Russel E. Rhodes: Shuttle Launch Development and Operations

Good morning. Audience: Good Morning
Russel Rhodes: Uh, I, uh,
will, uh,
separate through, I think. I thank
Tom, for a good introduction.
And, uh, before I get in
to the material I have here, I want to first
mention that when we transition
from one program to another
and especially the shuttle program
as it was that, uh, we
we, uh, we had, uh, budget constraints,
that was a severe thing. We had to
compromise it was pointed out.
And, one of the things that was
important was how much of the existing
systems could we reuse.
So, as you all can tell
we reused the VAV, we reused the
crawler, the WAVE
the crawler transporter
and, uh, the mobile launchers
became, uh, reuse of
the old, uh,
launch umbilical tower platforms.
And, even the towers
that were on those,
on those, uh, launch umbilical
towers, two of those were
removed and put on the pads.
So, that tower that sits at the pad
was the same tower
that was used in Apollo on the LUT.
So, we
used, reused, as much of this hardware
as we could to cut down the
front end or the development costs
as possible. And, I
think it's important that
you, uh, that we
make that known to you all and you realize
that, uh.
I'm at a point in my
life where I have to change my glasses
frequently. And, I'm right now at a point
where I look up here, it's pretty fuzzy.
So, uh,
I've got an appointment, but I'm
a little bit lagging in getting my glasses changed.
So,
at my notes to see if I can't
help that a little bit, uh.
The first thing when, uh,
when the program, uh, is
accepted as a, uh,
except in architecture,
is to look and see now
are we going to be able to
meet these requirements? And,
can you determine the functionality
through verification to
uh, mitigate the risk of a
catastrophic event? And, uh,
that's pretty important.
So, uh, when I look at
those, the first thing I look at it
is the, the fact on the Saturn I
we had, uh, a
system that supported the
vehicle that was totally active
and this system had to operate
and release
in order for it to be a good launch.
If we had a failure
of that system, it would have been a bad day.
There was no backup.
It was pure and simple,
a high risk system.
So, likely, we would want to
try to avoid this kind of
systems in the future.
Same thing is true of the Apollo.
We had, uh, multiple T-0
swings arms.
Well, uh, they all worked really nice; however,
you know, had one of them failed
during the launch,
I mean, it's a T-0,
that means you've already launched
the vehicle and if it fails,
it's gonna be a bad day.
Because, uh, the vehicle is not gonna
tolerate running into one of those swing arms.
So, another
design to try to stay away
from is T-0
swing arms.
Uh,
one of the things that
the early shuttle had, uh,
was a, uh,
an external tank that was a
lightweight balloon structure.
In fact, it couldn't support the
orbiter, and, therefore;
uh, the architecture had a, what they call,
a trolley. And this little trolley
sat underneath the
SSMEs, under the orbiter,
and actually supported the orbiter weight
to keep it from resting on the tank,
because the tank couldn't tolerate that.
So, uh, the idea was that,
that, uh, you would light the engines,
and as soon as it built up
thrust and carried the weight of the orbiter
this little trolley now would
move off and get into a
protective enclosure.
However, if you had an abort, and you
shut those engines down
before you launched, that little trolley
had to get back up under there and
receive that vehicle to support that weight
in order to avoid a
catastrophic event.
Uh, one of the
things, when I brought that to the attention of
our program managers, they said
"I think you're absolutely right,
that's something we can't tolerate." And so,
immediately got with Jim
Modem, and uh,
and Bob Thompson at JSC
and discussed this and
determined, you know, this was not
a good architecture to start
on. We needed to change that.
So, that evolved into a
freestanding external tank
that could support that weight
and we eliminated that whole trolley system.
Uh, another
item is a T-0 damper arm.
Again, if you have to have
a very lightweight vehicle
that's tall and slender
it's kind of like the CLV
was planned to be.
And, uh, it needed a damper arm to hold it
uh, there again
if that, that needs to not be a
T-0 function.
So, it would be a bad day
if it did not operate.
Uh,
at this point I'd like to correct something
that was said here about orbiter
or crew abort systems.
The, uh, original
shuttle architecture
had a crew abort
system. This seems to be a
lost identity here. But, there was
two solid rocket motors
attached to the orbiter, one on each side
of the fuselage.
And, that system was in the original
architecture. And, uh,
the, uh, program management
uh, deliberately removed those,
because when you
assess the, uh,
the probability of failure, uh,
of those
motors being on each side
of the crew vehicle,
that close, versus the style of motors
you're trying to get away from,
it didn't make a whole lot of sense.
So, as a result, they were,
they were removed. Needless to say,
it took off a lot of weight as well.
But, uh, and helped performance. But,
nevertheless, there was a
crew abort system originally,
and it was intentionally removed.
One of the things that, uh,
we, uh, we needed to address
of course was our natural environment.
It's always a
major function and, uh, design
in, uh, flight and ground systems
is that, you must be able
to tolerate rain, wind
humidity, and et cetera.
And, uh, one of the things
that, uh, is also a function
is the lightning.
Apollo 12 was a
case in point, where
the weather was looking pretty bad,
but we really weren't in a storm
or rain, and
we launched Apollo 12.
Well, unfortunately, it got struck by lightning.
Uh, the, uh,
Apollo crew capsule
had circuit breakers.
And, uh, I worked with--
the launch vehicle did not.
We were very fortunate the lightning
tripped the circuit breakers,
so the system was total
void of power for a period of time.
And, uh,
we had a very calm crew obviously.
They just simply reached down and
reset the circuit breakers,
and volia, they had power.
We were very fortunate.
So, uh another lesson is to make sure
you don't launch one of these vehicles when
there's imminent potential of lightning
in the area.
Another one that,
as Lewis said,
"This is kind of like being around
a Fireside chat." And, he told me,
"This is storytelling."
So, I try to live up to his advertisement,
you know, that, uh,
I'm here to tell you stories.
Well, one of the things that came late
in the program, in shuttle development, was
a concern about hail.
And, uh, it was
less than a year from launch time
and we had all this concern
about, well, "Have we covered this?
Have we covered that?
Have we forgotten anything?"
And, then, "Oh, my goodness, we don't have any hail protection."
And, uh, being
a part of the team that looked
into those requirements,
I started investigating that. And I said, "You know,
we don't have any evidence of ever having
hail at the
launch area here, or the Cape or
any place. Our meteorological station
had no evidence of it.
So, yes, there's hail at Daytona Beach, there's hail south,
it's everywhere, but this area just does not
seem to ever get hail.
So, I don't think that's a major
risk, we ought to put that off.
We don't need another delay
in our schedule just trying to build
hail protection. So, our program
accepted that and it wasn't but
a short time that I was, uh,
visited by the media
that wanted to talk to me about natural environments
and how, well,
it was affecting our
shuttle program development.
So, I
expounded on the fact that, this hail
story, and he said,
"Well, that was great." He was gonna get a write up
and put it in the paper, about our natural environments.
Anyway,
shortly thereafter, I expected to see it in the paper.
But, it didn't show up.
And I said, "Whew," I breathed a sigh of relief.
At least that's not going to show up in the paper
somewhere on the back page or something.
Anyway, lo and behold
on Labor Day weekend
here is it is front page,
big article, expounding on the fact that
I had assured him there would be no hail.
[ Audience laughter ] And, uh,
I think you already understand
where I'm going. [ Audience laughter ]
It wasn't long after that we had a
hail storm. Well, guess what.
The only place it hailed was at the pad.
[ Audience laughter ]
Really, now that sounds spooky,
but that's a fact. The only place
it hailed was the pad. It didn't hail up at the BAB
or anywhere else. Just on the launch vehicle.
And, uh, it did
minor damage to the
external tank, foam and the
orbiter tile, but nothing was
major and serious to hold us up.
But, uh, again
uh, don't ever
think you're in control of these things
and you know it all, because there's someone else
there that knows better. So, it keeps you
humble.
Ice, frost
and debris, uh.
We could tell immediately that
when we saw the architecture, that we were
gonna be dealing with fragile tile, that we couldn't
fly cryogenic vehicles with
uh, the way we had before, with
lots of frost and ice on them.
And, so
we worked at that
process deliberately and tried
to come up with ground solutions, but they
didn't appear to be anything practical whatsoever.
So, we eventually ended up with
a compromise to
uh, insulate the external
tank, and, uh,
as best we could. And, for those areas
that we could not cover with, uh,
with an insulation, we used
hot gas purge. So, the nose
cone is a hot gas purge.
The inner tank has a hot gas purge
to keep any from the exhaust
of that gases from being below freezing.
And, uh, and
the press lines that go up the side,
they're also purged with
hot gas. So, it was a
combination of purges, and, uh,
insulation of the tank
that allowed us to get a vehicle
that we could, uh,
fly for 30 years.
Um,
Hazardous commodities.
Uh,
of course in this business, we deal with a lot of these.
But, there again, this is
an area that we need to deliberately try
to avoid as much as possible.
Uh, toxic fluids are a
major concern. They drive
cost. They, uh, one of the
things that, uh, that
is behind the scenes that we have to do
is all the people that could be
exposed to hypergolic
toxic fumes
need to have,
um,
health exams every six months, so that
you have a record in the event that any of these
people get exposed. You know exactly
what their exposure level was and
how to treat them. So it's a very
costly process as well as the
suits they have to wear and
the team of people that maintain those suits.
So, it, uh,
there again,
we've been using these for years, but
it's a very expensive process.
Cryogenics, uh,
the main concern in cryogenics is
fire, water hammer and the
sensitivity to contamination
and geysering. Uh,
geysering is probably a phenomenon
that most of you are probably not really
familiar with, but you are familiar with
I'm sure a percolator coffee pot.
Well, percolator coffee pot
uses a geyser
process to operate.
The, uh, the little bowl in the
bottom is a heater. And, you
have a tube in there that has
a capture bulb
at the bottom that allows that water to boil
and producing gas
and that seam now pushes the liquid in that
tube up into the basket where the coffee
grounds are to make the coffee.
And, if you notice when it
when it percolates, you always hear
this thump. That thump
is the refilling of that system
with water. That's the water hammer
you're hearing. So, now think
about that in terms of something large,
like, uh, a cryogenic system
with, um, with
long vertical lines that are 50 to 100
feet long. Uh, you can
get the same kind of situation, as you
can see. If you had
gas accumulate at the bottom of the line
as it rises, it's going
to expand because the pressure is reduced
and causes it to, uh,
to gassify, and convert
the liquid to gas
and you're going to produce a coalescent bubble
as we call it, and it will force all the liquid
out of that line into the tank
and leave it void.
Well, that part could be a problem,
because when it spews into the tank
it can collapse the ullage and
collapse the tank because of the drop in pressure.
But, the worst thing is
that when you refill that line,
that water hammer will destroy
everything at the bottom. So,
you've gotta avoid geysers.
So, again, if you can avoid
those kinds designs, uh,
it's definitely recommended.
This was an area that we undertook some
changes in the tank, because we built
a thermal pumping loop on the tank
to help us solve that
initially. And finally evolved
to understand that by
using a helium boost system in there,
we could operate that thing safely,
without the research system
and we were able to take that other line off.
But, uh, again
it's a very difficult
development area and
you need to be aware of it.
Of course, contamination
uh, liquid oxygen
is highly sensitive to any kind of
hydrocarbon contamination, so,
any system that you use liquid oxygen or
gaseous oxygen, has to be
surgically cleaned.
Closed compartments, uh,
closed compartments again, are
a difficult area
and they drive costs.
They drive costs because they
require you to control that area.
If there's hazardous gases that can
be in there, you need to monitor those.
So, you have to have monitor systems.
Uh, you need, uh,
to, uh, put inert gases
in there, but if you need to go back
in those areas, you gotta have a conversion
system to get the inert gas out
and air back in.
Um, you may not remember, but
on the shuttle program
on STS-1, we were
preparing, and we went
through an abort situation.
Well, the guys, uh,
were ready to go back into the aft compartment
to, uh, perform work.
They pulled off the, uh, the
door and made an entry.
And, uh, and the
team did not, uh, have, uh,
good clearance to--
that system had changed over.
And, we lost two people in the back end
of the orbiter from the
nitrogen environment that was there.
So, uh, again, closed
compartments are very,
very difficult
uh, to control, because it's simply
a control process.
In order to manage the safety
working with those systems.
Um, we seem to have difficulty
getting rid of those things and I don't quite
understand that, because the Russians fly
uh, opens truss compartments.
So, I think
we, uh, could learn from the Russians
in that sense.
Um, excessive free
and trapped hydrogen.
I think that, uh, it was mentioned
ignition overpressure is
a concern of
there's a, we had a
6.4 percent model
that was built in Huntsville
to test these environments
and to understand this overpressure
as well as to
understand our acoustic environment so we'd know
how to design and what
design to put in for
acoustic suppression, or as they call it
sound suppression at the launch pad.
Uh, and, uh,
we understood that
there would be a lead
hydrogen that would be expelled
from that engine, uh,
that would, uh,
would ignite once
it, uh, became involved in the air.
And, it got its ignition
from the, uh, engine ignition itself.
So, the question was, how
much hydrogen would that be and would it
damage the heat shield?
Uh, we convinced the people that sent us
to run a test on the
MPTA to try to understand that
and sure enough, the pressure
levels were not going to be acceptable.
So, we, uh, were required
to design a ground system that would
burn off that hydrogen. You see
all those sparklers, or they
call them sparklers, at the bottom of the engines
every time before the engines light.
That's, uh, that's
zirconium particles,
50 micron zicronium particles that are expelled
from these little silent generators
that, uh, that blow them out
across there, like a little, uh,
firecracker. And,
uh, zirconium is pyrophoric,
so it ignites in the air
and burns and provides ignition
sources for burning off that hydrogen
and it's worked very successful.
The SRB,
again, uh,
we didn't really understand
uh, the, uh,
overpressure from SRB adequately
and STS-1.
It was pointed out, we received
damage on the, uh, shuttle
from that overpressure.
Um, we weren't
uh, fully appreciative of
exactly what caused that.
We determined that those sound waves
that comes from that are
about in the 25 Hertz range.
And, uh, I think the folks at JSC
determined that, uh, that
the fact that this was 25 Hertz
and we needed about one foot of water to
suppress it, that we could lace water
across that opening. And that's what
we do. We have, uh, we have
a system that allows us to fill
one foot of water over the entire SRB
holes so
that suppresses that ignition.
Now, what we learned from this was
that in trying to solve the problems
of uh, launching the Vandenberg
with closed launch, uh, ducts
out there, was that this
SRB has a 51 percent
excess fuel in the exhaust.
Uh,
27 percent of the, uh, excess fuel
is hydrogen.
Uh, that exhaust
comes out, you burn all of the oxygen
that was in the motor. So, it's not burning when
it comes out. However,
as soon as it, uh, starts
entraining air in the boundary layer,
you get,
whew, already?
Uh, you start getting,
uh, oxygen
available to
to, uh, cause this burning.
Well, five percent oxygen is required
to, uh, to burn hydrogen.
And, therefore
uh, when you get this penetration of
a mixing of five percent
you get ignition. It turns out
that's about the midpoint in the
mobile launcher itself, when that happens
and so we get a detonation, because
that hydrogen level is at
the lower end of the detonation level.
So, we actually have a hydrogen detonation
inside the mobile launcher,
every launch, from each one of
those SRB holes, and that
overpressure is suppressed by that one foot of water.
As you can see
there, the SRB exhaust
hydrogen level has
2,743 pounds a second
being exhausted, so
it's a tremendous, a lot of hydrogen.
Uh,
we need to, uh,
flow down the needs, goals and objectives.
That was one of our failures in the
shuttle program. Is that we didn't
understand that well. We thought that,
you know, if we had those,
that, uh, we could, we could take care of it.
But I'm flowing those down into
requirements and making it happen,
it just doesn't happen.
All those, when you have needs, goals
and objectives, unless they're flowed down
into hard requirements to
force the designs, uh,
they just won't uh, they won't
uh, be, uh,
visualized.
The, uh,
idea
or the goals on the
shuttle program was
160 hour turnaround, which is essentially
two weeks. We had a
24 hour notice to launch
from, uh, the roll out of the VAB.
And then, uh, it was added that
it would also accommodate a rescue,
which means you had to be able to change out
a payload at a pad,
in a two hour countdown.
And, uh, the assumption was that
if we did all that and we flew
40 launches a year at Kennedy,
that we could, uh, achieve about
$100 to $300 a pound to orbit, which was
going to be very economical.
The results
of focusing on performance and not have
flown down those needs, goals and
objectives can be seen here.
That the 160 hour turnaround turned
out to be a, uh,
126 day standard template.
That's a nine to one
ratio. And,
uh, 24 hour notice to launch
turned out to be 26 days.
So, that's a 26 to one.
Uh, the difficulty there
was front end money again. We needed
another $4 or $5 million dollars to
put in a new locks transfer
line, that would have allowed us to increase the
locks flow rate. And, we would have needed to
automate the crew ingress system
which would, uh, then change
us to the orbiter. And, uh,
the program just didn't have the money.
And, uh, so we
didn't accommodate it.
So,
our two hour countdown turns out to be
eight to nine hours.
Now, we typically
design for good
reliability and probably our
pedigree comes from building airplanes
there and using the same
kind of logic. However,
the application of space is
a lot more severe than
airplanes. No one
flies airplanes continuously for
seven days or 30 days.
And, uh, what we find is that
because of the
complexity of these vehicles and
so much hardware,
that about 40 percent of our
processing time in the OPF
is due to change
out of, or repairing problems
and changing out of failed
hardware. It's not
to say we got poor hardware. It's just
that there's so much, that the probability
drives you to have that kind of
a failure count.
And, in the future, we ought to be designing
for maintainability, which would drive
those requirements higher on the
reliability side and achieve
better, better life, because
it's that function alone
that causes, uh, uh,
a loss of productivity of that
orbiter. Instead of that orbiter
achieving a 10 flight a year,
it only achieved about two and a half flights a year
at our maximum
capability. And, uh,
this is all because of the
complexity of the system, as well as
the repairs
that are necessary.
Uh, there's
just as a matter of interest,
there's 102 interfaces on the
flight vehicle that required
ground servicing for every launch.
So you see the complexity I'm talking about.
Of this, there's
54 fluids in the, uh,
in the, uh, program
that has to be managed
that, uh, through a procurement process.
And, we need to know in detail
the commodity, uh, purity
and, uh,
characteristics. So, it requires
all of these things to be tested
and verified, and, uh,
maintain the cleanliness of these systems.
Uh, of that
54 fluids, 27 separate
fluids are serviced
in this vehicle every flight.
So, it's, uh, it's a huge
amount of servicing that is required.
And, again, we've got
to concentrate on better integration
of these functions so that we have
less interfaces,
less fluids to surface.
And that's the only way we're ever going to really
achieve higher affordability.
Um,
the resultant flight rate, as you can see,
was seven to eight. And, actually,
it's down to more like four or five these days.
So, uh, the
program labor increase and
I'm talking program, not just KSA,
is about a 21
per flight. So, it's a very
labor intensive, the way we
ended up. So, this is something we can learn from
the process.
I'm not shuttle bashing, believe me.
I think the shuttle is one of the greatest vehicles.
Well, it is the greatest vehicle
that the world has ever seen
fly in space. But,
we've got to learn from what
we've done and do better to
achieve affordability in
in a sustainable system. Uh,
the cost per pound
of course is a thirty to one factor.
Instead of bringing the cost down, like
we had planned to
one to three hundred, it's
much higher, and uh,
again, we did achieve cost reduction.
This is something that's not very much
told, but we did achieve a cost
reduction of about three to one
over the expendable process
during Apollo. So,
I do believe that we must
maintain our focus on
reusable systems if we ever
plan to achieve high affordability.
The message here is
the design trades that
we do, are focused on,
on the lowest subsystem.
And, we teach our people this, by the way.
If you take courses, and I'm not sure
about Apple, but I think we're still
teaching people to, uh,
to, uh, actually
optimize around the lowest
subsystem, and this will achieve you
the greatest performance
and this is something that we need to
uh, to change, because
we need to not only do that, but
then, as a last step, verify
that the overall system
has, uh, been optimized.
That's the, uh, the part that we miss.
Is, uh, overall optimization
of the system.
We would change many of these if
we were to do that and not end up with
the systems that we have.
If we are focusing on life cycle cost of course.
I wanted
to mention here that
as an example of complexity here,
the SSME is a
fantastic rocket engine, but
that engine requires 28 vehicles
support systems to satisfy it.
That's a huge number of subsystems
to try to support
that element. And, again
uh, we need to look at ways to
decrease that.
Um, and that's one of the
measurements we use to measure different
um, different propulsion systems.
We evaluate one system or another
that's, uh, that's one of the
characteristics we use to help
evaluate that.
Uh, the orbiter
hydraulic system is a distributed hydraulic system.
And, uh,
one of the choices in
that, because we didn't want to fly
the weight of a hydraulic
fluid in a line that runs from the system
to the mow-line,
requires us to go in and break the integrity
of that system every time we
we fly. We have to
go in with a ground system
in order to operate the landing gear. So,
we can't leave that system intact.
Again, that requires additional servicing
and, uh, additional hardware.
Uh, and
I'd like to point out that in
1960, we, uh,
we had a, uh, system
called the pershing system that started with a
distributing hydraulic system.
We went to, uh, unitized
components that eliminated
that distributing system and we reduced
our mechanical manpower
at the launch pad by 50 percent.
So, uh,
hydraulic systems are very, very
labor-intensive to maintain and operate.
Uh, the idea
behind a complement system, even though
it was still hydraulics, was that
if it failed, we could, uh,
we could remove that component,
take it to the lab and work on it in the lab
and get it off line. So it wouldn't, uh,
hold us up. Uh, that was a
real novel design, it was only
held together with one bolt, and
uh, and a simple electrical connection.
So, it was very easy to change,
and very quick.
And, the, uh,
hydraulic system as we have it in the shuttle requires
about 50 percent of the electrical power
just to keep the fluid warm so that the system
will be functional when you come,
return to Earth.
So, again, another reason why
uh, to avoid distributing
hydraulic systems.
Uh, the SRB
TVC has a
distributing hydraulic system that's powered
with a hydrazine
uh, system. Uh, very similar
to the ones that are in the
and the, uh,
uh, orbiter. And,
uh, we've, uh, made several
attempts to try to get that
simplified, but again, it's,
uh, we've frozen the design,
as Tom says. So, uh,
but, the next systems
ought to avoid these complex designs.
Unfortunately,
we hadn't accomplished that on
Constellation. To my demise,
it, uh, it, uh,
we seen the result of that.
Unless we get the costs down
and get affordability and sustainability,
we're not likely to proceed.
So, we, uh, we need to
get it right.
Um,
I
see that time's up, so
I'm gonna jump over the last slide.
And, you can, you'll
get my materials, I assume, so
you can read these over, and, uh,
gain from it. But,
I want to leave you with the bang,
the bang at the bottom there. The system
design must be accountable for
controlling the life cycle cost, including
its down time. Uh,
we, uh, we must change the
way we design our
design process. And, uh,
if we want affordability, that
we've got to take these things into account
and have a cost
control system in place.
Uh, we suggested it
be something like the way we control
weight. I know that's going to require manpower,
but, you either pay me now or
pay me later. So,
I advise that, uh, we have
cost control systems put in
place to help us manage
the cost drivers that go into these
designs, flight and ground,
such that we can achieve affordability.
Thank you.
[ Audience applause ]