NASA & World Space Exploration...News, Views, Photos & videos

Raptor is way more complicated than Merlin. More things that can go wrong in the engine.
It is all about probabilities. Even if the probability of a single engine failing is like 5% when you have a load of them it compounds.
I think around 28-29 engines is the limit with current engine technology in terms of reliability.
Another thing that people don't tell you is that engine reliability has scarcely improved since Apollo.

gelgoog said:

Raptor is way more complicated than Merlin. More things that can go wrong in the engine.
It is all about probabilities. Even if the probability of a single engine failing is like 5% when you have a load of them it compounds.
I think around 28-29 engines is the limit with current engine technology in terms of reliability.
Another thing that people don't tell you is that engine reliability has scarcely improved since Apollo.

Click to expand...

Why is more engines inherently unreliable? Doesn’t that mean that one engine going out won’t lead to as much of a loss of power?

You are assuming the engines will have a benign failure. Catastrophic losses are way more common than some people assume.
An exploding engine can take out other engines around it, or even damage the rocket internally, causing a total loss.
More engines means more parts and more things which can go wrong.

tacoburger said:

Have to say, NASA and Americans really have a lot of faith in Starship fulfilling all of it's promises. It's insane, people are acting like it's already launching daily and bringing launch costs to $100/kg and fully reusable.

Click to expand...

Let's just get the obvious out of the way: it's clearly an LEO optimized rocket that would make it very capable for launching Starlinks but makes little sense in the Artemis architecture.
HLS is a sketchy moon lander that will likely be canceled because lack of funds caused a sketchy bid to win.
The reason it needs so many refueling tankers because the mass fraction sucks. The dry mass is very very very big. It needs a huge number of tankers to have enough DV to do the mission: go to an elliptical departure orbit, do TLI, insert to NRHO, go to LLO, land, return to LLO, return to NRHO.
Gateway architecture is also heavily criticized by former NASA admin Michael Griffin and Apollo 11 astronaut Buzz Aldrin but that's on NASA not SpaceX at least.
The whole Artemis program is very poorly designed from the get-go and they finally admitted it was basically rushed as a response to China's lunar exploration program, as if that wasn't obvious to anyone who's been following it:

gelgoog said:

You are assuming the engines will have a benign failure. Catastrophic losses are way more common than some people assume.
An exploding engine can take out other engines around it, or even damage the rocket internally, causing a total loss.
More engines means more parts and more things which can go wrong.

Click to expand...

My understanding that the SpaceX engines all have an individual blast shield which prevents a single engine explosion from destroying its neighbours.

And that the Falcon 9 was designed for a single engine failure to still allow the overall mission to succeed.
Presumably the same applies to Starship.

With as many engines as Starship has you won't just get single engine failures.

H-3 reached orbit:

gelgoog said:

With as many engines as Starship has you won't just get single engine failures.

Click to expand...

If I use the same reliability figures for the Raptor engines as the Merlin engines (99.8%), I get the probability of one of the 33 Raptor Engines failing as 6.6%.

The probability of 2 Raptor Engines would be substantially lower

You have data available for the Raptor engine failures in actual use. So why not use what instead of assuming it will have the same reliability as a much simpler engine?

And I already told you an engine can damage neighbouring ones. So no, the probably of more than one engine failing isn't necessarily lower. You are assuming they will fail independently.

gelgoog said:

You have data available for the Raptor engine failures in actual use. So why not use what instead of assuming it will have the same reliability as a much simpler engine?

And I already told you an engine can damage neighbouring ones. So no, the probably of more than one engine failing isn't necessarily lower. You are assuming they will fail independently.

Click to expand...

No, we don't have Raptor engine data because it is still a new engine.

And I'm assuming SpaceX Raptor engineers fail independently because they each have a blast shield to prevent a catastrophic failure affecting the other engines.