SpaceX--Spacecraft, rockets, and Mars

[Ded As Ned]

THEY DID IT!  I realize money is the bottom line (but it doesn't really interest me greatly).  This is a big accomplishment!  Thank you @SpaceChampion for the links and info, always. 

[Free Northman Reborn]

10 hours ago, SpaceChampion said:

Post-Launch Press Briefing with Musk

At 12min: Successfully landed the payload fairing.  It has a parachute and small thruster system.  It's a $6 million savings when they reuse it.

At 18min: Falcon Heavy cores being manufactured, first demo launch will be on SpaceX's dime.  Expect launch in the mid summer.

At 20min: Spent about $1 billion on pursuing reusability, on their own dime, since no one was paying for it.  It'll take them a while to amortize that.

At 24min: Won't see Optimis Prime for a few months.

At 25min: The Interplanetary Transport System includes the propellant manufacturing capability on Mars as part of the system.  Will publish an update to the design on SpaceX's website at some point.

At 27min: (I think this means my calculations upthread were on target) Have to pay off reusability development costs so price has to be as much as the cost savings.  ie. profit would be ~$30 million per flight regardless if it's a $62M or $43M flight.

At 30min: Regarding refurbishment, just some paint where it's bubbled and replacing the thermal shield.  The grid fins take the most wear so they're upgrading to titanium alloy fins which will allow them to come in at a higher angle and therefore hotter.  Sounds like they don't really have much refurbishment at all.  They're going to apply more sensors I think he said earlier, so the rocket itself will tell them if it's ready to fly again.

At 39min: Sounds like this landed booster will stay at the Cape as a monument to the achievement.  Other landed boosters will stay at the Cape too but because they'll be flying a lot.

At 43min: Musk says the BFR part of ITS (ie. the lower stage) should be able to achieve a flight rate of once per HOUR. 

At 47min: Falcon Heavy will fly from pad 39A but not until they have pad 40 back in service so if FH blows up they don't want to be without a pad on the east coast.

Really nice summary. Thanks.

[SpaceChampion]

I realized that financially SpaceX should be in good shape.  I'd think their total development costs of about $2B over 15 years could be amortized in about 67 flights if each flight is making $30M in profit.  Since they've had about 30 successful flights so far (though the first couple were the discontinued Falcon 1) they are well on their way to amortization.  I'd say within two years they should reach it, maybe even 18 months.

Next launch is a secret military/security payload for the NRO.  After that, a bunch of commercial sats, and another CRS mission to ISS.  The should be launching about every 10-14 days.  I think they're going to be using up some of their Block 3 and Block 4 versions of F9.  Next reflown F9 is probably the next SES flight in June.

 

From: https://www.reddit.com/r/SpaceX/wiki/launches/manifest

Upcoming Falcon launches

NET Date [Launch window]	Vehicle	Launch site	Orbit	Payload mass (kg)	Payload(s)	Customer	Notes & Refs
2017 Apr 16	Falcon 9	LC-39A	?	?	NROL-76	National Reconnaissance Office	S, 1, 5, 37
2017 Apr 30	Falcon 9	LC-39A	GTO	6070	Inmarsat 5-F4	Inmarsat, UK	O, 1, 5
2017 May 14	Falcon 9 + Dragon	LC-39A	LEO	1737 / 1573	SpX CRS-11 & ROSA, MUSES, NICER	NASA	O, 5, 1, 50
2017 late May	Falcon 9	LC-39A	GTO	3669	BulgariaSat-1	Bulsatcom, Bulgaria	O, 5
2017 June	Falcon 9	LC-39A	GTO	~6000	Intelsat 35e	Intelsat	O, 5
2017 June	Falcon 9	LC-39A	GTO	5400	SES-11 (EchoStar 105)	SES, Luxembourg	O, 46, 5
2017 Jun 17	Falcon 9	LC-4E	PO	860 (x10)	Iridium 2 (21-30)	Iridium Communications	O, 11, 14, 3, 5
2017 Jul	Falcon 9	LC-39A	GTO	~3500	KoreaSat 5A	KT Corporation, Korea	O, 34
2017 July 22	Falcon 9	LC-4E	SSO	525	FORMOSAT 5	NSPO, Taiwan	O, 3, 44, 5
2017 Aug 1	Falcon 9 + Dragon	LC-39A	LEO	2349 / 961	SpX CRS-12 & CREAM	NASA	O, 1, 31, 30, 5
2017 Aug 24	Falcon 9	LC-4E	PO	860 (x10)	Iridium 3 (31-40)	Iridium Communications	O, 11, 14, 3, 5
2017 Oct	Falcon 9	SLC-40	GTO	~4000	SES-16 (GovSat-1)	SES, Luxembourg	O, 3, 46, 5
2017 Oct	Falcon 9	LC-4E	PO	860 (x10)	Iridium 4 (41-50)	Iridium Communications	O, 11, 14, 5
2017 Oct	Falcon 9	LC-4E	SSO	1600	SAOCOM 1A	CONAE, Argentina	O, 2, 16, 17, 5
2017 Oct	Falcon 9 + Dragon	LC-39A	LEO	2333 / 977	SpX CRS-13 & ASIM, TSIS, MISSE	NASA	O, 22, 30, 5
2017 Nov	Falcon Heavy	LC-39A	?	?	Demo Flight	SpaceX	O, T, 3
2017 Nov	Falcon 9 + Crew Dragon	LC-39A	LEO	?	SpX-DM1 (uncrewed)	NASA	O, T, 1, 30, 5
2017 Nov / Dec	Falcon 9	SLC-40 / LC-39A	GTO	~5000	Hispasat 30W-6 (Amazonas 5)	Hispasat, Spain	O, 5
2017 Dec 16	Falcon 9	SLC-40 / LC-39A	GTO	~3500	Bangabandhu-1	Bangladesh Gov	O, 6, 35
2017 Dec	Falcon 9	LC-4E	PO	860 (x10)	Iridium 5 (51-60)	Iridium Communications	O, 14
2017 Dec	Falcon Heavy	LC-39A	LEO	>2500	STP-2 (DCX, & 34 others)	U.S. Air Force	O 3
2017 Q4	Falcon 9	LC-4E	SSO	>1000	SSO-A	Spaceflight Industries, Inc	S, 39, 40
2017 Q4	Falcon 9	LC-4E	SSO	1341 / ?	Paz / ?	EADS Astrium / ?	S, 45, 49
2017 Q4	Falcon 9	SLC-40	GTO	4200	SES-14	SES, Luxembourg	O, 3, 46, 5
2017	Falcon 9	SLC-40 / LC-39A	GTO	~5000	PSN-6	Pasifik Satelit Nusantara	S
2017	Falcon 9	SLC-40 / LC-39A	GTO	?	ABS-8	Asia Broadcast Sat.	S
2017	Falcon Heavy	LC-39A	GTO	?	?	Intelsat, Luxembourg	O

[Ser Scot A Ellison]

So the first crewed launch is November?

[SpaceChampion]

First launch of the Dragon 2 crew vehicle, but launched uncrewed to prove it first.  The link has the complete table, I only quoted the 2017 portion.  The second demo flight in May 2018 would have the first crew flying aboard.

However, Musk said yesterday the Demo 1 mission would be in late-summer, so the November date might be wrong.  Or Musk is too optimistic and it'll slip to November.

[SpaceChampion]

As far as I heard, SpaceX has given up on 2nd stage reusability for Falcon 9, but this news for Falcon Heavy is not too surprising, while still amazing.  He does seem wistful at times doing it for F9; doing it for FH may be a lot more realistic.  If this works, I think he might push his customers to accept launching two satellites aboard a FH at a time, as co-payloads, and offer substantial price reduction if they can get the upper stage back routinely.

This probably means the 2nd stage would come back using the steerable parafoils like the fairing did yesterday, floating down to an ocean splash, rather than using it's single engine for retropropulsive death drop.  On the other hand..  who knows?

[Free Northman Reborn]

On 2017/03/31 at 2:12 AM, SpaceChampion said:

Replay of the hosted livestream:

 

The price offered to SES was in the range of 30% cheaper than a new booster, but cost is much less.  SpaceX still needs to make enough profit to amortize development costs, so while it could be priced lower, it won't for a while.  SES will do 3 launches with SpaceX this year, 2 of which would be on previously flown boosters at the 30% lower cost.  About ~$43 million price for the flight-proven hardware.

The first stage is about 75% of the cost of the whole rocket.  About only $200-300K is fuel .  Profit on the ~$61.2M price is thought by SpaceX observers to be about 50%, so cost is about $30M.  So the first stage costs about $22.5M, the second stage about $7.5M.  So total cost about $30M can be reduce to let say $9M for a flight-proven system.  Don't take those as fact, it's just good guesses.  So reusability actually seems to keep the profit similar to a new booster, about ~$33M compared to ~$30M-ish.  They could afford to reduce the profit more if flight rates increase substantially.

Could they get down to $15 million a launch or less, and still make 33% profit?  I think so.

Note also that today was an attempt (haven't heard yet if it was successful) of another first that probably saves a few million, that of recovery of the payload fairing, which would have a GPS transceiver attached to it and a parachute to bring it down for a splash landing.  We might hear tomorrow.

Third achievement was first use of the Optimis Prime bot to grab ahold of the landed Falcom booster on deck of the drone ship.  :-)

Reflight of a Dragon capsule is another milestone for later this year.

I don't think you are correct in your assessment of the current cost of SpaceX rockets.

Firstly, the 50% profit margin was floated in some articles from financial analysts who attempted to figure out SpaceX's potential share value should it decide to list. But it is unlikely to be true for a number of reasons. But before I get there, let me also point out that a 50% profit margin does not mean a $62m launch price represents a $31m dollar cost to SpaceX. No, because that would represent a 100% profit margin. A 50% profit margin on a $62m launch would imply roughly a $41m cost to SpaceX. So the $30m current cost estimate for an expendable Falcon 9 launch is definitely not accurate. At best, if they are indeed achieving a 50% profit margin, current costs are around $41m.

Now, as I mentioned, the 50% profit margin was really just an outsider's estimate, and many think it is not true at all. In fact, many of SpaceX's competitors have been suggesting that SpaceX might be making losses on current launches in order to capture market share as quickly as possible. And that they were using prepaid development funding from Nasa to fund their expansion to date.

Now, I don't know if I quite believe that. I think SpaceX is definitely making profit, despite only charging $62m for a launch that competitors would charge $100m for. That's due to their innovative and lean production processes. But, it is also quite obvious that the real cost savings will come not from SpaceX's lean production processes, but from reuse. Rapid and cheap reuse. Which this SES mission demonstrated for the first time. Maybe not the rapid or cheap part. But at least the conceptual part. Once Block 5 is operational, the rapid and cheap reuse part will come into play. That should be by the end of this year.

As for the likely current cost of a SpaceX rocket? Consider this:

Elon says that the first stage represents roughly 70% of the cost of a Falcon 9.

Let's start with say a $50m cost estimate (meaning they currently make about $12m gross profit per launch).

70% of that is $35m. Leaving $15m for the remainder.

However, he also said in the latest press briefing that the fairing alone costs $6m. If our above starting point of $50m is correct, then that leaves only $9m for the entire 2nd stage. Somehow I don't think that the fairing costs almost as much as the entire second stage. So the only way to leave more of the total to cover the 2nd stage, is to move the initial price point up.

So let's say a rocket in fact costs SpaceX $60m. 70% of that (first stage costs) equals $42m. That leaves $18m for the 2nd stage and the fairing. Since we know the fairing costs $6m, that leaves us $12m for the second stage. Which seems a bit more realistic.

So in my view, SpaceX is currently making a very tiny profit of maybe $2m per flight, but 1st stage reuse has the potential to drop their cost per launch by most of the 1st stage cost, other than refurbishment. So let's say by around $40m. So from $60m, their launch cost drops to around $20m. And potentially closer to $15m, if they can achieve fairing reuse, which they are now attempting.

That leaves 2nd stage recovery, which quite frankly is likely only achievable on Falcon Heavy's, with some major payload sacrifice. True full reusabilty will likely have to wait for ITS. Which will REALLY change the game completely.

That aside, the current achievement is enormous. If they can drop the cost per launch to around $20m (from the current $60m), that puts anyone other than Blue Origin's (as yet non-operational) rocket out of the business.

[SpaceChampion]

Yeah, it could be that, but I actually didn't originally assume 50% profit when I worked it out, just used that post-facto to do the calculations in the opposite way as a check.  I just noted Elon said the price drop would be kept the same as the cost savings, so the first stage had to cost around $19M minus extra costs for recovery and whatever refurb they do (which seems to be very little), since the price dropped $19M for SES.   And then I reasoned out the rest from there.

Competitors have said a lot of things that have turned out not to be true.  I highly doubt the claim of selling at a loss.

SpaceX however has always said their plans for Falcon 9 and Falcon Heavy did not depend on reusability.  It sure makes it sweet though.  ITS on the other hand entirely depends on the reusability factor, so that's why SpaceX pushed it.

But I think we've worked out the upper and lower bounds on the price, so the important thing is what this means going forward.  It shouldn't take more than 2 years to amortize their costs, in another 30-40 flights.  SpaceX will have a strong base to build their future plans on.

[SpaceChampion]

A wonderful chart from here displaying some of the physical characteristics of several Falcon 9 flight in comparison.

[SerHaHa]

Anyone else see the triangle lights + 1 forward light object appear in the rocket camera, and then they cut the feed right away?  The same object has been seen on video from the ISS - sure didn't look like "ice crystals" or "space junk", and was identical to previous objects seen on ISS feeds before, and there is proof of that on the net already.  Could be some classified observation/recon craft, like the x37 but actually classified in existence, and not just mission.  I'm not going to sully the thread with links, it's mostly the UFO types mentioning this, but I did find it interesting that the shape of lights/objects seen in the launch a couple days ago were identical to other light formations seen from the ISS cameras, and the feeds are always cut within a second of them showing up.

 

Such an amazing achievement, the faster SpaceX gets certified fully for manned space flight, the better.

[SpaceChampion]

The listed performance of Falcon Heavy to Low Earth Orbit has been upgraded by almost 10 metric tonnes to a total lift capacity of 64mT (if used as expendable).  That's about a 17.64% improvement, which is apparently due to structural improvements.

Also, a little bit more improvement to 70 mT (if that is at all possible) would officially cross it into the "super-heavy lift" class of rockets.

I think SpaceX will probably not fly it in expendable mode however, except when it's on it final flight (aiming for 99 reuses per booster).  With creative scheduling, they could launch their own sats 99 times and then sell the final expendable flight to a comsat for GEO orbit, to the Moon, Mars, or even the outer solar system.

Here's the landing video for the SES-10 mission.  It's on a loop, but it's three different views of the same landing.

 

[SerHaHa]

Those return landing videos are even more impressive when there is some scale to them - it's easy to underestimate just how big those things are from the videos.  Googling some pictures of people standing next to them, securing them after they've landed and cooled off/whatever - it's really an amazing thing something so large can do this.  What a fine balance between thrust/gravity those engines must keep to set something so large with so much mass down gently enough to not wipe out.

[SpaceChampion]

Found some numbers on how improved Falcon Heavy will be over its spec from prior to these latest improvenments:

Quote

Old / New numbers. (January 2017 / April 8 (today)) [expendable flights]

* LEO = 54,400kg / 63,800kg (+17%)
* GTO = 22,200kg / 26,700kg (+20%)
* Mars = 13,600kg / 16,800kg (+23%)
* Pluto = 2,900kg / 3,500kg (+20%)

Next flight of F9 is an NRO sat on NET April 30th, delayed a few weeks due to unspecified issues with the unknown payload.

 

Also, Italy could be sending a payload to Mars with SpaceX instead of the European Union's heavily subsidized rocket company Arianespace.

 

[SpaceChampion]

Launch scrubbed for today with caution regarding sensor issue.  Tomorrow, 7AM attempt to launch to NROL sat for the NRO.

 

Musk has an update to give regarding the Interplanetary Transport System that he says make the economics work out even better.  The public timeline for ITS is 8-10 for development, but Musk has said their internal timeline is much more aggressive.  Does that mean 6-8 years?  We'll never know until it's done.

Despite articles claiming SpaceX requires taypayer funding for the ITS, they don't.  $10 billion is an amount that SpaceX can generate itself from commercial contracts and investors, and if they successfully pull off their internet satellite constellation, it could be entirely self-funded.  SpaceX president Gwynne Shotwell has to be publicly more cautious than Musk,  She has said that they can do it in 10-15 years on their own, but with NASA help that would speed up to between 8-10.  SpaceX does want to continue providing services for NASA to send payloads any where in the solar system it wants, including Mars, so has to be mindful of not looking even more cocky than it does.  Once launch costs are reduced to a tiny percent of what they currently  are, the cost of developing payloads will be the limiting factor for all future space exploration & settlement.

[SpaceChampion]

Flip and burn:

 

Short webcast of the launch and landing:

 

[Altherion]

Here are some more details on SpaceX's satellite internet plans:

Quote

SpaceX today said its planned constellation of 4,425 broadband satellites will launch from the Falcon 9 rocket beginning in 2019 and continue launching in phases until reaching full capacity in 2024.

SpaceX gave the Senate Commerce Committee an update on its satellite plans during a broadband infrastructure hearing this morning via testimony by VP of satellite government affairs Patricia Cooper.

...

Cooper told senators:

"Later this year, SpaceX will begin the process of testing the satellites themselves, launching one prototype before the end of the year and another during the early months of 2018. Following successful demonstration of the technology, SpaceX intends to begin the operational satellite launch campaign in 2019. The remaining satellites in the constellation will be launched in phases through 2024, when the system will reach full capacity with the Ka- and Ku-Band satellites. SpaceX intends to launch the system onboard our Falcon 9 rocket, leveraging significant launch cost savings afforded by the first stage reusability now demonstrated with the vehicle."

The 4,425 satellites will "operat[e] in 83 orbital planes (at altitudes ranging from 1,110km to 1,325km)," and "require associated ground control facilities, gateway Earth stations, and end-user Earth stations," Cooper said. By contrast, the existing HughesNet satellite network has an altitude of about 35,400km, making for a much longer round-trip time than ground-based networks.

This differs from existing satellite internet in that these will not be in geostationary orbit. The good news about that is that the latency is dramatically lower (SpaceX hopes for 25ms vs. 600ms). The bad news is that no individual satellite is locked to a specific point on Earth so you need a whole lot of satellites and a mesh network. It will be awesome if they can make this work.

[SpaceChampion]

SpaceX did a successful static fire of a newly manufactured 1st stage today, so next launch is scheduled for the 15th, a 6-tonne commercial sat Inmarsat 5/F4.  No recovery of the first stage as this is a heavy beast.  Originally this was suppose to be a Falcon Heavy launch booked many years ago, but with the upgrades to F9 it can now loft it in expendable mode.

[Altherion]

Reddit has a transcript of a long interview with Tom Mueller, the CTO of SpaceX. There's a lot of interesting stuff there. For example:

Quote

The Block 5 Falcon rocket that we’re rolling out later this year is going to have a reusable thermal protection on it; so we don’t burn up the heat shielding on it. And it’s going to have a much better landing legs that just fold up and; just drop the rocket, fold the legs, ship it, fold the legs out when it lands. Making it turn very fast; our goal is; Elon asked us to do a twelve-hour turn. And we came back and said without some major redesigns to the rocket, with just the Block 5, we can get to a 24-hour turn, and he accepted that. A 24-hour turn time. And that doesn’t mean we want to fly the rocket, you know, once a day; although we could, if we really pushed it. What it does is, limits how much labor, how much <touch?> labor we can put into it. If we can turn a rocket in 24 hours with just a few people, you’re nuts. <inaudible> low cost, low opportunity cost in getting the rocket to fly again.

So the 24 hour turn around time is not some aspirational goal; the engineers actually looked at it and think they can do it with the Block 5 version.

[Ser Scot A Ellison]

I'm going to watch tonight's launch in Orlando.

[Ded As Ned]

8 minutes ago, Ser Scot A Ellison said:

I'm going to watch tonight's launch in Orlando.

In person?  Sweet!