General Question Simulate heat shield

[marcogavazzeni]

Hi guys
I have an idea,and want to know if i can achieve it.

Orbiter gives us the ability to use the spot light or point light by varying the intensity:

LightEmitter *le = AddPointLight (_V(0,-2.0,-29), 200, 1e-3, 0, 2e-3, col_d, col_s, col_a);
le->SetIntensityRef (&th_main_level);

It would be possible to transform the dynamic pressure in a level to simulate the heating of the shield?

double i20;

i20=GetDynPressure();

if (i20 < 0 ) i20= 0;
if (i20 > 100 ) i20= 1;

SpotLight *le1 = (SpotLight*)AddSpotLight (_V(0.0,-10.0,20), _V(0,0.3,-1.0), 150, 1e-3, 0, 1e-3, RAD*45, RAD*80, col_r, col_r, col_a);
le1->SetIntensity (i20);

But that does not work:facepalm:

 

[Urwumpe]

aerodynamic Heatflux in W/m² is the product of dynamic pressure and velocity.

The light emission is in the best case just a black body radiation case of the heat flux, assuming the heat shield to be in equilibrium.

 

[fausto]

Marco, una pressione dinamica di 110000 pascal equivale sull'Eridanus ad una temperatura dello scudo termicodi 1200 gradi circa. che è il limite.. l'avevo calcolato tempo fa .. quindi la temperatura è uguale alla pressione dinamica diviso 91.6666!

 

[Thundersnook]

That sounds awesome! Would be cool if anyone manages to solve this tricky stuff (unfortunatlay my programming-skills are just enough to install Orbiter :lol

 

[marcogavazzeni]

Marco, una pressione dinamica di 110000 pascal equivale sull'Eridanus ad una temperatura dello scudo termicodi 1200 gradi circa. che è il limite.. l'avevo calcolato tempo fa .. quindi la temperatura è uguale alla pressione dinamica diviso 91.6666!

Si ma il mio problema è trasformare un certo range di pressione dinamica ad uno che venga usato per settare l'intensità da lvl o a lvl 1 o 10 non sò

---------- Post added 11-03-10 at 01:51 AM ---------- Previous post was 11-02-10 at 03:52 PM ----------

double heat=GetDynPressure();
if(heat<1000)shield_proc=0;
if(heat>5500)shield_proc=0.05;
if(heat>11000)shield_proc=0.1;
if(heat>16500)shield_proc=0.15;
if(heat>22000)shield_proc=0.20;
if(heat>27500)shield_proc=0.25;
if(heat>33000)shield_proc=0.30;
if(heat>38500)shield_proc=0.35;
if(heat>44000)shield_proc=0.40;
if(heat>49500)shield_proc=0.45;
if(heat>55000)shield_proc=0.50;
if(heat>60500)shield_proc=0.55;
if(heat>66000)shield_proc=0.60;
if(heat>71500)shield_proc=0.65;
if(heat>77000)shield_proc=0.70;
if(heat>82500)shield_proc=0.75;
if(heat>88000)shield_proc=0.80;
if(heat>93500)shield_proc=0.85;
if(heat>99000)shield_proc=0.90;
if(heat>104500)shield_proc=0.95;
if(heat>110000)shield_proc=1.0;

SpotLight* le1 =(SpotLight*)AddSpotLight(_V(0.0,-12.0,20), _V(0,0.5,-1.0), 50, 1e-3, 0, 1e-3, RAD*75, RAD*80, col_r, col_d, col_d);
le1->SetIntensityRef (&shield_proc);

:feedback:

 

[tblaxland]

I don't understand your Italian, but instead of all those if statements, why not:

shield_proc = min(heat/110000.0, 1.0);

 

[Urwumpe]

Why don't you use a Q value (Enthalpy)?

This formula here:

gives you how much energy you get per second (J/s = Watt) per square meter (cross section), for a section of the heat shield. The variable [math]T_w[/math] is the outside wall temperature of the section. [math]T_\infty[/math] is the ambient air temperature, etc.

The complete heat balance of a volume cell (voxel) at the heat shield is:

Nothing surprising.

If you integrate the [math]\dot{Q}_{voxel}[/math] of the voxel over time (eg by RK4), you get the stored energy Q in the voxel, which you can then use for calculating temperature and thus, for calculating the spectrum of the radiative heat: from black body temperature, you can use a simple look-up table for getting the color and intensity of the radiation.

if you pay attention, the heating is not proportional to the dynamic pressure, but the heat flux.

hflux = 0.5*GetDynamicPressure() * velocity;

For the colors, see this one about the theory:

[ame="http://en.wikipedia.org/wiki/Black_body"]Black body - Wikipedia, the free encyclopedia[/ame]

The intensity of the light is not proportional to temperature, but proportional to [math]T_{w}^4[/math]

 

[marcogavazzeni]

I don't understand your Italian, but instead of all those if statements, why not:

shield_proc = min(heat/110000.0, 1.0);

This formula is much better!:thumbup:
If i want from a higher value than zero?

Why don't you use a Q value (Enthalpy)?

This formula here:

gives you how much energy you get per second (J/s = Watt) per square meter (cross section), for a section of the heat shield. The variable [math]T_w[/math] is the outside wall temperature of the section. [math]T_\infty[/math] is the ambient air temperature, etc.

The complete heat balance of a volume cell (voxel) at the heat shield is:

Nothing surprising.

If you integrate the [math]\dot{Q}_{voxel}[/math] of the voxel over time (eg by RK4), you get the stored energy Q in the voxel, which you can then use for calculating temperature and thus, for calculating the spectrum of the radiative heat: from black body temperature, you can use a simple look-up table for getting the color and intensity of the radiation.

if you pay attention, the heating is not proportional to the dynamic pressure, but the heat flux.

hflux = 0.5*GetDynamicPressure() * velocity;

For the colors, see this one about the theory:

Black body - Wikipedia, the free encyclopedia

The intensity of the light is not proportional to temperature, but proportional to [math]T_{w}^4[/math]

I try changing the texture alpha channel to be able to find a right shade


double heat = (0.5*GetDynPressure() * (GetAirspeed()))/100000;
shield_proc = min(heat/1200.0, 1.0);

SpotLight* le1 =(SpotLight*)AddSpotLight(_V(0.0,-12.0,20), _V(0,0.5,-1.0), 50, 1e-3, 0, 1e-3, RAD*75, RAD*80, col_r, col_d, col_d);
le1->SetIntensityRef (&shield_proc);

 

[Wishbone]

Alpha channel is just the transparency. You have to transform temperature to wavelength through the black body curve, and then wavelength to RGB triplet (see here http://www.magnetkern.de/spektrum.html)

EDIT: while the BBC gives out the whole emission spectrum, you can actually use the maximum by Wien's law.

 

[tblaxland]

This formula is much better!:thumbup:
If i want from a higher value than zero?

Many ways but an example:

const double max_proc = 1.0; // should always be 1.0
const double min_proc = 0.1; // range from 0 to max_proc
const double heat_at_min_proc = 1e3;
const double heat_at_max_proc = 1e5;

shield_proc = min_proc;
if (heat > heat_at_min_proc)
    shield_proc += min( (heat - heat_at_min_proc) * (max_proc - min_proc)/(heat_at_max_proc - heat_at_minproc), max_proc - min_proc;);

 

[marcogavazzeni]

Very thank tblaxland :tiphat:

 

[Ghostrider]

I don't understand your Italian, but instead of all those if statements, why not:

shield_proc = min(heat/110000.0, 1.0);

Because in Italian it sounds better. I mean, try whispering "una pressione dinamica di 110000 pascal equivale sull'Eridanus ad una temperatura dello scudo termico di 1200 gradi circa" in a girl's ear (provided she doesn't speak Italian). Success guaranteed!

... Or maybe you get your heatshield whacked. Whatever.

 

[marcogavazzeni]

Because in Italian it sounds better. I mean, try whispering "una pressione dinamica di 110000 pascal equivale sull'Eridanus ad una temperatura dello scudo termico di 1200 gradi circa" in a girl's ear (provided she doesn't speak Italian). Success guaranteed!

... Or maybe you get your heatshield whacked. Whatever.

:lol:

depends on what you do with the girl

double heat = (0.5*GetDynPressure() * (GetAirspeed()))/100000;
if (heat>300)shield_proc=0.01;
if (heat<1200) shield_proc=1.0;
shield_proc = min(heat/1200.0, 1.0);

 

[tblaxland]

:lol: I'll try it with my wife tonight. She does talk French, but not Italian

...code snipped...

You have a problem there for heat<300. I think you meant something like this...

if (heat<300) shield_proc=0.01;
else shield_proc = min(heat/1200.0, 1.0);

...or...

if (heat<300)shield_proc=0.01;
else if (heat>1200) shield_proc=1.0;
else shield_proc = heat/1200.0;

 

[RisingFury]

The light emission is in the best case just a black body radiation case of the heat flux, assuming the heat shield to be in equilibrium.

I haven't done any research into the matter, but off the top of the head I wouldn't just say it's a blackbody. The heatshild itself would be a blackbody due to obvious high thermal energies... but you have a lot of plasma around that might act like gasses that radiate when excited...

So I'd be careful with the guess...

 

[Wishbone]

The OP was talking about the heatshield. The dissociation of air and resultant nitrous oxides are quite well (for a sim) modelled by the orbiter itself...

 

[computerex]

hflux = 0.5*GetDynamicPressure() * velocity;

I thought velocity was cubed. In any case, for simulating "heat" in Orbiter you are going to end up with simplistic/semi-realistic results at best. Make good assumptions about your model and with fuzzy logic you should end up with a believable model to most.

 

[fausto]

i agree, computerex! I computed Antares capsule heat shield temperature by extracting it from a fraction of dynamic pressure!

 

[Urwumpe]

I thought velocity was cubed. In any case, for simulating "heat" in Orbiter you are going to end up with simplistic/semi-realistic results at best. Make good assumptions about your model and with fuzzy logic you should end up with a believable model to most.

Dynamic Pressure is already [math]p_d = \frac{1}{2} \cdot \rho \cdot v^2[/math]

 

[computerex]

Dynamic Pressure is already [math]p_d = \frac{1}{2} \cdot \rho \cdot v^2[/math]

Thanks for the correction. I made a big uh oh somewhere....