Not much credible information is yet available, as only a handful of official NASA photos have been released. There is one unofficial CBS news photo, whose digitally-enhanced zoom-in, fills-in some of the information gaps. But I do my best with what there is!
Figure 1 depicts the expected re-entry flow and
heating pattern for a lunar return by Orion,
which “flies” at angle-of-attack, to generate a small lift force that is used to
fine-tune the trajectory shape. It does
this by rolling about the wind vector,
to point the lift vector in the desired direction. This was also done on Apollo, and on the earlier Gemini missions.
Figure 1 – Expected Flow and Heating Pattern for an Orion
Lunar Return
The Orion capsule has now flown 3 times as of this writing, once before the Artemis program even existed! All 3 were re-entries at near escape
speed, which is what a lunar return
actually is! That first test was called
EFT-1, launched using a Delta-IV rocket
(now retired). It had the heatshield
fabricated from Avcoat, hand-gunned into
the cells of a fiberglass hex bonded to the capsule outer shell, just like Apollo! The other two Orion flights (so far to date) were
launched with the SLS rocket, and had
cast and machined Avcoat tiles bonded to their outer shells, unreinforced by any fiberglass hex!
Figure 2 below is a photo of what I believe to be the
Orion flown on EFT-1. I cannot vouch
for the pedigree of this photo, or that
some of the char has not been deliberately removed for investigating the
material beneath. But the presence of
the hex is clearly visible, so this cannot
be either the Artemis-1 or Artemis-2 Orion capsules! If an Orion at all, and I believe it is, this has to be EFT-1!
Note that a fair amount of ablation and erosion damage seems
to be evident on the lateral side. I
cannot say for sure, but this would
appear to be the lateral side nearest the offset stagnation point, where attached flow would feature heating
comparable to that at the rim of the base heat shield. The base heatshield itself is not visible in
this view.
Figure 2 – Photo,
Pedigree Unknown, of What is
Thought to Be EFT-1
NASA changed the manufacture of the Orion capsule heat
shield for Artemis, retaining the Avcoat
material, but deleting the reinforcing
hex, to save time, effort,
and money. They built both the
Artemis-1 and Artemis-2 heat shields this new way, without waiting for the results of the uncrewed
test flight that was Artemis-1 (and THAT was their fundamental mistake). They were surprised by the unpredicted
nature of the damage that Artemis-1 exhibited!
It shed both small and large chunks of char during re-entry, leaving alarming craters in the heat shield! Some of those are shown in Figure 3 below.
This Artemis-1 flight was a skip re-entry with two heating
episodes separated by a slight cooldown,
conducted at a very slightly shallower-than-normal angle, compared to a straight-in re-entry. Shallower reduces total heating a
little, but it does incur at least some
risk of bouncing off the atmosphere like a skipped rock, into an extended elliptical orbit, whose period (of 5-10 days) exceeds the
remaining crew life support duration!
Flying the capsule at angle of attack is a way to control
that skip effect. Early in the re-entry, you point the lift vector down, to stop any skipping-off. Late in the re-entry, you point the lift vector up, to keep the trajectory from “drooping”
downward too soon.
Figure 3 – The Official NASA Photo of the Artemis-1 Heat
Shield, After Recovery
After Artemis-1,
NASA spent nearly 2 years doing tests and calculations to “officially”
convince itself that if they deleted the skip and came in slightly steeper (at
higher heating), that the alarming chunk-shedding
would not occur! This was based
on the hypothesis that the pyrolysis gases produced during the second heating
pulse could not percolate through the char layer easily enough, and ended-up cracking it, and blowing-off chunks of it, with the obstructed gas pressure from beneath.
That hypothesis ignores the effects of fluid-scrubbing shear
action in attached flow, which would
want to peel chunks off from between any cracks. It also ignores any embrittlement and
weakening of the charred material, after
cooling down some, between the heating
pulses. Anyone who has ever dealt with
the fragile mantles of a Coleman gasoline lantern, would know exactly what embrittlement
effect I am talking about, and
that it is quite real!
So, NASA flew
Artemis-2 crewed, with the very same
heat shield as Artemis-1, just deleting
the skip in favor of a slightly-steeper straight-in re-entry. Their analyses said that would eliminate the
chunk-shedding. The real question
is, did that really work?
The few official photos released so far say that it did
work. However, there is an unofficial photo that says “maybe
not near as well as assumed”. You
judge for yourself.
Figure 4 is an official photo taken during crew
extraction from the capsule, floating in
the sea. Bear in mind that you cannot
see the capsule base heat shield at all in this view, and you can only see the lateral side where
the windows were located. Those have to
be away from the worst lateral heating,
to protect them from being destroyed by that heating.
Figure 4 – Official NASA Photo of Artemis-2 During Crew
Extraction
Figure 5 is an official NASA photo of mission
commander Reid Wiseman pointing at some kind of an eroded crater in the lateral
wall ablative insulation. You can see
the window behind his head, and you can
see there is no heavy charring (black) in this view. This view is of the windows-side of the
capsule, where heating and its effects
are greatly reduced by being in separated wake flow. You can see nothing of the base heat
shield at all, or the more highly-heated
opposite lateral side, where flow was attached.
Figure 5 – AI-Doctored Version of an Official NASA Photo of Mission Commander Reid
Wiseman Pointing to Damage Spot
Figure 6 is an official NASA photo taken by a Navy
diver, of the base heat shield, while the capsule was still in the
water. It looks rather pristine. However,
you can see by the streaks that the stagnation point was just out of
view, top center left of
photo. The odd feature lower left is the
mark or trace, left on the heat shield
from the flow about one of the tie-down pads.
We see no craters from lost chunks of char in this
view, which supports the conclusion that
NASA was right to delete the skip in the re-entry trajectory! But we cannot see the region nearest the
stagnation point, where heating is
higher, and where fluid shear is
higher. That would be because the
acceleration to sonic at the rim, takes
place over a much shorter distance. That
raises the surface shear forces that the heat shield material and its char
layer “feel”.
A suspicious person might say that only photos
supporting NASA’s hypothesis have been released. That is because we have not seen any photos
of the offset stagnation region on the base heat shield, or of the lateral wall on the side adjacent
to that stagnation region (opposite the windows), where heating is almost as high as on the
base heat shield. Know that
eventually, all the photos must
be released, in the report on their
heat shield investigation. That report must
be publicly released! But it may
be a year before we see it!
Figure 6 -- Official
NASA Photo of Artemis-2 Heat Shield Taken by a Navy Diver
As I said above,
there is an unofficial photo now circulating, that was taken by CBS News seconds before
splashdown, while the capsule was still
hanging from its parachutes. It was
distant and somewhat blurry, but it does
show some sort of white mark near the heat shield rim.
That white mark attracted a lot of attention, and led to the digital enhancement of that
photo, and a digital zoom-in to examine
that white mark more closely. But that
enhanced photo does indeed also show the near-stagnation region of the base
heat shield, and the lateral wall away
from the windows, where flow was
attached, and the heating much higher. That original blurry CBS News photo is Figure
7 below, and the zoomed-in
enhancement is Figure 8 below.
You cannot see very much in Figure 7, but you can in Figure 8! The white mark was left by the melting and
destruction of one of the tie-down pads.
This is something NASA says was expected, although it did not occur on Artemis-1, as you can see in Figure 3 above. It does seem to have left an
alarmingly-deep cavity eroded into the rim of the base heat shield! Expected or not, that cavity would appear to be of a depth
comparable the thickness of the heat shield.
And I do find that alarming!
Figure 7 – Unofficial,
Blurry Photo Taken by CBS News Seconds Before Splashdown
What nobody has been talking about are the other
things I see in Figure 8. I
have circled four places that I believe show where chunks of char were shed
from the base heat shield. These are
smaller chunk-shed craters, to be
sure, and scaling up from the limited
view, not anywhere near as numerous as
those seen on Artemis-1! Yet they are
there, and the revised
non-skip re-entry so very clearly did not entirely stop them from occurring! Which simply says that something else was
going on with that char chunk shedding,
besides pyrolysis gas percolation through the char!
There’s one other circled spot, and two arrows pointing to large locations, on the base heat shield in Figure 8, where I cannot tell what happened, but I can see that some sort of damage is
clearly there. It will take better
photos than this to evaluate those damages.
What I see on the lateral side adjacent to the stagnation
point are one unidentifiable small dark spot,
and two whitish bright spots of considerable size. The two bright spots are clearly places where
all the char was lost, exposing
bare metal to full heating! And that
metal looks distorted by that heating!
Very alarming indeed!
Those bright spots are not windows, those are the bare metal of the outer capsule shell, to which the Avcoat tiles were bonded! Simple thermal insulation separates it from the inner metal shell, which is the capsule crew cabin pressure vessel.
Figure 8 – Digital Enhancement and Zoom-In of Unofficial CBS
News Photo
There would seem to be four things going on here that
affect the damages seen, not
just the one thing that NASA hypothesized.
They are:
#1. Pyrolysis gas percolating out against permeability resistance, wanting to blow chunks off. (That is the NASA
hypothesis.)
#2. Fluid surface shear forces wanting to peel chunks out from
between cracks in the surface.
#3. Char layer shrinking,
cracking, and embrittlement upon
cooldown, between the heating pulses of
a skip-type re-entry. (Like the
fragility of a gasoline lantern mantle.)
#4. The presence of the reinforcing hex actually ties the
char layer tighter to the pyrolysis and virgin layers beneath, and it also acts to limit the spread of
cracks in the surface.
My conclusions about the rational things to do, depend
upon what response NASA takes to this Artemis-1 and -2 outcome:
#1. If NASA ever wants to resume flying skip-type
re-entries, then put the reinforcing hex
back into the Avcoat! Period! There actually is a way to do that, without hand-gunning Avcoat into every hex cell
like Apollo!
#2. Non-skip re-entries have higher peak heating. The Avcoat tile thickness is insufficient at
the attached-flow locations: near the
tie-down pads, and on the lateral wall
opposite the windows. NASA must thicken
it at those locations!
Avcoat tiles with reinforcing hex, but without hand-gunning:
Avcoat is an epoxy-novolac polymer loaded with solids. Those solids include some small amount of carbon
fibers, and a lot of tiny micro-balloons
made of phenolic resin. Higher
micro-balloon content lowers density,
raises ablation rate, and
increases the porosity and permeability of the char layer (also decreasing its
strength). It also greatly increases the
apparent uncured mix “viscosity”, which
is already extremely thixotropic (almost a solid).
The Avcoat mixture is so thick, that it is almost “crumbly” coming out of an
air-powered caulking gun, whose nozzle
matches the size and shape of the hex cells.
The hex is a fiberglass cloth with a phenolic resin matrix. The glass softens at a higher temperature (~
900 F) than that at which the polymer starts to pyrolyze (~ 300 F), which is why it is an effective char layer
stiffener and retention aid. The carbon
fibers add a bit to those reinforcing effects.
Apollo and Orion EFT-1 used hex panels bonded to the outer shell, into which cells the Avcoat was
hand-gunned. This was extremely
labor-intensive!
For the Artemis-1 and Artemis-2 Orion capsules, the Avcoat mixture was cast into blocks, from which tiles were machined. There was no reinforcing hex! The bonds and gap fillers of those tiles
are not in question here, those
performed just fine! The lack
of reinforcing hex is the question!
You can make tiles with hex in them, but only if you can stop being bound
by “either/or thinking”. That
would be either doing it the Apollo/Orion EFT-1 way, or doing it the Artemis-1/-2 way. However, there is a third path! Read on:
Put a chunk of the hex the size of the cast block you want
to make, into some tooling on the outlet
of a plastics extrusion press. Load your
Avcoat mixture into the press, and use
that press to force it through all the cells in the hex, all at once! Remove the loaded-hex tooling from the
press, put the bottom and top on that
tooling, and cure that block of Avcoat
that now contains the reinforcing hex! Then,
machine your tiles from those
blocks, and bond them to the Orion outer
shell, the same way as in Artemis-1 and
-2! Tiles that are hex-reinforced, but with NO hand-gunning!
I gave this idea to the NASA heat protection group in
Houston well over a year ago, as of this
writing, and again directly to the new
NASA Administrator Jared Isaacman only a few weeks ago. I was able to confirm (1) that the heat
protection group got it, and (2) that
they thought I was right. Nothing
confirms that Isaacman ever saw my letter to him.
But I never heard another word out of NASA about this
alternative, and NASA has not yet done
anything like it. So, I must conclude:
Money and schedule clearly still outweigh crew’s lives for the
decision-making NASA management levels.
And “not invented here” is still quite strong at NASA, as well as its contractors. Those two things are my real ongoing
reservations about NASA! And they have
been, ever since the first of two lost Space
Shuttle crews! Crews lost precisely
because of those very same two management culture flaws!
-----
Search code DDMMYYYY format: 28042026
Search keywords: aerothermo, space program
-----
Closely-Related Postings:
“About the Artemis-2 Mission” posted 31 March 2026
The updates to that article have these same photos in
them, just not as explicitly annotated
as here. And I have since added how the
flow and heating patterns vary. In
near-escape Earth entry with a blunt heat shield, plasma radiation heating is larger than
convective heating, and it decreases
with distance from stagnation less rapidly.
Search code DDMMYYYY format 31032026
Search keywords aerothermo, launch,
radiation, space program
“Ramjet Data Re:
Heat Shields” posted 1 March 2026
Shows how my old experiences with ablatives in solid rocket
motors, and especially ramjet
combustors, have strong overlap with
the re-entry phenomena involved here. Char
retention by the layers below, is a
crucial key, in both venues!
Search code DDMMYYYY format 01032026
Search keywords aerothermo, ramjet,
space program
-----
