In March 2024 I was an invited speaker at an American Carbon
Society meeting held at North Carolina State University in Raleigh, North Carolina. I brought 3 presentations to that
meeting, only one of which could
presented live. The other two were
converted on site to poster presentations,
so that all three were gen in one form or another.
My report
on the meeting is given in the posting on this site dated 23 April 2024 and titled “Presenter
at Workshop”. The live
presentation and associated text document was about using old-time by-hand
analyses for initial concept screening,
to enable a real brainstorming process,
without the expense of creating an actual design and multiple computer
models, for each and every concept. The example for this was by-hand calculations of re-entry
dynamics and conditions. That
presentation is also documented on this site as the posting dated 3 May 2024 and titled “Entry Concept
Screening”.
The other two presentations,
converted to poster presentations,
existed as potentially-live presentations and text documents when I went
to NCSU. One had to do with ablative
ramjet insulations that I tested in actual ramjet direct connect tests a few
decades ago. The other was an update to
a previous presentation and posting regarding a low density ceramic composite
burner liner that I tested a few decades ago.
That earlier
posting was dated 18 March 2013 and titled “Low-Density Non-Ablative Ceramic Heat
Shields”.
To quickly find any posting here on this site, all you need is its posting date and
its title. Use the blog
archive tool on the left side of this page. Click first on the year, then on the month, then on the title if need be (if more
than one posting was made that month).
To see an enlarged figure in any given posting, click on the figure. There is an “X-out” button top right, that takes you right back to the article.
-----
Presented here is the text document of the ramjet liner
presentation, based on those tests long
ago.
-----
Ramjet Ablative Liners GW
Johnson 12-24-2023
Abstract
For several years,
the author worked on the Hercules-McGregor plant’s “Airbreathing IR
& D” project, where “IR&D” means
“Independent Research and Development”, and is reimbursed by the government. The focus of that project was a fuel-rich
solid-propellant gas generator-fed ramjet,
which included an integral booster housed within its combustor.
During 1989-1992, a
series of ramjet tests were performed in the McGregor direct-connect test
facility, that addressed 3
objectives. First, the combustion performance of multiple
experimental fuel propellants was evaluated.
Second, a fuel rate control
approach with no moving parts was evaluated.
Third, multiple possible
alternate combustor ablative insulation materials were evaluated, as possible alternates for Dow Corning’s DC
93-104 silicone ablative, which is the
subject here.
Presented here are the heat protection results from that
testing series. Included are a unique
way of extending the time that heat protection can be obtained, and the means by which chemically
incompatible materials can be bonded together.
About the author:
The author had a 20 year career in aerospace defense doing
new product development design,
analysis, test, and evaluation, entering the workforce in the
slide rule days with a master’s degree in aerospace engineering. Transition to the then-expensive pocket
calculators was underway, but desktop
computers were still years in the future.
That career was mostly (but not entirely) in rocket and ramjet missile
propulsion. It ended with a plant
shutdown and layoff in 1994, just when
the industry was shrinking drastically. The
author then had a second 20 year career that was mostly in teaching (at all
levels from high school to university),
plus some civil engineering and aviation work. He earned a doctorate in general engineering late
in life, to support that second career. He is now retired.
Background:
Figure 1 illustrates how drastically-different the
conditions are in a ramjet, versus those
in a solid rocket. It was already known
that the types of insulation that work in rockets simply could not cope with
the heavy fluid shear “scrubbing” and very long burn times in the ramjet.
The materials commonly used in rocket cases are
randomly-oriented fibers reinforcing some type of rubber. In the old days, asbestos fibers were used, replaced in recent years with things like
Kevlar fibers. One common rubber was
EPDM (ethylene propylene diene monomer),
easily compatible with cast composite propellant binder systems. These are shown in Figure 2, along with the candidates being considered
for ramjet testing.
These materials were usually B-staged as partially-cured
sheets, so that a layered wrap upon a
mandrel could be inserted into a primed case,
that mandrel inflated to exert pressure against the case wall, and then heat-cured to vulcanize the
rubber. This was very cost-effective
processing, but the materials were inadequate
for ramjet application.
Figure 1 – Rocket Vs. Ramjet Conditions
Figure 2 – Candidate Materials
Replacing the randomly-oriented fibers with layers of
fully-woven cloth was the approach that was hoped to be adequate, while preserving the preferred cost-effective
processing. We already knew about Dow
Corning’s silicone material with the higher pyrolysis temperatures, so we switched to their poly dimethyl
silicone (PDMS) polymer, and attempted
carbon and silica fabric reinforcements.
Dow Corning’s DC 93-104 ablative is PDMS polymer loaded with
3 solids. Two are silica and silicon
carbide granulates, the third is carbon
fibers in random lengths. This is a very “thick” (viscous) material that can be
troweled, or pressure-cast around
installation tooling.
We had significant experience with it in the ASALM-PTV
program, where we insulated with DC
93-104 and then cast integral boosters, into 20-inch OD combustors, for ramjet propulsion designed by the
Marquardt Company. We applied this same experience
to our 7-inch OD ramjet engine.
We also identified a Japanese more-or-less equivalent to DC
93-104: Type 0 Shin Etsu. It processes
identically to DC 93-104. We also tested
it.
About the Ramjet Testing:
Over several years, we
also obtained a lot of experience and hardware for a 7-inch OD ramjet engine
being considered for an advanced propulsion replacement for the AIM-120 AMRAAM
missile. This used the same DC 93-104
liner that ASALM used. Figure 3 shows
the geometry and dimensions of the ramjet engine, and typical test conditions. There is even an approximate thermal analysis
plot of the temperature distributions expected from the flame. There are drops across the thermal boundary
layer, across the char, across the virgin, and finally a tiny one at the case.
Figure 3 – Typical 7” Test Conditions
Note that the thermal conductivity of the char is about
factor 3 times that of the virgin material,
leading to the strong temperature gradient slope change at the pyrolysis
zone. Despite this, the char is still more of an insulator than
any sort of thermal conductor. That is
an important result!
Such tests in full flight-weight 7” hardware usually burn
30-60 seconds, but sometimes can run
longer, which is challenging even for DC
93-104. That problem was solved on
ASALM, and that same solution was used
for the AMRAAM engine. The IR&D
tests being shorter-burn tests, we did
not need to use that solution in these IR&D tests. There is more information about that solution
below.
Figure 4 shows a color-highlighted table of some 8
tests conducted for 3 different reasons:
(1) to test experimental fuels,
(2) to test an experimental fuel control technique (unchoked-throat
self-throttling), and (3) to test and
compare experimental and “stock baseline” combustor insulations.
Figure 4 – 7” OD Test Results
Of these, the first
test was an unintended no-burn using the PDMS-carbon cloth insulation, leaving it intact. It was re-used “as-is” in the second test
(blue), a short-burn run with a “clean”
fuel.
The third test was another “clean” fuel short-burn test) with
the PDMS-silica fabric insulator (green.
It was in such good condition, we
re-used “as-is” in the fourth test (also green), another short-burn test of a
highly-metallized fuel.
The 5th test was a long-burn test of a low-boron
fuel, using DC 93-104 as the case
insulation (orange).
The 6th test was another low-boron fuel
test, this one a short burn, using the Shin Etsu “clone” of the Dow
Corning material (yellow). The 7th
test was a longer-burn test of an older baseline fuel with the unchoked
“throttle”, and a re-used Shin Etsu
insulator (again yellow) from test 6.
The 8th test (no color) did not give us any
useful data, it being
almost-a-no-burn, with a very
disappointing highly-experimental fuel,
on another DC 93-104 insulator,
not a re-used one.
Specific Test Results:
PDMS/Carbon Fabric
Figure 5 shows our hybrid flight-weight
combustor/heavyweight lab motor hardware,
as mounted on the thrust stand in the Hercules-McGregor direct-connect
test facility. This is a hybrid, using a flight-weight combustor with a
heavyweight lab motor as the fuel-rich solid propellant gas generator, in a gas generator-fed ramjet. (ASALM was a liquid fuel ramjet). IR&D stuff is never “pretty”, and this one certainly is ugly-looking. Figure 6 shows the post-burn
appearance of the insulator. It was
mostly used-up after only an 11 second burn. That was very disappointing, as we usually got over a minute out of DC
93-104, even without the very-long-burn
solution.
Figure 5 – Hybrid Test Hardware: Flight-weight Combustor and Heavyweight Lab
Motor Gas Generator
Figure 6 – Results From One PDMS/C Fabric Test
Specific Test Results:
PDMS/Silica Fabric
Figure 7 shows the white textured appearance of the
as-built PDMS/Silica fabric insulator. Figure
8 shows the post-burn appearance
after two short-burn tests, one after
the other, on the same combustor
insulation. The first was a 10-second
short burn repeat “clean” fuel test. The
second was a 15 second short burn of a very-highly metallized boron-titanium
fuel. Total accumulated burn for the two
tests was only 25 seconds. There was not
enough liner left to risk a third re-use.
Again, that was disappointing.
Figure 7 – Pre-Test Appearance of the PDMS/Silica Fabric
Insulator
Figure 8 – Post Test Appearance of the PDMS/Silica Fabric
Most of the 7-inch IR&D tests looked like what is shown
in Figure 9 for the boron-titanium test.
The fuel is 28% metallized boron-titanium. The gas generator throat is choked. The liner is the PDMS/Silica cloth being
reused “as is” from the previous test.
The sparklers are not from the fuel,
they are coming from the liner as it erodes. Otherwise,
the clean plume is quite amazing for such a highly-metallized fuel. Incidentally,
this was the very first time anyone ever burned high-percentage boron
efficiently, in a ramjet engine!
Figure 9 – Typical 7” Test Appearance: BTi/PDMS-SiO2
Specific Test Results:
DC 93-104
Figure 10 and Figure 11 are post-test, just two different views. This was a long-burn test of a low-boron
fuel, rate-controlled by the unchoked
throat. Note the typical “mud crack”
cracking pattern, which is what we
usually see with DC 93-104 liners. With
that kind of cracking, it is abundantly
clear that the char layer is very strongly held by the virgin material beneath. That would be the effect of the randomly-oriented
carbon fibers connecting the char layer to the virgin. There was enough left after this 40 second
test to possibly have risked a short-burn reuse, but we did not do that.
Figure 10 – DC 93-104 Post-Test Appearance, one view from
rear
Figure 11 – DC 93-104 Post-Test Appearance, another view
from the front
Specific Test Results:
Type 0 Shin Etsu
Figure 12 and Figure 13 show two post-burn
views of this insulator. The first one
is after a 15-second short burn with a low-boron fuel, unchoked.
The second one is after a 42-second long-burn unchoked-throttle test
with an old baseline fuel, in which the same
Shin Etsu liner was reused “as-is”.
Total accumulated burn was 57 seconds,
about what we usually get from DC 93-104.
Figure 12 – Type 0 Shin Etsu Post-Test Appearance, after low-boron short burn
Figure 13 – Type 0 Shin Etsu Test Appearance, after re-use in a long-burn unchoked throttle
test
There was some instability in the 42-second burn, resulting from a flow-control aero-grid
failure in one of the 2 inlets. Despite
this flow asymmetry, the burn was “good”
and the liner survived the pressure pulsations.
What we saw with Type 0 Shin Etsu was poorer char-retention
strength atop the virgin material, but
with a “slicker-looking” and apparently-harder char layer. It would generally shed most of the chunks of
char layer after the actual burn, during
air rundown, not during the burn itself! This was seen as a “puff” of black in the air-only
plume after the flame went out.
The Extended-Burn and Chemical-Compatibility
Solutions:
These liners (and the one in ASALM) were only about 0.20
inches thick. In the 7-inch
hardware, we would get about a minute’s
burn with DC 93-104 before the charring reached the case wall, releasing the char layer to break up into
chunks and fly downstream. Only the Type
0 Shin Etsu matched that duration performance.
Both are thick pressure-castables.
To get a longer protection duration, one has to mechanically retain the char, after the virgin material bonded to the case
wall has been fully pyrolyzed away. The
means to do this is illustrated in Figure 14, and was developed at Hercules-McGregor on the
20-inch ASALM program, working together
with Marquardt. If you take thin strips
of stainless steel about 0.1 inch wide,
kink them on a 1 inch spacing to extend up about halfway through the
liner, and spot-weld them to the case
spaced about an inch or so apart, this
will retain the char layer in-place,
long after the virgin has charred through. The kink tips are buried deep enough inside
the char so as not to overheat.
Figure 14 – Details Matter:
Long-Burn and Propellant Compatibility Solutions
This worked for up-to-15-minute burns in ASALM! It was adopted for the 7” AMRAAM engine as
well, on the contract programs. We think it might help address char chunk
loss with the Shin Etsu, as well.
The other issue with PDMS silicone is chemical
incompatibility with composite propellant binders, very important if integral boosters must be
packaged within the combustor! This
problem was also solved on ASALM. One
has to chemically isolate the materials with an inert separator sheet, yet still retain good adhesion to both the
ramjet liner and to the propellant. This
was achieved with thin Teflon film (essentially a large size “Saran
Wrap”), acid etched on both sides to
provide cleaned surfaces with rough texture.
This separator is bladdered onto a liner surface primed with
DC 1200 from Dow Corning, and cured in
place. Then the exposed inside surface
can be primed appropriately for the propellant binder system, before propellant is cast and cured. The bonds are strong, yet the chemical isolation is absolute.
Conclusions:
DC 93-104 is the best by far, compared to bladderable fabric-reinforced
rubbers
For
burns > ~ 1 minute, must use the
kinked strip retention system
Kinked
strip retention system tested in 20” diameter for burns up to 15 minutes
Type 0 Shin Etsu is almost as good
as DC 93-104
Char-virgin
strength is weaker
Sheds
char chunks during air rundown
Retention
strips might ease these troubles
As the conclusions indicate,
the greater pressure-cast processing and separator-sheet preparation efforts
are worth the better protection afforded by DC 93-104, especially with the post-char-through
protection afforded by use of the kinked retention strips. The kinked retention strip detail is not
something Dow Corning thought of, but it
makes their product perform very much longer!
The strips would likely enhance the performance of the
Japanese material, as well. It is not an exact clone of the Dow Corning
material, with the apparent weaker char layer
retention, but we believe the strips might
help correct that, based on these
IR&D test results.
References:
#1. Dow Corning Product Information Sheet “Dow Corning
93-104 Ablative Material” , available as
a pdf file from their website.
#2. MSDS (Material
Safety Data Sheet) for DC 93-104 kits,
MSDS number 000001189166, issued
2 April 2015, last revised 6 April
2015, available from Dow Corning as a
pdf file.
#3. US Patent 3,623,904 “Elastomeric Composition Containing
Silicon Carbide for Use as an Ablative Coating”, issued to James A. Ramseyer and assigned to
Dow Corning Corporation, 30 November
1971.
#4. Cheryl L. Resch,
“Ablation Models of Thermal Protection Materials”, published in the Johns Hopkins APL Technical
Digest, Volume 13, Number 3,
dated 1992.
