Size of engines

Jeff Findley wrote:
In article ,
says...
SpaceX just designed brand spanking new Raptor engines.
[...]
They're gonna putbetwene 37 and 40 on the big Super Heavy stage 1.
Big picture, what limits the size of engine you are building?

One of the issues with very big engines is combustion instability. That
was a huge issue during F-1 engine development. So, making many smaller
engines is a way to avoid having to develop very large engines.

There's also the Russian solution, RD-170 is an "F-1 class" engine
with four combustion chambers because they deemed it too hard to solve
the combustion stability. The Russians have a number of multi-chamber
engines in that lineage including the dual-chamber RD-180 used in the
Atlas V.

It MAY... be that combustion instability may be less of an issue these
days with all the advances in CFD but no one seems to want to go down
that path. Either way I suspect that with modern computer control it's
better to build and use multiple Raptor/BE4 sized engines unless the
rocket is bigger than even Super Heavy!


Or does growing beyond a certain size introduce fluid dynamic problems
of how liquid methane and o2 behave in pipes, behave as they flow
through large turbine with much mroe "space" to build, cause bubbles and
cavitate etc) ?

Would it be fair to state that the Raptors are sized to be easy to
build? (so they need more).

That was the philosophy with Merlin. Also, commonality between the
second stage engines and the first stage engines naturally means you
need far more engines for the first stage. Thanks rocket equation!

Musk has said that SpaceX is "tracking to well under $1M for V1.0" and
has a goal of $250k for V2.0 (250 ton thrust-optimized engine.

It's always been the plan that Raptor should be producable in large
numbers and that the per-unit cost should be lower than the smaller
Merlin, AFAIK they've actually below the much smaller Merlin's cost?
The benefit of starting with a clean sheet and thinking how to build
it during the entire process I guess.

They're still tweaking the design a lot but they're already producing
a lot of engines (SN20 was on Starship SN4 when it was consumed by
fire after ground equipment issues).

Still, not sure that the Merlin size was choosen for ease of
production reasons. Fundamentally only SpaceX knows ALL the reasons
for their decisions, it's not impossible it could have been involved
but there's so many other reasons why it makes sense for SpaceX to
build Raptor around that size that I doubt it.

I find it fascinating that there's so many rocket engines around the
1.6-2.4 MN (MegaNewton) range - Raptor, BE-4, RD-191, NK-33, RS-25 to
name some well-known ones. And the per *chamber* thrust for RD-170/180
(Energia and Atlas V respectively) is also in that range.

I can only think of one liquid fuelled rocket engine with more thrust
per chamber, the F-1 with 6.7 MN in a single chamber (RD-170 is more
powerful but needs 4 chambers). I discussed earlier why that might not
be a size path others may want to go down.

https://twitter.com/elonmusk/status/1179107539352313856

In article , lid says...

Jeff Findley wrote:

One of the issues with very big engines is combustion instability. That
was a huge issue during F-1 engine development. So, making many smaller
engines is a way to avoid having to develop very large engines.

There's also the Russian solution, RD-170 is an "F-1 class" engine
with four combustion chambers because they deemed it too hard to solve
the combustion stability. The Russians have a number of multi-chamber
engines in that lineage including the dual-chamber RD-180 used in the
Atlas V.

It MAY... be that combustion instability may be less of an issue these
days with all the advances in CFD but no one seems to want to go down
that path. Either way I suspect that with modern computer control it's
better to build and use multiple Raptor/BE4 sized engines unless the
rocket is bigger than even Super Heavy!

Agreed, that's an option. One set of turbopumps driving multiple
combustion chambers and nozzles.

Or does growing beyond a certain size introduce fluid dynamic problems
of how liquid methane and o2 behave in pipes, behave as they flow
through large turbine with much mroe "space" to build, cause bubbles and
cavitate etc) ?

Would it be fair to state that the Raptors are sized to be easy to
build? (so they need more).

That was the philosophy with Merlin. Also, commonality between the
second stage engines and the first stage engines naturally means you
need far more engines for the first stage. Thanks rocket equation!

Musk has said that SpaceX is "tracking to well under $1M for V1.0" and
has a goal of $250k for V2.0 (250 ton thrust-optimized engine.

It's always been the plan that Raptor should be producable in large
numbers and that the per-unit cost should be lower than the smaller
Merlin, AFAIK they've actually below the much smaller Merlin's cost?
The benefit of starting with a clean sheet and thinking how to build
it during the entire process I guess.

They're still tweaking the design a lot but they're already producing
a lot of engines (SN20 was on Starship SN4 when it was consumed by
fire after ground equipment issues).

Yep. Designing for mass production is a good thing. Starship and Super
Heavy are also intended to be designed for mass production as well.
Musk envisions sending a "fleet" of Starships to Mars every couple years
when the orbital mechanics makes the trip "affordable" in terms of
delta-V and travel time.

Still, not sure that the Merlin size was choosen for ease of
production reasons. Fundamentally only SpaceX knows ALL the reasons
for their decisions, it's not impossible it could have been involved
but there's so many other reasons why it makes sense for SpaceX to
build Raptor around that size that I doubt it.

I find it fascinating that there's so many rocket engines around the
1.6-2.4 MN (MegaNewton) range - Raptor, BE-4, RD-191, NK-33, RS-25 to
name some well-known ones. And the per *chamber* thrust for RD-170/180
(Energia and Atlas V respectively) is also in that range.

I can only think of one liquid fuelled rocket engine with more thrust
per chamber, the F-1 with 6.7 MN in a single chamber (RD-170 is more
powerful but needs 4 chambers). I discussed earlier why that might not
be a size path others may want to go down.

Yep. For some reason that size seems like a good "stopping point".
These days it seems easier to just add more engines to a design rather
than try to design an engine bigger than that.

One thing that was mentioned that the relatively smaller size of Merlin
made it easier and cheaper to test. When you're doing hundreds of test
firings, your engine needs to be small enough that the tests are
practical to do in the middle of Texas. I'm sure there is some truth to
that.

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.

"Jeff Findley" wrote in message
...

In article , lid says...

Jeff Findley wrote:

One of the issues with very big engines is combustion instability. That
was a huge issue during F-1 engine development. So, making many smaller
engines is a way to avoid having to develop very large engines.

There's also the Russian solution, RD-170 is an "F-1 class" engine
with four combustion chambers because they deemed it too hard to solve
the combustion stability. The Russians have a number of multi-chamber
engines in that lineage including the dual-chamber RD-180 used in the
Atlas V.

It MAY... be that combustion instability may be less of an issue these
days with all the advances in CFD but no one seems to want to go down
that path. Either way I suspect that with modern computer control it's
better to build and use multiple Raptor/BE4 sized engines unless the
rocket is bigger than even Super Heavy!

Agreed, that's an option. One set of turbopumps driving multiple
combustion chambers and nozzles.

Or does growing beyond a certain size introduce fluid dynamic problems
of how liquid methane and o2 behave in pipes, behave as they flow
through large turbine with much mroe "space" to build, cause bubbles
and
cavitate etc) ?

Would it be fair to state that the Raptors are sized to be easy to
build? (so they need more).

That was the philosophy with Merlin. Also, commonality between the
second stage engines and the first stage engines naturally means you
need far more engines for the first stage. Thanks rocket equation!

Musk has said that SpaceX is "tracking to well under $1M for V1.0" and
has a goal of $250k for V2.0 (250 ton thrust-optimized engine.

It's always been the plan that Raptor should be producable in large
numbers and that the per-unit cost should be lower than the smaller
Merlin, AFAIK they've actually below the much smaller Merlin's cost?
The benefit of starting with a clean sheet and thinking how to build
it during the entire process I guess.

They're still tweaking the design a lot but they're already producing
a lot of engines (SN20 was on Starship SN4 when it was consumed by
fire after ground equipment issues).

Yep. Designing for mass production is a good thing. Starship and Super
Heavy are also intended to be designed for mass production as well.
Musk envisions sending a "fleet" of Starships to Mars every couple years
when the orbital mechanics makes the trip "affordable" in terms of
delta-V and travel time.

Still, not sure that the Merlin size was choosen for ease of
production reasons. Fundamentally only SpaceX knows ALL the reasons
for their decisions, it's not impossible it could have been involved
but there's so many other reasons why it makes sense for SpaceX to
build Raptor around that size that I doubt it.

I find it fascinating that there's so many rocket engines around the
1.6-2.4 MN (MegaNewton) range - Raptor, BE-4, RD-191, NK-33, RS-25 to
name some well-known ones. And the per *chamber* thrust for RD-170/180
(Energia and Atlas V respectively) is also in that range.

I can only think of one liquid fuelled rocket engine with more thrust
per chamber, the F-1 with 6.7 MN in a single chamber (RD-170 is more
powerful but needs 4 chambers). I discussed earlier why that might not
be a size path others may want to go down.

Yep. For some reason that size seems like a good "stopping point".
These days it seems easier to just add more engines to a design rather
than try to design an engine bigger than that.

I think multiple engines gives more reliability, etc.

But I can definitely see a time, much like with jetliners where 3 and 4
engines became popular for a variety of reasons but now we are looking at
engines like the GE GENx that are larger than the 737 body, that rockets may
move in that direction.
That said, it took decades and literally millions of hours of flights before
we got there with jets, so I won't hold my breath for rockets.


One thing that was mentioned that the relatively smaller size of Merlin
made it easier and cheaper to test. When you're doing hundreds of test
firings, your engine needs to be small enough that the tests are
practical to do in the middle of Texas. I'm sure there is some truth to
that.

Eh, Texas is always bragging about size.

Jeff

--
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/

In article ,
says...

On 2020-06-15 20:25, Greg (Strider) Moore wrote:

I think multiple engines gives more reliability, etc.

Greater maintenance cost. Greater odds of an engine failing.

Possibly greater maintenance costs. Smaller engines are easier to work
on because most of the parts to pull and inspect wouldn't be so heavy as
to need cranes and hoists to remove.

A single engine plane as 37 times fewer chances of an engine failure
than a plane with 37 of the same engines. Or course, when you lose your
single engine, the impact is a tad more significant.

Early trans-Atlantic airliners had four engines. Why? Because early
turbofan engines were not so reliable back then and having four engines
meant you could lose one or two of them and still get to an airport.

Modern trans-Atlantic airliners have only two engines. Why? Because
modern turbofan engines are crazy reliable and losing one means you can
still get to an airport.

Guess where engine reliability is for liquid fueled rocket engines?
It's therefore advantageous to have extra engines so that an engine out
won't lead to loss of mission or loss of vehicle. Note that even the
space shuttle could handle the loss of an SSME during some parts of the
flight (if it's early enough, it will trigger an abort).

The type of engine failure is also important. Software detecting
something and performing a clean shutdown in flight is quite different
from a turbopump breaking apart and sending sharpnel all over the place,
damaging nearby engines and likely the tank above. So big kaboom.

That's why on Falcon 9 there are literally barriers between the engines
meant to contain an engine failure. I do not know if Starship and Super
Heavy will continue this practice or not. During the entire history of
Falcon 9, only *one* Merlin has "lost chamber pressure" during flight,
which meant it underwent some level of "kaboom". That flight completed
the primary mission. Redundancy worked even in this case.

Having 37 greater chances of en engine shutdown in flight just means
reduced thrust, so you brun longer so minor inconvenience with no real
impact compated to a single engine shutting down in flight.

Actually, when you have 37 engines, you can throttle up the remaining
engines a tiny bit and get the same thrust. The SSMEs had a bit of
reserve power to throttle up in an emergency, but because there were
only three of them, an engine out could not be fully compensated for by
the remaining two engines. When you have 37 engines, losing one of them
isn't nearly as big of a deal.

But having 37 greater chances of uncontained engine failure in flight is
37 time more dangerous than the single engine.

Again, that's why the Falcon 9 puts barriers between the engines. Super
Booster could do the same.

And you now have to introduce the issue of whether the smaller mass
produced engines are more reliable than the fewer handcrafted huge mega
engines.

I'd go with the mass produced version. Mass production typically means
automating many tasks that a human could screw up on a hand built
version. Your machines will never come in on a Morning hung over from
partying on the weekend.

In this thred, early on, it was pointed to me that having larger engines
introduced fuel flow issues with some rough combustion (forget exact
terminilogy). If this ends up with more complex engine and/or lwoer

You're missing the big point. A bigger engine is more likely to fail
because they're much harder to get right. Solutions to things like
combustion instability means more complexity. More complexity means
more things to go wrong. Smaller, simpler, engines are the way to go.

But I can definitely see a time, much like with jetliners where 3 and 4
engines became popular for a variety of reasons but now we are looking at
engines like the GE GENx that are larger than the 737 body, that rockets may
move in that direction.

The 3 or 4 engines were originally needed because of the power needed to
fly that jet since available engines didn't have the thrust to do the job.

Again, it wasn't just that, it was the reliability issue as well. It
wasn't until fairly recently that two engine aircraft were certified for
trans-ocean flights. They had to prove that modern turbofan engines
were up to the task in terms of reliability. That took a lot of effort
to certify to the FAA that the twin-jets were truly reliable enough.

What SpaceX is doing with engine re-usability has many similarities to
aviation.

https://en.wikipedia.org/wiki/ETOPS#History

Initially, all aircraft were restricted to routes where they were no
more than 100nm from an alternate airport in case they lost an engine
(irrespective of number of engines). (this was roughly 60 minutes on
piston propellor aircraft crossing Atlantic)

That's what I've been saying. I'm snipping the rest of your post
because I see a lot of rambling, but not much of a point.

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.

In article ,
says...

On 2020-06-17 07:24, Jeff Findley wrote:
Modern trans-Atlantic airliners have only two engines. Why?
Because modern turbofan engines are crazy reliable and losing
one means you can still get to an airport.

Thst is just on the surface. The rule is not so much about reliability,
but rather ability to complete a take off, remain at altitude and land
on a single engine. This means that each engine needs to have a surplus
of power available to compensate for loss of the other. So engine power
is what enabled twin engine widebodies. The first one being the A300 in
the 1970s.

The International Civil Aviation Organization (ICAO) has very strict
requirements for engine reliability in order for aircraft to be
certified for Extended Operations (ETOPS) over the ocean. Cite:

https://en.wikipedia.org/wiki/ETOPS

From above (emphasis mine):

The **cornerstone** of the ETOPS approach is the **statistics
showing that the turbine assembly of a modern jet engine is an
inherently reliable component**.

So your claim that it's "not so much about reliability" is simply not
true. It is, in fact the cornerstone of what it takes to earn the
required ETOPS ratings for long flights over the ocean.

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.