Entry thermal resilience modeling

Thorsten

Active member
Joined
Dec 7, 2013
Messages
785
Reaction score
56
Points
43
We all know that if you slam a spacecraft into the atmosphere on the wrong trajectory, it's going to break up in flames.

What I'm interested in here is modeling the 'when' of that, and I'm curious as to how people solve the problem in Orbiter, or whether anyone is aware of a documented professionally used model.

Assume I have a reasonably good model which gives me a figure of merit temperature on the heat shield (there's different temperatures at different points of the heat shield in reality, and to some degree I can estimate what they are as well, but let's keep it simple).

What I've used so far is a condition that the figure of merit temperature may not be larger than some maximum, otherwise the thermal protection system (TPS) fails. I believe however that is unrealistic and physically not correct (especially after studying temperature time curves for some off-nominal Shuttle trajectories which ought to be possible in reality but generate high temperature peaks for short times).

The actual failure mode that occurs when the heat shield fails or is thermally overstressed is that structure behind the heat shield comes to a temperature where it fails under stress.

Now, that condition is a line in a two-parameter plot - at sufficiently high stress the structure will fail even at room temperature, and at sufficiently high temperature it will melt, i.e. fail without any structural stresses. Somehow there's a line connecting the two conditions - beyond some (temperature/stress) the structure fails. But I don't really know what that line is.

The stress is fairly easy to get from the aerodynamics simulation, but the temperature less so.

The TPS works by (drastically) reducing the heat conduction, so most of the heat reaching the TPS is radiated away rather than conducted inward to the structure. A TPS failure hence means the heat flux to the structure is higher than normal.

Now, the structure itself is a heat sink (i.e. it will take some heat flux, increasing in temperature in the process) - but it will also conduct heat to other parts of the structure, and even potentially radiate heat when it can be conducted to a reasonably cool part.

Thus, even knowing the heat flux through the TPS, getting a temperature on the structure behind isn't easy because it again depends on factors which I do not know and find difficult to estimate.

So I can figure out easily how not to do it, but I don't really see an efficient way to do it closer to reality without guessing a lot of numbers (which never is a good idea...).

How do you do it for your Orbiter spacecraft? And how do professional spacecraft designers estimate it before going into the full numerics?
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,588
Reaction score
2,312
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
And how do professional spacecraft designers estimate it before going into the full numerics?

Well, they are likely modeling it the same way as they are doing it in numerical simulations: By creating a coarse FEM of the spacecraft and assigning a destruction enthalpy to each element. This is the energy that the element absorbs before failing. Usually, you can ignore special calculations and assume that about 5% of the heat generated by aerodynamic forces is absorbed into the structure, that conduction of heat happens not fast enough to matter and that emission of the heat as radiation is a plain function of temperature. In the simplest case, you just mark off energy and calculate how deep into the atmosphere you got before reaching the enthalpy of the next element. This way you can split the spacecraft FEM into fragments and tell which fragments are able to reach ground and which evaporate to dust at high atmosphere.

Of course, for spacecraft with a heat shield, you should also include this into the model, since it really matters. Conduction might also matter then.
 
Last edited:

Ravenous

Donator
Donator
Joined
Jul 19, 2013
Messages
275
Reaction score
0
Points
0
Location
sitting at the pointy end
There was some interest in monitoring re-entry of ESA's ATV-5. (Though I've never seen mention of any published results; I think the monitoring didn't go as planned.)

I found this though:

Preparations for the Airborne ATV-5 Re-Entry Observation Campaign_S. Loehle.pdf

(Apologies for the messy URL)

It describes some re-entry simulations using SCARAB software - no real detail given but the references might be worth looking at... can't tell from the paper if it goes into any structural detail.
 

Thorsten

Active member
Joined
Dec 7, 2013
Messages
785
Reaction score
56
Points
43
Some follow-up on this:

I've now started to record the heat shield temperatures the model spits out for a given trajectory.

Here's a comparison between a nominal profile (red) and a TAL profile after a second engine failure with droop guidance recovering trajectory altitude but insufficient propellant to reach the proper MECO condition - the latter ought to be survivable, because a late 3EO contingency green is similar in dynamics and not in a black zone.

temperature.gif


(vertical axis has reference temperature in K, horizontal axis has time in seconds, both curves arbitrarily start where temperature exceeds 500 K from below, recording ended at Mach 12)

So, for the 2EO scenario, there'a a short high-heat peak, but the integrated heat flux is quite a bit lower (which roughly makes sense, because also the kinetic energy is lower). The scenario is actually rather brutal (as evidenced by the phugoid seen in he temperature - it creates Nz of up to 3 g).

So I'm thinking integrated heat flux above some critical temperature is probably the measure I'm looking for. Creating a high heat-peak would still be bad in the sense that it would cause a disproportional weight in the computation, but it wouldn't be really crippling.
 
Last edited:
Top