogen has a boiling point of 20 K)

[8] M-III won't have a reusable second stage, but an M-IV probably would.

[9] It is implied that with BFR/MCT, they will transition to a single all-methane system in the future for simplicity reasons.

[10] hit "current" prices.

[11] Yeah, -you- calculated. SpaceX may have different factors.

[12] And yet they are planning a BFR anyway. The simple reason why M-III will not have a five-core/four-booster variant is because it makes recovery more complicated, which is the same reason why BFR will have a "single monster boost stage." So Falcon Heavy's three cores/two boosters are the upper limit for that.

Edited by Pipcard
Link to comment
Share on other sites

16 hours ago, Pipcard said:

(Okay this is annoying, pressing ctrl+z here reverts everything, and I can't redo)

XD

I know the feeling. Copy and paste your work onto word so you don't loose it. That's what I do, and I learned it the hard way...

16 hours ago, Pipcard said:

[7] Let us wait and see. If SpaceX unveils the BFR with unpainted insulation, I will change it. A speculative render of the BFR made by a NASASpaceflight member seems to show it painted.

So was Vulcan, until later, when it became unpainted. In fact, the BFR uses super-cooled Ch4, meaning the insulation is even more essential.

Practically everyone in the spaceflight community knew the Vulcan, SLS, and other rockets were going to eventually be a shade of orange due to insulation- they never initially show it with insulation due to the fact that it's bad for publicity.

 

However, if you want, you can get away with near-colourless unpainted insulation. Ariane 5 did it, and he texture looks pretty badass. ariane10.jpg

16 hours ago, Pipcard said:

[12] And yet they are planning a BFR anyway. The simple reason why M-III will not have a five-core/four-booster variant is because it makes recovery more complicated, which is the same reason why BFR will have a "single monster boost stage." So Falcon Heavy's three cores/two boosters are the upper limit for that.

...and there is speculation the BFR is 3-cored. But the real reason behind the BFR is that it's planned to carry "100T to the surface of Mars".

No F9 variant can do that, and in any case, it's not optimal due to a tiny payload fairing.

It makes recovery more complicated, but you would NEVER recover boosters on a rocket launching maybe 3 times a year (my M-III lunar version). You need as much performance as possible.

Please, stop comparing apples to oranges.

16 hours ago, Pipcard said:

[11] Yeah, -you- calculated. SpaceX may have different factors.

Not just me. '60s aerospace designers did too. Only NewSpace companies seem to care anything about CH4, and even them, Blue Origin is planning on a H2 upper stage for its reusable rocket, but a Ch4 lower stage.

 

http://space.stackexchange.com/questions/12170/why-did-it-take-so-long-for-methane-to-be-used-as-a-rocket-propellant

Quote

"Oberth had originally wanted to use methane as fuel, but as it was hard to come by in Berlin, their first work was with gasoline and oxygen. Johannes Winkler, however, picked up the idea, and working independently of the VfR, was able to fire a liquid oxygen-liquid methane motor before the end of 1930. This work led nowhere in particular, since, as methane has a performance only slightly superior to that of gasoline, and is much harder to handle, nobody could see any point to following it up."

http://space.stackexchange.com/questions/3161/why-is-spacex-considering-methane-as-fuel-for-their-next-engine-the-raptor

Looking at this link, the main reason that SpaceX would want a Ch4 upper stage is to use it for their Mars Missions. Mars is great with CH4, ISRU and all.

It's not useful otherwise, even for reusable Space tugs, as those can use Xenon/Argon.

 

The link states coking is an issue for Rp-1, but that isn't a major problem for the Merlin due to a Oxygen-rich burn (which is also more hot and efficient). Granted, you only really want to reuse a few times, as the launch rates don't justify 10> reuses. And SpaceX is fine with it on the F9R, otherwise they would have switched to CH4 on that rocket. http://spaceflight101.com/spacex-launch-vehicle-concepts-designs/

16 hours ago, Pipcard said:

[8] M-III won't have a reusable second stage, but an M-IV probably would.

Then that leaves plenty of time to work out the reuse kinks on a H2 rocket. After all, Blue Origin is using H2 on a reusable rocket, about the same size as a 2nd stage. And we have plenty of experience reusing H2 liquid engines.

16 hours ago, Pipcard said:

 

[9] It is implied that with BFR/MCT, they will transition to a single all-methane system in the future for simplicity reasons.

But is the F9/F9H transitioning? Those are the commercial rockets, and are the important ones, since those will be reused the most.

BTW, I addressed the Ch4 BFR reason above.

 

Lastly, I used to think SpaceX wanted to use a CH4 F9 earlier. Nope, apparently that's not happening.

Don't fall to the same trap I did.

16 hours ago, Pipcard said:
 

"Rockets with hydrogen upper stages are known for being expensive (Atlas, Delta, H-II, Ariane).  The low cost rockets (Falcon, Soyuz, Proton) do not use hydrogen in the upper stages." - LouScheffer

Ariane V is EXPENSIVE? Give me a break, they were the top dogs of the GEO sat launching industry before SpaceX came along.

Yes, it is more expensive, but the cost per kg is lower. 100% Solid rockets are actually more expensive despite lower motor cost for the same reason.

16 hours ago, Pipcard said:

 

"Higher efficiency means nothing if the entire system is not efficient.  For instance, the Shuttle [hydrolox] SSME's were very efficient, but on a $/kg to orbit basis the Shuttle system was pretty expensive.  And cost is the most important factor limiting us from doing more in space." - Coastal Ron

You don't need a hyper-efficient H2 engine. You can get away with a pump-fed H2 engine, and still get leaps and bounds of ISP over CH4.

16 hours ago, Pipcard said:

 

" Look, it's perfectly clear hydrogen has more energy per kg than kerolox, and hence allows a lighter first stage for the same performance.   That's simple physics and not in dispute.  But hydrogen has drawbacks as well, and hence may not be the most economical choice.  It's not a good first stage fuel (not dense enough).  So now you need a two-fuel system.  This implies different engines for the different stages, more specialists on your launch team, and now your second stage engine is produced in low volume.  All of these can be solved, but it costs money.  On the whole, is the hydrolox upper stage cheaper?  Like all engineering, it's a question of tradeoffs...

Overall, there can be no credible claim for a hydrolox upper stage reducing cost without running the numbers.  And the empirical evidence runs the other way - the hydrolox upper stages belong to the high cost vendors.  Why do you suppose that is, if a hydrolox upper stage should lead to a low system cost?" - LouScheffer, again

...Then why not use a 100% RP-1 rocket if you want to use that argument? The ISP difference is not very high, and a cheap 3rd stage (like a STAR motor http://www.astronautix.com/stages/pamd.htm)  (4.1 Million dollars, allow for expansion of rocket motor to a 10 Million dollar cost, and to 100 inches diameter, AKA the STAR 100) could increase payload to GEO if you really need it.

16 hours ago, Pipcard said:

" Now sure, if development cost doesn't matter (so you build multiple, totally different first stage engine(s) from upper stage engine(s)) and you ignore operations cost, it's clearly better to do both. But if you include development costs and operations cost, it's not at all clear that doing both is optimal...


From an economics perspective where money is actually limited, I'd say a full-kerolox rocket is far superior to a full-hydrolox rocket. Maybe methane/LOx (or another simple high-Isp and high-density hydrocarbon like propylene or propane) would be even better, I'm not sure." - Robotbeat

Development costs are much less of an issue if your company is supported and subsidized by the government (like ArianeSpace), which I'm assuming this one is. Ariane 6 has an enormous development cost- but it's not being transferred to ArianeSpace, as ESA is footing most of the bill.

16 hours ago, Pipcard said:

 

[1] During the start of Mirai's construction, there were no Canadarm-style robotic arms, and they couldn't have been launched with the module as they would not fit in the fairing. A Skylab-like mechanism is more mechanical complexity.

But you told Ian that the station had a Space station construction arm...

Either way, I see no reason why not. Mir didn't have one since its modules were docked, not berthed.

And Canadarm was first built by 1981. I doubt you would not be able to extend that to space station modules. After all, Space Station Freedom, a 1984 space station proposal that was continually revived until it became the ISS after merging with the Columbus Free-flyer and Mir-2, used a robotic arm to build the station.http://www.thelivingmoon.com/45jack_files/04images/Space_Station/ss86_fr1.jpg

Also, the IDA alone has a mass of 500 kg. A PMA is 1500 kg.

https://www.nasa.gov/externalflash/ISSRG/pdfs/pma.pdf

So you'd need 1.5T - 2T extra LV capacity, every launch. That's going to get expensive, and reduces margin considerably.

16 hours ago, Pipcard said:

[2] Because that would introduce extra complexity in avionics and detachment operations.

OK. I don't see how that's worse in complexity and safety than an extendable RCS boom, but assuming those can fold out (hopefully) it probably isn't that huge of a deal.

16 hours ago, Pipcard said:

[3] It can't be a reusable space tug because it is designed to carry modules with berthing ports and can't redock or refuel.

But you can design it to redock and refuel. In fact, I thought that was the point of these tugs.

And I never said it would be used as a space tug/repair station- I said it would WITH MODIFICATIONS. How many mods is a question, though.

16 hours ago, Pipcard said:

[4] But it is based on post-WWII Japanese society.

Ok, maybe it's a bad idea then.What's the payload capacity of Mu-V? If it's quite a bit bigger, then it won't be able to use the old SRBs of H-II. And in any case, it's not benefiting from mass-production anymore anyways.

 

Could you make the Mu-V to a Mu-VI, with the same dimensions, using higher ISP SRB propellant (if possible), a monolithic SRB as the 1st stage (saving money, as segmented SRBs are due to supplier locations), and a composite carbon motor instead of steel (like the GEM motors, or Black Knight), along with other mods to make it cheaper, like automated launch operations (like what Epsilon did)

http://www.globalsecurity.org/space/world/japan/m5.htm

16 hours ago, Pipcard said:

 

[5] Okay, so M-III's engines will be like those figurative "oranges." A significant improvement.

:huh:

I know you know what it means, stop trying to be funny please.

16 hours ago, Pipcard said:

[5a] and you need even larger tanks for hydrogen

Everything is relative. Hydrogen has a impressive ISP increase (almost 80s) over Methane, and 90-100s over RP-1. It's lower tank mass is more than made up for by the higher ISP- assuming use on an upper stage.

http://www.spacelaunchreport.com/vulcan.html

http://www.spacelaunchreport.com/delta4.html

Also, comparing Vulcan and Delta IV cores, it doesn't look like liquid hydrogen uses that much more volume than liquid methane, especially compared to the Atlas V's core of ~3.6 m.

 

I would like you to answer the question?

Why you would use CH4 as a common rocket propellant?

After everything I have told you.

Link to comment
Share on other sites

22 hours ago, fredinno said:

[1] So was Vulcan, until later, when it became unpainted. In fact, the BFR uses super-cooled Ch4, meaning the insulation is even more essential.

Practically everyone in the spaceflight community knew the Vulcan, SLS, and other rockets were going to eventually be a shade of orange due to insulation- they never initially show it with insulation due to the fact that it's bad for publicity.

However, if you want, you can get away with near-colourless unpainted insulation. Ariane 5 did it, and he texture looks pretty badass.

[2] ...and there is speculation the BFR is 3-cored. But the real reason behind the BFR is that it's planned to carry "100T to the surface of Mars".

No F9 variant can do that, and in any case, it's not optimal due to a tiny payload fairing.

[3] It makes recovery more complicated, but you would NEVER recover boosters on a rocket launching maybe 3 times a year (my M-III lunar version). You need as much performance as possible.

Please, stop comparing apples to oranges.

Not just me. '60s aerospace designers did too. Only NewSpace companies seem to care anything about CH4, and even them, Blue Origin is planning on a H2 upper stage for its reusable rocket, but a Ch4 lower stage.

http://space.stackexchange.com/questions/12170/why-did-it-take-so-long-for-methane-to-be-used-as-a-rocket-propellant

http://space.stackexchange.com/questions/3161/why-is-spacex-considering-methane-as-fuel-for-their-next-engine-the-raptor

[4] Looking at this link, the main reason that SpaceX would want a Ch4 upper stage is to use it for their Mars Missions. Mars is great with CH4, ISRU and all.

It's not useful otherwise, even for reusable Space tugs, as those can use Xenon/Argon.

The link states coking is an issue for Rp-1, but that isn't a major problem for the Merlin due to a Oxygen-rich burn (which is also more hot and efficient). Granted, you only really want to reuse a few times, as the launch rates don't justify 10> reuses. And SpaceX is fine with it on the F9R, otherwise they would have switched to CH4 on that rocket. http://spaceflight101.com/spacex-launch-vehicle-concepts-designs/

Then that leaves plenty of time to work out the reuse kinks on a H2 rocket. After all, Blue Origin is using H2 on a reusable rocket, about the same size as a 2nd stage. And we have plenty of experience reusing H2 liquid engines.

But is the F9/F9H transitioning? Those are the commercial rockets, and are the important ones, since those will be reused the most.

BTW, I addressed the Ch4 BFR reason above.

Lastly, I used to think SpaceX wanted to use a CH4 F9 earlier. Nope, apparently that's not happening.

Don't fall to the same trap I did.

[5] Ariane V is EXPENSIVE? Give me a break, they were the top dogs of the GEO sat launching industry before SpaceX came along.

Yes, it is more expensive, but the cost per kg is lower. 100% Solid rockets are actually more expensive despite lower motor cost for the same reason.

You don't need a hyper-efficient H2 engine. You can get away with a pump-fed H2 engine, and still get leaps and bounds of ISP over CH4.

[6] ...Then why not use a 100% RP-1 rocket if you want to use that argument? The ISP difference is not very high, and a cheap 3rd stage (like a STAR motor http://www.astronautix.com/stages/pamd.htm)  (4.1 Million dollars, allow for expansion of rocket motor to a 10 Million dollar cost, and to 100 inches diameter, AKA the STAR 100) could increase payload to GEO if you really need it.

Development costs are much less of an issue if your company is supported and subsidized by the government (like ArianeSpace), which I'm assuming this one is. Ariane 6 has an enormous development cost- but it's not being transferred to ArianeSpace, as ESA is footing most of the bill.

[7] But you told Ian that the station had a Space station construction arm...

Either way, I see no reason why not. Mir didn't have one since its modules were docked, not berthed.

And Canadarm was first built by 1981. I doubt you would not be able to extend that to space station modules. After all, Space Station Freedom, a 1984 space station proposal that was continually revived until it became the ISS after merging with the Columbus Free-flyer and Mir-2, used a robotic arm to build the station.http://www.thelivingmoon.com/45jack_files/04images/Space_Station/ss86_fr1.jpg

Also, the IDA alone has a mass of 500 kg. A PMA is 1500 kg.

https://www.nasa.gov/externalflash/ISSRG/pdfs/pma.pdf

[8] So you'd need 1.5T - 2T extra LV capacity, every launch. That's going to get expensive, and reduces margin considerably.

OK. I don't see how that's worse in complexity and safety than an extendable RCS boom, but assuming those can fold out (hopefully) it probably isn't that huge of a deal.

[9] But you can design it to redock and refuel. In fact, I thought that was the point of these tugs.

And I never said it would be used as a space tug/repair station- I said it would WITH MODIFICATIONS. How many mods is a question, though.

Ok, maybe it's a bad idea then.What's the payload capacity of Mu-V? If it's quite a bit bigger, then it won't be able to use the old SRBs of H-II. And in any case, it's not benefiting from mass-production anymore anyways.

[10] Could you make the Mu-V to a Mu-VI, with the same dimensions, using higher ISP SRB propellant (if possible), a monolithic SRB as the 1st stage (saving money, as segmented SRBs are due to supplier locations), and a composite carbon motor instead of steel (like the GEM motors, or Black Knight), along with other mods to make it cheaper, like automated launch operations (like what Epsilon did)

http://www.globalsecurity.org/space/world/japan/m5.htm

:huh:

[11] I know you know what it means, stop trying to be funny please.

Everything is relative. Hydrogen has a impressive ISP increase (almost 80s) over Methane, and 90-100s over RP-1. It's lower tank mass is more than made up for by the higher ISP- assuming use on an upper stage.

I would like you to answer the question?

[12] Why you would use CH4 as a common rocket propellant?

After everything I have told you.

[1] Actually, I'm thinking of doing a "frosted-over" effect in the texture. But I might not because it won't look like that when it is mostly empty of propellant (e.g. when landing). Although I've never downloaded the Real Solar System mod*, tanks with liquid oxygen don't seem to get the frosting effect either.

*because I'm waiting to get a better computer - KSP is already somewhat sluggish with a planet about 10 times smaller than Earth.

[2] Speculation that was denied when Musk said "At first, I was thinking we would just scale up Falcon Heavy, but it looks like it probably makes more sense just to have a single monster boost stage."

[3] It won't launch 3 times a year. In the future, there will be at least two manned lunar missions per year, with a reusable lunar lander going between the Earth-Moon L1 point and the lunar surface. Thus, each mission would need to bring up the RCV with extra habitation and propulsion modules (as the regular RCV can't last more than ~2 days by itself), Earth Departure Stages, as well as propellant tankers (propellant is relatively cheap) to refuel the lander. Also, it uses mostly-common hardware with the single-core variant of the M-III, so that isn't only being used for lunar missions.

[4] And Hatsunia will have manned Mars missions in the future as well.

[5] "before SpaceX came along." Now the hydrogen first+second stage approach won't be as competitive. The point that they're trying to say is that hydrogen will result in more expense for the same payload capacity. M-II would have been able to compete well during the 1990s and 2000s, but with the rise of a launch provider that follows a philosophy of "optimizing for total systems cost" instead of Isp, the next-generation rocket will have to adapt in order to stay competitive.

[6] Because even if you treat it as an "apples to apples" issue in terms of Isp, methane fuel doesn't leave as much residue in the engines, making for much better maintenance when reusing them. Methane is also being used instead of kerosene/RP-1 because it enables easier long-term storage in space (due to the oxidizer and fuel being close in temperature).

Don't forget about operational/recurring costs, and those tend to be higher with liquid hydrogen.

[7] It had to be brought up to the station on a separate mission that occurred after the launch of the core, using a HTV (UTV)-bus disposable tug.

[8] Doesn't matter if the rocket can still carry it. The total price of a launch is fixed, not determined by the mass of the spacecraft on each mission. The RCV adapter is also somewhat smaller and lighter than the PMA.

[9] It wasn't. It is similar to the disposable HTV propulsion and avionics bus.

[10] Negi-5 already had a monolithic SRB first stage, and is more like Epsilon although its first launch took place earlier, like Mu-V. And there will be no Negi-6 (although Hatsunia will have a different negi-themed vehicle in-universe, similar to this, but I'm not planning on making an Orbiter add-on for that any time soon), because the savings from reuse will allow M-III to supplant light-lift launch vehicles. An Epsilon launch: $38 million for 1.2 tonnes to LEO. SpaceX intends to sell F9 in reusable mode for a price of ~$40 million for ~10 tonnes to LEO (their advertised payload capacity has already taken reusability margins into account).

[11] Why do you keep wanting to make "apples to apples" comparisons to support your argument when SpaceX isn't doing that when considering Raptor development? They were considering hydrogen too, but realized that pad handling and engines would become more expensive due to hydrogen's deep cryogenic nature, so thus it was not worth the higher Isp.

[12] Didn't you read that whole array of quotes that say that it is better to have common rocket propellants than different fuel types in order to have optimization of recurring costs, and that hydrogen isn't a good common propellant because of low density and thrust in the first stage? Methane is the next best thing in terms of Isp, but with less of the "pain in the (butt)" factor (i.e. costly operations and engine manufacturing) that hydrolox has.

Here's another reason why: it makes reuse more effective at saving costs.

"The thing is, reusability the way SpaceX has been working on has been technologically possible for decades, but it requires the foresight and the daring to actually try it. More importantly, it requires building the rocket with reusability in mind from the outset. There are several aspects behind that. First, it's important that the majority of the cost of the vehicle be in the first stage, which is the easiest to reuse. This runs counter to conventional expendable rocket design optimization, which has resulted in a lineup of existing launchers where the upper stage is the majority of the cost (due to using light-weight alloys and materials and LOX/LH2 propellant, for example)." -rocketsocks

and here's another reason why M-III doesn't have a hydrogen upper stage:

"If you look at other rocket designs you see the influence of the siren song of "efficiency" or "advanced design" and so forth creeping in. One of the common historical routes to increasing rocket performance is to replace the upper stage with a more "advanced" one using higher ISP propellants (typically LOX/Hydrogen). This is possible because a LOX/Hydrogen stage will have a lower fueled mass than a LOX/Kerosene stage with better delta-V performance. But the end result of this is often that the upper stage becomes a much larger part of the overall cost of the rocket. So you've improved the performance of the launcher, but at the cost of inflating its price tag a great deal." - same person as above

Edited by Pipcard
Link to comment
Share on other sites

On 4/18/2016 at 3:46 PM, Pipcard said:

 

[11] Why do you keep wanting to make "apples to apples" comparisons to support your argument when SpaceX isn't doing that when considering Raptor development? They were considering hydrogen too, but realized that pad handling and engines would become more expensive due to hydrogen's deep cryogenic nature, so thus it was not worth the higher Isp.

Because "apples to apples" means that I am comparing the same thing to another, similar thing, with as few differing variables as possible.

I don't know why SpaceX chose a 100% methane rocket, for sure. No one does, and neither do you.

Also, Blue Origin is doing the same thing as SpaceX- except using 2 different propellants.

https://en.wikipedia.org/wiki/Blue_Origin#Orbital_launch_vehicle

 

It probably goes down to preference and how they did their calculations. Methane may only become viable in the insane launch rates SpaceX predicts.

On 4/18/2016 at 3:46 PM, Pipcard said:

[9] It wasn't. It is similar to the disposable HTV propulsion and avionics bus.

Oh, ok...

But if you can't reuse, or use it elsewhere, what's the point to a more complex, detachable SM bus? Why not integrate it into the spacecraft, like the Russians?

On 4/18/2016 at 3:46 PM, Pipcard said:

[6] Because even if you treat it as an "apples to apples" issue in terms of Isp, methane fuel doesn't leave as much residue in the engines, making for much better maintenance when reusing them. Methane is also being used instead of kerosene/RP-1 because it enables easier long-term storage in space (due to the oxidizer and fuel being close in temperature).

So? Long-term in space only matters, again, going to Mars.

MCT revolves around Mars.

 

The H-III and H-III Lunar rockets are NOT.

STOP MAKING ME REPEAT MYSELF.

Please, stop. Start reading my previous posts, please, and stop repeating the same points. I'm tired of it. I've addressed all of them.

I feel like I'm talking to a brick wall. Please, for my sake AND yours.

On 4/18/2016 at 3:46 PM, Pipcard said:

[10] Negi-5 already had a monolithic SRB first stage, and is more like Epsilon although its first launch took place earlier, like Mu-V. And there will be no Negi-6 (although Hatsunia will have a different negi-themed vehicle in-universe, similar to this, but I'm not planning on making an Orbiter add-on for that any time soon), because the savings from reuse will allow M-III to supplant light-lift launch vehicles. An Epsilon launch: $38 million for 1.2 tonnes to LEO. SpaceX intends to sell F9 in reusable mode for a price of ~$40 million for ~10 tonnes to LEO (their advertised payload capacity has already taken reusability margins into account).

... but a cheap smallsat launcher is cheaper than even the F9. And it's not practical to launch IXV on F9.

https://en.wikipedia.org/wiki/Vega_%28rocket%29

Quote

Developments costs for the Vega rocket were €710 million, with ESA spending an additional €400 million to sponsor five development flights between 2012 and 2014.[72] Commercial launch cost have been estimated at €32 million including Arianespace's marketing and service costs or €25 million for a rocket alone, assuming launch rate of 2 per year. By increasing flight rate up to 4 per year price of an each individual launch vehicle will drop to €22 million.[73]

Keep in mind, the average future launch rate for VEGA is 3-4, so I'm going to use the 22 Mil. Euros number, or 25 Mil. Dollars US.

Besides, the Negi-6 is pretty much a Negi-5A, so it's only textural changes and performance adjustments needed.

On 4/18/2016 at 3:46 PM, Pipcard said:

[12] Didn't you read that whole array of quotes that say that it is better to have common rocket propellants than different fuel types in order to have optimization of recurring costs, and that hydrogen isn't a good common propellant because of low density and thrust in the first stage? Methane is the next best thing in terms of Isp, but with less of the "pain in the (butt)" factor (i.e. costly operations and engine manufacturing) that hydrolox has.

Here's another reason why: it makes reuse more effective at saving costs.

"The thing is, reusability the way SpaceX has been working on has been technologically possible for decades, but it requires the foresight and the daring to actually try it. More importantly, it requires building the rocket with reusability in mind from the outset. There are several aspects behind that. First, it's important that the majority of the cost of the vehicle be in the first stage, which is the easiest to reuse. This runs counter to conventional expendable rocket design optimization, which has resulted in a lineup of existing launchers where the upper stage is the majority of the cost (due to using light-weight alloys and materials and LOX/LH2 propellant, for example)." -rocketsocks

and here's another reason why M-III doesn't have a hydrogen upper stage:

"If you look at other rocket designs you see the influence of the siren song of "efficiency" or "advanced design" and so forth creeping in. One of the common historical routes to increasing rocket performance is to replace the upper stage with a more "advanced" one using higher ISP propellants (typically LOX/Hydrogen). This is possible because a LOX/Hydrogen stage will have a lower fueled mass than a LOX/Kerosene stage with better delta-V performance. But the end result of this is often that the upper stage becomes a much larger part of the overall cost of the rocket. So you've improved the performance of the launcher, but at the cost of inflating its price tag a great deal." - same person as above

Did you see what I addressed them with? I answered all of them.

YOU are ignoring what I've been saying.


Also, I was recommending 2 designs: One with Rp-1 common propellant, and one with Lh2/RP-1 propellant.

If you are against using H2, then use RP-1. I (and Blue Origin, ArianeSpace, and almost everyone else in the rocket industry except SpaceX and the Russians) prefer 2 types of propellant in their rockets, as shown by their rocket designs, so yes.

This discussion started, as I was against the use of Ch4 due to only marginal higher ISP and much lower density and higher R+D costs.
Keep that in mind. Don't attack hydrogen specifically, I'm attacking Methane.
 

Also, I love how you cited a Reddit user. Such a great source :)

I'll let it slide though.

 

PS:Let's compare 2 rockets, one with common propellants (RP-1), and one with 2 types of propellant (solid and H2). Both are expendable LV made by government- controlled/sponsored companies, and both are under development, designed to be cheaper to compete in the commercial market. (Apples to Apples)

http://www.russianspaceweb.com/angara5.html

Quote

In 2015, a total cost of the Angara-5 development was estimated at 150 billion rubles (744) and around the same time, each launch was expected to cost from $95 to $105 million.

Quote
Angara-5 with Block DM upper stage
-
-
Geostationary transfer orbit (GTO)
6.5 tons
7.0 tons

https://en.wikipedia.org/wiki/Ariane_6

Quote
Cost per launch € 90 million (Ariane 64), 75 million (Ariane 62)[2] (2014 est.)
Quote
Payload to GTO A62: 5,000 kg (11,000 lb)
A64: 11,000 kg (24,000 lb) or 10,000 kg (22,000 lb) with dual payload

:)

Checkmate.

On 4/18/2016 at 3:46 PM, Pipcard said:

[8] Doesn't matter if the rocket can still carry it. The total price of a launch is fixed, not determined by the mass of the spacecraft on each mission. The RCV adapter is also somewhat smaller and lighter than the PMA.

But you said:

Quote

16. About 10-11 tonnes.

https://en.wikipedia.org/wiki/H-IIA

The latter number needs an entirely new rocket variant for H-II (and for my H-III as well)

And engineers will always choose the larger number when given a mass LV decision. Better to have too much LV capacity than too little.

 

Quote

[1] Actually, I'm thinking of doing a "frosted-over" effect in the texture. But I might not because it won't look like that when it is mostly empty of propellant (e.g. when landing). Although I've never downloaded the Real Solar System mod*, tanks with liquid oxygen don't seem to get the frosting effect either.

Tanks with lox do get frosting over, at least enough to prevent soot from covering the area of the lox tank.

But you still need an insulated tank. Also, wait for KSP 1.1, it is supposed to have better performance.

 

On 4/18/2016 at 3:46 PM, Pipcard said:

OK, makes sense. Bigger fairings = Better when it comes to interplanetary landers.

On 4/18/2016 at 3:46 PM, Pipcard said:

 

[3] It won't launch 3 times a year. In the future, there will be at least two manned lunar missions per year, with a reusable lunar lander going between the Earth-Moon L1 point and the lunar surface. Thus, each mission would need to bring up the RCV with extra habitation and propulsion modules (as the regular RCV can't last more than ~2 days by itself), Earth Departure Stages, as well as propellant tankers (propellant is relatively cheap) to refuel the lander. Also, it uses mostly-common hardware with the single-core variant of the M-III, so that isn't only being used for lunar missions.

That's what I meant by launching 3 times a year.

Also, RCV? Couldn't you develop a Capsule? I mean, the mods you added (and others needed, like a buffed up heat shield, and a proper SM to leave Lunar Orbit- that is the min. amount of hypergol propellant needed in case something bad happens, you can always escape with a reliable hyoergol engine no matter what) to the vehicle seems like it would be better to build a new system.)

I can modify my car to become a plane. But it'd be cheaper, safer, and better to buy a personal plane instead.

On 4/18/2016 at 3:46 PM, Pipcard said:

[4] And Hatsunia will have manned Mars missions in the future as well.

But it's not designed for Mars, is it (like MCT)? It's designed for commercial use, which even SpaceX prefers Rp-1 with.

On 4/18/2016 at 3:46 PM, Pipcard said:

[5] "before SpaceX came along." Now the hydrogen first+second stage approach won't be as competitive. The point that they're trying to say is that hydrogen will result in more expense for the same payload capacity. M-II would have been able to compete well during the 1990s and 2000s, but with the rise of a launch provider that follows a philosophy of "optimizing for total systems cost" instead of Isp, the next-generation rocket will have to adapt in order to stay competitive.

Yeah, and ArianeSpace is pretty darned sure it's going to compete with SpaceX using 2 propellants types, including H2, and 0 reuse.

Also, Ariane 6 does a lot of things SpaceX didn't do, due to high R+D Costs (H-III and Ariane 6 have an advantage here in that the R+D cost will be generously subsidized, making those costs much less relevant). For example, single engines on both stages, instead of 9 engines on the 1st stage is simpler and more cheap (COST PER LAUNCH) (Developing engines cost an arm and a leg- leading to the Atlas V RD-180 decision that bit Lockheed/ULA in the poophole). Reuse means that there will be no cost savings in mass production, thus there is little reason to cluster engines (except the aforementioned development costs)

 

 

 

 

Tl;DR, My recommendations:

1. If you are against using H2, then use RP-1. I (and Blue Origin, ArianeSpace, and almost everyone else in the rocket industry except SpaceX and the Russians) prefer 2 types of propellant in their rockets, as shown by their rocket designs, so yes.

This discussion started, as I was against the use of Ch4 due to only marginal higher ISP and much lower density and higher R+D costs.
Keep that in mind. Don't attack hydrogen specifically, I'm attacking Methane.

2. Could you make the Mu-V to a Mu-VI, with the same dimensions, using higher ISP SRB propellant (if possible), a monolithic SRB as the 1st stage (like Mu-V), and a composite carbon motor instead of steel (like the GEM motors, or Black Knight), along with other mods to make it cheaper, like automated launch operations (like what Epsilon did).

This is mainly a textural and programming adjustment.

 

3. My recommendation for the rocket design (applicable to H2 and Rp-1 common propellant solution):

The basic core + upper stage takes 8T to LEO under propulsive Barge landings. 5T or so under RTLS.

Adding one core (asymmetric) increases payload to ~15T to LEO under barge landings. 13T or so under RTLS.

Adding another core (symmetric) increases payload to ~22T to LEO under barge landings, and 20T or so with RTLS. ~30T to LEO or so in expendable mode, which is plenty for space station modules.

Each stage uses a single engine (justification above).

Only the core for 1st version, and boosters for the others, are reused, at first. If doing booster + core reuse + high-speed barge landings for version 1 and 2, use the RLTS payload numbers.

For all three, a STAR motor is a option to do direct to GEO or MEO transfers (like what was done on the Delta II)

You can use whatever core diameter you want.

4. Have your H2/CH4 rocket stage insulated, like all real H2 or Ch4 rocket stages do.

(I have no clue why you haven't accepted this already)

 

I'm still working on the SHLV recommendation.

 

Got it? Let's stop everything else and stick to these 4 topics.

Explain why you don't/do want them (and make sure I have not addressed it, or if you have a question/objection to it).

Let's do this for the sake of brevity. Let's cut down our focus and make decisions and/or compromises NOW.

Thank you, Pipcard, and sorry for this.

Link to comment
Share on other sites

HLV recommendation #1:

Lunar version of my proposed M-III with 4 cores as boosters could carry 55T to LEO, according to delta_iv_heavy_lift.jpg

an expendable version would probably be more like ~65T to LEO, and a potential 6-booster core version could carry even more, if needed.

The RL-60 (aka MB-60 has 266kN thrust. 3 are needed to meet F9 2nd stage thrust). However, structural mods are needed for 6-booster configs, and should only be done when necessary.

https://www.nasaspaceflight.com/2013/08/dual-sls-required-nasas-lunar-landing-option/

3 launches of my M-III would be required, assuming an SLS-style lunar mission . Less than optimal, but not bad. Considering you could reduce mass even more if you used a Soyuz-style config with separate orbital and descent modules (this shaves off lots of CM mass), and also use a smaller SM to compensate for the smaller CM, plus a H2lox lunar lander with IVF, you might even get away with 2 launches, plus another launch of a smaller LV to carry only the crew to LEO to dock with the lunar transfer vehicle.

 

Pros: Minimal development costs, commonality with H-III.

Cons: Tiny Payload fairing heavily limits overall payload to LEO, preventing a Earth -Orbit Rendezvous solution. And forget about Mars Landers.

Small Payload Capacity. (65-70T To LEO)

A possible RP-1/CH4 upper stage limits BLEO capacity.

 

HLV Recommendation #2:

http://www.astronautix.com/lvs/vulkan.htm

Vulcan-Herceles type design.

H-III core as LRBs, H2 or RP-1 Propellant in the 2nd and 3rd stages.

 

Pros: Optimal solution in performance, high payload capacity (175 T to LEO? max), Large payload fairing makes Mars and Lunar Missions "easily" feasible.

Cons: High development cost, new pads and ground support needed.

 

HLV Recommendation #3:

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

7 expendable H-III cores attached to a upper stage of 7 down-stretched H-III cores in a Saturn IB first stage-style arrangement.

 

Pros: Fairly Large Payload capacity (over 100T to LEO payload capacity?), H-III commonality, Large payload fairing makes Mars and Lunar Missions "easily" feasible, low development cost.

Cons: Less commonality than Option one, and higher development cost than Option #1 (but not as bad as Option #2.)

Link to comment
Share on other sites

Quote

I don't know why SpaceX chose a 100% methane rocket, for sure. No one does, and neither do you.

Everyone does, because Musk has already stated why.
Musk thinks the "slight" Isp increase is worth it, and considers it along with "energy cost" to be the "main driver" for that choice. Methane could be made using carbon from Mars's atmosphere (+ hydrogen that was brought along or mined from water ice deposits), and you might be able to make methane on the Moon as LCROSS detected carbon (in carbon monoxide) in the impact ejecta. You cannot make RP-1 using ISRU on Mars or the Moon.

Quote

But if you can't reuse, or use it elsewhere, what's the point to a more complex, detachable SM bus? Why not integrate it into the spacecraft, like the Russians?

Because it provides commonality with the UTV (HTV analog).

Quote

So? Long-term in space only matters, again, going to Mars.

MCT revolves around Mars.

The H-III and H-III Lunar rockets are NOT.

Long term storage also applies to a depot at Earth-Moon L1, as well as a lander staying at a lunar base for several months. A reusable lunar lander will also have less coking problems when it is using methane instead of RP-1. And please say "M-III," as we are not talking about the H-III.

M-III revolves around taking humans to many places, including Earth orbit, the Moon, and even Mars (although a successor to the M-III might be used for the human Mars missions. But still, even if that happens, M-III is being used to gain experience with methalox propulsion technology.)

Quote

STOP MAKING ME REPEAT MYSELF.

Please, stop. Start reading my previous posts, please, and stop repeating the same points. I'm tired of it. I've addressed all of them.

I feel like I'm talking to a brick wall. Please, for my sake AND yours.

You know what, I feel the exact same way towards you. I'm tired of arguing too, and I just wanted this thread to be something to show my Orbiter add-on developments in, and now this page has gotten really long because we both keep having to respond to each other.

Quote

But it's not designed for Mars, is it (like MCT)? It's designed for commercial use, which even SpaceX prefers Rp-1 with.

It is designed for commercial use and government-led human activities beyond Earth's orbit (which may become commercialized as well if companies such as SpaceX have their way).

Vulcan is using methane (liquified natural gas is almost entirely comprised of methane) as well, even though it does not revolve around Mars missions, because it is better for engine reuse. SpaceX only preferred RP-1 because they were just starting out.

Quote

For example, single engines on both stages, instead of 9 engines on the 1st stage is simpler and more cheap (COST PER LAUNCH) (Developing engines cost an arm and a leg- leading to the Atlas V RD-180 decision that bit Lockheed/ULA in the poophole). Reuse means that there will be no cost savings in mass production, thus there is little reason to cluster engines (except the aforementioned development costs)

"except the aforementioned development costs" - And that's exactly why. Developing a single larger engine costs an arm and a leg compared to developing a smaller engine that can be clustered multiple times. Having 7-9 engines also means more mass production of that engine rather than a single large engine, as well as engine-out capability (although engine failures won't be simulated in the Orbiter add-on). And you can have a vacuum-optimized variant of that same first stage engine for the second stage.

Quote

This discussion started, as I was against the use of Ch4 due to only marginal higher ISP and much lower density and higher R+D costs.
Keep that in mind. Don't attack hydrogen specifically, I'm attacking Methane.

Okay, let me say this for the last time: the benefits (even with "apples to apples" engine comparisons) are more synergy and commonality with ISRU systems in future Mars missions (and possibly lunar propellant mining operations as well), in addition to space-storability (RP-1 is harder to store due to greater temperature difference between oxidizer and fuel), and little-to-no coking (better maintenance when reusing engines). HASDA will be going to many places, and it is important to plan for the future. You are only thinking short-term while I am thinking about the long-term development of space.

Quote

PS:Let's compare 2 rockets, one with common propellants (RP-1), and one with 2 types of propellant (solid and H2). Both are expendable LV made by government- controlled/sponsored companies, and both are under development, designed to be cheaper to compete in the commercial market. (Apples to Apples)

Not apples to apples, as Ariane 6 launches much closer to the equator than Angara-5 (so it gets the speed boost from the Earth's rotation, and the upper stage and/or satellite need less delta-V to change inclination in order to get to a geosynchronous orbit of 0 degrees).

Quote

The latter number needs an entirely new rocket variant for H-II (and for my H-III as well)

And engineers will always choose the larger number when given a mass LV decision. Better to have too much LV capacity than too little.

M-II could carry RCV without needing boosters. As you can see here, the RCV adapter is substantially smaller than the PMA (and yes, I know the stock ISS in Orbiter has the wrong kind of docking petals). And that statement in bold is why M-III single core will have a capacity of 14 tonnes to LEO in RTLS mode, compared to your "5 tonnes in RTLS mode."

Quote

But you still need an insulated tank.

Insulated tanks add more mass. Although this animation for ULA's Vulcan is sub-par quality for professional work, it shows the entire first stage of Vulcan frosting up (for kerolox, only the oxidizer section frosts), which only happens when it isn't insulated like a hydrolox stage. This is the only official depiction of a orbital launch vehicle using methalox propellant. (until SpaceX unveils the BFR)

Quote

Also, RCV? Couldn't you develop a Capsule? I mean, the mods you added (and others needed, like a buffed up heat shield, and a proper SM to leave Lunar Orbit- that is the min. amount of hypergol propellant needed in case something bad happens, you can always escape with a reliable hyoergol engine no matter what) to the vehicle seems like it would be better to build a new system.

RCV has PICA-X style heatshields, which are capable of re-entry from the Moon and even Mars. An entirely separate re-entry vehicle to develop and operate sounds much more costly than an extra appended habitation and service module that doesn't need to re-enter intact or operate separately from the RCV.

Quote

Could you make the Mu-V to a Mu-VI, with the same dimensions, using higher ISP SRB propellant (if possible), a monolithic SRB as the 1st stage (like Mu-V), and a composite carbon motor instead of steel (like the GEM motors, or Black Knight), along with other mods to make it cheaper, like automated launch operations (like what Epsilon did).

Negi-5 was that. It did have a monolithic SRB like Epsilon as the first stage and I already told you that but you didn't listen. Now it is being replaced by M-III for the same reasons that Falcon 9 replaced Falcon 1, and Falcon 9R will be the same price as Epsilon for a lot more payload capacity (and could spread launch costs even further by carrying multiple small payloads, thus undercutting Vega). Using the same launcher means less separate stage production lines for less manufacturing costs.

Quote

However, structural mods are needed for 6-booster configs, and should only be done when necessary.

And that is one of the reasons why M-III has a maximum of three cores, as you increase the amount of cores, more structural mods will be needed to support them. For HLV option 3, having 7 cores and 7 second stages joined up like that results in diminishing returns in performance (because the mass of the outer skin of each core adds up, launchers in the real world aren't designed the same way as in KSP). HLV option 2 is not preferred due to lack of commonality. 175-tonnes to LEO = lower flight rate than multiple launches of a 60-tonne launcher (actually ~30 tonnes with two boosters RTLS & central core landing on barge, or ~40-50 tonnes with two boosters RTLS & and central core expended) that also uses common components with a 20-tonne launcher.

For M-III, there will be 3 cores/2 boosters (simpler [i.e. less costly] integration and handling process in the VAB than a 7-core or 5-core version) to handle 60-tonne payloads (like Falcon Heavy, but slightly better); the single core will provide launches of ~4-tonne satellites to GTO in RTLS mode and ~6-tonne satellites in barge landing mode (haven't really tested barge landings yet, but the fully expendable GTO capacity is ~8 tonnes, so the barge mode capacity is somewhere in between 4 and 8).

With the 5.39-m diameter, the rocket would also be too short if it were to carry only 8-10 tonnes to LEO.

Quote

Thank you, Pipcard, and sorry for this.

I am sorry, too.

Edited by Pipcard
Link to comment
Share on other sites

11 hours ago, Pipcard said:

Everyone does, because Musk has already stated why.
Musk thinks the "slight" Isp increase is worth it, and considers it along with "energy cost" to be the "main driver" for that choice. Methane could be made using carbon from Mars's atmosphere (+ hydrogen that was brought along or mined from water ice deposits), and you might be able to make methane on the Moon as LCROSS detected carbon (in carbon monoxide) in the impact ejecta. You cannot make RP-1 using ISRU on Mars or the Moon.

Because it provides commonality with the UTV (HTV analog).

Long term storage also applies to a depot at Earth-Moon L1, as well as a lander staying at a lunar base for several months. A reusable lunar lander will also have less coking problems when it is using methane instead of RP-1. And please say "M-III," as we are not talking about the H-III.

M-III revolves around taking humans to many places, including Earth orbit, the Moon, and even Mars (although a successor to the M-III might be used for the human Mars missions. But still, even if that happens, M-III is being used to gain experience with methalox propulsion technology.)

You know what, I feel the exact same way towards you. I'm tired of arguing too, and I just wanted this thread to be something to show my Orbiter add-on developments in, and now this page has gotten really long because we both keep having to respond to each other.

Just because the majority of rocket manufacturers are doing that doesn't mean it is the most cost-optimized choice. The choice to use a single propellant and engine type is based on first principles regarding manufacturing and operating costs, which are increased when there are separate engine designs and propellant types.

It is designed for commercial use and government-led human activities beyond Earth's orbit (which may become commercialized as well if companies such as SpaceX have their way).

Vulcan is using methane (liquified natural gas is almost entirely comprised of methane) as well, even though it does not revolve around Mars missions. SpaceX only preferred RP-1 because they were just starting out.

"except the aforementioned development costs" - And that's exactly why. Developing a single larger engine costs an arm and a leg compared to developing a smaller engine that can be clustered multiple times. Having 7-9 engines also means more mass production of that engine rather than a single large engine, as well as engine-out capability (although engine failures won't be simulated in the Orbiter add-on). And you can have a vacuum-optimized variant of that same first stage engine for the second stage.

Okay, let me say this for the last time: the benefits (even with "apples to apples" engine comparisons) are more synergy and commonality with ISRU systems in future Mars missions (and possibly lunar propellant mining operations as well), in addition to space-storability (RP-1 is harder to store due to greater temperature difference between oxidizer and fuel), and little-to-no coking (better maintenance when reusing engines). HASDA will be going to many places, and it is important to plan for the future. You are only thinking short-term while I am thinking about the long-term development of space.

Not apples to apples, as Ariane 6 launches much closer to the equator than Angara-5 (so it gets the speed boost from the Earth's rotation, and the upper stage and/or satellite need less delta-V to change inclination in order to get to a geosynchronous orbit of 0 degrees).

M-II could carry RCV without needing boosters. As you can see here, the RCV adapter is substantially smaller than the PMA (and yes, I know the stock ISS in Orbiter has the wrong kind of docking petals). And that statement in bold is why M-III single core will have a capacity of 14 tonnes to LEO in RTLS mode, compared to your "5 tonnes in RTLS mode."

Insulated tanks add more mass. Although this animation for ULA's Vulcan is sub-par quality for professional work, it shows the entire first stage of Vulcan frosting up (for kerolox, only the oxidizer section frosts), which only happens when it isn't insulated like a hydrolox stage. This is the only official depiction of a orbital launch vehicle using methalox propellant. (until SpaceX unveils the BFR)

RCV has PICA-X style heatshields, which are capable of re-entry from the Moon and even Mars. An entirely separate re-entry vehicle to develop and operate sounds much more costly than an extra appended habitation and service module that doesn't need to re-enter intact or operate separately from the RCV.

Negi-5 was that. It did have a monolithic SRB like Epsilon as the first stage and I already told you that but you didn't listen. Now it is being replaced by M-III for the same reasons that Falcon 9 replaced Falcon 1, and Falcon 9R will be the same price as Epsilon for a lot more payload capacity (and could spread launch costs even further by carrying multiple small payloads, thus undercutting Vega). Using the same launcher means less separate stage production lines for less manufacturing costs.

And that is one of the reasons why M-III has a maximum of three cores, as you increase the amount of cores, more structural mods will be needed to support them. For HLV option 3, having 7 cores and 7 second stages joined up like that results in diminishing returns in performance (because the mass of the outer skin of each core adds up, launchers in the real world aren't designed the same way as in KSP). HLV option 2 is not preferred due to lack of commonality. 175-tonnes to LEO = lower flight rate than multiple launches of a 60-tonne launcher (actually ~30 tonnes with two boosters RTLS & central core landing on barge, or ~40-50 tonnes with two boosters RTLS & and central core expended) that also uses common components with a 20-tonne launcher.

For M-III, there will be 3 cores/2 boosters (simpler [i.e. less costly] integration and handling process in the VAB than a 7-core or 5-core version) to handle 60-tonne payloads (like Falcon Heavy, but slightly better); the single core will provide launches of ~4-tonne satellites to GTO in RTLS mode and ~6-tonne satellites in barge landing mode (haven't really tested barge landings yet, but the fully expendable GTO capacity is ~8 tonnes, so the barge mode capacity is somewhere in between 4 and 8).

I am sorry, too.

Screw this, this is a waste of time. 

 

I'm out.

You're not going to be convinced either way.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...