A Space & astronomy forum. SpaceBanter.com

Go Back   Home » SpaceBanter.com forum » Space Science » Policy
Site Map Home Authors List Search Today's Posts Mark Forums Read Web Partners

Were liquid boosters on Shuttle ever realistic?



 
 
Thread Tools Display Modes
  #1  
Old October 21st 17, 05:36 AM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.


If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?


Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.


Better ISP means more thrust per pound of fuel, does it not?


Yes.

Or is there something magical which makes LH2 engines perform not as
well at sea level? Isn't the difference in tuning engines for sea level
vs vacuum more the engine bell shape/size?


It is very difficult to pump enough of LH2 to the combustion
chamber in a very short time. You end up needing more engines
or bigger engines to have enough thrust.


RS-68A seems sufficiently 'high thrust' at over 700,000 lbs of thrust.
Using two of them gives you F-1 levels of thrust with almost half
again the Isp. Two of them will weigh (dry weight) about half again
what an F-1 weighs, but that difference is lost in the noise when you
look at the difference in fuel weight you get from the Isp advantage.

Does 14.7 difference in PSI outdoors make a difference when it comes to
the turbines and combustion chambers which are assume are a tad more
pressuzised ?

is the only drawback of LH2 engines (from engineering) the aerodynamics
of the bigger tanks, which for the first stage, must endure max Q ?


The bigger tanks giving greater aerodynamic pressure is a problem. But
it is not as important as the problem of getting enough thrust. You
can get enough thrust, but it will cost you dearly.


Or not so much. RS-68A cost is significantly less than SSME and it's
designed to be thrown away after one use. It gives up some
performance over the SSME but the high thrust and lower cost make that
more than worthwhile. An RS-68A costs about 20% of what an RS-25
costs.

As for aerodynamics, R.H. Coates, lead propulsion engineer for SLS,
seems to disagree with you. When asked why RP-1/LOX is better for
first stage he said, "Refined petroleum is not the most efficient
thrust-producing fuel for rockets, but what it lacks in thrust
production it makes up for in density. It takes less volume of RP-1 to
impart the same thrust force on a vehicle, and less volume equates to
reduced stage size. A smaller booster stage means much less
aerodynamic drag as the vehicle lifts off from near sea-level and
accelerates up through the more dense (thicker) part of the atmosphere
near the earth. The result of a smaller booster stage is it allows a
more efficient ascent through the thickest part of the atmosphere
which helps improve the net mass lifted to orbit."

Given a choice between him as an authority and you, well, I'm going to
go with him.

From a cost point of view, would RP1 engines be much easier and simpler
to make or do they represent same level of complexity and thus cost of
manufacture ?


Bingo! Pumping a tonne of RP1 per second is much easier than pumping
a tonne of LH2 per second. Both because the LH2 will have a much bigger
volume and because it is very cold. To get the same thrust from a LH2
engine will cost you much more than from an RP1 engine.


Both F-1 and RS-68 are gas generator fed engines. RS-68A reportedly
costs on the order of $10-$12 million each. I don't find a cost for
what an F-1B (the current version) would cost, but I'd bet they're not
cheap. SpaceX Merlin, also a gas generator fed engine, but with much
less thrust than an F-1, costs around 20% of what an RS-68 costs. Yes,
an LH2/LOX engine is going to cost more than an RP1/LOX engine of
similar thrust, but don't let RS-25 cost lead you to exaggerate the
difference.


--
"The reasonable man adapts himself to the world; the unreasonable
man persists in trying to adapt the world to himself. Therefore,
all progress depends on the unreasonable man."
--George Bernard Shaw
  #2  
Old October 21st 17, 11:22 AM posted to sci.space.policy
Alain Fournier[_3_]
external usenet poster
 
Posts: 548
Default Were liquid boosters on Shuttle ever realistic?

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?


Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.


No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.

The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


Alain Fournier
  #3  
Old October 21st 17, 03:56 PM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

Alain Fournier wrote:

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?

Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.


No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.


Yes, you can always rig the numbers by the problem statement.

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.


It's not 'nothing'. It's more payload to orbit, which is sort of the
goal of the thing (and what you've ignored with your rigged numbers).
Explain Delta IV.


The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


And because you rigged the problem by assuming you only have a low
thrust engine that you use everywhere.


--
"Millions for defense, but not one cent for tribute."
-- Charles Pinckney
  #4  
Old October 22nd 17, 10:04 AM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

JF Mezei wrote:

On 2017-10-21 00:36, Fred J. McCall wrote:

reduced stage size. A smaller booster stage means much less
aerodynamic drag as the vehicle lifts off from near sea-level and
accelerates up through the more dense (thicker) part of the atmosphere
near the earth. The result of a smaller booster stage is it allows a
more efficient ascent through the thickest part of the atmosphere
which helps improve the net mass lifted to orbit."


While this sounds logical, the problem with the text is that it lacks
numbers or orders or magnitudes.

What is the cost of aerodynamics, and what are the weight savings from
using more efficient fuel?


Jesus, Mayfly, do SOME of your own homework. The fuel weight part is
easy once you know Isp.


And the cost of aerodynamics can vary depending on whether you use a
short and fast booster or long and skinny one.


You're going to eat yourself alive on dry weight if you make a 'long
and skinny' booster, since it takes a lot more structure to keep the
thing from breaking into pieces.


And in the case of the shuttle, what if say 25% of a booster' fuel came
from a taller (but same diametre) ET, allowing for smaller aerodynakic
drag on the smaller boster tanks ?


What if it was all powered by magic unicorn farts? You just
complicated the **** out of the engineering, what with the taller
tank, needing to worry about crossfeeding the boosters during PART of
the flight, etc.


The problem with the expert arguments as stated above is that they can
be used as spin to pusgh for one architecture without providing numbers
to support that suggestion because the numebrs may or may not give the
proposed solution such an edge.


So stop asking and believe what you like.


Also remember that if the boosters need less fuel mass, it means that
you need fewer booster engines and that also lowers fuel consumption and
thus lowers aerodynamic drag.


Uh, no, because you aren't using the same engines (or even engines
with the same thrust).


With regards to cost of engines, I am not sure it is fair to compare the
costs of the SpaceX merlin engines with cost of engines by bloated
companies who life comfortably on pork.


You're an idiot.


If SpaceX had built both types, then yeah, you could compare costs since
they both would have been made by a lean and mean company. It could be
that the cost difference would be huge, or minimal. We don't know
because that hasn't happened.


And why do you think that is?


As a newbie into the rocket business, SpaceX likely started with the
simplest and cheapest, and that means RP1. What is not known is how far
is LH2 from being competitive if it were done by SpaceX or another
company that is as lean and mean as SpaceX.


There's a reason SpaceX has no interest in LH2/LOX engines. If they
wanted to go 'simplest' they'd be using hypergolic fuels (although
they're a little ugly to handle during loading), since those engines
are MUCH simpler (which is why Draco and Super Draco use hypergolic
fuels). Note that SpaceX's next big engine is CH4/LOX rather than
LH2/LOX.

An LH2/LOX engine will always cost a lot more than an RP1/LOX engine.
This is obvious with very little thought, which I guess explains why
you don't get it. The most expensive parts of the engine are the
turbopumps. To get the same thrust out of a lower density fuel
(actually a lower density exhaust, but it correlates) you must pump a
higher volume of fuel in a given time. This means that your
turbopumps need to be bigger and more powerful (and thus more
expensive). In addition, with LH2 engines you're pumping a deeply
cryogenic fuel, which means the materials your pump is made out of
need to account for that as well as standing up to the higher power
required.

Musk is a bright guy, but he can't change the physical laws of
material science. A Merlin engine costs $2 million and change, as
near as can be determined. RD-180, which is a more aggressive design
with higher operating pressures and over four times the thrust costs
around $20 million. Before the Russians jacked up the prices they
were $10 million. That's actually for two engines hooked together,
though, so call it around $5 million or so per engine. So even a
simple LH2/LOX engine costs at least twice as much.


--
"Millions for defense, but not one cent for tribute."
-- Charles Pinckney
  #5  
Old October 22nd 17, 06:51 PM posted to sci.space.policy
Alain Fournier[_3_]
external usenet poster
 
Posts: 548
Default Were liquid boosters on Shuttle ever realistic?

On Oct/21/2017 at 10:56 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?

Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.


No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.


Yes, you can always rig the numbers by the problem statement.


Which number was rigged? The only slightly dubious thing in my
example is that I was assuming the same size for the second stage
as for the first stage, which would be unusual for a rocket. But
it doesn't change the idea of the outcome if you divide by two
the size of the tanks in the second stage of my example. It only
makes the computations a little more complicated.

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.


It's not 'nothing'. It's more payload to orbit, which is sort of the
goal of the thing (and what you've ignored with your rigged numbers).
Explain Delta IV.


In my example I was assuming the same functionality for the stages.
Meaning, the stages would put the same mass at the same altitude
and same speed whether it was the RP1 variant or the LH2 variant.

For the Delta IV, there are several variants. Let's look at the
one with one CBC as a first stage and the 5m diameter second stage.
The first stage has one RS-68A engine with a dry weight of the
engine of 6740 kg and 3137 kN thrust, the gross mass of the booster
is 226400 kg. The second stage has one RL10B-2 engine weighing
277 kg and providing 110 kN thrust, the gross mass of the second
stage is 30710 kg.

The example I gave above applies with only minor modifications.
Instead of putting 9 times more engines on the first stage, they
put only one engine as for the second stage but that engine
weighs 24 times more. Saving on the mass and cost of the RL10B-2 by
replacing it with a RP-1 engine isn't much worth the trouble even
if you include the weight saved on the dry weight of the tank.
The added weight of the fuel would approximately counter balance
your gains. Different people will come to different conclusions
on this but the gains or losses by going to RP-1 wouldn't be great.
But for the first stage, you have *much* more to gain by going
to RP-1, the engine is 24 times bigger, I don't know how many times
more expensive it is but it should also be much more expensive
than the second stage engine (which is more important than the
fact that the engine is heavier). Even if you take into account
the fact the the first stage is bigger than the second stage,
the first stage's engine is comparatively bigger. The gains you
can get by going to RP-1 would be more important on the first
stage.

Why didn't they go to RP-1 for the first stage of the Delta IV? I
don't know. But what I have said before is that if you use LH2
for the first stage, it will work but it will cost you. I
don't consider Delta IV to be a low cost launcher.

The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


And because you rigged the problem by assuming you only have a low
thrust engine that you use everywhere.


No, I was assuming the same engine everywhere like on the Falcon 9.
The Merlin engine on the Falcon 9 has a thrust to weight ratio of 180
which is much more than what you have on the Delta IV.


Alain Fournier
  #6  
Old October 23rd 17, 10:01 AM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

Alain Fournier wrote:

On Oct/21/2017 at 10:56 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?

Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.

No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.


Yes, you can always rig the numbers by the problem statement.


Which number was rigged? The only slightly dubious thing in my
example is that I was assuming the same size for the second stage
as for the first stage, which would be unusual for a rocket. But
it doesn't change the idea of the outcome if you divide by two
the size of the tanks in the second stage of my example. It only
makes the computations a little more complicated.


Why does the first stage need nine engines?

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.


It's not 'nothing'. It's more payload to orbit, which is sort of the
goal of the thing (and what you've ignored with your rigged numbers).
Explain Delta IV.


In my example I was assuming the same functionality for the stages.
Meaning, the stages would put the same mass at the same altitude
and same speed whether it was the RP1 variant or the LH2 variant.

For the Delta IV, there are several variants. Let's look at the
one with one CBC as a first stage and the 5m diameter second stage.
The first stage has one RS-68A engine with a dry weight of the
engine of 6740 kg and 3137 kN thrust, the gross mass of the booster
is 226400 kg. The second stage has one RL10B-2 engine weighing
277 kg and providing 110 kN thrust, the gross mass of the second
stage is 30710 kg.

The example I gave above applies with only minor modifications.
Instead of putting 9 times more engines on the first stage, they
put only one engine as for the second stage but that engine
weighs 24 times more. Saving on the mass and cost of the RL10B-2 by
replacing it with a RP-1 engine isn't much worth the trouble even
if you include the weight saved on the dry weight of the tank.
The added weight of the fuel would approximately counter balance
your gains. Different people will come to different conclusions
on this but the gains or losses by going to RP-1 wouldn't be great.
But for the first stage, you have *much* more to gain by going
to RP-1, the engine is 24 times bigger, I don't know how many times
more expensive it is but it should also be much more expensive
than the second stage engine (which is more important than the
fact that the engine is heavier). Even if you take into account
the fact the the first stage is bigger than the second stage,
the first stage's engine is comparatively bigger. The gains you
can get by going to RP-1 would be more important on the first
stage.


And an RP1/LOX engine with equivalent thrust will weigh about the same
as the LH2/LOX RS-68A, so you save nothing so far as weight goes. The
fuel to get equivalent performance will weigh much more, so your
RP1/LOX engine needs higher thrust because it has to lift more mass in
fuel.


Why didn't they go to RP-1 for the first stage of the Delta IV? I
don't know. But what I have said before is that if you use LH2
for the first stage, it will work but it will cost you. I
don't consider Delta IV to be a low cost launcher.


But it's comparable to most other launchers of its generation. Both
Atlas V and Delta IV cost about $13k/kg to get stuff into orbit. Atlas
V is RP1/LOX while Delta IV is LH2/LOX for the 'core stage'.

The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


And because you rigged the problem by assuming you only have a low
thrust engine that you use everywhere.


No, I was assuming the same engine everywhere like on the Falcon 9.


Which rigs your result.


The Merlin engine on the Falcon 9 has a thrust to weight ratio of 180
which is much more than what you have on the Delta IV.


And it has a lower Isp and lower thrust. So what? Engine weight is
trivial unless you rig the numbers.


--
"False words are not only evil in themselves, but they infect the
soul with evil."
-- Socrates
  #7  
Old October 24th 17, 01:20 AM posted to sci.space.policy
Alain Fournier[_3_]
external usenet poster
 
Posts: 548
Default Were liquid boosters on Shuttle ever realistic?

On Oct/23/2017 at 5:01 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 10:56 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?

Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.

No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.


Yes, you can always rig the numbers by the problem statement.


Which number was rigged? The only slightly dubious thing in my
example is that I was assuming the same size for the second stage
as for the first stage, which would be unusual for a rocket. But
it doesn't change the idea of the outcome if you divide by two
the size of the tanks in the second stage of my example. It only
makes the computations a little more complicated.


Why does the first stage need nine engines?


You should ask SpaceX why they put nine engines on their first stage.

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.


It's not 'nothing'. It's more payload to orbit, which is sort of the
goal of the thing (and what you've ignored with your rigged numbers).
Explain Delta IV.


In my example I was assuming the same functionality for the stages.
Meaning, the stages would put the same mass at the same altitude
and same speed whether it was the RP1 variant or the LH2 variant.

For the Delta IV, there are several variants. Let's look at the
one with one CBC as a first stage and the 5m diameter second stage.
The first stage has one RS-68A engine with a dry weight of the
engine of 6740 kg and 3137 kN thrust, the gross mass of the booster
is 226400 kg. The second stage has one RL10B-2 engine weighing
277 kg and providing 110 kN thrust, the gross mass of the second
stage is 30710 kg.

The example I gave above applies with only minor modifications.
Instead of putting 9 times more engines on the first stage, they
put only one engine as for the second stage but that engine
weighs 24 times more. Saving on the mass and cost of the RL10B-2 by
replacing it with a RP-1 engine isn't much worth the trouble even
if you include the weight saved on the dry weight of the tank.
The added weight of the fuel would approximately counter balance
your gains. Different people will come to different conclusions
on this but the gains or losses by going to RP-1 wouldn't be great.
But for the first stage, you have *much* more to gain by going
to RP-1, the engine is 24 times bigger, I don't know how many times
more expensive it is but it should also be much more expensive
than the second stage engine (which is more important than the
fact that the engine is heavier). Even if you take into account
the fact the the first stage is bigger than the second stage,
the first stage's engine is comparatively bigger. The gains you
can get by going to RP-1 would be more important on the first
stage.


And an RP1/LOX engine with equivalent thrust will weigh about the same
as the LH2/LOX RS-68A, so you save nothing so far as weight goes.


Four Merlin 1D engines will give more thrust (4x845 = 3380 kN) than one
RS-68A (3137 kN) but they will weigh 1880 kg which is much less than
the 6740 kg of the RS-68A. But that is not really very important here.

The
fuel to get equivalent performance will weigh much more, so your
RP1/LOX engine needs higher thrust because it has to lift more mass in
fuel.


Yes. The fuel will weigh much more. You don't use RP1/LOX engines
instead of LH2/LOX engines to save weight. I'm glad to see that we
now agree on this.

Why didn't they go to RP-1 for the first stage of the Delta IV? I
don't know. But what I have said before is that if you use LH2
for the first stage, it will work but it will cost you. I
don't consider Delta IV to be a low cost launcher.


But it's comparable to most other launchers of its generation. Both
Atlas V and Delta IV cost about $13k/kg to get stuff into orbit. Atlas
V is RP1/LOX while Delta IV is LH2/LOX for the 'core stage'.

The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


And because you rigged the problem by assuming you only have a low
thrust engine that you use everywhere.


No, I was assuming the same engine everywhere like on the Falcon 9.


Which rigs your result.


Hum! are you claiming that SpaceX designed their Falcon 9 that way
to rig the results of this conversation? I can assure you that I
did not collude with SpaceX on that (nor on anything else for that
matter).

The Merlin engine on the Falcon 9 has a thrust to weight ratio of 180
which is much more than what you have on the Delta IV.


And it has a lower Isp and lower thrust. So what? Engine weight is
trivial unless you rig the numbers.


Engine weight isn't the important thing here. I'm glad to see we agree
on this. But the cost of the engines is important. If you need nine
on the first stage against only one on the second stage, you can see
that saving on engines on the first stage is more important than on
the second stage.

And if you have a design such as the Delta IV, it isn't as obvious but
the same argument applies. The RS-68A on the first stage weighs 24 times
the RL10B-2 on the second stage. Also the RS-68A has 28.5 times more
thrust than the RL10B-2. I don't know what's the price tag of neither
the RS-68A nor the RL10B-2, but I would expect it to cost many times
more. So once again, you can see that saving on the first stage engines
is more important than on the second stage engines.

This argument still holds even if you take into account that the second
stage is smaller than the first. The second stage of the Falcon isn't
nine times smaller than the first stage and the second stage of the
Delta IV isn't 28 times smaller than the first. You want to have
proportionally more thrust on the first stage than on the second stage
and you have to pay for that one way or another.


Alain Fournier
  #8  
Old October 24th 17, 07:08 AM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

Alain Fournier wrote:

On Oct/23/2017 at 5:01 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 10:56 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/21/2017 at 12:36 AM, Fred J. McCall wrote :
Alain Fournier wrote:

On Oct/19/2017 at 9:48 PM, JF Mezei wrote :
On 2017-10-19 20:15, Alain Fournier wrote:

It was my claim from the very start that LH2/LOX does NOT offer better
performance for the first stage.

If discussing engine performance only, would it be correct to state that
SSMEs with higher ISP would offer better performance then RP1 engines or
SRBs since it has better Isp?

Performance is a somewhat vague term. Higher ISP shouldn't be your
measure of performance. If you measure performance by ISP, SSME
is better, but it will cost you dearly if you measure performance
by cost in dollars.


That applies to all stages, though, so if LH2/LOX has poor performance
(in dollars) on the first stage, it will have equally poor performance
on all other stages.

No. Assume a rocket where the first stage and the second stage are
identical except that the first stage has 9 engines and the second
stage has only 1, and that the rocket uses RP-1. Somewhat like the
Falcon 9, the Falcon 9 isn't exactly like that but assume a rocket
that is like that.


Yes, you can always rig the numbers by the problem statement.

Which number was rigged? The only slightly dubious thing in my
example is that I was assuming the same size for the second stage
as for the first stage, which would be unusual for a rocket. But
it doesn't change the idea of the outcome if you divide by two
the size of the tanks in the second stage of my example. It only
makes the computations a little more complicated.


Why does the first stage need nine engines?


You should ask SpaceX why they put nine engines on their first stage.


You should ask everyone else why they didn't.

Let's divide the cost of a stage into 3 parts. E = cost of an engine,
T = cost of empty tank and F cost of fuel. For the fuel, the cost
should be in kg, the dollar cost of the fuel is not important, for
the engines and tank using dollars or weight for costs doesn't
make much difference. So the costs of the the first stage is
S1 = T + F + 9E
and the cost of the second stage is
S2 = T + F + E.

Now someone comes along and says, we could save by using LH2 on
the second stage. Now we have to see if you really save overall.
The tank would cost more, the fuel would be lighter and the engine
would cost more. So cost of stage with LH2 becomes
S2LH2 = T + delta T + F - delta F + E + delta E.
So the difference in cost between the RP-1 second stage and the
LH2 second stage is
S2LH2 - S2 = delta T - delta F + delta E.
People at SpaceX looked at that and figured it's not worth it
meaning at SpaceX they think that
delta F delta T + delta E.
Other rocket people looked at that and said yes we would save
by having LH2 for the second stage, so others think that
delta F delta T + delta E.
What is important to know is that it isn't obvious which is
true. That is because we have approximately
delta F = delta T + delta E.

Now let's look at the first stage. If we go to LH2 we get the
cost for the first stage with LH2
S1LH2 = T + delta T + F - delta F + 9(E + delta E).
And the difference between a first stage with RP-1 is
S1LH2 - S1 = delta T - delta F + 9(delta E).
We just said that we have approximately
delta F = delta T + delta E. So we have approximately
S1LH2 - S1 = 8(delta E).
And no sane person would pay 8(delta E) for nothing.


It's not 'nothing'. It's more payload to orbit, which is sort of the
goal of the thing (and what you've ignored with your rigged numbers).
Explain Delta IV.

In my example I was assuming the same functionality for the stages.
Meaning, the stages would put the same mass at the same altitude
and same speed whether it was the RP1 variant or the LH2 variant.

For the Delta IV, there are several variants. Let's look at the
one with one CBC as a first stage and the 5m diameter second stage.
The first stage has one RS-68A engine with a dry weight of the
engine of 6740 kg and 3137 kN thrust, the gross mass of the booster
is 226400 kg. The second stage has one RL10B-2 engine weighing
277 kg and providing 110 kN thrust, the gross mass of the second
stage is 30710 kg.

The example I gave above applies with only minor modifications.
Instead of putting 9 times more engines on the first stage, they
put only one engine as for the second stage but that engine
weighs 24 times more. Saving on the mass and cost of the RL10B-2 by
replacing it with a RP-1 engine isn't much worth the trouble even
if you include the weight saved on the dry weight of the tank.
The added weight of the fuel would approximately counter balance
your gains. Different people will come to different conclusions
on this but the gains or losses by going to RP-1 wouldn't be great.
But for the first stage, you have *much* more to gain by going
to RP-1, the engine is 24 times bigger, I don't know how many times
more expensive it is but it should also be much more expensive
than the second stage engine (which is more important than the
fact that the engine is heavier). Even if you take into account
the fact the the first stage is bigger than the second stage,
the first stage's engine is comparatively bigger. The gains you
can get by going to RP-1 would be more important on the first
stage.


And an RP1/LOX engine with equivalent thrust will weigh about the same
as the LH2/LOX RS-68A, so you save nothing so far as weight goes.


Four Merlin 1D engines will give more thrust (4x845 = 3380 kN) than one
RS-68A (3137 kN) but they will weigh 1880 kg which is much less than
the 6740 kg of the RS-68A. But that is not really very important here.


Nice of you to finally figure that out. Falcon 9 Expendable (no
recovery of first stage) can manage just short of 23 tonnes to LEO.
Delta IV Heavy (3 engines first stage) can put just short of 29 tonnes
in the same orbit. The differences become even larger (percentage
wise) when you start looking at GEO (a little over 8 tonnes vs a
little over 14 tonnes) and polar orbits (4 tonnes vs 23.5 tonnes).

The
fuel to get equivalent performance will weigh much more, so your
RP1/LOX engine needs higher thrust because it has to lift more mass in
fuel.


Yes. The fuel will weigh much more. You don't use RP1/LOX engines
instead of LH2/LOX engines to save weight. I'm glad to see that we
now agree on this.


'Now agree on this'? Like we didn't before? I'm glad to see you're
sharpening your reading skills and have finally realized this. So the
thrust comparison is bogus, since you have to lift more mass of fuel
with an RP1/LOX rocket so you NEED more thrust to get the same
performance. True, this difference is LOWER for the first stage,
since it expends its fuel early and then drops off, but it isn't
inconsequential.

Why didn't they go to RP-1 for the first stage of the Delta IV? I
don't know. But what I have said before is that if you use LH2
for the first stage, it will work but it will cost you. I
don't consider Delta IV to be a low cost launcher.


But it's comparable to most other launchers of its generation. Both
Atlas V and Delta IV cost about $13k/kg to get stuff into orbit. Atlas
V is RP1/LOX while Delta IV is LH2/LOX for the 'core stage'.

The difference comes from the fact that the first stage
has more engines because you need more thrust on the
first stage.


And because you rigged the problem by assuming you only have a low
thrust engine that you use everywhere.

No, I was assuming the same engine everywhere like on the Falcon 9.


Which rigs your result.


Hum! are you claiming that SpaceX designed their Falcon 9 that way
to rig the results of this conversation? I can assure you that I
did not collude with SpaceX on that (nor on anything else for that
matter).


No, I'm claiming you picked the Falcon configuration to rig your
result so as to require more engines on the LH2/LOX rocket to drive up
the cost. Compare Atlas V (RP1/LOX) to Delta IV (LH2/LOX).

The Merlin engine on the Falcon 9 has a thrust to weight ratio of 180
which is much more than what you have on the Delta IV.


And it has a lower Isp and lower thrust. So what? Engine weight is
trivial unless you rig the numbers.


Engine weight isn't the important thing here. I'm glad to see we agree
on this. But the cost of the engines is important. If you need nine
on the first stage against only one on the second stage, you can see
that saving on engines on the first stage is more important than on
the second stage.


Yes, if you rig the numbers by choosing a configuration that drives up
engine cost, you can get any result you want. Hey, let's assume 40
engines on the first stage!


And if you have a design such as the Delta IV, it isn't as obvious but
the same argument applies. The RS-68A on the first stage weighs 24 times
the RL10B-2 on the second stage. Also the RS-68A has 28.5 times more
thrust than the RL10B-2. I don't know what's the price tag of neither
the RS-68A nor the RL10B-2, but I would expect it to cost many times
more. So once again, you can see that saving on the first stage engines
is more important than on the second stage engines.


Yes, which is why you save by using fewer larger engines. That's why
most rockets (Falcon 9 and Falcon Heavy are exceptions) use different
engines on the first stage than on the second stage. I've told you
what both an RS-68A and a Merlin 1D cost. Best estimates say RS-68A
is around $10-$12 million and Merlin 1D is around 2 million and
change. So Delta IV Heavy has $30-$36 million worth of engines on the
first stage. Falcon 9 has around $18-$20 million worth of engines on
the first stage. So yes, Delta IV Heavy costs more for engines but it
has more performance. LH2 engines are more expensive, but they're not
as ridiculously more expensive as RS-25 costs or assuming large
numbers of engines being required would lead you to believe.


This argument still holds even if you take into account that the second
stage is smaller than the first. The second stage of the Falcon isn't
nine times smaller than the first stage and the second stage of the
Delta IV isn't 28 times smaller than the first. You want to have
proportionally more thrust on the first stage than on the second stage
and you have to pay for that one way or another.


Yes, but the difference isn't as preposterous as your case makes it
sound. LH2/LOX is going to be around 50% more expensive for engines
for 'similar' performance. The ability to stage higher and faster
because of Isp differences makes up for some of that (because you're
either carrying a smaller mass of fuel or because you're carrying the
same mass of fuel and it burns longer). You then have to adjust that
for tank weight and aerodynamic drag from the need for bigger volume
tanks. There are 'sweet spots' in there where LH2/LOX is competitive
and there are hypothetical cases that make LH2/LOX look even worse
than one would expect.

GENERALLY, a higher density exhaust (from a denser fuel) makes sense
for a first stage, but that's not always the case or Delta IV wouldn't
look like it does.


--
"The reasonable man adapts himself to the world; the unreasonable
man persists in trying to adapt the world to himself. Therefore,
all progress depends on the unreasonable man."
--George Bernard Shaw
  #9  
Old October 24th 17, 08:00 PM posted to sci.space.policy
Fred J. McCall[_3_]
external usenet poster
 
Posts: 10,018
Default Were liquid boosters on Shuttle ever realistic?

JF Mezei wrote:

On 2017-10-24 02:08, Fred J. McCall wrote:

Nice of you to finally figure that out. Falcon 9 Expendable (no
recovery of first stage) can manage just short of 23 tonnes to LEO.


There is one problem with Falcon 9 arguments: unlike NASA, SpaceX is
driven by business decisions. Get a limited payload at lowest possible
cost up there, and that includes getting up and running with the lowest
R%D costs possible.


A payload of 23 tonnes is not exactly considered 'limited'.


Musk admitted that at the time of their first succesful launch, they had
runned out of money and it would have been their last launch had it failed.


Yeah, that was Falcon 1 with a lot more limited payload, too.


And once you have an assembly line spewing out Merlin engines, and your
demand for engines goes down as you start to re-use Flacon 9s, it makes
business sense to use Merlins for your Falcon Heavy since you already
not only have a tested design, but also the assembly lines runnning.


There is no such thing as assembly line production 'spewing out'
rocket engines. It made sense to use Merlin engines because the idea
was to have three identical cores for Falcon Heavy, sort of like what
Delta IV Heavy does. In the event, Musk found they couldn't do that
and that they couldn't just use three Falcon 9 cores for Falcon Heavy.
The side boosters are now different from the central 'core'.


It does not necessarily mean that it is the optimal design from an
engineering point of view. It's the design that makes the most business
sense.


It may not even be that. Certainly no one here is advocating 'optimal
engineering design' as a GOAL, although that usually makes the most
business sense.


My sense is that the Shuttle was started, much like the Apollo program
with a "the sky is the limit, innovate, don't worry about budgets" only
to find that budgets were limited and had to compromise.


Sense seems to be something you lack. Like most big technical
projects, the original estimates were woefully low. Everyone tends to
underestimate the cost of dealing with the 'known unknowns' and of
course there's no budget at all for the 'unknown unknowns'.


The reason I asked the original question was to get a feel for what
could have been possible back in the 1970s had such budgets not been
limited and NASA be able to develop a Shuttle with engineering in control.


Find some of the discussions of the original Shuttle design.


--
"The reasonable man adapts himself to the world; the unreasonable
man persists in trying to adapt the world to himself. Therefore,
all progress depends on the unreasonable man."
--George Bernard Shaw
  #10  
Old October 25th 17, 04:02 AM posted to sci.space.policy
Greg \(Strider\) Moore
external usenet poster
 
Posts: 752
Default Were liquid boosters on Shuttle ever realistic?

"Fred J. McCall" wrote in message
...

There is no such thing as assembly line production 'spewing out'
rocket engines. It made sense to use Merlin engines because the idea
was to have three identical cores for Falcon Heavy, sort of like what
Delta IV Heavy does. In the event, Musk found they couldn't do that
and that they couldn't just use three Falcon 9 cores for Falcon Heavy.
The side boosters are now different from the central 'core'.


What are the differences? Last I knew the side boosters for the primary
flight are B1023.2 and B1025.2 (i.e 2nd flight for those two boosters)



--
Greg D. Moore http://greenmountainsoftware.wordpress.com/
CEO QuiCR: Quick, Crowdsourced Responses. http://www.quicr.net
IT Disaster Response -
https://www.amazon.com/Disaster-Resp...dp/1484221834/

 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Are rotating stations realistic ? John Doe Space Station 2 May 19th 10 10:15 AM
"Boeing To Study Liquid Fly Back Shuttle Boosters For NASA" gaetanomarano Policy 19 November 27th 07 06:59 AM
shuttle, tank and boosters on its crawler Rich Space Shuttle 37 September 11th 06 09:09 AM
Shuttle Liquid Fly-Back Booster to save money, improve safety(flashback) Bob Wilson Space Shuttle 0 July 16th 06 02:12 AM
Space Shuttle Boosters and Launch Pad Revell Model Kit on eBay TB Space Shuttle 2 February 1st 05 08:00 AM


All times are GMT +1. The time now is 09:12 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 SpaceBanter.com.
The comments are property of their posters.