"Heavy lift: examining the requirements"

"Ed Kyle" wrote in news:1110339171.176808.196650
@g14g2000cwa.googlegroups.com:

Will McLean wrote:

I would argue that segmented solids with thrust vector control have a
higher failure rate than the liquid engines used in the core stages
of
Titan, Delta II, Atlas II and STS during the same period.

I count a premature SSME shutdown and a Titan first stage propulsion
problem, vs. three SRB catastrophic failures on STS and Titan. Am I
missing anything?


Two liquid failures in a combined 234 core
liquid stage cycles (113 STS + 121 Titan)
is a 0.0085 failure rate.

Three solid failures in a combined 468
booster cycles (234*2) is a 0.0064 failure
rate, so solids had a lower realized failure
rate in this comparison.

Wasn't one of those liquid failures an in-flight shutdown
of a SSME with a resulting abort-to-orbit? That might
tip the statistics in favor of liquids.

--Damon

"Jon S. Berndt" wrote in message
oups.com...
(somewhat tongue-in-cheek): The STS SRBs are flown on a manned vehicle.
Therefore, they are man-rated.

Not really. Just because people ride them doesn't make them man-rated, at
least if one applies NASA's man-rating standards to the vehicle. Fact is,
Shuttle doesn't meet NASA's own standards for man-rating.

-Kim-

"Jon S. Berndt" wrote in message
oups.com...
FWIW:

"Human-Rating Requirements, JSC - 28354"
http://www.hq.nasa.gov/office/codea/...documentd.html

That's been replaced by NPR 8705.2A. Look it up he

https://ice.exploration.nasa.gov/Windchill/gov/nasa/esmd/web/jsp/bidders/BiddersFolderContents.jsp?folder=/Default/Bidder's%20Library/Reference%20Documents

-Kim-

"Jon S. Berndt" jsb.at.hal-pc-dot.org wrote in message
...
Yes, I thought about that, too. Seems to me that roll control could be
done
with RCS quads on the CEV "service module" itself, or mounted on the upper
stage?

Current ELV practice is to use RCS for second-stage attitude control. I
doubt RCS sized for vacuum operations would have much effect deep in the
atmosphere.

-Kim-

Ed Kyle wrote:
Will McLean wrote:

I would argue that segmented solids with thrust vector control have
a
higher failure rate than the liquid engines used in the core stages
of
Titan, Delta II, Atlas II and STS during the same period.

I count a premature SSME shutdown and a Titan first stage
propulsion
problem, vs. three SRB catastrophic failures on STS and Titan. Am I
missing anything?


Two liquid failures in a combined 234 core
liquid stage cycles (113 STS + 121 Titan)
is a 0.0085 failure rate.


But if you count the Delta II and Atlas II cores, that changes.

Will McLean

Three solid failures in a combined 468
booster cycles (234*2) is a 0.0064 failure
rate, so solids had a lower realized failure
rate in this comparison.

- Ed Kyle

"Kim Keller" wrote in message

"Jon S. Berndt" jsb.at.hal-pc-dot.org wrote in message
Yes, I thought about that, too. Seems to me that roll control could be
done with RCS quads on the CEV "service module" itself, or mounted on
the upper stage?

Current ELV practice is to use RCS for second-stage attitude control. I
doubt RCS sized for vacuum operations would have much effect deep in the
atmosphere.

-Kim-

Whups. Overlooked that. I guess that's where some kind of fins might help.

Just curious: is a little bit of roll rate OK? Some solid rocket-propelled
ICBMs operate the same way, don't they - that is, no roll control?

Jon

Will McLean wrote:
Ed Kyle wrote:
Will McLean wrote:
I would argue that segmented solids with thrust vector control
have

Two liquid failures in a combined 234 core
liquid stage cycles (113 STS + 121 Titan)
is a 0.0085 failure rate.

But if you count the Delta II and Atlas II cores, that changes.


We were looking at segmented solids, but nonsegmented
solids can provide an interesting comparison too.

Delta II has suffered two failures in 118 missions.
One was due to an SRB case burn through. One was
caused by a failed SRB separation. I'm not sure
how you classify the latter failure, but it was
not an SRB propulsion problem. If we assume an
average of six solids per launch (a guess), the
realized solid failure rate is 1/708 = 0.0014.
No first stage propulsion problems have occurred,
but there have only been 118 "samples". I suspect
that the sampling error is greater than 0.0014.

There were only 30 Atlas IIAS flights, with four
solids each. And no failures. So while we can't
very accurately predict the failure probability
for a hypothetical 31st launch, we can say that
we have four times as many successful solid booster
samples as first stage samples for Atlas IIAS.

Looking back futher into the Delta flight history
to get more samples, consider the 238 Delta flights
since the introduction of the Long Tank Thor first
stage in 1968. There have been 10 or so failures,
including two outright SRB failures and two failures
during first stage flight. One of the failures, in
1968, involved a control system failure at 108 seconds.
The other, in 1986, involved a premature shutdown
of the RS-27A engine due to an electrical glitch.

Even if we only assume that the 1986 failure was
the only first stage problem, we get a realized first
stage failure rate of 0.0042. If we again assume an
average of 6 SRBs per flight, we get an SRB failure
rate of 2/1428 = 0.0014.

- Ed Kyle

Ed Kyle wrote:
Will McLean wrote:
Ed Kyle wrote:
Will McLean wrote:
I would argue that segmented solids with thrust vector control
have

Two liquid failures in a combined 234 core
liquid stage cycles (113 STS + 121 Titan)
is a 0.0085 failure rate.

But if you count the Delta II and Atlas II cores, that changes.


We were looking at segmented solids, but nonsegmented
solids can provide an interesting comparison too.

Delta II has suffered two failures in 118 missions.
One was due to an SRB case burn through. One was
caused by a failed SRB separation. I'm not sure
how you classify the latter failure, but it was
not an SRB propulsion problem. If we assume an
average of six solids per launch (a guess), the
realized solid failure rate is 1/708 = 0.0014.
No first stage propulsion problems have occurred,
but there have only been 118 "samples". I suspect
that the sampling error is greater than 0.0014.

There were only 30 Atlas IIAS flights, with four
solids each. And no failures. So while we can't
very accurately predict the failure probability
for a hypothetical 31st launch, we can say that
we have four times as many successful solid booster
samples as first stage samples for Atlas IIAS.

Looking back futher into the Delta flight history
to get more samples, consider the 238 Delta flights
since the introduction of the Long Tank Thor first
stage in 1968. There have been 10 or so failures,
including two outright SRB failures and two failures
during first stage flight. One of the failures, in
1968, involved a control system failure at 108 seconds.
The other, in 1986, involved a premature shutdown
of the RS-27A engine due to an electrical glitch.

Even if we only assume that the 1986 failure was
the only first stage problem, we get a realized first
stage failure rate of 0.0042. If we again assume an
average of 6 SRBs per flight, we get an SRB failure
rate of 2/1428 = 0.0014.

- Ed Kyle


So counting the other Atlas II models, the liquid first stage failure
rate is either 2/415, or if you include the earlier Delta, 3/535, or
..0048-.0056.

I think you need to include the separation failure for the smalller
strap-ons, since it's a failure mode that isn't present without the
booster, and it killed the mission. It then becomes .0021, which is
still pretty good relative to the liquid stage.

The interesting question is whether a failure of one of a pair of
segmented solids should be considered one failure and one success out
of two trials. The segment that failed had a twin poured from the same
batch on the other booster, and range safety destroys it before you
find out if it would also have failed later in the burn.

Will McLean

"Murray Anderson" wrote:

:A shuttle SRB would require modification to be used as an ordinary first
:stage. At minimum it would need roll control and new guidance software. The
:configuration would put a very dense first stage (specific gravity about
:1.25) under a large hydrogen-burning upper stage and large payload and
:fairing, so the current gimballing system might need hardware modifications
:for faster response. You'd need more hydraulic fluid.
:The current SRB reliability wouldn't carry over to the new configuration. If
:the steering ever failed, you'd need to get the payload away very fast.

It would also need a new thrust termination system, since I believe
the current one is to blow the nose off. Not what you want to do if
there's something sitting up there.

--
"Millions for defense, but not one cent for tribute."
-- Charles Pinckney

On Thu, 10 Mar 2005 15:25:41 GMT, in a place far, far away, Fred J.
McCall made the phosphor on my monitor glow in
such a way as to indicate that:

:The current SRB reliability wouldn't carry over to the new configuration. If
:the steering ever failed, you'd need to get the payload away very fast.

It would also need a new thrust termination system, since I believe
the current one is to blow the nose off. Not what you want to do if
there's something sitting up there.

The current SRB has no thrust-termination system.