Author Topic: Speculation thread: intermediate-lift Raptor-derived RLV  (Read 31076 times)

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #40 on: 04/25/2017 07:54 PM »
Payload to LEO with expendable Falcon Heavy boosters is just a "what-if" measuring stick since most rockets are compared by LEO payload.

More interested in just the TMI number for the expendable FH scenario, as if one were trying to land volume+mass on Mars as soon as possible it would probably be the fastest path forward.
The scenario where expendable booster 3-cores of Falcon Heavy might launch this mini-ITS direct to Mars would be where: in space prop transfer isn't developed yet, but one really wants to get a test of the system to land on Mars with at least some meaningful payload, so you expend some used FH cores.
Right, makes perfect sense.

Here's the table showing payloads. Everything is listed in tonnes.

_LEO|GTO|TMI
Falcon 9 FT (RTLS)18.3|5.7|2.7
Falcon 9 FT (droneship)19.2|6.6|3.3
Falcon 9 FT (expended)22.3|8.1|4.4
Falcon H (recovered)31.6|11.7|6.7
Falcon H (core expended)39.3|15.4|9.3
Falcon H (expended)51.5|20.6|12.9
Falcon 9M (RTLS)19.9|7.4|4.0
Falcon 9M (droneship)21.9|8.4|4.7
Falcon 9M (expended)25.5|10.2|6.1
Falcon HM (recovered)37.3|14.5|8.9
Falcon HM (core expended)46.4|19.0|12.1
Falcon HM (expended)59.51|25.1|16.3
Falcon 9M+ (RTLS)22.0|7.8|1.4
Falcon 9M+ (droneship)23.8|8.8|2.1
Falcon 9M+ (expended)27.0|10.4|3.3
Falcon HM+ (recovered)42.4|16.9|7.7
Falcon HM+ (core expended)50.2|20.8|10.6
Falcon HM+ (expended)64.8|27.0|14.9

"Falcon 9 FT" and "Falcon H" represent the current incarnations with expendable kerolox upper stages. Standard recovery for Falcon Heavy is RTLS for the boosters and droneship for the core; "core expended" means droneship recovery for the boosters.

"Falcon 9M" and "Falcon HM" represent the current configuration with an expendable methalox upper stage powered by a single vacuum-optimized devRaptor. Upper stage material and dimensions are the same, with a shifted common bulkhead.

"Falcon 9M+" and "Falcon HM+" represent the replacement of the current upper stage with my two-engine reusable composite upper stage, including landing thrusters and recovery margins. LEO missions have enough margin for bringing the full payload back to Earth (e.g., crewed missions), GTO missions have enough margin for bringing the stage back empty, and TMI missions have enough margin to land the full payload on Mars.

Not shown in the table is the theoretical maximum performance to LEO (e.g., upper stage fully expended, with plan to refuel in LEO); this is 65.7 tonnes. However, these are all Falcon 9 FT numbers, with Falcon Heavy based on this. SpaceX is currently projecting Falcon Heavy (based on Block 5 cores) at an expendable payload of 63.8 tonnes to LEO, while my table has it at 51.5 tonnes to LEO. So the true theoretical maximum performance could be significantly higher, maybe as high as 75 tonnes.

Not sure how Block 5 FH can make 63.8 tonnes to orbit, though. That's a jump far larger than what thrust increases alone can manage. Maybe specific impulse is coming up too?

Mars is 6,000 m/s past LEO; if you bring the upper stage tanks and propulsion along all the way to the surface the payload drops to about what Red Dragon can do with a fully expendable FH - roughly 2,000 kg.

Cryo prop transfer is really, really important if we want to go to Mars or anywhere else. It's not that hard, compared to landing on Mars and making fuel there. It's something that could easily be flight-ready before the rest of a mini-ITS system.
Actually, I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.

But you're right, orbital prop transfer is super important. A reusable methalox upper stage for Falcon 9 and Falcon Heavy would be a fantastic test bed for this.
« Last Edit: 04/25/2017 08:03 PM by sevenperforce »

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1157
  • Likes Given: 731
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #41 on: 04/25/2017 08:26 PM »
Mars is 6,000 m/s past LEO; if you bring the upper stage tanks and propulsion along all the way to the surface the payload drops to about what Red Dragon can do with a fully expendable FH - roughly 2,000 kg.

Cryo prop transfer is really, really important if we want to go to Mars or anywhere else. It's not that hard, compared to landing on Mars and making fuel there. It's something that could easily be flight-ready before the rest of a mini-ITS system.
Actually, I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.
Your mass fractions are also extremely optimistic. 6.6/(6.6+141) = 4.5%. SpaceX is claiming 7.1% dry mass for the  ITS ship, which presumably includes a pressure vessel for crew and ECLSS equipment but no consumables. The ITS tanker is at 3.5%, but I have seen no indication that it has anything needed for flight beyond LEO like a high velocity heatshield, solar panels, payload adapter, or cargo doors.

I was assuming 4.5 km/s for TMI and 1.5 km/s for landing, which is probably conservative. A slower transfer could be done for 5 km/s total, which even with more realistic mass fractions does get 5 to 10 tonnes to the surface.

But with refueling and very conservative mass fractions it's straightforward to get 40 or 50 tonnes landed. There's just no comparison to direct launch, so LEO refueling is going to be one of the first things validated by any serious Mars architecture.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #42 on: 04/25/2017 08:27 PM »
Here's the table showing payloads. Everything is listed in tonnes.

Falcon H (expended)51.5|20.6|12.9

Falcon heavy expedable is 63.8 tonnes to LEO, 26.7 tonnes to GTO and 16.8 tonnes to TMI. So your numbers are way off.
My numbers are based on the Full Thrust Falcon Heavy, before the planned Block 5 uprating. We don't yet have enough numbers to model the performance of Block 5 Falcon Heavy. SpaceX does, and that's why SpaceX's website has those higher numbers.

As I said before, I'm not sure how they are able to get those numbers. Shotwell said the primary change is more powerful engines but they would have to be hella powerful to get that much of a boost.

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1157
  • Likes Given: 731
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #43 on: 04/25/2017 08:33 PM »
Here's the table showing payloads. Everything is listed in tonnes.

Falcon H (expended)51.5|20.6|12.9

Falcon heavy expedable is 63.8 tonnes to LEO, 26.7 tonnes to GTO and 16.8 tonnes to TMI. So your numbers are way off.
My numbers are based on the Full Thrust Falcon Heavy, before the planned Block 5 uprating. We don't yet have enough numbers to model the performance of Block 5 Falcon Heavy. SpaceX does, and that's why SpaceX's website has those higher numbers.

As I said before, I'm not sure how they are able to get those numbers. Shotwell said the primary change is more powerful engines but they would have to be hella powerful to get that much of a boost.

Maybe they are bringing crossfeed back :D

Offline gospacex

  • Senior Member
  • *****
  • Posts: 3030
  • Liked: 532
  • Likes Given: 604
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #44 on: 04/25/2017 08:33 PM »
At which point you ask yourself, "maybe it's easier to just build a separate factory and don't mess with the existing one". "And while we are at it, why not build this new factory in a location where road transport is not a limitation".
............
IOW: a small diameter change probably is not worth it. If you go for a diameter change, got for a big one.

That would be true for a new booster stage. If you want a new upper stage that can be launched on mostly existing ground support hardware you go for the diameter that can be supported on a TEL. 4m is exactly what I was thinking about. Enough volume increase to have at least the same propellant weight, making good use of methalox propellant. Prove a lot of technology by using carbon composite, Raptor, self pressurization, even refuelling in orbit. As was shown in this thread the system would be very capable.

I' not disputing the capability. I'm disputing than going to 4m core makes most sense. Why not just go for a new factory and 6m core, for example?

Online GWH

That chart is phenomenal work.

Mars is 6,000 m/s past LEO; if you bring the upper stage tanks and propulsion along all the way to the surface the payload drops to about what Red Dragon can do with a fully expendable FH - roughly 2,000 kg.
Actually, I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.

The mini-ITS can deliver something that a Dragon can't however: volume. And if the hinged nose cone is used mobile equipment can be deployed direct to surface without the added mass of a crane.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #46 on: 04/25/2017 08:38 PM »
I' not disputing the capability. I'm disputing than going to 4m core makes most sense. Why not just go for a new factory and 6m core, for example?
There's a limit to what is readily road-transportable that's right around 4-5 meters.

But in any case, as the numbers above demonstrate, there is really no need for a new booster core. Falcon Heavy can take the methalox upper stage anywhere it needs to go. Having to pipe in both methane and kerosene kinda sucks, but oh well.

The upper stage size is flexible; I made it 4 meters to match the 4-meter booster core I was proposing, but since that's not gonna happen, it could be larger if necessary. My guess is that the ideal size will depend on what will have the best drag characteristics for mating to Falcon Heavy. You don't want it to be too much taller than the existing stack.
« Last Edit: 04/25/2017 08:41 PM by sevenperforce »

Offline gospacex

  • Senior Member
  • *****
  • Posts: 3030
  • Liked: 532
  • Likes Given: 604
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #47 on: 04/25/2017 08:44 PM »
I' not disputing the capability. I'm disputing than going to 4m core makes most sense. Why not just go for a new factory and 6m core, for example?

There's a limit to what is readily road-transportable that's right around 4-5 meters.

If you go for a new factory, you plan to have it where water transportation is possible. And/or build it close to the pad.

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1157
  • Likes Given: 731
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #48 on: 04/25/2017 08:51 PM »
That chart is phenomenal work.

Mars is 6,000 m/s past LEO; if you bring the upper stage tanks and propulsion along all the way to the surface the payload drops to about what Red Dragon can do with a fully expendable FH - roughly 2,000 kg.
Actually, I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.

The mini-ITS can deliver something that a Dragon can't however: volume. And if the hinged nose cone is used mobile equipment can be deployed direct to surface without the added mass of a crane.

Agreed that volume and landed height are useful, but I don't think such a vehicle is worth designing and building to launch direct with no refueling. If you're going through that much effort to build a vehicle that capable, it's a very small additional  step to refuel it in LEO.

And I'll pit the mass of a crane against the additional structure and propulsion required for horizontal landing any day. A simple unloading ramp probably has more mass than a crane. The operational simplicity is definitely nice though.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #49 on: 04/25/2017 09:01 PM »
As I said before, I'm not sure how they are able to get those numbers. Shotwell said the primary change is more powerful engines but they would have to be hella powerful to get that much of a boost.

Maybe they are bringing crossfeed back :D
Even if I switch up the numbers to factor in 6-engine crossfeed, it only goes up to 57.6 tonnes.

Of course, it might be that the calculator I've been using is not handling the boosters properly.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #50 on: 04/25/2017 09:30 PM »
I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.
Your mass fractions are also extremely optimistic. 6.6/(6.6+141) = 4.5%. SpaceX is claiming 7.1% dry mass for the  ITS ship, which presumably includes a pressure vessel for crew and ECLSS equipment but no consumables. The ITS tanker is at 3.5%, but I have seen no indication that it has anything needed for flight beyond LEO like a high velocity heatshield, solar panels, payload adapter, or cargo doors.
Here, I can break down my numbers.

I'm putting the Raptor's SL TWR at 200:1, which makes its mass 1.55 tonnes. There are nine of these on the ITS Tanker and Spaceship, but I'll go ahead and kick the engine mass up to a total of 15.5 tonnes to account for the mass of the vacuum nozzle extensions.

The dry mass of the ITS tanker is projected at 90 tonnes, which includes the monocoque tanks/body, the intertanks, the RCS thrusters, the TPS, the engines, and the landing legs. Subtract the engine mass, and that's 74.5 tonnes for the tanker's 2,500 tonnes of propellant, or a mass ratio of 33.6:1.

To account for some square-cube losses, I projected my structural mass ratio at 29:1 to get a base structural mass of 4.88 tonnes for the 141 tonnes of propellant I wanted. I added back 1.28 tonnes for the two dev Raptor engines and 405 kg for the eight landing thrusters.

I didn't include payload encapsulation in my overall dry mass because I'd want a monocoque pressure vessel for the crewed version. The payload fairings on the Falcon 9 mass 1.75 tonnes total so that's a good estimate for the cargo bay mass. The actual downmass to Mars on a FH-expendable direct ascent is probably closer to 13 tonnes, then. That's reserving 2.9 tonnes of propellant for the landing burn, which gives about 448 m/s of dV. That should be enough, I think; if I need more it would decrease the downmass slightly but not by much.

Agreed that volume and landed height are useful, but I don't think such a vehicle is worth designing and building to launch direct with no refueling. If you're going through that much effort to build a vehicle that capable, it's a very small additional  step to refuel it in LEO.

And I'll pit the mass of a crane against the additional structure and propulsion required for horizontal landing any day. A simple unloading ramp probably has more mass than a crane. The operational simplicity is definitely nice though.
I would certainly support refueling. But there is really very little additional structure or propulsion required for horizontal landing. Composites have better biaxial rigidity than aluminum, you need biaxial strength for the biconic re-entry anyway, and you need the thrusters regardless because of TWR issues.

You might still need an arm or crane or some other mechanism for getting the payload out, but it is a LOT easier than trying to do it with a capsule.
« Last Edit: 04/25/2017 09:33 PM by sevenperforce »

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1157
  • Likes Given: 731
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #51 on: 04/26/2017 01:04 AM »
I got 15 tonnes to the Martian surface with a direct ascent on fully-expendable Falcon Heavy. And this number, like the rest of the numbers, is based on the FT Heavy, not the Block 5 Heavy.
Your mass fractions are also extremely optimistic. 6.6/(6.6+141) = 4.5%. SpaceX is claiming 7.1% dry mass for the  ITS ship, which presumably includes a pressure vessel for crew and ECLSS equipment but no consumables. The ITS tanker is at 3.5%, but I have seen no indication that it has anything needed for flight beyond LEO like a high velocity heatshield, solar panels, payload adapter, or cargo doors.
Here, I can break down my numbers.

I'm putting the Raptor's SL TWR at 200:1, which makes its mass 1.55 tonnes. There are nine of these on the ITS Tanker and Spaceship, but I'll go ahead and kick the engine mass up to a total of 15.5 tonnes to account for the mass of the vacuum nozzle extensions.

The dry mass of the ITS tanker is projected at 90 tonnes, which includes the monocoque tanks/body, the intertanks, the RCS thrusters, the TPS, the engines, and the landing legs. Subtract the engine mass, and that's 74.5 tonnes for the tanker's 2,500 tonnes of propellant, or a mass ratio of 33.6:1.

To account for some square-cube losses, I projected my structural mass ratio at 29:1 to get a base structural mass of 4.88 tonnes for the 141 tonnes of propellant I wanted. I added back 1.28 tonnes for the two dev Raptor engines and 405 kg for the eight landing thrusters.

I didn't include payload encapsulation in my overall dry mass because I'd want a monocoque pressure vessel for the crewed version. The payload fairings on the Falcon 9 mass 1.75 tonnes total so that's a good estimate for the cargo bay mass. The actual downmass to Mars on a FH-expendable direct ascent is probably closer to 13 tonnes, then. That's reserving 2.9 tonnes of propellant for the landing burn, which gives about 448 m/s of dV. That should be enough, I think; if I need more it would decrease the downmass slightly but not by much.

This makes sense, except that you're comparing to the ITS tanker. As I pointed out already, the tanker isn't designed to fly to Mars. It most likely doesn't have TPS for an interplanetary entry, which will probably mass about twice as much as TPS for Earth LEO entry. It probably doesn't have significant solar power, operating mostly off batteries for the few hours it's in LEO. And it probably doesn't have active or passive cooling to keep the landing fuel from boiling off during a 6 month transit, which is how long a mass-optimized transfer to Mars takes. I'd figure dry mass of at least 5% of wet, before payload.

And it will need at least 1000 m/s to land. Terminal velocity is going to be Mach 2-ish.

Quote
Agreed that volume and landed height are useful, but I don't think such a vehicle is worth designing and building to launch direct with no refueling. If you're going through that much effort to build a vehicle that capable, it's a very small additional  step to refuel it in LEO.

And I'll pit the mass of a crane against the additional structure and propulsion required for horizontal landing any day. A simple unloading ramp probably has more mass than a crane. The operational simplicity is definitely nice though.
I would certainly support refueling. But there is really very little additional structure or propulsion required for horizontal landing. Composites have better biaxial rigidity than aluminum, you need biaxial strength for the biconic re-entry anyway, and you need the thrusters regardless because of TWR issues.

You might still need an arm or crane or some other mechanism for getting the payload out, but it is a LOT easier than trying to do it with a capsule.
What Isp and TWR are you assuming for the thrusters? I haven't seen any reliable info on them other than fuel and thrust. I'd be surprised if it's over 340 sec for a low pressure methalox RCS thruster.

Vertical landing allows the more efficient Raptors to null 99% of the velocity with thrusters handling only terminal thrust. And you only need about 1 g worth of terminal thrust instead of 3 to 4 gees. So with vertical landing you some structure mass, some thruster mass, and some fuel mass thanks to being lighter AND more efficient.

Offline TheKutKu

  • Member
  • Posts: 15
  • France
  • Liked: 5
  • Likes Given: 15
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #52 on: 04/26/2017 01:42 PM »
Why does the second stage need to have 2 engines? One single Vacuum Raptor-dev (assuming it is at 1/3rd of the thrust, and doesn't have much less TWR compared to the full sized one) could push the full second stage well enough, you don't need a 1 TWR (on earth) for a second stage or even a mars/moon lander.

Quote
And it will need at least 1000 m/s to land. Terminal velocity is going to be Mach 2-ish.

I may be wrong but since it is both a lifting body and it is quite light (<15 tons), i've read that in this case mars EDL delta v could be as low as 800 m/s.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #53 on: 04/26/2017 03:06 PM »
I'm putting the Raptor's SL TWR at 200:1, which makes its mass 1.55 tonnes. There are nine of these on the ITS Tanker and Spaceship, but I'll go ahead and kick the engine mass up to a total of 15.5 tonnes to account for the mass of the vacuum nozzle extensions.

The dry mass of the ITS tanker is projected at 90 tonnes, which includes the monocoque tanks/body, the intertanks, the RCS thrusters, the TPS, the engines, and the landing legs. Subtract the engine mass, and that's 74.5 tonnes for the tanker's 2,500 tonnes of propellant, or a mass ratio of 33.6:1.

To account for some square-cube losses, I projected my structural mass ratio at 29:1 to get a base structural mass of 4.88 tonnes for the 141 tonnes of propellant I wanted. I added back 1.28 tonnes for the two dev Raptor engines and 405 kg for the eight landing thrusters.

I didn't include payload encapsulation in my overall dry mass because I'd want a monocoque pressure vessel for the crewed version. The payload fairings on the Falcon 9 mass 1.75 tonnes total so that's a good estimate for the cargo bay mass. The actual downmass to Mars on a FH-expendable direct ascent is probably closer to 13 tonnes, then. That's reserving 2.9 tonnes of propellant for the landing burn, which gives about 448 m/s of dV. That should be enough, I think; if I need more it would decrease the downmass slightly but not by much.

This makes sense, except that you're comparing to the ITS tanker. As I pointed out already, the tanker isn't designed to fly to Mars. It most likely doesn't have TPS for an interplanetary entry, which will probably mass about twice as much as TPS for Earth LEO entry. It probably doesn't have significant solar power, operating mostly off batteries for the few hours it's in LEO. And it probably doesn't have active or passive cooling to keep the landing fuel from boiling off during a 6 month transit, which is how long a mass-optimized transfer to Mars takes. I'd figure dry mass of at least 5% of wet, before payload.
See, this is just the sort of engineering counter-analysis I was hoping for with this thread! Thanks. The points regarding high-velocity TPS, cooling, and power are good ones. I hadn't considered that. However, even if there is some dry mass growth on my upper stage, it's not catastrophic. Payload with full reuse is already very high, and this is not a SSTO where a 10% increase in dry mass results in negative payload fraction.

It was difficult to get a good estimate of tankage/structural fraction on the ITS Spaceship, because I don't know how much to subtract for the crew cabin and cargo hold. But after some consideration, I think we have a way we can at least place an upper bound on it.

The ITS Tanker and ITS Spaceship have identical external form lines, so we can say with relative confidence that if the tanker DID have all the TPS and cooling and power systems of the spaceship, it would still have a lower dry mass than the spaceship, since the crew cabin and ECLSS and cargo hold are heavier than monocoque tanks of equivalent size. So we can divide the structural dry mass of the ITS Spaceship (150 tonnes minus 15 tonnes of engine) by the propellant capacity of the ITS Tanker (2,500 tonnes) to get a structural/tankage/body fraction of 5.4%. This, I think, is generous and conservative. Even if there are square-cube losses to factor in, they are greatly outweighed by the fact that we are including the crew cabin and cargo hold and ECLSS dry mass, which we definitely don't need. And the ITS Spaceship dry mass already includes the additional mass of landing legs, control surfaces, RCS, and additional outer mold lines, so I don't need to add anything in to compensate for that.

Add it up, and the dry mass of my second-stage vehicle climbs from 6.56 tonnes to 9.32 tonnes, which means payload is simply reduced by a little under three tonnes. However, my reusable upper stage already beats the Falcon 9 upper stage for all LEO payloads by more than 3 tonnes when lofted on F9FT and it beats the Falcon 9 upper stage for all payloads period when lofted on Falcon Heavy...and that, again, is with full reuse. So it is by no means a dealbreaker.

Quote
Quote
I would certainly support refueling. But there is really very little additional structure or propulsion required for horizontal landing. Composites have better biaxial rigidity than aluminum, you need biaxial strength for the biconic re-entry anyway, and you need the thrusters regardless because of TWR issues.

You might still need an arm or crane or some other mechanism for getting the payload out, but it is a LOT easier than trying to do it with a capsule.
What Isp and TWR are you assuming for the thrusters? I haven't seen any reliable info on them other than fuel and thrust. I'd be surprised if it's over 340 sec for a low pressure methalox RCS thruster.

Vertical landing allows the more efficient Raptors to null 99% of the velocity with thrusters handling only terminal thrust. And you only need about 1 g worth of terminal thrust instead of 3 to 4 gees. So with vertical landing you some structure mass, some thruster mass, and some fuel mass thanks to being lighter AND more efficient.
Ah, but here's the problem: we're dealing with an upper stage, not a first stage; we can't use vacuum-expanded dev Raptors for landing at sea level. So for Earth entry we must have the thrusters handle the landing burn.

Not to mention that terminal velocity on Earth will be far lower for a blunt lifting body than for a tail-first descent, and the structural mass really doesn't go up because, again, you're already using a biconic re-entry.

Good catch about the isp of the landing engines. They are ten-metric-tonne thrusters, which is overpowered for RCS on a craft of this size, but great for landing. I did already factor in a decreased thrust due to SL expansion. Ten metric tonnes of vacuum thrust is 98.1 kN, but since these are going to need to be sea-level-expanded, underexpansion will decrease vacuum thrust to around 92 kN and SL thrust will be around 86 kN. 

I had been projecting vacuum isp for the landing engines at 360 seconds and SL isp at 330 seconds, forgetting that pressure-feeding decreases their efficiency. But by how much? The pressure-fed methalox thruster tested by NASA recently had a projected vacuum specific impulse of 348 seconds; for SL-expanded thrusters this translates to a vacuum isp of around 326 s and a SL isp of 305 s.

So that's going to increase landing fuel somewhat. But again, I have a LOT of margin, so it's not catastrophic.
Why does the second stage need to have 2 engines? One single Vacuum Raptor-dev (assuming it is at 1/3rd of the thrust, and doesn't have much less TWR compared to the full sized one) could push the full second stage well enough, you don't need a 1 TWR (on earth) for a second stage or even a mars/moon lander.
I tried it with a single dev Raptor first, but I didn't like it. TWR would be under half a gee for crewed missions, which is not really ideal; you end up needing an aggressively lofted trajectory, which really hurts your ability to safely abort. And FFSC gives you a really good TWR; removing one of the dev Raptors would only save you 640 kg of dry mass while cutting your vehicle TWR in half.

Quote
Quote
And it will need at least 1000 m/s to land. Terminal velocity is going to be Mach 2-ish.

I may be wrong but since it is both a lifting body and it is quite light (<15 tons), i've read that in this case mars EDL delta v could be as low as 800 m/s.
You need 1000 m/s for a capsule, but not for a biconic lifting body. Even 800 m/s is an overestimate, I think. The Mid-L/D MAV concept allocates 650 m/s for the landing burn, and my upper vehicle would be much fluffier (higher-drag, lower-mass) than the Mid-L/D. Speaking of which, the re-entry profile is quite similar:



I designed this from the ground up as an Earth-based reusable upper stage, but the adjustments to make it work equally well for Mars or the moon are really, really minor. The few tweaks merely make it a more capable spacecraft for Earth missions.
« Last Edit: 04/26/2017 03:22 PM by sevenperforce »

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1157
  • Likes Given: 731
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #54 on: 04/26/2017 03:58 PM »
... a structural/tankage/body fraction of 5.4%. ... second-stage vehicle climbs from 6.56 tonnes to 9.32 tonnes...

Sounds about right.

Ah, but here's the problem: we're dealing with an upper stage, not a first stage; we can't use vacuum-expanded dev Raptors for landing at sea level. So for Earth entry we must have the thrusters handle the landing burn.

Not to mention that terminal velocity on Earth will be far lower for a blunt lifting body than for a tail-first descent, and the structural mass really doesn't go up because, again, you're already using a biconic re-entry.

Musk said they plan to test the dev Raptor with a 150:1 nozzle, so it can probably operate at max thrust in Earth's atmosphere and still get about 375s Isp in vacuum.

On Earth, the terminal velocity is low enough to permit the flip to tail-first at a low altitude, so the final velocity is roughly the same. But the terminal velocity is also low enough that the Isp difference is almost negligible.

But you are right that you have margins, especially if refueled. Landing at Mars on 326s vs 375s engines eats only a tonne of the 40t or more payload capacity. Since this design is likely volume-limited, that's not a issue - and horizontal exit to the Mars surface is definitely worth the payload difference.

You need 1000 m/s for a capsule, but not for a biconic lifting body. Even 800 m/s is an overestimate, I think. The Mid-L/D MAV concept allocates 650 m/s for the landing burn, and my upper vehicle would be much fluffier (higher-drag, lower-mass) than the Mid-L/D. Speaking of which, the re-entry profile is quite similar:




Interesting that they have a supersonic drag coefficient over 3.0 at a 55 degree AoA, I'd expected that to be half that, at best. The higher drag can get the terminal velocity under 500 m/s, allowing the landing burn to be under 650 m/s. Nice.

I'd love to see some simulations of this vehicle doing EDL at Earth and Mars. Maybe OneSpeed can work something up similar to his ITS sims.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #55 on: 04/26/2017 04:49 PM »
To follow up on the one devRaptor vs two devRaptor question:

Assume launching on a Falcon 9 FT with droneship recovery of the first stage and recovery of the upper stage:

engines|LEO payload|LEO payload up/down (crew)|GTO payload
one vacuum dev Raptor|27 tonnes|25 tonnes|8.8 tonnes
two vacuum dev Raptors|25 tonnes|22 tonnes|9.0 tonnes

Dropping to a single engine can give you an advantage for lower-payload GTO flights, because your vehicle TWR isn't so bad at staging, but the GTO gains are very small and the LEO losses are large. Plus, it may no longer be able to SSTO from Mars. So I don't think it is a good tradeoff, especially because of the abort mode problems I mentioned before.

That's just for reference; the above numbers are based on the old dry mass and thruster impulse. Going to go ahead and generate new numbers to get an idea of what sort of performance it would really have with the additional conservatism suggested by envy.

Ah, but here's the problem: we're dealing with an upper stage, not a first stage; we can't use vacuum-expanded dev Raptors for landing at sea level. So for Earth entry we must have the thrusters handle the landing burn.
Musk said they plan to test the dev Raptor with a 150:1 nozzle, so it can probably operate at max thrust in Earth's atmosphere and still get about 375s Isp in vacuum.

On Earth, the terminal velocity is low enough to permit the flip to tail-first at a low altitude, so the final velocity is roughly the same. But the terminal velocity is also low enough that the Isp difference is almost negligible.

But you are right that you have margins, especially if refueled. Landing at Mars on 326s vs 375s engines eats only a tonne of the 40t or more payload capacity. Since this design is likely volume-limited, that's not a issue - and horizontal exit to the Mars surface is definitely worth the payload difference.
Since we are dealing with the main propulsion engines, we want to squeeze out all the vacuum specific impulse we can get, especially if this is going BLEO. And the landing thrusters are simple; I'm projecting their mass at just 45 kg each or thereabouts, so we'd be talking about a savings of maybe 200 kg max if we went to vertical, main-engine-assisted landing instead of horizontal-attitude landing. You're gonna lose more than 200 kg of payload by cutting isp from 381 to 375.

Horizontal attitude on Mars is definitely a big advantage. If the nose opens up as I showed above, then the payload could be an ISRU unit on an arm that extends a few meters out of the nose and then drops a drill straight down. Hit rock? No worries; simply take off on remaining fuel, fly to another site, and try again. One of the possible Mars mission architectures I'm looking at is using an ISRU-equipped mini-ITS to fill up on the Martian surface and then head to orbit. The crewed vehicle would aerocapture and rendezvous, then refuel fully in Martian orbit before descending; this avoids the need to do propellant transfer on the surface and allows you to abort to Earth if there's a problem.

I would argue that horizontal attitude on Earth is also worth the payload difference, particularly when you're dealing with a manned vehicle. Tipover or hard landing is way way less of a problem.

Quote
Interesting that they have a supersonic drag coefficient over 3.0 at a 55 degree AoA, I'd expected that to be half that, at best. The higher drag can get the terminal velocity under 500 m/s, allowing the landing burn to be under 650 m/s. Nice.

I'd love to see some simulations of this vehicle doing EDL at Earth and Mars. Maybe OneSpeed can work something up similar to his ITS sims.
I'm going to try to get someone with KSP+RSS and procedural parts to do a mockup and simulation, though it wouldn't be as nice as other sims.

Offline sevenperforce

  • Full Member
  • ****
  • Posts: 746
  • Liked: 175
  • Likes Given: 220
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #56 on: 04/26/2017 09:51 PM »
Here's a build request I did over at the KSP forums -- I'll see what I can get people to do.

I promised I would get some revised payload numbers based on the adjusted dry mass and corrected landing isp, so here we go! First, some landing fuel fractions:

Earth (255 m/s @ 305 s): 8.12%
Mars (650 m/s @ 325 s): 18.46%
Luna (1600 m/s @ 381 s, then 200 m/s @ 326 s): 38.81%

This gives me a starting point.

_|LEO|GTO|ISS|
Falcon 9FT (RTLS)|20.6|4.0|5.5|
Falcon 9FT (OCISLY)|22.4|5.0|7.2|
Falcon 9FT (expended)|25.7|6.7|10.1|
Falcon H (recovered)|41.0|13.1|23.0|
Falcon H (core expended)|49.1|16.2|30.3|
Falcon H (expended)|61.4|23.2|41.0|

LEO and GTO are for payload launches only, factoring in the weight of the cargo bay, cargo bay doors, and deployment system. All launches are with full upper stage recovery, as recovery penalty for an upper stage is about the same regardless of whether it is coming in from GTO or coming in from LEO.

ISS represents the payload capabilities of a manned shuttle-style variant, with an ejectable crew cabin containing seats for up to 10 passengers, ECLSS, LES, and independent re-entry/lifeboat capability. Passenger weight, equipment, and consumables must be deducted from listed payload. In all cases, the vehicle has sufficient reserves to return at least 5.5 tonnes of its payload (crew, cargo, or both) to the surface. Dry mass of the crew cabin variant is 20.5 tonnes.

A crewed flight to LEO without going to the ISS will have slightly higher payload margins.

Tomorrow, if I get a chance, I'll run some numbers to see what kind of direct-ascent cislunar, lunar-surface, TMI, and Martian-surface performance I can get. Then I'll look into what it can do with LEO refueling.

One neat idea I saw somewhere was that if SpaceX started flying something like this, it could fly comsat missions with more reserves than necessary (e.g., flying with FH instead of F9, or flying F9 with droneship recovery rather than RTLS recovery) and rendezvous with a tanker variant in LEO before deploying the payload. In this way, the tanker would end up accumulating propellant without requiring dedicated launches.

For reference, a tanker variant the same size as the standard vehicle would have a dry mass of 11.1 tonnes and be able to carry 174 tonnes of propellant. Launched on Falcon Heavy with full recovery, it would reach LEO with a little under 40 tonnes of propellant, plus its own landing reserves.

Offline hkultala

  • Full Member
  • ****
  • Posts: 615
  • Liked: 156
  • Likes Given: 92
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #57 on: 04/27/2017 04:36 AM »
Before running more (incorrect) numbers for different trajectories, you should try to match your numbers with the ACTUAL numbers released by spaceX.

Online dror

  • Full Member
  • ****
  • Posts: 541
  • Israel
  • Liked: 119
  • Likes Given: 336
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #58 on: 04/27/2017 06:50 AM »

This makes sense, except that you're comparing to the ITS tanker. As I pointed out already, the tanker isn't designed to fly to Mars. It most likely doesn't have TPS for an interplanetary entry, which will probably mass about twice as much as TPS for Earth LEO entry. It probably doesn't have significant solar power, operating mostly off batteries for the few hours it's in LEO. And it probably doesn't have active or passive cooling to keep the landing fuel from boiling off during a 6 month transit, which is how long a mass-optimized transfer to Mars takes. I'd figure dry mass of at least 5% of wet, before payload.

A little OT -
I figured that orbital refuel will be done first to one of the tankers on orbit, and than from the refueled tanker to a crewed spaceship. That means the tanker will stay longer on orbit and will act as a fuel depot for the spaceship.
The spaceship will have it's fuel ready on orbit when it is launched thus increasing crew safty.
For that to happen, at least one of three tankers, if not all, need to have built in cooling and power to act as a depot.
"If we crave some cosmic purpose, then let us find ourselves a worthy goal. "
Carl Sagan, Pale Blue Dot

Offline JamesH65

  • Full Member
  • ****
  • Posts: 600
  • Liked: 351
  • Likes Given: 8
Re: Speculation thread: intermediate-lift Raptor-derived RLV
« Reply #59 on: 04/27/2017 10:36 AM »
I'm putting the Raptor's SL TWR at 200:1, which makes its mass 1.55 tonnes. There are nine of these on the ITS Tanker and Spaceship, but I'll go ahead and kick the engine mass up to a total of 15.5 tonnes to account for the mass of the vacuum nozzle extensions.

The dry mass of the ITS tanker is projected at 90 tonnes, which includes the monocoque tanks/body, the intertanks, the RCS thrusters, the TPS, the engines, and the landing legs. Subtract the engine mass, and that's 74.5 tonnes for the tanker's 2,500 tonnes of propellant, or a mass ratio of 33.6:1.

To account for some square-cube losses, I projected my structural mass ratio at 29:1 to get a base structural mass of 4.88 tonnes for the 141 tonnes of propellant I wanted. I added back 1.28 tonnes for the two dev Raptor engines and 405 kg for the eight landing thrusters.

I didn't include payload encapsulation in my overall dry mass because I'd want a monocoque pressure vessel for the crewed version. The payload fairings on the Falcon 9 mass 1.75 tonnes total so that's a good estimate for the cargo bay mass. The actual downmass to Mars on a FH-expendable direct ascent is probably closer to 13 tonnes, then. That's reserving 2.9 tonnes of propellant for the landing burn, which gives about 448 m/s of dV. That should be enough, I think; if I need more it would decrease the downmass slightly but not by much.

This makes sense, except that you're comparing to the ITS tanker. As I pointed out already, the tanker isn't designed to fly to Mars. It most likely doesn't have TPS for an interplanetary entry, which will probably mass about twice as much as TPS for Earth LEO entry. It probably doesn't have significant solar power, operating mostly off batteries for the few hours it's in LEO. And it probably doesn't have active or passive cooling to keep the landing fuel from boiling off during a 6 month transit, which is how long a mass-optimized transfer to Mars takes. I'd figure dry mass of at least 5% of wet, before payload.
See, this is just the sort of engineering counter-analysis I was hoping for with this thread! Thanks. The points regarding high-velocity TPS, cooling, and power are good ones. I hadn't considered that. However, even if there is some dry mass growth on my upper stage, it's not catastrophic. Payload with full reuse is already very high, and this is not a SSTO where a 10% increase in dry mass results in negative payload fraction.

It was difficult to get a good estimate of tankage/structural fraction on the ITS Spaceship, because I don't know how much to subtract for the crew cabin and cargo hold. But after some consideration, I think we have a way we can at least place an upper bound on it.

The ITS Tanker and ITS Spaceship have identical external form lines, so we can say with relative confidence that if the tanker DID have all the TPS and cooling and power systems of the spaceship, it would still have a lower dry mass than the spaceship, since the crew cabin and ECLSS and cargo hold are heavier than monocoque tanks of equivalent size. So we can divide the structural dry mass of the ITS Spaceship (150 tonnes minus 15 tonnes of engine) by the propellant capacity of the ITS Tanker (2,500 tonnes) to get a structural/tankage/body fraction of 5.4%. This, I think, is generous and conservative. Even if there are square-cube losses to factor in, they are greatly outweighed by the fact that we are including the crew cabin and cargo hold and ECLSS dry mass, which we definitely don't need. And the ITS Spaceship dry mass already includes the additional mass of landing legs, control surfaces, RCS, and additional outer mold lines, so I don't need to add anything in to compensate for that.

Add it up, and the dry mass of my second-stage vehicle climbs from 6.56 tonnes to 9.32 tonnes, which means payload is simply reduced by a little under three tonnes. However, my reusable upper stage already beats the Falcon 9 upper stage for all LEO payloads by more than 3 tonnes when lofted on F9FT and it beats the Falcon 9 upper stage for all payloads period when lofted on Falcon Heavy...and that, again, is with full reuse. So it is by no means a dealbreaker.

Quote
Quote
I would certainly support refueling. But there is really very little additional structure or propulsion required for horizontal landing. Composites have better biaxial rigidity than aluminum, you need biaxial strength for the biconic re-entry anyway, and you need the thrusters regardless because of TWR issues.

You might still need an arm or crane or some other mechanism for getting the payload out, but it is a LOT easier than trying to do it with a capsule.
What Isp and TWR are you assuming for the thrusters? I haven't seen any reliable info on them other than fuel and thrust. I'd be surprised if it's over 340 sec for a low pressure methalox RCS thruster.

Vertical landing allows the more efficient Raptors to null 99% of the velocity with thrusters handling only terminal thrust. And you only need about 1 g worth of terminal thrust instead of 3 to 4 gees. So with vertical landing you some structure mass, some thruster mass, and some fuel mass thanks to being lighter AND more efficient.
Ah, but here's the problem: we're dealing with an upper stage, not a first stage; we can't use vacuum-expanded dev Raptors for landing at sea level. So for Earth entry we must have the thrusters handle the landing burn.

Not to mention that terminal velocity on Earth will be far lower for a blunt lifting body than for a tail-first descent, and the structural mass really doesn't go up because, again, you're already using a biconic re-entry.

Good catch about the isp of the landing engines. They are ten-metric-tonne thrusters, which is overpowered for RCS on a craft of this size, but great for landing. I did already factor in a decreased thrust due to SL expansion. Ten metric tonnes of vacuum thrust is 98.1 kN, but since these are going to need to be sea-level-expanded, underexpansion will decrease vacuum thrust to around 92 kN and SL thrust will be around 86 kN. 

I had been projecting vacuum isp for the landing engines at 360 seconds and SL isp at 330 seconds, forgetting that pressure-feeding decreases their efficiency. But by how much? The pressure-fed methalox thruster tested by NASA recently had a projected vacuum specific impulse of 348 seconds; for SL-expanded thrusters this translates to a vacuum isp of around 326 s and a SL isp of 305 s.

So that's going to increase landing fuel somewhat. But again, I have a LOT of margin, so it's not catastrophic.
Why does the second stage need to have 2 engines? One single Vacuum Raptor-dev (assuming it is at 1/3rd of the thrust, and doesn't have much less TWR compared to the full sized one) could push the full second stage well enough, you don't need a 1 TWR (on earth) for a second stage or even a mars/moon lander.
I tried it with a single dev Raptor first, but I didn't like it. TWR would be under half a gee for crewed missions, which is not really ideal; you end up needing an aggressively lofted trajectory, which really hurts your ability to safely abort. And FFSC gives you a really good TWR; removing one of the dev Raptors would only save you 640 kg of dry mass while cutting your vehicle TWR in half.

Quote
Quote
And it will need at least 1000 m/s to land. Terminal velocity is going to be Mach 2-ish.

I may be wrong but since it is both a lifting body and it is quite light (<15 tons), i've read that in this case mars EDL delta v could be as low as 800 m/s.
You need 1000 m/s for a capsule, but not for a biconic lifting body. Even 800 m/s is an overestimate, I think. The Mid-L/D MAV concept allocates 650 m/s for the landing burn, and my upper vehicle would be much fluffier (higher-drag, lower-mass) than the Mid-L/D. Speaking of which, the re-entry profile is quite similar:



I designed this from the ground up as an Earth-based reusable upper stage, but the adjustments to make it work equally well for Mars or the moon are really, really minor. The few tweaks merely make it a more capable spacecraft for Earth missions.

How does the lander get vertical again, for takeoff? Once fueled the landing engines will not be enough to get it off the ground will they?

Tags: