Latest / Elon Musk Podcast / Starship V3 Launch update

Transcript

Starship V3 Launch update · Elon Musk Podcast

0:00
A super heavy booster with a completely redesigned and
0:04
partially missing engine layout just rolled out to a brand new
0:07
launchpad while it's upper stage finished testing to survive
0:11
being caught midair by massive robotic.
0:13
Arms the sheer mechanical scale of these vehicles is genuinely
0:16
difficult to wrap your head around.
0:18
I mean, you are looking at machines the size of
0:20
skyscrapers, and the current state of this enormous
0:25
engineering operation is shifting heavily.
0:27
They're moving away from just, you know, basic flight
0:30
capability into rigorous, highly specific stress testing.
0:35
They are pushing both the vehicle hardware and the
0:37
physical ground infrastructure supporting it to the absolute
0:40
limit. So how does testing a partially
0:43
built rocket on a brand new pad actually solve the bottleneck
0:47
for putting humans on the moon? Well, to answer that, we have to
0:50
look at what is physically sitting on the launch site right
0:52
now. We are seeing Booster 19 rolling
0:55
out to the newly constructed pad too, and this is a staggering
0:58
launch facility required a very long construction period to
1:01
complete. We are talking about pouring
1:04
thousands of tons of concrete, routing incredibly complex
1:08
plumbing, building this towering launch mount, and now it is
1:13
finally seeing action. Getting a vehicle onto this
1:17
specific pad is just a massive step forward for their entire
1:21
operational flow. Right, but the hardware rolling
1:23
out to that pad looks completely wrong to anyone who has been
1:26
paying attention. Yeah, it really does.
1:28
Booster 19 rolled out with only ten of its 33 Raptor engines
1:32
installed. Just 10?
1:33
Yeah, they placed two engines in the center, 4 in the inner ring
1:37
and four in the outer ring. The rest of the booster is just
1:39
empty slots where engines should be.
1:41
Oh wait, back up. Why only 10 engines?
1:43
It looks ridiculous. I know.
1:44
It looks like a car missing half its engine block.
1:46
Yeah. But they are deliberately
1:48
holding back. They're running very specific
1:50
validation tests on the internal plumbing and the newer gas
1:54
manifolds, and they're doing it without risking the entire
1:57
vehicle. OK, before we go further, what
1:59
exactly is a gas manifold in this context?
2:01
Because I think for you listening, that might just sound
2:03
like technical jargon. Sure.
2:05
Think of a manifold as a complex highway system of pipes.
2:11
It distributes the highly pressurized liquid oxygen and
2:13
methane fuel directly to the engines, right?
2:16
And when you ignite those engines, you really want to
2:19
avoid what engineers call a spin.
2:21
Boom A. Spin boom, that sounds bad.
2:23
It is very bad, Yeah. Imagine turning on a gas grill,
2:28
but the igniter is delayed, the gas pools under the lid, and
2:31
when the spark finally hits you get a huge fireball right in
2:35
your face. Oh wow.
2:36
A spin boom is essentially that, but on a massive scale.
2:39
It's an incident where oxygen and methane gas ignite and
2:43
explode directly under the launch mount.
2:45
That actually happened to a previous booster I.
2:47
Remember that. Yeah, and it caused significant
2:49
structural damage to the rocket and the pad.
2:52
So they are intentionally holding back the engines just to
2:54
make sure the launchpad itself doesn't blow up during a test.
2:57
Exactly this deliberate partial engine layout changes how they
3:01
validate the pads, deflector and Ridge cap.
3:04
All right. Explain the deflector and the
3:05
Ridge cap for me. We are getting into some heavy
3:07
structural engineering terms here.
3:09
And so when you ignite rocket engines, you are basically
3:13
unleashing the power of a controlled bomb.
3:16
The acoustic energy, literally the sound waves themselves and
3:19
the physical pressure of the exhaust gas, have to be directed
3:22
away from the vehicle quickly and safely.
3:25
Or else what? Or the vibration will literally
3:28
tear the rocket apart. The deflector buckets and the
3:31
Ridge cap are these giant reinforced steel structures
3:35
underneath the launch mount. They take the brunt of that
3:37
force and channel the exhaust outward.
3:40
Got it. So by only firing 10 engines in
3:43
this specific sort of sparse pattern, they severely limit the
3:47
risk of a catastrophic explosion before they commit to testing a
3:50
fully stacked, fully fueled vehicle.
3:53
That makes perfect sense. Yeah, you're essentially using a
3:56
partially built rocket as a diagnostic tool for the concrete
3:59
and steel structure holding it up.
4:00
They are testing the pads ability to handle the extreme
4:03
physics of a launch without risking all 33 engines at once.
4:06
And while the booster is sitting on the new pad acting as a
4:09
diagnostic tool, the upper stage ship 39 is going through its own
4:14
extreme physical trials. Oh yeah, it just went through
4:18
ambient pressure tests, cryo proof tests, flap testing, and
4:22
chopstick simulator squeeze testing.
4:25
Stop right there. What is a cryoproof test?
4:27
So a cryoproof test is when they pump the entire vehicle full of
4:31
ultra cold liquid nitrogen. OK, they want to see if the
4:34
steel hole shrinks, groans or cracks under the freezing
4:37
temperatures before they ever introduce explosive, super
4:41
chilled rocket fuel into the mix.
4:43
That sounds intense. It is.
4:45
It is a critical stress test for the welds and the metal itself.
4:48
If a Weld is going to pop, you want it to pop with inert
4:51
nitrogen, not methane. Right, of course.
4:53
And what about this chopstick simulator squeeze testing?
4:56
Because that sounds like a bizarre arcade game.
4:58
The squeeze testing is absolutely fascinating when you
5:01
consider what they're preparing this vehicle to do for you.
5:04
Listening. Imagine trying to catch a
5:06
falling empty soda can with a pair of metal kitchen tongs.
5:09
OK, I'm picturing it. If you squeeze too hard, the can
5:12
freshes immediately. Yeah, if you don't squeeze hard
5:15
enough, it slips right through and hits the floor.
5:18
Now scale that soda can up to a towering steel spacecraft
5:21
returning from orbit, and scale the tongs up to mechanical arms
5:25
on a launch tower. That is just wild to think
5:28
about. Right, they built a giant
5:30
simulator to literally apply immense physical pressure to the
5:34
hull to ensure it doesn't crush when those arms grab it out of
5:37
the sky. The structural integrity
5:39
required to survive that lateral squeeze right after and during
5:43
the extreme heat of reentry is staggering.
5:46
And they are making physical changes to the booster to help
5:49
with that exact recovery process, right?
5:51
Yes, Booster 19 has new hardware.
5:54
They installed three grid fins that are 50% larger than the
5:57
previous generation. 50% larger. That's a massive aerodynamic
6:01
change. It is, and they're mounted lower
6:04
on the hall and equipped with specific lifting pins.
6:07
Those lifting pins are the contact points for the
6:09
mechanical arms. The entire system is being
6:12
optimized for rapid capture and turn around, but there's a huge
6:16
constraint here. Yeah, there is.
6:18
The ultimate goal is catching the upper stage over land.
6:21
However, this action is strictly limited.
6:24
The engineers will only attempt a tower catch after executing 2
6:27
perfect soft landings in the ocean right?
6:30
This limits immediate rapid recovery, but completely opens
6:33
up a safer long term path to reusability by minimizing the
6:37
risk of the ship breaking up over land.
6:40
I have to be honest though, 2 ocean landings seems like it's a
6:44
surprisingly low threshold of proof before flying an object
6:47
that size over a populated landmass.
6:50
You think so? Well yeah, we're talking about
6:53
catching an orbital class vehicle with mechanical arms.
6:56
Proving it works twice the ocean feels like an incredibly small
6:59
sample size for such a high stakes maneuver.
7:01
I see where you're coming from. What happens if a sudden cross
7:04
wind hits during the land attempt?
7:06
Basically a skyscraper falling from space.
7:08
I hear what you were saying, but it represents an aggressive but
7:11
calculated threshold standard for modern rocketry.
7:15
You have to look at the volume of telemetry data they gather
7:18
during those two flights. Fairpoint.
7:19
They aren't just looking at the window to see whether it's
7:21
splashed down softly. Right, they've sensors
7:23
everywhere. Exactly.
7:25
They are measuring thousands of different data points on
7:27
structural stress, aerodynamic control and software response in
7:31
real time. If those two flights return
7:34
absolutely perfect data across every sensor, 1/3 flight over
7:39
land is statistically justified within their engineering
7:42
framework. We'll see if that statistical
7:44
gamble pays off when they actually fly it over land and
7:48
test that software in real time. But Speaking of gathering
7:51
incredible amounts of data and getting paid for your time.
7:53
If you've got a decent mic and laptop and some free time,
7:56
Babble Audio is paying people to record speech data and annotate
8:00
audio for AI training. No minimums, no fixed hours.
8:03
You work when you want and get paid weekly via PayPal, Venmo,
8:06
or bank transfer. They pay per recorded or
8:09
annotated hour, plus bonus challenges for hitting weekly
8:11
goals. And if you sign up through our
8:13
link, you get priority processing and a $15 bonus.
8:16
Links in the show notes. So, bringing it back to the
8:18
schedule, the target for the next launch has slipped from
8:22
previous estimates to a short upcoming window.
8:25
Yeah, the timeline keeps adjusting.
8:26
And that slip schedule connects directly to NASA.
8:31
NASA is actively pressuring the primary contractors to
8:34
accelerate their lunar Lander work for the Artemis program.
8:38
The pressure's coming straight from the top down to get the
8:41
hardware ready for actual missions to the moon.
8:43
And there has been a recent shift in the Artemis
8:46
architecture. The Artemis 3 mission is no
8:48
longer a landing, right? It will be a low Earth orbit
8:52
mission where the Orion capsule docks with the Landers.
8:55
The actual lunar landing attempts are pushed back to
8:58
Artemis 4 and Artemis 5. Yeah, that's a huge shift.
9:01
Can you explain what that architecture shift actually
9:04
means for the program? Basically they are putting off
9:07
the difficult part. Artemis 3 was supposed to put
9:09
boots on the lunar surface. Now it is just a practice run in
9:13
low Earth orbit. A dress rehearsal.
9:15
Exactly. The Orion capsule will fly up,
9:18
dock with the lunar Lander in space.
9:20
The astronauts will check the systems and then they will come
9:23
home. They're not going down to the
9:25
surface on that first one. Which brings us to the
9:27
consequence of all this testing. NASA stated that the key to
9:31
unlocking the capability to live on the moon is simply the launch
9:34
rate. Launch rate, It's all about
9:36
frequency. Therefore, the weird partial
9:38
engine tests on the pad and the ship squeezing simulator aren't
9:42
just isolated engineering experiments, they are the direct
9:45
bottleneck. Yes, this limits NASA's lunar
9:49
presence entirely based on how fast these specific hardware
9:53
tests can be finished. Because space exploration is
9:55
fundamentally A logistics problem.
9:57
Explain that. To build a sustained presence on
9:59
the moon, you need thousands of tons of cargo, fuel and habitat
10:04
materials. You cannot achieve that with a
10:05
vehicle that launches once a year.
10:07
You need a vehicle that can launch, land, refuel and launch
10:10
again within days. So that is why they are testing
10:13
the limits of the launchpad with 10 engines.
10:16
That is why they were physically squeezing the hull of ship 39
10:19
with a giant metal vise. Exactly.
10:22
Every single test is designed to eliminate a barrier to a high
10:26
launch rate. If they cannot catch the rocket,
10:29
they cannot reuse it quickly. And if they can't reuse it
10:32
quickly, the logistics chain breaks down.
10:34
And think about the pad. If the launchpad blows up during
10:37
ignition because the deflective buckets fail, they face a long
10:40
construction period to rebuild. It and Artemis sits on hold.
10:44
Right. The entire lunar program is
10:46
tethered to the structural integrity of those lifting pins
10:49
and the thermal resistance of that launchpad deflector.
10:52
NASA's ability to execute Artemis 4 and Artemis 5 is
10:56
completely dependent on how quickly these physical stress
10:59
tests at the launch site yield successful data.
11:02
It's all connected, but we need to zoom out to the financial
11:04
machinery driving this engineering.
11:06
The money behind the Rockets? Right.
11:08
The parent company is reportedly considering a NASDAQ listing
11:12
targeting a $1.75 trillion valuation to gain early entry
11:17
into the Top 40 index. It's staggering money.
11:20
Wall Street usually hates exploding prototypes, but this
11:23
specific CEO has trained investors to see explosions as
11:27
fast data collection. You see this corporate scale
11:30
directly tied to the speed and parallel ventures led by the
11:33
same CEO? Absolutely.
11:35
A giant AI supercomputer cluster was built in Memphis with
11:39
100,000 GPU's operational in an extremely short build period.
11:44
Additionally, a tunnel boring machine just completed a record
11:48
2.28 mile underground loop segment.
11:51
Can you explain why getting into the Top 40 index of the NASDAQ
11:54
matters for blowing up rockets and digging tunnels?
11:57
It creates an infinite foundation of capital.
12:00
How does that work? If a company gets listed in the
12:02
Top 40 index, massive institutional investment firms
12:06
and index funds are mathematically required to buy
12:08
the stock to balance their portfolios.
12:10
Oh, because they track the index.
12:12
Exactly. They don't have a choice.
12:13
It guarantees billions of dollars of permanent capital.
12:15
That capital funds the lunar Lander development, the AI
12:18
supercomputer expansion, and the tunneling projects all at once.
12:21
The brute force engineering approach observed on the
12:24
Launchpad is the exact same mechanism driving these
12:27
corporate valuations and scaling those parallel tech projects.
12:31
It opens up unprecedented funding pathways but also
12:34
creates immense pressure to maintain an aggressive testing
12:38
pace across all industries. Because when you look at the
12:41
Memphis supercomputer, getting 100,000 GPUs installed, powered,
12:46
and cooled in an extremely short build period requires A
12:50
logistics chain that ignores traditional industry pacing.
12:55
It requires running multiple construction and integration
12:57
phases simultaneously. You push until something breaks,
13:01
you fix it, and you push again. We are seeing that exact same
13:04
methodology applied to the Rockets.
13:06
They are not building 1 vehicle, testing it, and then carefully
13:09
building the next one. No, not at all.
13:10
They are building multiple vehicles, testing components of
13:13
them simultaneously across different facilities, and
13:16
accepting a high level of hardware loss as a natural cost
13:19
of gaining data quickly. Which would terrify most
13:21
traditional aerospace companies. Completely.
13:24
But whether it is moving 68,000 cubic yards of dirt for a 2.28
13:30
mile underground tunnel or pushing thousands of gallons of
13:33
liquid oxygen through a partially completed rocket
13:36
engine manifold, the goal is always to find the failure point
13:39
as quickly as possible, reinforce it, and move forward.
13:43
The data is the product right now, yes.
13:45
And the valuation target of $1.75 trillion provides the
13:50
capital required to sustain that high rate of hardware loss,
13:53
right? If you blow up a rocket or melt
13:55
a launchpad, you need the financial runway to immediately
13:58
replace it without slowing down the overall program.
14:01
The meticulous engine spacing and physical stress tests
14:04
happening on the pad right now are the mandatory physical steps
14:07
needed to satisfy NASA's demand for a higher launch rate.
14:11
Every test clears a very specific hurdle toward
14:14
consistent lunar flights. With the potential pressure of a
14:16
trillion dollar public market valuation looming, you have to
14:20
wonder if that financial weight will accelerate this aggressive
14:23
testing process or eventually force them to play it safe.
14:26
If you're not subscribed yet, take a second and hit follow on
14:29
whatever app you're using. It helps us keep making this.
14:32
We appreciate you being here.