In article ,
says...

On 2020-01-19 09:57, Jeff Findley wrote:

Firstly, Merlin engines are designed to throttle.

Also, in an abort situation, you don't care one bit if you destroy the
engines (i.e. turbopumps) in the process of quickly shutting them down.

I would assume that the mechanical design of the valves was done to
ensure that engine shutdown command was survivable. Adding a new mode of
shutdown that is quickler but destructive would introduce an unnenssary
failure cause.

I'm quite sure Merlin relies on software to throttle the engine. This
isn't an engine designed in the 1940s or 1950s with a mechanical engine
controller.

Again, in order to have a fast throttle response, for things like
landing, Merlin no doubt has very fast actuators on the valves leading
to the gas generator. Slamming those valves shut is something which can
be done in software.

And in the video, there was a fair gap in time between the crowds
counting down to 0, a progressive shrinkage of flame bhind engines till
no flame. Not like an on-off of a light switch.

I've not gone through the video frame by frame, but when I watched it,
it was on the order of seconds at most.

But Musk stated in post test press conference that the command to Dragon
is sent milliseconds after the "shutdown" command to engines.

A delay measured in milliseconds is still quite fast. It's an order or
magnitude faster than any human could ever react (this is why when you
drive a car you should have a two second following distance).

He also
stated that Dragon has enough power to escape even with Falcon9 still
producing thrust. Mentions theoretical capability up to 6Gs.

"Theoretical capability" is an interesting way to word it.

In this test, both he and NASA were impressed that Dragon experienced no
worse than 3.5Gs during escape.

I've literally been on "thrill rides" at the Kings Island theme park
with higher G loads than that. One of the steel coasters had a
manufacturers plaque on it that, among other facts, listed the maximum G
load. At the time I noted that was quite a bit higher than max G during
a space shuttle launch.

Cite:
http://www.tecumseh.k12.oh.us/Downlo...rPoint_9_3.pdf

So, far less than any military ejection seat. Should be quite
survivable, especially considering the orientation of the seats in
Dragon 2 is far more favorable than an ejection seat which puts an
incredible compressive load on the pilot's spine.

If you're in an abort situation, you're *already* having a "bad day" and
are trying to save the crew's lives!

If the abort is for reason other than engines exploding (aka: just off
course or enough engines shutdown) , you may not wish to cause an
emergency "abrupt" shutdown that initiates engines exploding.

Making the "decision tree" complex introduces its own possible failure
modes. Keeping it simple is almost always better, which means slamming
the valves shut as fast as possible. Again, the lives of the astronauts
are always more important than trying to save the Falcon 9 first stage.
Best not to take chances with complex logic, IMHO.

SpaceX and NASA would have the gory details. I'm just saying what makes
the most sense from an engineering point of view. #1 requirement is to
save the crew in an abort. Nothing else even comes close.

Regardless of the situation, Dragon 2 will leave *immediately*. Which
is to say it will start pressurizing the tanks for the Super Dracos the
instant that it receives the abort signal.

I thought those were pressurized at the time lauch abort is armed on the
pad, before launch and in case of abort they just needed to open a
valve to let fuel and oxydizer flow , meet, fall in love and create
fireworks.

No.

But at the press conference, Musk mentioned that the tanks are
pressurized as first step in an abort.

Correct. You normally don't want the tanks at abort pressure because
that's far higher than needed for the Dracos.

The human body can
take that jerk to 4 Gs. Look up the USAF ejection seat tests done
decades ago. This acceleration is quite tame compared to what e-seats
expose USAF pilots to.

I've worked out the logic (I think). Assuming fixed thrust on Falcon9
and progressively increasing thrust on Super Dracos. As Super Dracos
increase thrust, the weight Dragon exerts on Falcon9 is reduced,
allowing Merlin engines to accelete the stack at higher rate.

Eventually, Dragon accelerates faster than Flacon9 and separation happens.

Yes, basic physics here.

So, if Super Dracos go from 0 to 4G acceleration "instantly", the crew
will go from existing 3G to 4G acceleration. So escaping from a running
Falcon9 wouldn't result in a big jolt.

It's actually called jerk. Jerk is defined as the rate of increase in
acceleration. In this scenario, jerk is essentially caused by the ramp
up in thrust of the Super Dracos. The faster they ramp up in thrust,
the higher the jerk.

But if they wait for engines to shut down, they woudl go from 3G to 0G
to 4G at which point it is a big jolt.

The pressurization and firing of the Super Dracos takes less than a
second. That is happening simultaneously with Merlin shutdown. Would
be interesting to see two graphs:

1. the overlay of normalized thrust (0-100%) of the Merlins overlayed
with a graph of normalized thrust (0-100%) of Super Dracos. This would
show what I'm talking about above.

2. the overlay of acceleration (actually deceleration) of Falcon
overlayed with a graph of acceleration of Dragon 2.

However, in the press conference, Musk alluded to the thrust from Super
Dracos is adjusted based on conditions. (Crew safety mentioned). So if
Falcon has shut its engines, Dragon wouldn't have thrust has high from
Super Dracos than if the Falcon9 engines are still running.

So it appears to be a fairly smart system.

I saw that. Still, Dragon 2 won't wait for the Merlins to shut down.
They're going to fire as quickly as they can. What they will change is
the thrust of the Super Dracos. If the computers think the engines can
be safely shut down, the Gs on Dragon 2 won't be as high as the scenario
where Dragon 2 determines that the Falcon went "boom". In the case of a
"boom", I'd expect Dragon 2 to GTFO as quickly as possible.

safety comes next. But in many cases, like this abort test, the abort
will result in the destruction of the Falcon 9 launch vehicle before the
range safety system attempts to fire.

An abort in the pad prior to launch does not necessarily result in the
Stage1 being destroyed.

Prior to launch if the Falcon isn't in the process of being destroyed,
why are you aborting? Aborting Dragon 2 is dangerous. Like ejection
seats, you only do it because it's literally attempted suicide to avoid
certain death.

This situation simply *can't* happen. Dragon 2 would abort *long*
before "imminent danger" to anyone on the ground. Also, this is why
Falcon 9 flies *east* out over the ocean. Nothing launched from KSC
flies towards Orlando.

I realize that. But NASA stll feels it necessary to have range safety.
(and it will be very interesting to see how that policy is adapted to
Starship)

You're not getting it. Range safety provides for an area around the
launch pad and along the flight path where debris will drop in case the
launch vehicle destroys itself or is destroyed by range safety. As
such, you don't have to destroy Falcon 9 at the first sign of trouble.
You have to first abort, which means shutting down the engines and
telling Dragon 2 to GTFO. Deliberate destruction of Falcon 9 will
therefore happen after the abort.

Now, there is always the chance Falcon 9 goes "boom", in which case
destruction of Falcon 9 becomes moot (impossible, actually since it
would already be in pieces).

Consider the Shuttle. They had range safety on it with crew on it. So
obviously NASA had considered the issue of having to initiate range
safety with crew still on board.

But also note that the shuttle orbiter itself had no destruct charges.
Only the SRBs (to unzipper them and terminate thrust) and the ET (to
disperse propellant) had destruct charges. Range safety is to protect
people on the ground, so putting charges on the orbiter, or Dragon 2,
would make no sense because they need to protect the crew.

This is why both Dragon 2 and Starliner have already performed pad abort
tests (both passed).


Were those tests done from an actual rocket at the pad, or were they
done on the ground with the Dragon/Starliner on a cradle and it just
fired its engines to pop up, deploy parachutes and come back down?

On the ground. There was simply no reason to stack either of them on
top of an expensive launch vehicle to perform this test.

It is a simple answer. Either your sensors indicate the engine is
working properly, or they don't.


Was hoping for answer one whate xactly do those sensors sense to gauge
thrust being produced by an engine. For now, I will take explanation by
someone selse of messuring pressure in combustion chamber. (I assume
there is a pipe that leads to a sensor far enough to not burn).

On something as complex as a liquid fueled rocket engine, there are
actually lots of pressure sensors, temperature sensors, accelerometers
(to measure vibration), and etc. There are also sensors in the tank
(certainly pressure and sometimes a "low level" sensor as well). I
simply don't want to go down the rat hole of talking about every
possible sensor and what it would mean if it read "lower" or "higher"
than expected.

BTW, during the press conference, Musk mentioned that they are looking
at p]reventing Dragon 2 from landing in water by catching it i flight.
They are precticing with fairings and if that ends up being reliable,
may try with a Dragon (not sure if mentioned Dragon or Dragon 2).

I heard about that. I think NASA would nix that right away. SpaceX
would certainly want to test this with Cargo Dragon and Cargo Dragon 2
before trying it with a Crew Dragon 2. The problem here is, what if you
barely miss it? Say you smack the capsule with one of the steel "arms"
on the fairing recovery ship as it's falling. What kind of damage would
happen?

Just like testing propulsive landing with Cargo Dragon 2, NASA won't
like the idea of things like essentially irreplaceable EMUs being lost
at sea due to testing of this sort.

Not sure how realistic this is, but I guess this really means SpaceX
won't try to convert Dragon 2 to powered landings.

That was nixed by NASA long ago for the reasons stated above.

Also, for those who are interested, here's a link to the transcript:

IFA Post-Presser Presser
2020-01-19
https://gist.github.com/theinternetf...25229fb6a17195

From above:

Q2: Elon, does the press deserve a dance today? [room laughs, groans]

Elon Musk: I can't take these expectations of dancing constantly. I'm
not that good of a dancer. I haven't ever worked on my skills.

Q3: Dance!

Elon Musk: No! I am not your dancing puppet!

Jeff
--
All opinions posted by me on Usenet News are mine, and mine alone.
These posts do not reflect the opinions of my family, friends,
employer, or any organization that I am a member of.