Does Starship Have Too Many Engines?
HTML-код
- Опубликовано: 19 апр 2024
- Today we're diving into the age-old question: "How many engines is too many?" From Saturn-V to SpaceX's Starship, we explore the evolution of rocket design and the surprising statistics behind engine reliability.
TL;DW: More engines mean more failures, but more engines also mean more redundancy. Through super-advanced analysis (random numbers in Matlab) I figure out where the cut-off is, and when adding more engines ends up being a net benefit. Spoiler: SpaceX is a competent engineering company... go figure. 
Наука
I recall Everyday Astronaut came to the same conclusion regarding the number of engines vs. reliability. Plus he points out that many small engines means many more engines manufactured, which allows for a faster cycle of improvement.
Is it really about redundancy for SpaceX or is it about cheap rapid manufacture and iteration? It is obviously both, but which aspect is important.
Cost is absolutely a huge factor, I didn’t talk about it in the video but the F1 is massive, to launch starship with 5 engines you’d have to roughly double that. Just machining the combustion chamber of that engine would require massive custom machines, not to mention the nozzle, and transportation would be like moving around an entire falcon 9 fairing. Small engines like raptor can use cheaper off the shelf machines and are moved by forklifts
I would be interested to see how raptor performs under real mission parameters. We’ve seen it fly but not under full load and the mission failed (for various reasons) every time. If we fully loaded starship would you expect more raptors to fail? If starship is meant to be reusable should raptor failures be considered more likely to cause a mission failure?
Yeah, they keep increasing the maximum thrust so I wonder where the limit is. So far they seem to be working well despite thrust being increased on each flight.
Reuse would mean more wear on the engines, but if anything is that means you’d want the engine out capability as a backup. Landing might be harder, but I wonder if the booster is able to land with multiple engine configurations
Insufficient data to come to a conclusion, how many times have we lit starship in space? What's it's reliability to relight in space, after sitting for a week? We don't know.
True, but that doesn’t really weigh in to the number of engines you use, just how those engines need to be designed
Still, there are obviously many more bridges to cross before Starship can become man-rated. To be fair, even Apollo was iterated upon as each test launch was performed. And the whole program advanced in stages that added to its reliability as time went along.
I understand that now Elon Musk is saying that starship launch 40 to 50 metric tons into orbit with 30 rocket engines, where is once he was saying it could launch 100 tons. Blue origin is saying that their New Glenn will be able to launch 45 metric tons into orbit, with 7 BE-4 engines. We will find out in September the validity of the New Glenn rocket.
Subbed
Ur hair looks good
The critical factor isnt engine count, its parts count, and especially moving parts count.
And Elon has a long ways to go counting parts before he gets the numbers as low as is good.
Part of the problem is equating combustion chambers as engines.
its not a valid perspective, as the inevitable destination of minimum parts count design would see engines with multiple combustion chambers running off a single pumping system, if you had say six chabers to one pump, talking about reducing parts count by nearly a factor of 6.
The Russian R-180 had two chambers to one pumping system.
which should have been a clue.
Basically what you want is a combustion chamber below instability size, ganged to a herky pumping system that can easily run every chamber.
Elon knows about this logic, and has since parts count was first proposed, right after his memo about potential bankruptcy from the over complicated Raptor 1 engines.
He hasnt used it yet. But that doesnt mean he wont eventually.
My understanding is that multi chamber mostly helps with combustion instability which doesn’t seem to be destroying raptors. It’s an interesting concept but I’m not sure dual chamber raptors would be any less failure prone? But it would reduce your redundancy and thus engine out capability. Probably fine but it would technically make a launch failure more likely because each engine failure removes twice the thrust
@@ConHathy
I would consider dual chambers only a starting point. What you need is the pumps to be strong, simple, and highly unlikely to fail.
I suspect there is an upper number feasible for combustion chambers. Which is why I usually mention six per pump.
Its not an insane number. nor would the pumps need to be too huge.
Pumps usually are spinning circles, with circles if you double the diameter they are essentially 4X bigger. So what? about 3x the diameter on turbine pumps would easily run six combustion chambers?
In steam power plants there are huge turbines, some running since the early 1900's without even an oil change for the bearings... i.e. turbines are pretty tough.
And what I have seen of raptor pumps they look like notches in large shafts. which would be incredibly strong. Way stronger than vaned turbines.
The main thing about parts count and simplification is reduction in moving parts, they are most prone to failure.
Usually heavy cases or blocks almost never fail.
So a tested and stong enough pump or turbine casing, or a combustion chamber should be low problem.
The other thing besides simplification, is durability factor.
And in the original writs presenting parts count it was mentioned that Mars engines should be the simplest most reliable rocket engines ever made by humans..
Believe it or not that statement got down voted, as did parts count. And everything else.
But it all made it to Elon, and the rest is history.
The answer to failures is being very strong. I have heard rumor that they may push the Raptors even higher, like 3x current levels. I suspect because they tested and found out it wouldnt blow up. The first thing I read about Raptor 3 test was they had run it long duration, expecting to learn its limits & blow it up, and it ran them out of test fuel mix or something. Basically they are working towards a continuous duty rocket engine.
I do believe that everything space, should be designed with triple redundency.
ie the ships be able to fly on ⅔rds of their engines no problem, and ⅓ getting thru a crisis.
But, you have to admit, things have come a long way since Raptor 1's, way way more reliable and safer than those Spaghetti Monsters...
More engines, higher likelihood of failure. Starshit is a piece of crap.
Good. Now go and select your lottery numbers.
I think you cheat here. Starship needs refueling in space, so it needs at least five times the number of engines (five launches or more)
Interesting point, but even then above 99% engine reliability it’s still less likely to fail than a 5 engine, 1 engine out capable vehicle
Eyeballing the chart, at 99% engine reliability, the 1-engine case is obviously 99%, Saturn V is maybe 99.95%, Falcon 9 is 99.6%, and Starship is maybe 99.99%.
So if we take Starship's reliability of 0.9999 and raise it to the power of 5 for 5 launches, we get 0.9995, or 99.95%, so about the same as Saturn V.
However it's worth noting that those 5 refuels gets a whopping 240 tonnes of payload to TLI. To match Saturn V's 50 tonnes to TLI only takes 1 refuel, or 2 launches total.
That puts the reliability at 99.98%. Also worth noting that Starship in expendable config should be able to do at least 50 tonnes direct to TLI, in which case it'd be 99.99%.
@@ConHathy I don't think the reliability of engines with those turbopumps is that high in space
@@user-jp1qt8ut3s well we will have a better idea after flight 4, that will bring it to 150 with the upper stage engines (for reference only 65 F1 engines ever flew). If they have another flight without failures it’s a good sign
@@ConHathy do they bring some dummy load this time?