Centuri used the elastic cords, which I preferred to straight
rubber, but I see that Apogee offers Kevlar shock cords. Are these
impossible to burn, and what are the most effect methods to attach them
to the rockets? Would they be fireproof enough to attach to a standard
engine clip on a typical engine mount inside the main body tube?
Burning with curiosity,
--Jay Goemmer
Sent via Deja.com http://www.deja.com/
Before you buy.
Yes. They're what Quest uses; my Ol' Reliable scratchbuilt (with more
launches on it than I remember) uses a Quest motor mount/shock cord
and it hasn't failed yet. The kevlar will fail eventually (mine is
starting to look fairly ratty at this point), but not many rockets
last long enough for it to be a problem.
An idea I'm toying with is building a rocket with the body tube
removable from the fin can (just put a coupler at the top of the fin
can, and put some small screws through the body tube to hold it
together; then the shock cord would be replaceable.
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
VL 2000 Homepage: http://www.cs.orst.edu/~burnett/vl2000/
Another way to do this is to use a *loop* of kevlar line, pass it
through a small hole in each centering ring, and loop it around the
protruding motor mount tube at the aft end of the rocket (secure it there
with a drop of yellow glue). When the kevlar loop needs replacing you can
slide it out through the holes, and put a new one in with the help of a thin
wire slid through the holes to pull it through.
I did this on one of my cluster scratch-builts and it's worked well.
For a permanent shock cord mount, I like the small loop of kevlar
epoxied in the tube where the crappy Estes style mount would normally go. I
learned that one (as well as the one above) from someone here on RMR.
--
Jim Z in Vermont
JimZ...@adelphia.net
That's *nice*. thanks.
If you will use an appropriate size hole in the C-rings (that would allow a
tight knot to pass thru), you should be able (when replacement time comes
around) to tie the new line to the old and use *it* to help 'thread' the new
line thru the rings.
I would also protect that exposed Kevlar (at the aft end) -- the radiant
energy of a White Lightning motor could compromise the strength of the
material (which decomposes at a mere 700-800 degrees). Some of that (all)
aluminum HVAC tape would do fine (have used this for other things on the 'hot
end' and it's done well).
-- john.
>
> Jim Z in VT wrote
> > Another way to do this is to use a *loop* of kevlar line, pass it
> >through a small hole in each centering ring, and loop it around the
> >protruding motor mount tube at the aft end of the rocket (secure it there
> >with a drop of yellow glue). When the kevlar loop needs replacing you can
> >slide it out through the holes, and put a new one in with the help of a
thin
> >wire slid through the holes to pull it through.
>
>
>
> If you will use an appropriate size hole in the C-rings (that would allow
a
> tight knot to pass thru), you should be able (when replacement time comes
> around) to tie the new line to the old and use *it* to help 'thread' the
new
> line thru the rings.
excellent suggestion.
>
> I would also protect that exposed Kevlar (at the aft end) -- the radiant
> energy of a White Lightning motor could compromise the strength of the
> material (which decomposes at a mere 700-800 degrees). Some of that (all)
> aluminum HVAC tape would do fine (have used this for other things on the
'hot
> end' and it's done well).
I haven't had any problems with that. But I've only used this method
with clusters of BP motors, so I can't really comment on how it works with
composite motors without commiting a tai fu........
>
> -- john.
>
>
>
jim z (VT version):
> I haven't had any problems with that. But I've only used this method
>with clusters of BP motors, so I can't really comment on how it works with
>composite motors without commiting a tai fu........
What about fabbing a small ring of SST tie wire (or monel, etc) of the size
needed (to clear the motor mount tube) and then 'looping' the loop (of
Kevlar) around this ring. Set in place (with the knot of Kevlar line
(obviously) directly over the hole in the centering ring) and then placing a
smallish square of aluminum tape right over this point. This would protect
the knot and that would be the only exposure of the textile material on the
aft C-ring. MIght need another or two (at 120° spacing) to hold the metal
ring in place. Might could even just use one of those split rings here.
Admittedly, a little heavy for ModRoc -- but at the LMR step, this would seem
to work.
Jim, I *like* your idea -- but I would use a little caution with that exposed
Kevlar and any kind of 'visible' (flame) composite propellant -- I've seen
brand new painted 'aft ends' (of 5"+ rockets) just blistered to hell and gone
after ONE flight (mostly with White Lightning). It is clear that there is a
LOT of radiant energy 'bouncing' off that aft end -- and I would be skittish
of exposing ANY textile material (including Kevlar or Nomex) to that kind of
energy. These aramids can't handle more than about 800° (and Nomex is an
aramid). Fiberglass? Well, maybe here it might be OK -- but a simple metal
ring might just do the trick.
Just some 'out loud thinking'.
-- john.
I have some old Estes rockets that are 30+ years old now. The original shock
cords are worthless, so I would need to replace them in order to fly these
rockets. I thought it would be good to have a method where the actual shock
cord could be easily replaced.
The LOC Precision attachment method: They supply a piece of nylon braided
string which you form into a long narrow loop, and glue one end into the body
tube. Then you tie the shock cord to the end of this loop. Could be done
easily in a smaller rocket with some nylon, or maybe kevlar cord.
In my LOC kits now, I modify the engine mount "ala Kaplow" and install an
eye-bolt for my shock cord attachment.
...
>What about fabbing a small ring of SST tie wire (or monel, etc) of the size
>needed (to clear the motor mount tube) and then 'looping' the loop (of
>Kevlar) around this ring. Set in place (with the knot of Kevlar line
>(obviously) directly over the hole in the centering ring) and then placing a
>smallish square of aluminum tape right over this point. This would protect
>the knot and that would be the only exposure of the textile material on the
>aft C-ring. MIght need another or two (at 120° spacing) to hold the metal
>ring in place. Might could even just use one of those split rings here.
>
>Admittedly, a little heavy for ModRoc -- but at the LMR step, this would seem
>to work.
Or this: Get a length of stainless steel cable and make a loop at each
end with swedge fittings (for LMR or higher) or use 100# steel fishing
leaders (for modrocs). During building, leave a CR hole large enough to
pass a loop. Also during construction, put a loop through the hole,
around the MMT, back up through the hole, and pass the free end through
the loop. Attach the SC to that free end. Size the cable so that the loop
ends up just below the top of the BT.
You can replace the steel cable, if need be, by attaching a new one to the
free end and threading it through the hole, around the BT, and back
through the hole by pulling on the bottom end of the old cable.
>
>Jim, I *like* your idea -- but I would use a little caution with that exposed
>Kevlar and any kind of 'visible' (flame) composite propellant -- I've seen
>brand new painted 'aft ends' (of 5"+ rockets) just blistered to hell and gone
>after ONE flight (mostly with White Lightning). It is clear that there is a
>LOT of radiant energy 'bouncing' off that aft end -- and I would be skittish
>of exposing ANY textile material (including Kevlar or Nomex) to that kind of
>energy.
I'd love to see someone instrument this; my own theory is that 99.9% of
this sort of damage happens during ignition and liftoff due to reflected
energy from blast deflectors. People put rockets above flat blast
deflectors see this a lot, I bet :-)
--tc
My opinions only.
<snip John's stuff>
>Or this: Get a length of stainless steel cable and make a loop at each
>end with swedge fittings (for LMR or higher) or use 100# steel fishing
>leaders (for modrocs). During building, leave a CR hole large enough to
>pass a loop. Also during construction, put a loop through the hole,
>around the MMT, back up through the hole, and pass the free end through
>the loop. Attach the SC to that free end. Size the cable so that the loop
>ends up just below the top of the BT.
This is roughly what I've done with everything from Python sized
models on up, and I credit the original idea to John. I picked up
some fine SS cable from McMaster-Carr awhile back (50 feet for about 7
bucks), which I've used in my Phoenix, Merc Atlas, Python, AT
Strongarm, and Cosmo BB II.
In the Phoenix, for example, I placed holes in the upper CR, like Ted
mentions, and I ran the cable down, and around the MMT once. After
the ends are crimped, I end up with a loop of wire in the upper BT
about an inch below the top. A short piece of electrical shrink
tubing divides that loop into two, and aids in attaching additional
shock cord. The "crimp" connector is located between the two CR's in
order to prevent 'chute fouling. The cable's not replaceable because
of this, but due to the fact that it's SS with a 500#+ breaking
strength, I'm sure it'll outlast the rocket! <bg>
It's worked flawlessly over the Phoenix's last 22 flights...
FWIW..
tah
Tod A. Hilty NAR #72099
Hilty Information Systems
Member MTMA, NAR Section #606
Mantua Township Missile Agency
http://web.raex.com/~markndeb/rockets/mtma/
"I'm going to put the wheels of the bus back on... just in case"
- BlankReg, Max Headroom: 20 Minutes Into the Future
"I speak for myself _and_ my corporation! Deal with it!"
- blankreg
- remove nospam.ever, and replace with apk for reply
...
>In the Phoenix, for example, I placed holes in the upper CR, like Ted
>mentions, and I ran the cable down, and around the MMT once.
You mean the cable is all the way around the CR, and then some, to go out
the holes, right? Otherwise, with a "one loop, two holes" method, you
have to be careful that the holes are far enough apart, and the CR strong
enough, to prevent the cable from ripping out half of a CR--like happened
to me once :-(
You could also expose the crimp connector to more tension than it is
designed for, perhaps.
>The "crimp" connector is located between the two CR's in
>order to prevent 'chute fouling. The cable's not replaceable because
>of this, but due to the fact that it's SS with a 500#+ breaking
>strength, I'm sure it'll outlast the rocket! <bg>
Seems like if one hole is big enough to let the crimp connector through,
you could fish the connector up, cut the cable, attach a new cable to one
end, fish the new cable down, around, and up, crimp connect the ends of
the new cable, and then fish the connector back down.
Agreed, this is likely to be a rare event if you use quality cable, but
even SS cable is corroded by AP and BP byproducts, so it could happen...
As Tod says, any method along these lines is pretty likely to far outlast
the rest of the rocket......
FWIW,
--tc
My opinions only.
I just stepped into this thread, but given that you're talking about 5+"
rockets I might have an idea...
I've never had *any* scorching on the back of my rockets using 38mm
motors. However, I have two rockets that have each been launched by
a single 54mm motor and their back ends are charred... no way that was
done during the moment of ignition, especially considering that there was
*no* blast deflector on the pad.
The theory seems to be that the culprit is Aerotech's Medusa nozzle,
originally designed to give the flame a larger appearance from what I
understand.
I've seen this happen 3 times on Bill Wagstaff's PML Aurora, using K550
motors. His beautifully-finished boattail *always* ends up getting
badly scorched.
--
----------------------------------------------------------------------
Marcus Leech Mail: Dept 8M70, MS 012, FITZ
Advisor Phone: (ESN) 393-9145 +1 613 763 9145
Security Architecture and Planning Fax: (ESN) 395-1407 +1 613 763 9435
>In article <39a533f0...@news.apk.net>, blan...@nospam.ever.net
>(Hilty Information Systems) wrote:
>
>...
>
>>In the Phoenix, for example, I placed holes in the upper CR, like Ted
>>mentions, and I ran the cable down, and around the MMT once.
>
>You mean the cable is all the way around the CR, and then some, to go out
>the holes, right?
Yup. The holes are 180 degrees apart. In one hole, one complete
wrap, and then I bring the other end of the cable down through the
other in order to make the connection.
>Otherwise, with a "one loop, two holes" method, you
>have to be careful that the holes are far enough apart, and the CR strong
>enough, to prevent the cable from ripping out half of a CR--like happened
>to me once :-(
Ouch. What happened?
My Cosmo BB II had "one loop, two holes, right next to each other" in
the CR for cable attachment. I filled those two with a couple of
blobs of epoxy, and redrilled two others 180 degrees apart, and then
installed new cable as mentioned above. The Cosmo design looked to me
like it could pull through.
>You could also expose the crimp connector to more tension than it is
>designed for, perhaps.
Dunno. I tested the crimp connector by crimping a loop, standing in
it, and then pulling as hard as I could (mostly with my back). I
don't know how much force I applied, but dang, that cable will cut
skin easily! No movement of the cable in the crimp. My logic is that
if that kind of force were applied during ejection, the rocket would
most likely be trashed for other reasons anyway.
>>The "crimp" connector is located between the two CR's in
>>order to prevent 'chute fouling. The cable's not replaceable because
>>of this, but due to the fact that it's SS with a 500#+ breaking
>>strength, I'm sure it'll outlast the rocket! <bg>
>
>Seems like if one hole is big enough to let the crimp connector through,
>you could fish the connector up, cut the cable, attach a new cable to one
>end, fish the new cable down, around, and up, crimp connect the ends of
>the new cable, and then fish the connector back down.
The connector I used ended up being about 3/8" wide after I crimped
it. I thought about using the method you just described, but I didn't
want that large of a hole in the CR. Dunno, maybe I'm a little anal
with regard to strength... <bg>
I made the holes just large enough to allow the cable to pass through.
>Agreed, this is likely to be a rare event if you use quality cable, but
>even SS cable is corroded by AP and BP byproducts, so it could happen...
>
>As Tod says, any method along these lines is pretty likely to far outlast
>the rest of the rocket......
Yep! So far so good!
Tod "BB II's getting her first coat of black paint as we speak!" Hilty
I suspect that 99% of the tail end thermal damage we encounter is from the
blast deflector deflecting the rocket exhaust right back at the rocket, and
not from thermal heating via the motor casing.
We need angled blast deflectors. IIRC, the original MMI launcher had an
angled deflector, as did some of the old Estes and Centuri pads. The neatest
was the parabolic cone used by MPC. The Aerotech Mantis pad has an anled
deflector, but I've seen this try to "hop" the whole pad, especially when a
fat model like a G-Force on a high thrust G motor is flown. THis pad needs
to be anchored to the ground, and is still inadequate for the largest LMR
stuff.
Bob Kaplow NAR # 18L TRA # "Ctrl-Alt-Del"
Kaplow Klips: http://members.aol.com/myhprcato/KaplowKlips.html (baffle too!)
NIRA: http://www.nira.chicago.il.us NAR: http://www.nar.org
SPAM: spamr...@ChooseYourmail.com u...@ftc.gov postm...@127.0.0.1
> > >Jim, I *like* your idea -- but I would use a little caution with that
> exposed
> > >Kevlar and any kind of 'visible' (flame) composite propellant -- I've
> seen
> > >brand new painted 'aft ends' (of 5"+ rockets) just blistered to hell and
> gone
> > >after ONE flight (mostly with White Lightning). It is clear that there
> is a
> > >LOT of radiant energy 'bouncing' off that aft end -- and I would be
> skittish
> > >of exposing ANY textile material (including Kevlar or Nomex) to that kind
> of
> > >energy.
> >
>
For modrocks (BT50 and larger, anyway), I epoxy an eye screw or
eyebolt to the top centering ring, then connect a ss fishing leader to
that. For paper rings, I back up the cr with a small piece of 1/8"
aircraft ply. Seems to be plenty strong, and it's a cinch to change out the
leader if necessary.
--
David Wallis
Again, the K550 uses the Medusa nozzle. Try it with any 38mm motor,
even the J570 and I'll lay down money that it doesn't get scorched.
(Are there any 54mm motors that don't use the Medusa nozzle?)
Matt
Well, I'll add a data point to this -- it isn't just White Lightning
where this happens, and it isn't reflection from the blast deflector.
I have a modified Fat Boy with six fins and six motor mounts -- a 24 mm
core and five 18 mm outboards, so I can fly with core only, or 2, 3, or
4 outboards. The outboards are (were, before they burned through)
plugged at the forward end by the forward centering ring; the rear
centering ring was left off because it was too much work to cut it to
fit around the six mounts.
On this rocket, every time I've launched with a cluster, I've burned
some of the unused outboard tubes and/or some of the exposed fin roots
-- and this is on Estes and Quest black powder motors, with a curved
deflector that directs exhaust away from the rocket.
I subscribe to the base turbulence theory, with the outer edges of the
exhaust streams being diverted by swirling air into the hollows in the
base of the rocket.
The next rocket I build with BT-80 and a cluster mount will have a full
rear centering ring, minimal exposed outboard tube, and the entire aft
end soaked in thin CA.
Another point in favor of base swirl: I also have a repro Astron Cobra,
and have no burning around the exposed mounts (this has no rear
centering ring, and uses a forward bulkhead instead of the "putty" seal
from the original instructions) even after it's been flown on some flat
deflectors; that suggests to me that either a) the Cobra has much less
base turbulence than the Fat Boy (which is certainly possible given the
smaller tube and no large area without nozzles in it), or b) the
occupied tubes sink heat enough better to prevent burning them. My
money is on a combination of both, as the occupied tubes in the Fat Fat
Boy were also less affected, but fin tabs alongside them were burned,
while the Cobra shows no scorching inside the exposed airframe tube,
either.
--
WARNING!! This area has been designated an official DOPE FREE ZONE!!
If you're going to be a dope, please do it somewhere else!
Donald Qualls, aka The Silent Observer NAR # 70141-SR Insured
Rocket Pages http://silent1.home.netcom.com/launches.htm
Telescope Pages http://silent1.home.netcom.com/astronomy.htm
Opinions expressed are my own -- take them for what they're worth
and don't expect them to be perfect.
There is a very good chance you are correct. Base drag can be rather
severe.
: I've seen this happen 3 times on Bill Wagstaff's PML Aurora, using K550
: motors. His beautifully-finished boattail *always* ends up getting
: badly scorched.
Even a boattail
: --
: ----------------------------------------------------------------------
: Marcus Leech Mail: Dept 8M70, MS 012, FITZ
: Advisor Phone: (ESN) 393-9145 +1 613 763 9145
: Security Architecture and Planning Fax: (ESN) 395-1407 +1 613 763 9435
--
Bill Nelson (bi...@peak.org)
>I'd love to see someone instrument this; my own theory is that 99.9% of
>this sort of damage happens during ignition and liftoff due to reflected
>energy from blast deflectors. People put rockets above flat blast
>deflectors see this a lot, I bet :-)
You've seen other responses, Ted, but I'd lay my money on 'base drag' and the
'radiant energy' thing. Admittedly, I didn't mention 'base drag' and that is
probably more the situation - but I wouldn't rule out radiant energy (on
even smaller rockets where the base drag would be considerably less).
I would state this with authority based on our experiences here -- I came up
with a launcher design that had a 6" square 'flame hole' in the base (12" sq.
SST plate) -- and an inverted 'V' deflector (i.e. kind of like that main
deflector at LC 39 for the Saturn / Shuttle). I discovered (long time ago)
that a deflector that only deflected ONE way was not a good thing -- as it
creates a pretty hefty lateral force on the launcher (on the bigger
flights)... and, my failing memory seems to recall in incident where a really
high thrust motor kicked a smallish launcher over one time. At any rate --
'splitting' the exhaust TWO ways (horizonally opposed) cancels out this
effect -- and this is where my design came from.
Coming back on track, this design spurred several more -- and we had (at one
time) quite a few of these kinds of launchers in use here -- and several
'brand new' rockets flew off of them. That design would almost guarantee
that there would be no 'flashback' from launch -- so any damage to the aft
end came from the flight proper. And we had some nice, blistered paint on
those brand new models.
-- john.
The scorching does seem to be more prevalent among cluster rockets.
--
Kurt Kesler
>Cochran Ted wrote
>
>>I'd love to see someone instrument this; my own theory is that 99.9% of
>>this sort of damage happens during ignition and liftoff due to reflected
>>energy from blast deflectors. People put rockets above flat blast
>>deflectors see this a lot, I bet :-)
>
>
>You've seen other responses, Ted, but I'd lay my money on 'base drag' and the
>'radiant energy' thing. Admittedly, I didn't mention 'base drag' and that is
>probably more the situation - but I wouldn't rule out radiant energy (on
>even smaller rockets where the base drag would be considerably less).
>
Most of the replies talk about blistered aft ends on cluster models.
That effect might involve radiant energy (the exhaust plume, being located
off center, gets to radiate against a larger portion of one side of the
base; a combination of motors get to gang up on the same spots on the far
side of the base.
Another effect, which I'd guess is as large or larger than radiation, is
that the interactions between adjacent exhaust plumes generate eddies
which transport high temperature exhaust back onto the base of the
rocket. I'm guessing that Silent's experience with cooked unused motor
tubes in a six-around-one cluster flying in a 3 around 1 motor
configuration was the result of that sort of plume interaction.
I'd still like to see an R&D project involving instrumented flights of
single motor rockets of various diameters to look at the impact of radiant
energy and base drag. I'm sure there are examples of cooked single motor
rocket tails without evidence of blast reflection, but I'm not convinced
that it is all that common-- FPODs and the like, with huge glowing
exhausts attached to their bases, don't seem to be much the worse for that
effect.
One more thing to look at might be the effects that occur near apogee,
with a slowly moving rocket (perhaps even tail sliding) and a
still-very-hot, if not high speed, motor exhaust.
We've all seen rockets coming down under chute, with significant smoke
still emitting from the motors. The base of the rocket is in effect still
sitting right in that fire! [Come to think of it, that would explain
Silent's data as well, right?]
--tc
My opinions only.
This needs a name. I propose the "Mantis Hop"
> flights)... and, my failing memory seems to recall in incident where a really
> high thrust motor kicked a smallish launcher over one time. At any rate --
> 'splitting' the exhaust TWO ways (horizonally opposed) cancels out this
> effect -- and this is where my design came from.
>
> Coming back on track, this design spurred several more -- and we had (at one
> time) quite a few of these kinds of launchers in use here -- and several
> 'brand new' rockets flew off of them. That design would almost guarantee
> that there would be no 'flashback' from launch -- so any damage to the aft
> end came from the flight proper. And we had some nice, blistered paint on
> those brand new models.
Here's an interestng test. Mount a rocket upside down on a fence post or
other sturdy object, and static fire it. See what happens to the back end.
Maybe -- especially since there are sometimes some discrepancies in
delay, so (say) a core motor might still be burning delay/tracking smoke
while the outboards have fired ejection and (due to the burned through
bulkhead) deployed the recovery.
Generally, though, this phenomenon is limited to single-use composite
motors, where the delay is adjusted during assembly by drilling, and
continues to burn for a few seconds after ejection firing. BP motors
seldom have any significant delay burn remaining after ejection.
I was thinking that the base turbulence would be worse during coast than
boost, with less entrainment of the entering air to make it travel with
the thrust plumes, and more tendency to bring the (still hot) delay
smoke into the base area. Of course, that smoke isn't hot enough to
burn paper, or the sulfur particles that are the primary component of
the smoke would also ignite (leaving little or no smoke).
No, in BP motors and reloads I have to plump for the damage occurring
during boost. There's just not enough happening after ejection.
David Erbas-White
Mike Howie wrote:
>
> >snip.... but I'd lay my money on 'base drag' and the
> > >'radiant energy' thing. Admittedly, I didn't mention 'base drag' and that is
Hmmm... somehow my (admittedly uninformed) impression was that the
"Medusa" design was intended to provide a selection of total throat
areas from a single set of mold tooling for the phenolic nozzle
blank, by opening a variable number of the holes.
-dave w
Sent via Deja.com http://www.deja.com/
Before you buy.
Hot gas recirculation is apparently a problem on the space shuttle as well.
Folks on sci.space.shuttle have mentioned seeing this on launch footage,
particularly visible on night launches (I don't have cable so I will take
their word for it), and the ET is supposed to have thicker foam on the back
end to deal with this situation.
>I was thinking that the base turbulence would be worse during coast than
>boost, with less entrainment of the entering air to make it travel with
>the thrust plumes, and more tendency to bring the (still hot) delay
>smoke into the base area. Of course, that smoke isn't hot enough to
>burn paper, or the sulfur particles that are the primary component of
>the smoke would also ignite (leaving little or no smoke).
Silent, as a 'data point', the 98mm RMS motors produce a flame about 3 feet
long during delay burn (one of the advantages of being a Motor Test chair
<g>). And, of course, if you remember my comments about these 'extra extra
long' delays (circa 30 seconds), we had plenty of time to look at it <g>.
These kinds of things the average joe rocketeer doesn't see -- since that 'M'
motor is several thousand feet away (i.e. up). From 50 feet, it's quite a
show.
Point being -- these composite delay columns are a not too different
composition from the actual propellant -- and, thus, they DO produce a pretty
good flame.
The example was from the 98mm -- but even a 54mm does a pretty good job as
well (oh, maybe 8in to a foot).
-- john.
>Um, Mike, not to start a grammar police thread, but it's 'missile', not
>'missle' (I thought it was a typo in your email until I looked at the
>site and saw the same error).
Oooo... I'll just leave that one alone!
<bg>
tah