Who will compete with SpaceX? The last two and next two years.

[envy887]

OneWeb is expecting to pay $3.5B for 900 satellites delivered, and only $450M of that is the satellites themselves. $10B for 4425 satellites isn't outrageous, particularly since SpaceX can launch them at cost on the cheapest available LV.

[Lar]

My guess at that split of 2.5M is ~.5M per bird as built cost and 2M per bird launch/checkout. I'm willing to bet on that.

[envy887]

And you believe it?

How much would you have paid for a Merlin class turbopump before 2008, or the best 200,000 pound class kerlox engine ever made?

What says it is.

Best TWR ever, 100% engine flight success rate on 300+ engines, restartable, reusable, and actually reused. What says it isn't?

[AncientU]

And you believe it?

Yes.

And do you believe that their price will rise to match current market prices of Boeing, LM, et al as you did (or still do) the launch vehicles?

[gongora]

My guess at that split of 2.5M is ~.5M per bird as built cost and 2M per bird launch/checkout. I'm willing to bet on that.

Will be interesting to see how that plays out.  The more you move production of overpriced components in-house the cheaper you can build your constellation, but the longer it takes to design all of those parts.  I don't think they build the constellation if they can't get the price under $2M per satellite.  Would be nice to know how many of them fit in a F9 fairing.

[Jim]

And you believe it?

Yes.

And do you believe that their price will rise to match current market prices of Boeing, LM, et al as you did (or still do) the launch vehicles?

yes, and they are climbing

[Jim]

 100% engine flight success rate

that is wrong

[Lar]

We're a bit off topic, and we mods haven't been helping, sorry. Maybe we should move the satellite cost discussions to  "SpaceX, now a satellite vendor?" https://forum.nasaspaceflight.com/index.php?topic=36552 ??

[Lar]

 100% engine flight success rate

that is wrong

Wrong.

He worded it carefully to be able to claim 100%, I think  
There was an engine failure, true, but the primary mission was a success

(queue up Ed saying "no, the mission wasn't a success because a secondary payload wasn't 100% successful').

And do you believe that their price will rise to match current market prices of Boeing, LM, et al as you did (or still do) the launch vehicles?

yes, and they are climbing

Wrong.

Earlier in the thread it was clearly shown that adjusted for inflation SpaceX prices are declining.

Jim's right on this one. Failure of an engine on CRS-1.

Wrong.

Envy didn't say 100% engine success, he said engine FLIGHT success. He weaseled, and is therefore correct. (although wrong in spirit)

Wow, I'm on a roll today!

[Jim]

We're a bit off topic, and we mods haven't been helping, sorry. Maybe we should move the satellite cost discussions to  "SpaceX, now a satellite vendor?" https://forum.nasaspaceflight.com/index.php?topic=36552 ??

Yeah, just call this the "I believe all Spacex spin" thread

[Space Ghost 1962]

 100% engine flight success rate

that is wrong

Jim's right on this one. Failure of an engine on CRS-1.

[Space Ghost 1962]

We're a bit off topic, and we mods haven't been helping, sorry. Maybe we should move the satellite cost discussions to  "SpaceX, now a satellite vendor?" https://forum.nasaspaceflight.com/index.php?topic=36552 ??

Yeah, just call this the "I believe all Spacex spin" thread

Sounds like sour grapes.

[envy887]

 100% engine flight success rate

that is wrong

Jim's right on this one. Failure of an engine on CRS-1.

That was a M1C, which wasn't close to 200 klbf class (more like 85-110k) and was a quite different engine.

M1D is 100% on 300+ flights

[Robotbeat]

 100% engine flight success rate

that is wrong

Not wrong for Merlin 1D, the only engine made entirely in house.

[ZachF]

And you believe it?

How much would you have paid for a Merlin class turbopump before 2008, or the best 200,000 pound class kerlox engine ever made?

What says it is.

Best TWR ever, 100% engine flight success rate on 300+ engines, restartable, reusable, and actually reused. What says it isn't?

And cheap...

[Lar]

My guess at that split of 2.5M is ~.5M per bird as built cost and 2M per bird launch/checkout. I'm willing to bet on that.

Will be interesting to see how that plays out.  The more you move production of overpriced components in-house the cheaper you can build your constellation, but the longer it takes to design all of those parts.  I don't think they build the constellation if they can't get the price under $2M per satellite.  Would be nice to know how many of them fit in a F9 fairing.

see https://forum.nasaspaceflight.com/index.php?topic=36552.msg1654158#msg1654158 ...

Just relocated these old quotes from EM on number of launches to support a 4,025 sat (early days) constellation and mass of sats:

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

and

"Musk: Smaller satellites, few hundred kg, but capability of much larger satellites."
- my guess is optimum size is somewhere in 300-600 kg range, driven mainly by antenna size.

https://forum.nasaspaceflight.com/index.php?topic=36552.msg1317801#msg1317801

'Hundreds of launches' -- 4000 sats/200 launches = 20/launch
'50 per year' -- 4000 sats, 4 year lifetime --> 20/launch
'300-600kg' (450kg nominally) 6t-12t (9t nominally) payload without dispenser, so F9 has sufficient capability.

Appears the original concept was F9, lots of launches per year to keep 4,000 sat ConnX aloft. 
 

So 20.

[abaddon]

 100% engine flight success rate

that is wrong

Not wrong for Merlin 1D, the only engine made entirely in house.

To pile on here, the failure on the 1C engine was due to a manufacturing flaw introduced by a process that doesn't exist on the 1D.  There have been no 1D failures in flight to date.

[Robotbeat]

RD-117/118 (107/108) can probably be considered more reliable just due to longer flight history and more consecutive launchers between failures. Last time one failed was in 2002, due to debris ingestion. Five engines on every launch, so is that about 800 consecutive successes lately?

Lots of earlier failures, of course, so Merlin 1D maybe about the same or a little better overall. But hard to tell.

Of course, M1D has much higher performance which means it can have fewer stages, which should improve reliability overall.

[Robotbeat]

My guess at that split of 2.5M is ~.5M per bird as built cost and 2M per bird launch/checkout. I'm willing to bet on that.

Will be interesting to see how that plays out.  The more you move production of overpriced components in-house the cheaper you can build your constellation, but the longer it takes to design all of those parts.  I don't think they build the constellation if they can't get the price under $2M per satellite.  Would be nice to know how many of them fit in a F9 fairing.

see https://forum.nasaspaceflight.com/index.php?topic=36552.msg1654158#msg1654158 ...

Just relocated these old quotes from EM on number of launches to support a 4,025 sat (early days) constellation and mass of sats:

"Musk: Talking about around 4000 satellites. 4025 exactly in current design."
- that is probably 100s of launches, maybe 50/year.

and

"Musk: Smaller satellites, few hundred kg, but capability of much larger satellites."
- my guess is optimum size is somewhere in 300-600 kg range, driven mainly by antenna size.

https://forum.nasaspaceflight.com/index.php?topic=36552.msg1317801#msg1317801

'Hundreds of launches' -- 4000 sats/200 launches = 20/launch
'50 per year' -- 4000 sats, 4 year lifetime --> 20/launch
'300-600kg' (450kg nominally) 6t-12t (9t nominally) payload without dispenser, so F9 has sufficient capability.

Appears the original concept was F9, lots of launches per year to keep 4,000 sat ConnX aloft. 
 

So 20.

20 per F9. Original concept was to replace each bird every 4 years. That's a consistent 50 F9 launches per year on top of probably 20-40 additional launches.

But the constellation has gotten bigger, 12000 birds. That'd require 150 F9 launches per year (total of nearly 200 F9 launches per year). And they say they expect the satellites to grow, so let's say 300 or 600 F9 launches. 1500 F9 launches if they become the size of typical GSO birds. Mueller imagines larger satellites, so multiple thousands.


Short of it: even their short-term constellation makes reuse not just worthwhile but essential. And long-term their constellation goals far outgrow the Falcon family.

Which is partly the point: by setting the constellation goals toward something that would be much too great for Falcon, they can justify using outside investor capital for ITS development. And that's the key take away, even if you don't think they'll ever be able to grow their constellation that big:

By setting such a goal, they turn a fun-money pipe dream rocket into something that could be funded by investment capital.

[Space Ghost 1962]

 100% engine flight success rate

that is wrong

Jim's right on this one. Failure of an engine on CRS-1.

That was a M1C, which wasn't close to 200 klbf class (more like 85-110k) and was a quite different engine.

M1D is 100% on 300+ flights

Suggest that the inconsistent subjective context here opens up ambiguity that some attempt to drive a truck through.

Yes M1D is 100% AFAIK. But "Merlin" ... isn't. And yeah its kinda silly.

After a while its a "two bit" behavior. And the constant Musk version of the Simpson's "Malibu Stacy has a new hat" does get on the nerves, like jeesh it's a rocket of course and why should it matter for just about any little thing.

Things that matter though climb above this.