Hacker News Reader

Comment by johnyzee | Hacker News Reader

Yeah, breaking up in-air during re-entry (at 65 km going by last telemetry) seems like a potential big issue to fix.

OP's point is that the tests become a bit cheaper and a bit easier to get licenses for since they can get to orbit and can deploy Starlinks. The reentry problems of course have to be fixed, but the FAA mishap investigation will involve fewer delays, just like how Falcon 9 was able to keep flying, attempting landings without having to wait for a mishap investigation to finish every time a landing failed.

Iirc F9 had its main challenge in that final suicide burn. Usually the boosters made it there just fine and then "only" failed to get the parameters aligned just right.

Based on what we saw today, the main chokepoint for Starship might be right in that hot reentry telemetry blind spot. Debugging without logs.

It sounded like they might have had flight data recorders on-board which might have survived through reentry with telemetry. If so, they'd have some logs, but yes, figuring out reentry will likely be the main challenge for now, not only is it hard to debug, even getting to it requires everything else to go more or less perfectly.

With all the solid state gizmos we have today, is it really that difficult to make a blackbox that can survive serious impact G-forces, especially if it just needs to work and doesn't need to be safety rated?

Unlike with F9, in this case they most probably had high-bandwidth telemetry all the way till the vehicle broke up (if the high-def video stream is anything to go by). So, they probably have a lot more data this time. Even the video can yield information - for example, given that the cameras were mounted on the flaps, they can probably back out the actuation angles from that.

Video feed interruption was very much expected, and went through a series of on/off before the final signal. From what we saw in the live presentation, both scenarios are possible: breakup while transmission was up, transmitting as long as latency allows, with hopefully good explanations in the log, or a disconnect from the challenging radio situation in the reentry plasma long before something went wrong.

Unless they do find some failure buildup in the logs they did receive, it will be difficult to tell disconnect due to breakup from breakup (long) after disconnect. Gradual signal loss on the receiving side would indicate that it's likely the latter, but I'm not sure a signal loss unrelated to breakup would necessarily be gradual. I'd guess atmospheric properties can change fast at reentry velocity.

In the livestream they mentioned that they lost signal on both Starlink and TDRS simultaneously, which is why they figured it had broken up at that point. They were prioritizing the telemetry over video frames on Starlink, and TDRS can only handle telemetry.

Thanks, must have missed that detail in the stream. Hope they find a good explanation candidate in the data!

> Yeah, breaking up in-air during re-entry (at 65 km going by last telemetry) seems like a potential big issue to fix.

That's what pretty much every non-SpaceX rocket does today with very few exceptions.

And indeed every spacex rocket. Starship is the second stage -- and a second stage on a F9 burns up in the atmosphere just as much as Starship from IFT3 did.

The big things to fix seem to be

1) The roll rates for starship (which prevented the inflight relighting test)

2) The Booster relights, which didn't have enough lighting to soft land.

Both of those feel like minor problems.

I'm amazed that the propellant transfer demo seems to have worked first time, and of course they managed to get the tihng up there in the first place.

The reusability of the Starship part of the system is a much bigger unknown, but that doesn't seem to be necessary for the next launch.

Wouldn't surprise me if the next try is by the end of April

The point they were trying to make is that getting up is MVP, getting back down can be figured out later

Why? Saturn V broke up in-air during re-entry on every "successful" launch as does every ULA rocket.

Because that is their stated design goal for Starship.

And you don't need to meet that goal to start launching (risk-tolerant) payloads to orbit during their test flights. Especially with such a hardware-rich development program.

it's the expected outcome of an imperfect (literally not precisely perfect) reentry. something went wrong. it could be a very minor thing like a stuck valve somewhere. we'll know in the next test if they figured it out.

bright side: booster made it all the way to splashdown

haha, it was travelling at 1100 km/h...

Not so much a splashdown but a smashdown :-D

to have been a fly on the wall... was surely a site to behold :)

Reply on news.ycombinator.com