Armchair analysis of Delta performance shortfall

How can the Delta performance shortfall be explained?

First, the rockets fired for too short of a time. Either (a) they did
not start with a full load, or (b) they ate it too fast, or (c) they
shut off while fuel still remained.

But (a) is unlikely since the same CBC have flown before, so they know
the exact capacity. (b) seems unlikely since they adjust the thrust by
measuring the fuel flow, I'd assume, and since the Delta 4 medium with
the same tanks and engines fired for the expected amount of time.
Also, if the engines were using extra fuel they would have
proportionally higher thrust while firing, leading to little or no net
loss of performance.

We are left with (c), they shut down with fuel remaining. But why?
Watching the replay, for both the outer boosters and the core,
everything is normal until they try to throttle back. When the
throttle back time comes, they shut off completely instead.

The side boosters are supposed to run for 10 sec at reduced thrust; the
core booster for 16 seconds. If cutoff is due to running out of fuel,
then the side boosters are missing about 10sec times 58% thrust = 5.8
seconds of (full throttle) fuel. The core stage is missing 16sec times
58% = 9.3 seconds of (full throttle) fuel. This would be quite some
coincidence if it was fuel starvation, which is further evidence the
fuel amount is not the problem.

On the other hand, the center engine throttled down OK at about 1
minute on the heavy flight. Also, on the regular delta 4 flights, the
same engine has always throttled down OK about 40 seconds from the end
of the first stage firing.

So what's different about the Heavy? Two things stand out; the
acceleration is less, since the second stage is heavier, and the amount
of fuel in the tanks is less. Both of these lead to less pressure at
the pump inlets.

So my guess is that when they reduced thrust on the side engines, the
combination of less fuel in the tank, low acceleration, and maybe
structural rebound from the removal of full thrust, caused the pressure
at one or both engines to get too low. This caused at least one engine
to go out (on the video it looks like the left one goes out first) and
hence the other to get shut off. The center engine was unaffected at
that point (and when it throttled down earlier) because it still had
plenty of fuel in the tank.

Then the same thing then happened when the center engine tried to
throttle. Compared to a Delta 4 medium flight, there was less fuel in
the tank (16 sec vs 40 sec), and less acceleration (heavier second
stage and payload). The combination caused less pressure at the inlet,
so this engine too ate a bubble or cavitated or whatever it does when
it gets too little pressure, and went out. The rest, as they say, is
sci.space.history.
My guess, worth exactly what you paid for it,

Lou Scheffer

wrote in message
ups.com...
How can the Delta performance shortfall be explained?

First, the rockets fired for too short of a time. Either (a) they did
not start with a full load, or (b) they ate it too fast, or (c) they
shut off while fuel still remained.

Actually there is another perspective, which appeared with the Ariene 5
first launch.
Flight software not taking into proper account flight profile differences
between Medium / Heavy

But (a) is unlikely since the same CBC have flown before, so they know
the exact capacity. (b) seems unlikely since they adjust the thrust by
measuring the fuel flow, I'd assume, and since the Delta 4 medium with
the same tanks and engines fired for the expected amount of time.
Also, if the engines were using extra fuel they would have
proportionally higher thrust while firing, leading to little or no net
loss of performance.

We are left with (c), they shut down with fuel remaining. But why?
Watching the replay, for both the outer boosters and the core,
everything is normal until they try to throttle back. When the
throttle back time comes, they shut off completely instead.

It will be interesting to see if the flight software had this contingency
covered and that sufficient allowances were fully developed to maximize the
situational parameters actually experienced on this flight.

The side boosters are supposed to run for 10 sec at reduced thrust; the
core booster for 16 seconds. If cutoff is due to running out of fuel,
then the side boosters are missing about 10sec times 58% thrust = 5.8
seconds of (full throttle) fuel. The core stage is missing 16sec times
58% = 9.3 seconds of (full throttle) fuel. This would be quite some
coincidence if it was fuel starvation, which is further evidence the
fuel amount is not the problem.

On the other hand, the center engine throttled down OK at about 1
minute on the heavy flight. Also, on the regular delta 4 flights, the
same engine has always throttled down OK about 40 seconds from the end
of the first stage firing.

So what's different about the Heavy? Two things stand out; the
acceleration is less, since the second stage is heavier, and the amount
of fuel in the tanks is less. Both of these lead to less pressure at
the pump inlets.

Good points and the flight software must take these into account.

So my guess is that when they reduced thrust on the side engines, the
combination of less fuel in the tank, low acceleration, and maybe
structural rebound from the removal of full thrust, caused the pressure
at one or both engines to get too low. This caused at least one engine
to go out (on the video it looks like the left one goes out first) and
hence the other to get shut off. The center engine was unaffected at
that point (and when it throttled down earlier) because it still had
plenty of fuel in the tank.

Then the same thing then happened when the center engine tried to
throttle. Compared to a Delta 4 medium flight, there was less fuel in
the tank (16 sec vs 40 sec), and less acceleration (heavier second
stage and payload). The combination caused less pressure at the inlet,
so this engine too ate a bubble or cavitated or whatever it does when
it gets too little pressure, and went out. The rest, as they say, is
sci.space.history.

This highlights a difference (and disadvantage in this case) of the 3 core
configuration versus the Saturn V or even the Space Shuttle clustered engine
approach. (I can't remember the final Saturn I-C configuration). You lose
an engine, you still have all of the fuel available for the remaining
engines -- more options (burn longer on remaining engines).

If this was a manned mission, then this would have been an abort scenario.

gb

"w9gb" wrote in message
news:hXKAd.586273$wV.480123@attbi_s54...
If this was a manned mission, then this would have been an abort scenario.

Probably not. NASA preference is for insertion into orbit, even if it is the
wrong orbit, rather than subject the crew to the hazards of a first-stage
abort.

-Kim-

Kim Keller wrote:
"w9gb" wrote in message
news:hXKAd.586273$wV.480123@attbi_s54...
If this was a manned mission, then this would have been an abort
scenario.

Probably not. NASA preference is for insertion into orbit, even if it
is the
wrong orbit, rather than subject the crew to the hazards of a
first-stage
abort.


But maybe you have no choice. If it's a LEO mission with a near
maximum payload, a serious performance shortfall means no orbit (at
least no orbit high enough to avoid hitting the atmosphere). If you go
as high as you can, perhaps you'll come down in mid Indian ocean, or
mid-Pacific (like the shuttle tank), and it could take a long time to
find you, and you are taking a chance on the weather there. It's
probably better to wait till the first stage stops, or turn it off if
you can, and abort as soon as possible. I suspect you'll just be a few
hundred miles from the Florida coast, where there are lots of resources
to find and recover you.

Lou Scheffer

wrote in message
oups.com...
But maybe you have no choice. If it's a LEO mission with a near
maximum payload, a serious performance shortfall means no orbit (at
least no orbit high enough to avoid hitting the atmosphere). If you go
as high as you can, perhaps you'll come down in mid Indian ocean, or
mid-Pacific (like the shuttle tank), and it could take a long time to
find you, and you are taking a chance on the weather there. It's
probably better to wait till the first stage stops, or turn it off if
you can, and abort as soon as possible. I suspect you'll just be a few
hundred miles from the Florida coast, where there are lots of resources
to find and recover you.

Of course there are plenty of instances during first stage when an immediate
abort would be required. I was addressing the instance of the Delta IV
underperform - in a case like that, where the crew is in no immediate
danger, the decision would be to do what STS calls an abort-to-orbit.

-Kim-

I'm not sure how Delta IV fuel-level sensing works, but
in the past, on other vehicles, early engine shut downs
have occurred when fuel level sensors were fooled into
sensing "empty" tanks after fuel sloshing exposed them
to the non-liquid environment side of the tank. This
started the normal commanded engine shutdown,
similar to what appears to have happed to Delta 310.

In addition, liquid hydrogen acts a little different than
denser fuels under such conditions. This was the first
time a hydrogen booster/core combination has ever
flown a combined, parallel boost profile complete with
major throttling and staging events. There could be
some tricky fluid/structural dynamics happening here
that Boeing hasn't figured out yet.

On the other hand, the explanation might be simple.
Has anyone else noticed that the launch photos posted
by Boeing have disappeared?

- Ed Kyle

I'm not sure how Delta IV fuel-level sensing works, but
in the past, on other vehicles, early engine shut downs
have occurred when fuel level sensors were fooled into
sensing "empty" tanks after fuel sloshing exposed them
to the non-liquid environment side of the tank. This
started the normal commanded engine shutdown,
similar to what appears to have happed to Delta 310.

In addition, liquid hydrogen acts a little different than
denser fuels under such conditions. This was the first
time a hydrogen booster/core combination has ever
flown a combined, parallel boost profile complete with
major throttling and staging events. There could be
some tricky fluid/structural dynamics happening here
that Boeing hasn't figured out yet.

On the other hand, the explanation might be simple.
Has anyone else noticed that the launch photos posted
by Boeing have disappeared?

- Ed Kyle

Kim Keller wrote:
wrote in message
oups.com...
But maybe you have no choice. If it's a LEO mission with a near
maximum payload, a serious performance shortfall means no orbit (at
least no orbit high enough to avoid hitting the atmosphere). If
you go
as high as you can, perhaps you'll come down in mid Indian ocean,
or
mid-Pacific (like the shuttle tank), and it could take a long time
to
find you, and you are taking a chance on the weather there. [...]I


Of course there are plenty of instances during first stage when an
immediate
abort would be required. I was addressing the instance of the Delta
IV
underperform - in a case like that, where the crew is in no immediate

danger, the decision would be to do what STS calls an abort-to-orbit.

But if you are carrying a heavy load to LEO, and have underperformance
of the first stage, there is no abort-to-orbit option - it's not
physically possible no matter how much you would prefer it. You need
all the normal delta V just to achieve orbit. If the first stage
seriously underperforms, you won't have enough delta-V to reach even a
low orbit, even if all other stages work perfectly. Your only choices
might be to re-enter now or get nearly to orbit and come back down.

With the shuttle, the abort-to-as-high-as-you-can-get might make sense
- the shuttle can use the height to reach a broader range of landing
sites. If instead you have a capsule with limited maneuverability, you
might really be taking your chances if you just continue - if you can't
get to orbit, you'll just land whereever the capsule takes you. In
this case it might be better to abort earlier, and land where the
support is better....

Lou Scheffer

In article . com,
wrote:
...in a case like that, where the crew is in no immediate
danger, the decision would be to do what STS calls an abort-to-orbit.

But if you are carrying a heavy load to LEO, and have underperformance
of the first stage, there is no abort-to-orbit option - it's not
physically possible no matter how much you would prefer it. You need
all the normal delta V just to achieve orbit.

Not so, for two reasons. First, there *will* be fuel margins against
slight underperformance of the engines. (One of the ways the Saturn V
was improved for the J-series Apollos was to reduce those margins, based
on better understanding of typical performance variations.) Second, the
orbit you need to carry out the mission typically will not be the lowest
possible orbit, so some small performance shortfall is still compatible
with achieving a temporary orbit (especially since abandoning the primary
mission frees up some OMS fuel etc.).

...Your only choices
might be to re-enter now or get nearly to orbit and come back down.

Given a sufficiently large performance shortfall, that's certainly
possible. But the second option might still be desirable, for the sake of
better reentry conditions (suborbital reentries tend to be harsh because
of the steep entry angle) and better landing sites (including the option
of maneuvering later in ascent to improve the ground track). Notably, I
can easily see a once-around abort into the Gulf of Mexico being preferred
over a suborbital abort into the North Atlantic.
--
"Think outside the box -- the box isn't our friend." | Henry Spencer
-- George Herbert |

Henry Spencer wrote:
In article . com,
wrote:
...in a case like that, where the crew is in no immediate
danger, the decision would be to do what STS calls an
abort-to-orbit.

But if you are carrying a heavy load to LEO, and have
underperformance
of the first stage, there is no abort-to-orbit option - it's not
physically possible no matter how much you would prefer it. You
need
all the normal delta V just to achieve orbit.

Not so, for two reasons.

There are at least 3 classes of underperformance:
(a) out of spec, but covered by magins
(b) not covered by margins, but at least you can reach *some* orbit
(c) you can't get to any orbit
First, there *will* be fuel margins against
slight underperformance of the engines. [...]

That's case (a). An example was the SeaLaunch flight, and some Apollo
missions.

Second, the
orbit you need to carry out the mission typically will not be the
lowest
possible orbit, so some small performance shortfall is still
compatible
with achieving a temporary orbit (especially since abandoning the
primary
mission frees up some OMS fuel etc.).

That's case (b), which was the Delta Heavy flight. This is the 'abort
to orbit' option.

...Your only choices
might be to re-enter now or get nearly to orbit and come back down.

Given a sufficiently large performance shortfall, that's certainly
possible.

This is case (c). If the Delta heavy had been boosting astronauts into
LEO, I think this would have been the case. The second stage appears
to have been about a minute short of fuel. Assuming 1G acceleration
for that minute, that's about a 600 m/s (about 2000 ft/sec) shortfall,
even after all reserves were used. That's enough that you can't reach
*any* stable orbit, and any likely OMS can't make up the difference.
So you are going to come down - the only question is where...

But the second option [continuing rather than aborting when you
detect
the problem] might still be desirable, for the sake of
better reentry conditions (suborbital reentries tend to be harsh
because
of the steep entry angle) and better landing sites (including the
option
of maneuvering later in ascent to improve the ground track).

These are the only options you get, so certainly pick the best of what
you have.

Notably, I
can easily see a once-around abort into the Gulf of Mexico being
preferred
over a suborbital abort into the North Atlantic.

Maybe, if you are just a little short. In this case we are discussing,
this option would not be possible, I think. The shuttle uses a 300-400
fps re-entry burn, then comes down in half an orbit, and it's a good
glider compared to a capsule. For a hypothetical Delta Heavy mission
with the same performance shortfall, your choices will range from a few
hundred miles downrange (if you abort as soon as you see you won't make
orbit), to a few thousand miles. In this case you might reasonably
choose the few hundred...

Some modern airline navigation gadgets, and I'd guess the Shuttle, have
real time predictive displays (how much runway will I use/how far can I
get, given measured performance so far and predicted performance for
the rest) exactly to help in making this type of decision. I assume
any new space system would have this as well. This should be better
than strict time based rules (i.e. before X seconds, abort now, after X
seconds abort long) since it can cover more cases.

Lou Scheffer