Discussion Should SpaceX create a larger engine to reduce the number of engines of the Falcon 9?

[Urwumpe]

The N-1 failures actually had one simple cause: The stage was never tested as whole because there was no sufficiently large test stand in the USSR. It was intentionally flight tested, but the N-1 proved to be very expensive for flight testing and the number of tests flights had been limited by costs. All in all, a very bad mixture.

Safety in spaceflight means just that, based on what you know what can go wrong, all that you can control has to stay controlled. Perfect safety is not possible, never. What you can do is staying in control as long as possible. The Shuttle is a good example why entering uncontrollable regimes is a poor choice - the first minutes of flight had no abort options, thus the options to stay in control had been limited.

Safety is not about preventing every accident. A launch abort is a mission failure, but no failure of safety. If nobody is harmed in the process, the launch abort was working perfectly and the situation remained under control.

The goal is simply that only a few missions fail. The less missions fail, the better it is. But if the safety measures for a set success goal cost more than the additional successful missions gained, you have to be cold-hearted and not implement such expensive safety measures - and that is only the design. Safety itself is a duty for every person in the organisation. As long as the crew is not harmed by preventable accidents (The Space Shuttle had two such preventable accidents), the safety management process in your company or spaceflight agency works. If one small step there does not work properly, you won't have safety.

Fewer big engines won't increase safety at all, if the stage as system is not properly tested - not just during R&D, but during the full rocket (and rocket engine) lifecycle. And big engines require much more testing and much more costs. A small rocket engine can be tested with a small test stand and requires not much instrumentation. A big one has complex combustion processes and chamber acoustics, that you need to monitor and tune during the process. No computer simulation exists that can handle such phenomena properly - they are pretty chaotic, small changes in the inputs mean big changes in the resulting engine behavior.

A Falcon 9 with 5 new engines would not be cheaper or more reliable by design.

 

[Hlynkacg]

A Falcon 9 with 5 new engines would not be cheaper or more reliable by design.

A larger engine running at higher a thrust/pressure will by nature be more complex and require tighter engineering tolerances.

Couple this with the costs of testing and re-tooling being spread over a smaller production run means that the cost per engine will, in all likelyhood, be increased by an order of magnitude.

This is why a single SSME costs more than an entire Falcon 9 ($80 million vs. $50 million)

 

[Mader Levap]

However, the experience with the Falcon 9 is that one engine has failed in each launch, so it can't really be said they have higher reliability.

This got me interested. Can you give sources for that claim?

 

[RGClark]

This got me interested. Can you give sources for that claim?

Other engine anomalies described here:

Falcon 9 Drops Orbcomm Satellite in Wrong Orbit.
Posted by Amy Svitak 2:17 PM on Oct 08, 2012
http://www.aviationweek.com/Blogs.a...f385Post:fdf0d27c-fdf2-4efb-a71f-8272017dbfc3

For the flight in May, I still consider that as having an engine failure because it had to be stopped and repaired on the pad even though after the repair there was not an in flight failure.


Bob Clark

 

[Mader Levap]

http://www.aviationweek.com/Blogs.a...f385Post:fdf0d27c-fdf2-4efb-a71f-8272017dbfc3

In this article, these things was mentioned:
- Well known spectacular RUD on CRS-1 flight. This indeed counts as engine failure.
- Oxygen-rich shutdown was off-nominal shutdown of engine after fulfilling its mission. If we are speaking in black/white categories, I do not count is as engine failure.
- Roll on inagural launch was not caused by engine failure.

For the flight in May, I still consider that as having an engine failure because it had to be stopped and repaired on the pad even though after the repair there was not an in flight failure.

I am afraid I cannot agree with such definition of engine failure.

Summarizing, there was one flight of F9, where engine failure happened. At most two, if counting off-nominal shutdown.

 

[Hlynkacg]

In this article, these things was mentioned:
- Well known spectacular RUD on CRS-1 flight. This indeed counts as engine failure.
- Oxygen-rich shutdown was off-nominal shutdown of engine after fulfilling its mission. If we are speaking in black/white categories, I do not count is as engine failure.
- Roll on inagural launch was not caused by engine failure.


I am afraid I cannot agree with such definition of engine failure.

Summarizing, there was one flight of F9, where engine failure happened. At most two, if counting off-nominal shutdown.

Furthermore none of those failures have or would have resulted in a loss of crew or vehicle so I fail to see how they can be considered to be a critical safety issue.

Does having redundent flight controls in an aircraft make it inherently less safe?

Afterall, a second set of flight controls DOUBLES the chance of a FCS failure so clearly it is must be safer to have only one of any critical component or system. :uhh:

 

[T.Neo]

If I understand correctly, the roll in second stage ascent on the first F9 flight was caused by a stuck roll nozzle (Mvac uses GG exhaust for roll control, if I recall correctly). That can hardly be used as an example of the unreliability of the nine-engine design, as the upper stage of F9 has only one engine...

 

[FADEC]

IF it really is the case that in this latest launch the combustion chamber was breached there is no way that NASA would certify the Falcon 9 for manned launches.

Which is not the NASA I know :shifty:

They even certify vehicles which are launched with an entirely unprotected thermal "protection" system. But their best job, from the management point of view, definitely was the fireworks above the Cape on Jan. 28, 1986. Kudos. Way more spectacular than the manned test flight in 1981 (as long as one doesn't take a look at the anomalies). And I'm sure that riding on top of a giant, solid rocket booster, would be the smoothest and most reliable and safe thing to do.

Seriously, I think that Falcon 9 is doing fine. If something is wrong with the combustion chamber, or something else, then SpaceX certainly will do its best to overcome the challange.

 

[RGClark]

If I understand correctly, the roll in second stage ascent on the first F9 flight was caused by a stuck roll nozzle (Mvac uses GG exhaust for roll control, if I recall correctly). That can hardly be used as an example of the unreliability of the nine-engine design, as the upper stage of F9 has only one engine...

But there was an unexpected roll as well on the first stage. That may also have well been caused by a stuck roll nozzle on one or more first stage engines. SpaceX has not released the cause for it.
And in any case 10 engines of very similar design just as 9 engines have a higher probability of at least one engine failing.


Bob Clark

---------- Post added at 02:28 PM ---------- Previous post was at 02:20 PM ----------

Which is not the NASA I know :shifty:

They even certify vehicles which are launched with an entirely unprotected thermal "protection" system. But their best job, from the management point of view, definitely was the fireworks above the Cape on Jan. 28, 1986. Kudos. Way more spectacular than the manned test flight in 1981 (as long as one doesn't take a look at the anomalies). And I'm sure that riding on top of a giant, solid rocket booster, would be the smoothest and most reliable and safe thing to do.

Seriously, I think that Falcon 9 is doing fine. If something is wrong with the combustion chamber, or something else, then SpaceX certainly will do its best to overcome the challange.

I understand you're being humorous but it has to kept in mind how potentially dangerous a combustion chamber breach is. Remember the Challenger accident was a combustion chamber breach in the SRB with a hot jet of gases pointing toward the external tank igniting the fuel.
IF this was a combustion chamber breach at the top then you would also have a hot jet of gas aimed at the fuel tank. The web site where this was offered as an explanation, NasaSpaceFlight, is a well read one in the industry. Many NASA engineers read and contribute to it as well.
I can't believe SpaceX would let that stay out there as a legitimate explanation for what happened unless at this point they believe it still could be the valid explanation for what actually happened.


Bob Clark

 

[Urwumpe]

I understand you're being humorous but it has to kept in mind how potentially dangerous a combustion chamber breach is.

There is just one problem that means you are producing much ado about nothing: It was no combustion chamber breach.

The breach happened near the MCC, but outside it. It happened at a higher pressure, but at lower energy. And the engine was successfully shutdown despite it, delivering even telemetry. The SSME fared much worse in its history despite being man-rated.

 

[T.Neo]

But there was an unexpected roll as well on the first stage. That may also have well been caused by a stuck roll nozzle on one or more first stage engines. SpaceX has not released the cause for it.

As far as I know- and I may be wrong here- the F9 does not use the GG exhaust to perform roll control on the first stage; that is done by gimballing the engines themselves.

And in any case 10 engines of very similar design just as 9 engines have a higher probability of at least one engine failing.

The higher production runs and increased flight experience increases reliability, rather than the opposite. And regardless of whether you have five or nine engines on the first stage, you only have one on the second- assuming a set engine reliability level, the chance of second stage failure is the same regardless of the number of engines on the first stage.

Of course, creating an all-new engine for first stage use on F9 presents other problems; you lose commonality with the MVac on the second stage, which will cause price to increase...

 

[Urwumpe]

It has to be said that one Soyuz launch uses 5 similar engines with 20 similar combustion chambers - the differences between first and second stage engines are small. The high production numbers are part of the success.

 

[N_Molson]

It has to be said that one Soyuz launch uses 5 similar engines with 20 similar combustion chambers - the differences between first and second stage engines are small. The high production numbers are part of the success.

And it has to be underlined there are technically only 5 engines, or, more accurately, 5*1 engines, each one at the bottom of a booster. Please don't use the R-7 to find arguments in favour of the almost-free-almighty-completely-perfectly-safe-and-best-mankind-engineering-design-ever Falcon 9, those are quite a different design. Also the former has +50 years of proven heavy-duty flight experience. And I don't want to hear the "but the R-7 had catastrophic failures in the late 50's / early 60's" : they had no real computers, primitive radars, experimental test chambers, B&W cameras, and were basically creating rocketry in the form we know it today. Today Kero/Lox engines are a common and completely proven technology. There is a huge database of engineering data about them. It's a completely different situation in a completely different world.

(and S.P. Korolev)

 

[RGClark]

There is just one problem that means you are producing much ado about nothing: It was no combustion chamber breach.
The breach happened near the MCC, but outside it. It happened at a higher pressure, but at lower energy. And the engine was successfully shutdown despite it, delivering even telemetry. The SSME fared much worse in its history despite being man-rated.

Do you have a reference for that? On the NasaSpaceFlight forum, discussion is still centering on a rupture of the fuel dome.


Bob Clark

 

[Urwumpe]

Do you have a reference for that? On the NasaSpaceFlight forum, discussion is still centering on a rupture of the fuel dome.


Bob Clark

Fuel dome is not combustion chamber. It is above upstream of the injector.

 

[RGClark]

The aborted first launch attempt in May was due to a valve failure in one of the engines that led to too high pressure in the engine. That conceivably could have had an effect on the engine failure on the October launch as well.
I was reminded of this:

Is SpaceX changing the rocket equation?
1 visionary + 3 launchers + 1,500 employees = ?
By Andrew Chaikin
Air & Space magazine, January 2012

Significantly, the Merlin engines—like roughly 80 percent of the components for Falcon and Dragon, including even the flight computers—are made in-house. That’s something SpaceX didn’t originally set out to do, but was driven to by suppliers’ high prices. Mueller recalls asking a vendor for an estimate on a particular engine valve. “They came back [requesting] like a year and a half in development and hundreds of thousands of dollars. Just way out of whack. And we’re like, ‘No, we need it by this summer, for much, much less money.’ They go, ‘Good luck with that,’ and kind of smirked and left.” Mueller’s people made the valve themselves, and by summer they had qualified it for use with cryogenic propellants.
http://www.airspacemag.com/space-exploration/Visionary-Launchers-Employees.html?c=y&page=2

Hmmm...

Bob Clark

 

[Urwumpe]

Yes, not really a surprise there. Just look at the tiny fact that a tiny RAT for an aircraft costs at least 250,000 USD - and that is usually the lowest bidder.

Most of the costs is testing to meet a specified certification level. If your customer specification says "MTBF of 20,000 hours", you have to prove that this number is achieved.

 

[Hlynkacg]

Most of the costs is testing to meet a specified certification level. If your customer specification says "MTBF of 20,000 hours", you have to prove that this number is achieved.

Which of coures requires at least 20,000 hours worth of testing and time, as they say, is money.

 

[T.Neo]

And it has to be underlined there are technically only 5 engines, or, more accurately, 5*1 engines, each one at the bottom of a booster.

This does not degrade Urwumpe's original point, which is that the R-7 has a total of 20 combustion chambers (not turbopumps, engines or boosters) burning together at launch. Simply because some components on the engines (such as turbopumps) are not multiplied does not change the fact others are.

the almost-free-almighty-completely-perfectly-safe-and-best-mankind-engineering-design-ever Falcon 9

I cannot find any instance anywhere in this thread where someone has claimed such a thing. There's quite a large difference between providing a counterargument to a claim that Falcon 9 is hideously unsafe, and claiming that it's the best piece of engineering to ever walk the Earth.

 

[N_Molson]

This does not degrade Urwumpe's original point, which is that the R-7 has a total of 20 combustion chambers (not turbopumps, engines or boosters) burning together at launch. Simply because some components on the engines (such as turbopumps) are not multiplied does not change the fact others are.

The fact is that turbopumps and such components include a lot of moving parts and are the most prone to failure. Combustion chambers problems are often related to combustion unstability, and this is something more "predictible". In fact the original R-series engines had one big combustion chamber, but the result was unstable and deceptive in terms of thrust. The quad-chambers design solved that and actually increased thrust.

In general, the more moving parts, the more potential problems, and then increased costs to make sure those problems don't happen. So a lot of "cheap" engines might actually have been a false-good-idea. Which doesn't mean it is a fatal flaw ; but that operational costs are going to be higher than expected.

...to a claim that Falcon 9 is hideously unsafe...

I cannot find any instance anywhere in this thread where someone has claimed such a thing.