How Do You Deal with Inerplanetary Aerocapturing in 1.0.4?

[regex]

Do you guys think Jool's atmosphere values should be tweaked to allow aerocapture?. TBH, I never tried the tylo assist, aerocapture at Jool was the easiest option until 1.0.3 and I never tried another.

Any atmosphere in the game that is represented by a floatcurve that terminates in what is essentially a brick wall should be changed.

[Brotoro]

By what mechanism does the heat from the compressed air get into the ship? It's not convection, certainly--that's just totally the wrong mechanic for this. There would be some radiation, but as the radiation goes in all directions, most of it is heating the surrounding atmosphere, not the ship. And no individual air molecule is in contact with the ship long enough to do much conduction. However, that air molecule is hitting the ship at Mach-whatever, which generates heat all by itself at the point of impact. Thus, I'm pretty sure the main way of heating the ship is by mechanical friction. I suppose any radiant heat it absorbs is indirectly caused by the mechanical friction of compressing the air, but I don't see that in itself as the major heating cause.

Instead, I believe most of the ship's kinetic energy loss does indeed go into compressing the air, which heats it, and then that heat is released and transfers to the surroundings as the air decompresses after the momentary passage of the shockwave. But most of that heat radiates in directions other than towards the ship.

The air is compressed, forming a shock wave (the ship is coming in faster than the air can get out of the way). The compression heats the thin layer of air in front of the ship, which radiates about half of its energy into the ship (since it's quite close). You also get a sheath of hot, ionized air streaming back around the ship, and I would expect about half if the radiated energy from the sheath to radiate inwards (although I wouldn't expect this gas to be as hot as the layer in front of the ship).

It doesn't matter if any particular atom quickly slides past the ship along the shock front...you still have a continuous stream of excited atoms radiating energy inward.

But I too don't understand what "convection" has to do with it. I would think that energy radiated from the shock front and energy conducted in from the shock front through the boundary layers of air around the ship would be the source of heating the ship.

- - - Updated - - -

Any atmosphere in the game that is represented by a floatcurve that terminates in what is essentially a brick wall should be changed.

This is certainly true.

[DoToH]

@Geschosskopf:

https://en.wikipedia.org/wiki/Aerodynamic_heating

Interesting reading... seems like we both were wrong

Edit: Ninja'd

[regex]

But I too don't understand what "convection" has to do with it. I would think that energy radiated from the shock front and energy conducted in from the shock front through the boundary layers of air around the ship would be the source of heating the ship.

As I understand it the convection (in this case) is from the ablation of material from whatever is being heated and would generally only apply as a form of heating in the case of a heat shield.

The shape of the object entering the atmosphere should have a large impact on how the atmosphere affects it. For instance, a large, open structure might experience much more frictional heating than a monolithic blunt body that compresses the atmosphere ahead of it (this is a big reason why I never enter the atmosphere with a craft that isn't designed for it, the only "roleplaying" I really indulge in). I'm not sure whether KSP models that.

[moogoob]

I don't. I avoid it, until it's fixed in stock. Then i wouldn't mind playing around with it.

[Superfluous J]

I don't. I avoid it, until it's fixed in stock. Then i wouldn't mind playing around with it.

Waiting for it to be fixed is presuming it's broken. Many don't think it's broken, and it's pretty likely Squad is in that camp. It may therefore never - to your satisfaction - be "fixed."

[juanml82]

Setting up a Tylo or Laythe gravity capture into Jool orbit is not as difficult as you might think. The trip out to Jool takes a long time compared to the orbital periods of those moons, so just increasing or decreasing your travel time by altering your trans-Jool trajectory can get you there at the proper time for the encounter you want. You just need to tweak you prograde/retrograde in combination with your radial outward/inward components (while focusing on the Jool system as you tweak the distant maneuver node in the background) to see ho this changes things. I usually do this targeting burn just after leaving Kerbin's SOI. The prograde/retrograde component will shift your arrival time, and the radial component can be used to target the encounter. You want to arrange for the encounter to be tangential to the moon's orbit...and the distance you pass by the outside of the the moon will determine the resulting capture orbit around Jool. I play with it until I get an acceptable result (I like the Precise Node mod for the fine tweaking)

If you have one, could you share a screenshot of such a trajectory for those of us who are reading this half asleep? A picture can be worth a thousand words for the sleepy mind

[Geschosskopf]

@Geschosskopf:

https://en.wikipedia.org/wiki/Aerodynamic_heating

Interesting reading... seems like we both were wrong

Well, the stationary boundary layer (which is always present at any speed unless you've got massive turbulence like in a stall), if it gets hot, is transferring heat by conduction, not convection. It's effectively part of the ship for heat transfer purposes. So using the name "convection" is technically incorrect.

Also note the part at the end about reentry vehicles, how with a heat shield, most of the heat released as the shockwave decompresses radiates away from the ship like I was saying. And the shuttle, it was basically coated in heatsink. It worked on the same principle as wood-frame buildings with gypsum board on the walls covering the wooden studs of the structure. You put a material with a high specific heat between the fire and the easily damaged stuff. Specific heat is the amount of energy required to raise the temperature of a unit mass 1 degree. The more energy this requires, the less hot the material gets in a given time, so the less heat it conducts to the underlying structure. Of course, houses still burn down because the gypsum board is pierced for light figures, HVAC ducts, and the very low specific heat nails holding it to the studs, but that's the theory anyway .

The same thing also works with the materials the other way around. A thin layer of low-specific heat, heat-conductive material on top of a mass of high-specific heat material is also protected. Again, think of the plastic bottle of water exposed to the flame of a match. So full fuel tanks protect the skin of the tank.

The shape of the object entering the atmosphere should have a large impact on how the atmosphere affects it. For instance, a large, open structure might experience much more frictional heating than a monolithic blunt body that compresses the atmosphere ahead of it (this is a big reason why I never enter the atmosphere with a craft that isn't designed for it, the only "roleplaying" I really indulge in). I'm not sure whether KSP models that.

An open structure also has a higher surface-to-mass ratio. The bigger this number is, the more of its mass is directly exposed to heat, so if the heat source totally envelopes the object, it will heat up faster regardless of its specific heat (although it still takes the same amount of energy to heat up 1 kg of the material 1 degree).

This gets into what is really the all-important thing about heat transfer, which is heat flux. Or as we say in the fire service, heat release rate. Watts instead of Joules. This is way more important than temperature. For instance, suppose you have some sort of candle that burns at 1000^F and produces 1 Watt. If you put 10 candles side-by-side, the temperature remains the same but now you have 10 Watts, which will heat things up faster (although the kJ/kg is still the same).

Waiting for it to be fixed is presuming it's broken. Many don't think it's broken, and it's pretty likely Squad is in that camp. It may therefore never - to your satisfaction - be "fixed."

It's demonstrably broken if there is no thin, wispy upper layer to the atmosphere. That's a hydrosphere instead.

[NathanKell]

In pretty much all the books and journal articles I've seen on reentry heating, it's broken down into convective heating (the shock/friction/hot-air-against-fuselage heating) and radiative heating (black-body radiation from air molecule temperature). That's my $0.02.

[Superfluous J]

It's demonstrably broken if there is no thin, wispy upper layer to the atmosphere. That's a hydrosphere instead.

Aha. That part. As you did not specify (at least not in the post I quoted) I thought you were just talking about the increased difficulty, like most people do.

I do agree that that should be corrected, though unlike in reality there has to be a hard limit somewhere in the game. The density curve should go down to zero at that point, though, and not just cut off at it.

[Xeiki]

I just never touch the athmosphere.

Since in my book looks are as am important as functionality, heatshields are out of the question. Ships with one heatshield have no field of view forwards, and ships that need more than one to cover all parts are just plain ugly.

Also, those strange athmospheres are no fun. Kerbin is kinda ok, Laythe is of no use and Jool is just plain stupid. Dipped 1m into the athmosphere, half of my ship exploded. Half a second later the rest followed suit.

[NathanKell]

It does go down smoothly to zero. Do any test you like, measure density at any arbitrarily small altitude below max atmosphere altitude.

It's just that it increases above zero too fast, which can appear nearly wallish when you're moving super-fast but (unlike .90) is not actually a wall.

[Brotoro]

Setting up a Tylo or Laythe gravity capture into Jool orbit is not as difficult as you might think. The trip out to Jool takes a long time compared to the orbital periods of those moons, so just increasing or decreasing your travel time by altering your trans-Jool trajectory can get you there at the proper time for the encounter you want. You just need to tweak you prograde/retrograde in combination with your radial outward/inward components (while focusing on the Jool system as you tweak the distant maneuver node in the background) to see ho this changes things. I usually do this targeting burn just after leaving Kerbin's SOI. The prograde/retrograde component will shift your arrival time, and the radial component can be used to target the encounter. You want to arrange for the encounter to be tangential to the moon's orbit...and the distance you pass by the outside of the the moon will determine the resulting capture orbit around Jool. I play with it until I get an acceptable result (I like the Precise Node mod for the fine tweaking)

If you have one, could you share a screenshot of such a trajectory for those of us who are reading this half asleep? A picture can be worth a thousand words for the sleepy mind

OK. Below is a trajectory plot for a Jool encounter. I already did the fine targeting maneuver at the descending node here before I quicksaved, so I have put in another maneuver node to show you the prograde/retrograde & radial/antiradial tweaking.

For the next image below, I shifted my point of view to Jool by pressing the Tab key repeatedly until I was focused on Jool, and then I oriented the view so that I could see the far-off maneuver node in the background. You can see that my trajectory is currently set up for an encounter with Laythe tangentially to its orbit. If it wasn't set up already, I could tweak the prograde/retrograde and radial/antiradial and normal/antinormal handles on the maneuver node to get the trajectory into about the right area. OR, since I have the Precise Node mod installed, I could directly edit the maneuver components in the fields of the dialog box (or left- or right-click the buttons to increment the values by the amount selected in the dialog box).

The next image below shows the effect of decreasing the prograde by 0.01 m/s...the dotted line would be the result of that change. This has the effect of swinging the trajectory out a little, and it also has the effect of causing the ship to arrive at Jool a little bit later so that Laythe is further along in its orbit:

In the next image below, I show the result of instead INCREASING the prograde by 0.04 m/s. The dotted line shows that the resulting trajectory would swing inward a little, and would also result in the ship getting to Jool a little sooner so that Laythe would not be as far around in its orbit. So even small changes in the prograde/retrograde component can make a big difference in the arrival time and where Laythe will be in its orbit (because this maneuver is being done a LONG TIME before the encounter). The maneuver can certainly be done closer to Jool (even within Jool's SOI), but at a higher cost in delta-V.

So if you tweak the prograde component, you can get your ship to arrive when Laythe is at the right place in its orbit. BUT your trajectory might not be hitting that point. SO you need to adjust the radial/antiradial component to swing your trajectory in or out as needed to hit the spot. It requires making many changes to the prograde/retrograde and radial/antiradial components to arrive at the right place at the right time...but it's fairly easy to do with Precise Node. The next image below show the effect of INCREASING the RADIAL component by 0.04 m/s. The trajectory swings inward a little, and the encounter with Laythe occurs a little earlier.

And the final image shows the effect on the trajectory of DECREASING the radial component by 0.04 m/s. The trajectory swings outward a little, and the ship encounters Laythe a little further ahead in its orbit.

And as shown in a previous post, if you miss Laythe's atmosphere (even on the side further from Jool), you can get a gravitational capture (although at a periapsis of 21 km, this would result in an aerocapture -- I later tweaked the aerocapture altitude to 43 km to get the capture orbit I wanted around Laythe).

Edited July 3, 2015 by Brotoro

[Geschosskopf]

In pretty much all the books and journal articles I've seen on reentry heating, it's broken down into convective heating (the shock/friction/hot-air-against-fuselage heating) and radiative heating (black-body radiation from air molecule temperature). That's my $0.02.

Hmm. Well, I guess that's just the convention but the mechanism is much more akin to conduction than conventional convection (geez, it's hard to keep all these "con...tion" words straight in 1 sentence ).

Anyway, I hope the heat model has 2 separate "convection" routines, one of this reentry thing and the other as a cooling mechanism when the vehicle is at rest (or moving relatively slowly) in an atmosphere or ocean. They work in entirely different ways and do entirely different things. If the heat model uses 1 routine for both meanings of "convection", then it will probably give undesirable results in 1 case or the other.

It does go down smoothly to zero. Do any test you like, measure density at any arbitrarily small altitude below max atmosphere altitude. It's just that it increases above zero too fast, which can appear nearly wallish when you're moving super-fast but (unlike .90) is not actually a wall.

It might not be a vertical drop, but it's close enough . This really should be changed.

OK. Below is a trajectory plot for a Jool encounter. I already did the fine targeting maneuver at the descending node here before I quicksaved, so I have put in another maneuver node to show you the prograde/retrograde & radial/antiradial tweaking.

Very informative, thanks.

[NathanKell]

Yes, there are two models, see Physics.cfg -- one is the newtonian model, and one is the hypersonic model. Well, 2.5, since the newtonian model behaves a bit differently when you're splashed.

And I don't disagree in the slightest that the slope should be fixed.

[Technical Ben]

I just never touch the athmosphere.

Since in my book looks are as am important as functionality, heatshields are out of the question. Ships with one heatshield have no field of view forwards, and ships that need more than one to cover all parts are just plain ugly.

Also, those strange athmospheres are no fun. Kerbin is kinda ok, Laythe is of no use and Jool is just plain stupid. Dipped 1m into the athmosphere, half of my ship exploded. Half a second later the rest followed suit.

Wait. I'm making my craft look cool WITH heatshields.

[Geschosskopf]

And I don't disagree in the slightest that the slope should be fixed.

Is this something that DRE, FAR, or a combination of them can fix or do we have to wait on Squad?

[Superfluous J]

Is this something that DRE, FAR, or a combination of them can fix or do we have to wait on Squad?

I'd prefer a non-FAR fix. I was hoping this curve was defined in a config file somewhere but either it's not or I'm not smart enough to find it.

[regex]

I'd prefer a non-FAR fix. I was hoping this curve was defined in a config file somewhere but either it's not or I'm not smart enough to find it.

Like most (or all) of the planetary data it's in binary data.

[NathanKell]

It is, but a very simple mod and cfg can change it. I'll post the code in the dev forum, anyone who wants can make a release of it.

[Delrey]

I just grazed Eve's atmosphere and did a retrograde burn. I was at about 85km and had some scary heating going on. Its dangerous as hell now. My attempt at aerocapturing at 76km lead to destruction. You have to have a heat shield.

[Blacks]

It is, but a very simple mod and cfg can change it. I'll post the code in the dev forum, anyone who wants can make a release of it.

sound delicious (golum model activate), special if that code > cfg < it will bring a model or a reference at how to play with value, coz change it i bet is easy but make it correct is hard part of problem, speaking from perspective of a person who was done yesterday soo many test special for see how heating it can be adjust

[NathanKell]

Blacks: Um, I was talking about modifying body properties. RSS already has the correct body properties. What you're looking at is the global physics properties, i.e. changing them to be appropriate for a real-size universe not the KSP one.

Anyway, here's the thread. It even has an example for Jool.

http://forum.kerbalspaceprogram.com/threads/127755-Bodyloader-A-lightweight-body-modifier

[Richy teh space man]

I've found interplanetary aerocapturing to be a great way of doing fireworks, but seriously its a lot more realistic and will take some redesigns to be able to do it for me.

For less experienced players, I think it will be another learning curve if they plan to do aerocaptures.

Edit: Wow I didn't realise how broken the Jool aerobrake is, even going from a 201km x 199km means certain death for the vessels that I've been testing so far.

Edited July 5, 2015 by Richy teh space man

[Sharpy]

Missing option: Use 100% heat and *touch* the atmosphere, without heatshields, taking exactly as much heat as the (unshielded) craft can get, burn away the rest of speed with engines.

With 90km atmospheric height, I performed my Eve orbital insertion at 87km periapsis. I believe I lost about a 1000m/s to airbraking, burned away the rest.