Craig Fink wrote:"So, with Columbia, part of the bow shock entered
the wing. But the flow around the hole that didn't, still had some
expansion to go through. I imagine that might have kept the boundary
layer somewhat well behaved"


I find volume V part 13 very of the caibs report helpful, you might to
please see below.

REENTRY ANALYSIS:
Caib report vol v part 13 page 106 par 6
Chapter 4 Aerodynamics
4.4.2 Damage Progression Theory and Supporting Aero
"Based on the damage assessment and timeline period correlations
covered in Section 4.4.1, the following is a postulated damage
progression theory based on the results of the aerodynamic
investigation. This damage progression, approached from an aerodynamic
perspective, is consistent with the working scenario and attempts to
maintain consistency with other data from the investigation. References
are made to figures which include a combination of aerodynamic
extraction results and the major timeline events noted.
An initial WLE breach (small hole or slot) in an RCC panel exists at
entry interface. By EI + 300 sec thermal events are occurring internal
to the WLE cavity, however no identifiable aerodynamic increments are
observed."



Caib report vol v part 13 page 189 par 2
Chapter 5 Aerothermodynamics
5.2.3.3 Results for Mach 6 Air
"Holes Through Wing
Limited parametric studies of simulated damage in the form of a wing
breach from the windward surface to the leeward surface were attempted
in this facility and were primarily associated with aerodynamic testing
(see Section 4.3.1). Initially, circular holes dimensionally consistent
with the width of a carrier panel (approximately 4 inches full scale)
were placed at the interfaces for carrier panels 5, 9, 12, and 16. The
holes were found to force boundary layer transition on the windward
surface to the damage site. The model and IR setup for the aerodynamic
tests at this point in time precluded imaging the side fuselage. Since
the model also incorporated damage in the form of missing RCC panel 6,
it is believed that effects (if any) from the carrier panel holes would
have been dominated by the disturbance from the missing RCC panel. TPS
damage in the form of a much larger breach through the wing was
attempted, but the side fuselage heating measurements were considered
qualitative due to compromised phosphor coatings on the models that
were used. The holes were orientated normal to the wing chord and were
located near the left main landing gear door. One hole location was
approximately located at the center of the forward bulkhead
(X=1040-inches in Orbiter coordinates) and the second location was near
the center of the outboard bulkhead (Y=167-inches in Orbiter
coordinates). At each location, the wing hole diameter was
systematically changed from 0.0625 to 0.125 and 0.25-inch at wind
tunnel scale (8.3, 16.7, and 33.3-inchfull scale). While the
compromised phosphor coating considerably degraded the image quality,
it was evident that no change in side surface heating was apparent for
any tested combination of location or diameter."


Caib report vol v part 13 page 305 par 2
Chapter 5 Aerothermodynamics
"5.3.2.5 Properties of Flow into Breach Holes
The engineering method was exercised for four hole sizes in leading
edge Panel 6 - 1, 2, 4, and 6 inches in diameter. Figure 5.3.2-5
presents the predicted fraction of the boundary layer that is ingested
into the hole Both the 4 and 6-inch diameter holes result in the
ingestion of the entire boundary layer at later times in the
trajectory. Figure 5.3.2-6 and Figure 5.3.2-7 present the predicted
bulk gas temperature and enthalpy of the ingested gas. The gas
temperatures are mostly between 9,000 and 10,000 R. Figure 5.3.2-8 and
Figure 5.3.2-9 present the predicted mass flow rate and energy flux and
compare the results to the values obtained from the CFD solutions for
the holes in Panel 6. For the smaller holes - 2 and 4-inch diameter
- the engineering method provides a very good estimate of the
ingested flow rate and energy flux. However, as the hole is increased
in size to the 6-inch diameter, the engineering methods tend to under
predict the values
Breach holes in Panel 8 were also analyzed with the engineering method.
In this case, the external pressure used for the computation (Cp=1.46)
was for a location slightly different from the location where the
curve-fits were obtained. Figure 5.3.2-10 and Figure 5.3.2-11 present
the predicted mass flux and energy flux into the holes. The values for
the CFD solution with a 10-inch diameter hole are also plotted and
indicate that the engineering method is under predicting the energy
ingested into large breach holes".


Caib report vol v part 13 page 311 par 3
Chapter 5 Aerothermodynamics
5.3.3.1 Basic internal plume physics and impingement issues
The nature and structure of a free jet issuing into the Orbiter
interior through a breach in the thermal protection system is dictated
by the total pressure difference across the surface, the hole diameter,
boundary layer properties, and local boundary layer edge Mach number.
For small orifices, on the order of one inch, the jet may be assumed to
enter normal to the internal surface with a sonic condition. For such a
case the governing
parameter that dictates shock structure and mixing of the jet is the
ratio of external riving pressure to internal pressure. Bulk fluid
properties are a function of the percentage of external boundary layer
drawn off into the hole. For very small hole, the properties will be
near wall conditions, whereas larger holes can produce internal flows
with enthalpy levels approaching free stream total values. As the
high-speed gas enters the cavity, it immediately starts mixing with the
ambient fluid at the boundaries of the jet. This mixing zone gets
larger as the jet progresses, until finally the core portion of the jet
has been consumed and the jet has reached a fully developed condition.
Depending on the conditions driving the jet, this may not occur until
ten's of diameters downstream. In the highly under-expanded state,
the jet shock structure is dominated by a normal shock downstream of
the initial expansion called the Mach disk. Immediately downstream of
the Mach disk the flow is subsonic, though it may re-expand to
supersonic flow. Figure 5.3.3-1 displays variations in free-jet
structure and with varying pressure ratios. Figure 5.3.3-2 shows the
impact of pressure ratios in the range expected for Orbiter
penetrations on computed flow structure. Larger hole diameters display
significant departure from this relatively simple structure as larger
percentages of the highly energetic boundary layer are ingested and
increased transverse momentum bends the jet over in the direction of
the boundary layer edge flow. This effect was discovered with the first
fully coupled,
internal/exterior flow solution performed for a two-inch breach into
RCC panel 6. Complete results are presented in Section 5.3.6.1.4.
Larger diameter penetrations tend to carry highly supersonic, high
temperature gases directly to the interior surfaces and produce highly
complex shock/impingement structures that can significantly impact
local heat transfer rates."


Caib report vol v part 13 page 535 par 7
Chapter 6 Thermal
6.7 Damaged Wing Leading Edge Coupled Aero-Thermal-Structural Analysis
"A comparison of the times at which these critical events occur
during the entry is shown in Table 6-7. As expected, failure times are
accelerated for the 10 inch case compared with the 6 inch due to the
higher levels of internal heating. Thermal response of instrumentation
within the left wing of STS-107 have suggested the initial breach
through the spar occurred at 491 seconds after entry interface. With a
predicted spar breach time of 470 seconds, the 6 inch provides a better
comparison to flight data than the 10 inch case. As shown in Figure
6-82, better agreement for the 6 inch damage case can also be seen by
comparing the temperature response of V09T9895 (panel 9 spar rear
facesheet thermocouple) from the OEX flight data with the model
predictions at an analogous location on panel 8 (in this case panel 8
in the model is used as a surrogate for panel 9 as noted previously).
The average predicted temperature of two nodes on the rear facesheet
are used in the comparison for each damage case. Up to the flight
estimated time of spar breach at approximately 490 seconds the
predicted thermal response for the 6 inch case is in reasonable
agreement. After this point, the predicted temperature rise rates are
much slower than flight data, indicating the effect of convective
heating experienced during flight in this area from the hot gas jet
expanding into the wing interior. Modeling of such heating was not
included in this analysis."

Caib report vol V page 539, par 7
Chapter 6 Thermal
AeroAerothermalThermalStructuresTeamFinalReport
"6.11 Leading Edge Reinforced Carbon-Carbon (RCC) Hole Growth Thermal
Analysis
A prediction of RCC hole growth was performed using JSC arc jet test
data obtained from hypervelocity impacted RCC test specimens when
subject to a high temperature entry environment. The objective of the
arc jet testing was to establish the oxidation characteristics of RCC
with thru holes obtained from hypervelocity impacts. The specimens were
exposed to constant heating conditions at temperatures of 2500 and
2800F and pressures of 50 to 180 psf. Correlations were developed from
the data for use in trajectory simulations to predict hole growth and
hot gas flow through an enlarging hole into the wing leading edge
cavity.
A 0.75 inch diameter hole in the RCC was assumed for analysis purposes.
Figure 6-97 shows the heat flux and pressure environment at the hole
while Figure 6-98 shows the resulting RCC surface temperature as a
function of time. The predicted RCC temperature of approximately
4800°F is assumed to be consistent with a diffusion-limited erosion
regime for bare or uncoated RCC. With this assumption, the erosion or
hole growth rate measured for the 2800°F arc jet tests can be used for
erosion rate estimates here. The erosion rate in this flight
environment and regime is .0032 in/sec. Figure 6-99 reveals the results
of the analysis and shows the predicted growth to a final OML diameter
of 4.0 inches. The predicted IML (back-face) diameter is slightly
smaller at 3.0 inches. Extrapolation of this analysis to higher RCC
temperatures (sublimation regime) or larger initial hole diameter is
not recommended since the data base is very limited."


Open sharing of information is crucial to improving everybody's
understanding of the universe around us.
Tom