Author Topic: Would it have been a good idea for FH to be like Angara (i.e. 5-core)?  (Read 39492 times)

Offline Lemurion

I just don't see the proposed F5x5 as being anything more than a misstep in comparison to FH.

Five smaller cores means all the necessary components to run each core are duplicated five times instead of three; smaller cores also run afoul of the cube-square law so they're most likely proportionately heavier for the volume of propellant carried.

F5x5 also has fewer engines than FH, meaning less total thrust.

I don't see the benefits of increasing complexity to reduce performance.

Offline fredinno

  • Member
  • Posts: 20
  • Liked: 0
  • Likes Given: 5
I just don't see the proposed F5x5 as being anything more than a misstep in comparison to FH.

Five smaller cores means all the necessary components to run each core are duplicated five times instead of three; smaller cores also run afoul of the cube-square law so they're most likely proportionately heavier for the volume of propellant carried.

F5x5 also has fewer engines than FH, meaning less total thrust.

I don't see the benefits of increasing complexity to reduce performance.
So you'd rather have 2 non-modular rockets, instead? (likely meaning 2 rocket diameters)

Offline Pipcard

  • Full Member
  • ****
  • Posts: 622
  • Liked: 275
  • Likes Given: 130
I just don't see the proposed F5x5 as being anything more than a misstep in comparison to FH.

Five smaller cores means all the necessary components to run each core are duplicated five times instead of three; smaller cores also run afoul of the cube-square law so they're most likely proportionately heavier for the volume of propellant carried.

F5x5 also has fewer engines than FH, meaning less total thrust.

I don't see the benefits of increasing complexity to reduce performance.
So you'd rather have 2 non-modular rockets, instead? (likely meaning 2 rocket diameters)
No, they'd rather have a 3-core instead of a 5-core for equivalent capacity, it's the
trade-off between modularity/commonality and reduced complexity.
« Last Edit: 06/14/2016 06:40 pm by Pipcard »

Offline fredinno

  • Member
  • Posts: 20
  • Liked: 0
  • Likes Given: 5
Quote
And yet you suggested creating a complex set of various stages and an H2 stage on a Falcon rocket earlier. You've completely ignored all of SpaceX's strong points and made suggestions that all run counter to them.
I suggested it, AS A POSSIBILITY.
I wasn't ENDORSING IT.
(though to be fair, Blue Origin is using Ch4 and H2 on its orbital rocket- so it's not a completely stupid idea, and definitely worth looking into).
I don't think its the best of ideas (see below for reasoning), though (unless you have a solid lower stage config.)

A better idea is probably to have a 6-core config, to capture that last stretch of payload capacity from 5.5 to 7T to GTO.
The last stretch from 7-10 to GTO would need an expendable version of the rocket, or a 7-core config.
(7 core is impossible without redesigning tank structure to allow for pumping fuel through the tanks instead of around- not impossible, Delta II did it, but it doesn't seem to be as cheap.)

However, that side of the payload equation only gets 1 or 2 launches, and is growing far more slowly than the small side of the payload equation. Modifications for 7-core F9 would be done if the need arises (ION drives on Satellites make that unlikely however as they vastly reduce satellite mass, see below- and potentially Space tugs in the future, with fuel launched by HLVs like MCT).

BTW, what's complex? I'd argue the 6-core variant has LESS complexity than Soyuz (considered a fairly cheap launcher), which has the same number of stages, but numerous different core diameters.
It also has less complexity than Ariane 6, with 6 staging events, but using H2.

Also, F9 isn't the pinnacle of simplicity
. Replacing the 1st stage with a single Rp-1 engine would probably reduce costs per launch, since the reuse cancels the majority of the "mass production" benefit using 9 engines on the 1st stage had.

 
Quote
Yes, a significantly less lucrative one. And a market full of "paper rockets" that never make it past development. Falcon 1 made it, and despite having invested in its design and manufacture, SpaceX ditched it, and with good reason. Investing in the Falcon 9 has been a smart long-term move. If the smallsat business was really all that lucrative, SpaceX would have kept the Falcon 1 in production.
It obviously was a good idea when Elon canned the Falcon 1 and 5. Now, it's a little different.
http://www.parabolicarc.com/wp-content/uploads/2015/02/Smallsat_chart_SpaceWorks2015.jpg
There is a vast increase in 1-50 kg satellites in the past few years, and that trend is expected to continue.

Also:http://www.satellitetoday.com/telecom/2013/11/08/new-euroconsult-report-forecasts-exciting-changes-in-the-satellite-industry/
Quote
“Many small satellites are being developed, requiring shorter development time and lower launch costs,” said Rachel Villain, principle advisor at Euroconsult. “On average each satellite is smaller at 2.1 tons (vs. 2.2 Tons). This average aggregates a growing number of small satellites (< 500 kg) from newcomer countries and smaller replacement satellites in LEO in established space countries.”
Miniaturization is causing smaller satellites to become more and more viable, with more and more launches of these satellites expected.

F1 is still abandoned right now, because reviving it would bring it in competition with 3 other operators. However, F1 capability could have been kept under a modular F9 solution, and thus F1 would have a competitive edge. They would been the "innovators" of the smallsat launch industry- and industry which will quickly escape from SpaceX's grasp.

Also, it would have made replacing satellites in SpaceX's internet sat constellation much cheaper, without having to launch a dedicated F9 for a few replacement spares.

SpaceX might be avoiding costs now by ignoring the smallsat market, but the more it grows, the more that decision will bite SpaceX's rear. Considering it's expected to get pretty big, I would have taken the market before everyone else did, meaning SpaceX would get the max. $$ from that market.

It's easier to keep a market than it is to gain it. But that's just my 2 cents.

Offline fredinno

  • Member
  • Posts: 20
  • Liked: 0
  • Likes Given: 5
I just don't see the proposed F5x5 as being anything more than a misstep in comparison to FH.

Five smaller cores means all the necessary components to run each core are duplicated five times instead of three; smaller cores also run afoul of the cube-square law so they're most likely proportionately heavier for the volume of propellant carried.

F5x5 also has fewer engines than FH, meaning less total thrust.

I don't see the benefits of increasing complexity to reduce performance.
So you'd rather have 2 non-modular rockets, instead? (likely meaning 2 rocket diameters)
No, they'd rather have a 3-core instead of a 5-core for equivalent capacity, it's the
trade-off between modularity/commonality and reduced complexity.

How do you know?

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 37441
  • Cape Canaveral Spaceport
  • Liked: 21451
  • Likes Given: 428
Considering it's expected to get pretty big, I would have taken the market before everyone else did, meaning SpaceX would get the max. $$ from that market.


Where is the data to support that claim?   Spacex is ignoring it because there is no money in it.

Offline fredinno

  • Member
  • Posts: 20
  • Liked: 0
  • Likes Given: 5
Considering it's expected to get pretty big, I would have taken the market before everyone else did, meaning SpaceX would get the max. $$ from that market.


Where is the data to support that claim?   Spacex is ignoring it because there is no money in it.
I already gave you that link. Look above. If not, here again:

http://www.euroconsult-ec.com/shop/space-industry/64-prospects-for-the-small-satellite-market.html
http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3018&context=smallsat

There IS money in it. Even SpaceX knows, they got that 27 satellite or so Oorbcomm contract for F1 right before they killed that rocket.

The cumulative weight of all those satellites would have fit on a Delta II. In other words, F9 was way too big for them.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 37441
  • Cape Canaveral Spaceport
  • Liked: 21451
  • Likes Given: 428

1.  http://www.euroconsult-ec.com/shop/space-industry/64-prospects-for-the-small-satellite-market.html
http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=3018&context=smallsat

2.  There IS money in it. Even SpaceX knows, they got that 27 satellite or so Oorbcomm contract for F1 right before they killed that rocket.


1.  What says that is a relevant source?

2.  Not really an example.  That is only one constellation and it has existed for years.  That does not mean it is a trend.   There are other choices for them, they didn't have to stick with Spacex.
« Last Edit: 06/14/2016 09:11 pm by Jim »

Offline Pipcard

  • Full Member
  • ****
  • Posts: 622
  • Liked: 275
  • Likes Given: 130
How do you know?
They directly said that the three-core FH was preferable to a five-core equivalent because it was less complex. But the three-core also benefits from modular commonality. Thus, it is a trade-off/compromise. They did not say they wanted a different core diameter.

BTW, what's complex? I'd argue the 6-core variant has LESS complexity than Soyuz (considered a fairly cheap launcher), which has the same number of stages, but numerous different core diameters.
It also has less complexity than Ariane 6, with 6 staging events, but using H2.

Also, F9 isn't the pinnacle of simplicity
. Replacing the 1st stage with a single Rp-1 engine would probably reduce costs per launch, since the reuse cancels the majority of the "mass production" benefit using 9 engines on the 1st stage had.
Soyuz only has 4 liquid rocket boosters, Ariane 6 has a maximum of 4 solid boosters (but solid rocket motors have a relatively simpler design compared to liquid rocket boosters), you are proposing something with 5 or 6 liquid boosters.

And if reuse cancels out the mass production benefit, it would be canceled even further if there was a single engine per core. Clustered engines with one in the center also makes throttling the t/w ratio when landing easier. It can also share common components with the second stage engine.
« Last Edit: 06/14/2016 11:52 pm by Pipcard »

Offline Toast

...A better idea is probably to have a 6-core config, to capture that last stretch of payload capacity from 5.5 to 7T to GTO.... or a 7-core config...

You've officially jumped the shark here. The existing Falcon 9 (NOT Falcon Heavy) can lift 8.3 mT to GTO. You're talking about a six to seven core rocket to get the same performance. That is a laughably bad idea.

BTW, what's complex? I'd argue the 6-core variant has LESS complexity than Soyuz (considered a fairly cheap launcher), which has the same number of stages, but numerous different core diameters.

Soyuz is reliable in large part because it's a design that has fifty years of launch history behind it that's allowed all the kinks to be worked out. Advocating a complexly staged vehicle when an alternate with fewer stages and similar performance is ill advised. Russia keeps using Soyuz because they already have the manufacturing infrastructure and a robust design, starting from the ground up would be expensive. Note that you don't see any other launch providers copying Soyuz's iconic design, and not without reason.

Also, F9 isn't the pinnacle of simplicity. Replacing the 1st stage with a single Rp-1 engine would probably reduce costs per launch, since the reuse cancels the majority of the "mass production" benefit using 9 engines on the 1st stage had.

Sure, I never said it was. The key is that any decision that removes simplicity must balance that loss out elsewhere. The Falcon 9 uses multiple engines because it improves reliability via engine-out capability and because it enables the deep-throttling needed for propulsive landing. Recovering the stages offsets the increased complexity. Creating a six-core rocket does not provide enough offsetting positive factors (or any, in my opinion) to compensate for the added infrastructure, testing, reliability and logistics costs.

Also, it would have made replacing satellites in SpaceX's internet sat constellation much cheaper, without having to launch a dedicated F9 for a few replacement spares.

From what we know of SpaceX's intentions with their satellite cluster, the goal is to launch mass numbers at once. They have no intentions or need for replacing individual satellites when they are expected to launch to a very low orbit and replace satellites very frequently.

Offline Pipcard

  • Full Member
  • ****
  • Posts: 622
  • Liked: 275
  • Likes Given: 130
You've officially jumped the shark here. The existing Falcon 9 (NOT Falcon Heavy) can lift 8.3 mT to GTO. You're talking about a six to seven core rocket to get the same performance. That is a laughably bad idea.
Again, just to clarify his position, all his estimated launch capacities are for reusable mode, not expendable mode.

Offline Lemurion

I just don't see the proposed F5x5 as being anything more than a misstep in comparison to FH.

Five smaller cores means all the necessary components to run each core are duplicated five times instead of three; smaller cores also run afoul of the cube-square law so they're most likely proportionately heavier for the volume of propellant carried.

F5x5 also has fewer engines than FH, meaning less total thrust.

I don't see the benefits of increasing complexity to reduce performance.
So you'd rather have 2 non-modular rockets, instead? (likely meaning 2 rocket diameters)
No, they'd rather have a 3-core instead of a 5-core for equivalent capacity, it's the
trade-off between modularity/commonality and reduced complexity.

How do you know?

Because that's what I said.

The five-core rocket assembly is more complex and has less total thrust than the three-core assembly. I also think that the combination of smaller fuel tanks, and additional components would result in a worse thrust to weight ratio for the F5-based assembly than the F9-based Falcon Heavy.


Offline Toast

You've officially jumped the shark here. The existing Falcon 9 (NOT Falcon Heavy) can lift 8.3 mT to GTO. You're talking about a six to seven core rocket to get the same performance. That is a laughably bad idea.
Again, just to clarify his position, all his estimated launch capacities are for reusable mode, not expendable mode.

Even fully recovered, the Falcon 9 can launch almost that amount of mass to GTO. They've already demonstrated that they can handle 4.7mT to GTO and recover, and they're expecting to add another ton to that later this year with a thrust upgrade. That's not even touching the Falcon Heavy. I am strongly of the opinion that a n x F5 core modular rocket is a terrible idea. SpaceX ditched the F5 and F1 for a reason, and modular multi-core rockets are nowhere near as simple as fredinno wants to believe.

Offline fredinno

  • Member
  • Posts: 20
  • Liked: 0
  • Likes Given: 5
Ok, what is so complex about multi-core configs?
None of it is unknown. Really, it's not even crossfeed.
5 cores are burning independently (potentially, the core might actually be started in flight- something required for SpaceX reuse anyways), with structural supports to the core, and a separation mechanism.

Whoopie, worst case scenario, the rocket is a bit more costly initially to develop. But it benefits from mass production of the cores, especially at the rosy regular launch scenarios Elon wants.

It also provides a back in case reuse doesn't pan out for some reason, and ends up being less economical than predicted.

Right now, the F9 only really hits the 4-5T to GTO market, and the big, 6-8T to GTO market. Meanwhile, literally the entire LEO and MEO market needs to waste fuel and payload to launch on Falcon 9.

Including Dragon, SpaceX's own vehicle. Reusable F9 carries ~13T to LEO under barge landings (slow).  Dragons max out at 10T.

I think I can safely conclude that I come from a very different community. This discussion is pretty much going in circles, and I don't want that to happen again, like with Pipcard a while ago.

I'm not convincing you guys, and you guys aren't convincing me.

I'm going to agree to disagree. No one will truely know until we get a dedicated report to it. Hopefully we do someday.

Offline IainMcClatchie

  • Full Member
  • ***
  • Posts: 394
  • San Francisco Bay Area
  • Liked: 279
  • Likes Given: 411
The advantage of a smaller rocket that can be configured with more cores is that you have more configurations that can be better matched to the mass and orbit requirements.

SpaceX has this too, except they configure the recovery rather than the rocket.  RTLS launches less mass more cheaply than barge landing.

The next big cost savings will come from reducing the amount of testing when reusing a first stage.  My guess is that to reduce testing, you have to reduce handling.  For instance, if you don't strip the legs off the booster in port, you eventually won't have to confirm they were reinstalled correctly before launch.

Offline hkultala

  • Full Member
  • ****
  • Posts: 1199
  • Liked: 748
  • Likes Given: 945
Ok, what is so complex about multi-core configs?

1) Designing the cores so that they(or practically the center core) can carry the load of the vehicle, and drag forces etc while being accelerated by the powerful engines.

In Kerbal you do't have to care about this or you just add more struts, in real life it's not so simple.

The loads also mean that the cores are NOT identical OR that the side cores are much heavier than they need to be

2) Putting tho multiple cores together before the launch.
Just integrating two stages horizontally is qute easy/cheap. integrating 2 boosters in addition to these requires bigger hall and much more equiment, but it can till be done on 2D for reasonable price.
Integrating 5 core on +-fomation makes this a 3D job, much more complex lifters etc are needed

3) Also the pad and the erector needs to be more complex and expensive to be




Quote
None of it is unknown. Really, it's not even crossfeed.
5 cores are burning independently (potentially, the core might actually be started in flight- something required for SpaceX reuse anyways), with structural supports to the core, and a separation mechanism.

Starting the center core at mid-air would be even stupider. Maximum thrust is needed at liftoff and immediately after liftoff to make gravity losses smaller.

Benefits of parallel staging are 1) Better thrust at liftoff 2) Ability to ground-start the engine of the higher stage. Pitfalls of parallel staging is higher drag and a comphromize nozzle which cannot be optimized for vacuum usage to prevent flow separation at sea level.

By having parallel stage but starting the center core at mid-air you lose all the benefits or parallel staging while still getting all the pitfalls (assuming you have similar engines with similar nozzles on all cores, and you have to have that because there is no space for multiple vacuum nozzles)

Quote

Whoopie, worst case scenario, the rocket is a bit more costly initially to develop. But it benefits from mass production of the cores, especially at the rosy regular launch scenarios Elon wants.


Wrong, as I explained. Either your booster cores weight much more than necessary, OR you have two or three different kind of boosters (separate core, 2-booster core, 4-booster core)

Quote
It also provides a back in case reuse doesn't pan out for some reason, and ends up being less economical than predicted.

Right now, the F9 only really hits the 4-5T to GTO market, and the big, 6-8T to GTO market. Meanwhile, literally the entire LEO and MEO market needs to waste fuel and payload to launch on Falcon 9.

Fuel costs are less than one percent of the launch costs. Saving fuel by making the launch vehicle much more costly and expensive does not make any sense.

And "unused capasity" is not wasted. It makes the rocket much safer. For example, when a engine breaks down, the T/W  of the rocket gets worse and the rocket has to fight gravity longer, meaning worse gravity losses. So in case of engine failure, this "wasted capasity" is used to prevent accident and deliver the payload to correct orbit. And if there is more extra capasity, the rocket can survive loss of multiple engines.


And F9 is still the cheapest laucher on the market for 3-4 and 5-6 tonne to GTO market.
« Last Edit: 06/15/2016 07:02 am by hkultala »

Offline Lemurion

Ok, what is so complex about multi-core configs?
None of it is unknown. Really, it's not even crossfeed.
5 cores are burning independently (potentially, the core might actually be started in flight- something required for SpaceX reuse anyways), with structural supports to the core, and a separation mechanism.

Whoopie, worst case scenario, the rocket is a bit more costly initially to develop. But it benefits from mass production of the cores, especially at the rosy regular launch scenarios Elon wants.

It also provides a back in case reuse doesn't pan out for some reason, and ends up being less economical than predicted.

Right now, the F9 only really hits the 4-5T to GTO market, and the big, 6-8T to GTO market. Meanwhile, literally the entire LEO and MEO market needs to waste fuel and payload to launch on Falcon 9.

Including Dragon, SpaceX's own vehicle. Reusable F9 carries ~13T to LEO under barge landings (slow).  Dragons max out at 10T.

I think I can safely conclude that I come from a very different community. This discussion is pretty much going in circles, and I don't want that to happen again, like with Pipcard a while ago.

I'm not convincing you guys, and you guys aren't convincing me.

I'm going to agree to disagree. No one will truely know until we get a dedicated report to it. Hopefully we do someday.

Part of the complexity lies in operations:

SpaceX fuels at the pad right before liftoff. For the FH, they just run 2 additional fueling systems in a line right beside the center one. For a F5x5 they would need to run 2  more, and the plumbing would be much more complex, especially for the center core. That means a more expensive launch facility.

Then there's the landing. Five boosters means five landing sites, and right now SpaceX has two.

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39271
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Why do these kinds of threads always go on so long?
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Pipcard

  • Full Member
  • ****
  • Posts: 622
  • Liked: 275
  • Likes Given: 130
The question fredinno was always wondering about was, "if it's so complex and expensive to integrate and fuel a five core launch vehicle, why didn't that stop the development of Angara-5?"
« Last Edit: 06/15/2016 03:10 pm by Pipcard »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 37441
  • Cape Canaveral Spaceport
  • Liked: 21451
  • Likes Given: 428

I'm not convincing you guys, and you guys aren't convincing me.

I'm going to agree to disagree. No one will truely know until we get a dedicated report to it. Hopefully we do someday.
there is no such thing as " agree to disagree".  That is what the party in the wrong usually say.


Right now, the F9 only really hits the 4-5T to GTO market, and the big, 6-8T to GTO market. Meanwhile, literally the entire LEO and MEO market needs to waste fuel and payload to launch on Falcon 9.

Who cares?  Propellant is the less expensive commodity on a launch vehicle.  If the F9 is cheapest out there then it doesn't if it is "oversized".   The point isn't to build the most optimally performing vehicle but to operate the least costly vehicle.  And if this vehicle flies missions missions that don't use all the capability, it doesn't matter because it allows for cost optimization.

F9 is sized so it can fly the most missions in the most optimal and hence cheapest configuration, the single core stick.  It can cover the lucrative market of GSO comsats as well as most NASA and DOD missions.  And hence with this vehicle and its lower costs, it can be used for smaller spacecraft and not take a loss.

« Last Edit: 06/15/2016 03:37 pm by Jim »

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0