Looking for physics

[Adoniram]

Hello Everyone,

I'm a physics PhD student from Dallas, TX.  I've been planning a rather ambitious rocket project with some friends, and it was suggested to me to check out KSP too to help with planning (currently I'm working to solve a number of difficult differential equations to plan mass vs propellant requirements).  I was hoping someone here could point me to where I could learn more about the physics taken into account in KSP, or perhaps which forum I should post my physics questions under.

 

Thanks!

Matt

[Ohm is Futile]

Hello and welcome!

I would think this post would be most helpful to you:

Otherwise, this section of the forums may appeal to you more: http://forum.kerbalspaceprogram.com/index.php?/forum/44-science-spaceflight/

As for having a good idea of what KSP takes into account, you may want to visit the wiki: http://wiki.kerbalspaceprogram.com/wiki/Main_Page

Now, I am not the most science literate guy around, but I can tell you off the top of my head that the physics in KSP are much simpler than real life, although they do a good job of reproducing real hurdles of rocket science. Things you will encounter in KSP: aerodynamics/atmospheric drag and heat effects, Oberth effect, Hohmann transfers and such. Things that are especially unrealistic: clear cut sphere of influences, celestial bodies are fixed on their orbit and don't influence one another, no need to worry about wear and tear (dust, micro-meteorites, etc.) and probably more things I cannot think of at this moment.

[Magzimum]

My top-tip: Just play with KSP to "get a feeling" for rocketry. I am not sure that this game is the best to really design the rocket engines (if you plan to do that). Nozzle shapes and chamber pressures are probably not a part of the game engine. I think that the flames out of the rockets are just graphics, and the engine just 'magically' moves the rocket up in the game engine. (Not sure, correct me if this game is even more awesome than I thought). However, ratios such as fuel/dry weight and the ratio of the mass of subsequent stages can be tested quite well. 

On 7-2-2017 at 7:43 PM, Ohm is Futile said:

Things that are especially unrealistic: clear cut sphere of influences, celestial bodies are fixed on their orbit and don't influence one another, no need to worry about wear and tear (dust, micro-meteorites, etc.) and probably more things I cannot think of at this moment.

Hehe... The OP is a rocketry amateur (i.e. probably not NASA) who wants to build a rocket. I am not sure that the unrealistic clear cut spheres of influence will be a major issue. In fact, I am not sure it should be any consideration... I mean, that would be a REALLY ambitious rocketry project!

According to a quick Google search, the highest amateur rocket (ever) reached 117 km altitude. You need at least few additional kilometers before you need to start counting the gravity of other celestial bodies!

[Adoniram]

Thanks for the replies, everyone.

 

Yeah I am shooting for 70,000ft (~21km) as the ultimate goal, not 117km or anywhere near that record.  My goal isn't fastest or highest, but something completely different: rocket launches to 21km, ditches engine, deploys retractable wings, takes sweet video on the way down as a self-guided glider, lands gently on the ground.  

 

However... there's going to be something like 4 evolutions of the project before we get there.  My job is to figure out the science (how big of an engine do we need to reach 21km? what body shape optimizes lift on the way down, but minimizes drag on the way up? How much energy do we save by stabilizing the rocket on the way up instead of letting it spin like a football? etc)

 

A few people told me about KSP, so I checked it out but it seems like it isn't quite as customizable as people thought, or maybe they thought I was being purely hypothetical.  Also I couldn't find any information on air drag (for launch).  I know it's taken into account (sort of) for heating materials on the way into an atmosphere, but I think that's it.  Air drag is a VERY complicated process from what I've gathered:

Drag depends on cross sectional area, and density of air, which depends on the local pressure of air, which depends on the altitude and temperature (which depends on altitude up to the tropopause, and is ~fixed thereafter for my purposes).  And it also depends on the square of the velocity.  

 

So... solving for velocity of a rocket after burnout is really going to be a 2nd order, non-linear, non-homogeneous differential equation.    I've done it for the assumption of linear drag (which is not correct, but more correct than 0 drag), but I wanted to improve on that result, which is why I was hoping KSP could help.  Oh well, back to the drawing board (i.e. textbooks, mathematica, etc)

[Zeiss Ikon]

When I was last doing model and high power rocketry (ca. 1996), there were early attempts to put Computed Flight Dynamics into a package that could run on (then) a reasonably current home computer.  For reference, as I recall I had an early Pentium-class AMD processor, either 8 or 16 MB (not GB) RAM, and a 350 MB hard disk at the time.  IOW, this is stuff that, now, should be well within the capability of even a bottom end smart phone, if you can get the software compiled/packaged for Android.

CFD was about calculating both stability and drag based solely on the shapes of the parts -- no "balance formula", no rules of thumb; you'd build up a simulation of your actual rocket (nose cone, body, fins, transitions, tail cones) and the program would optionally add an exhaust plume (which does have a measurable effect on stability and drag), and spit out a number of results -- many of which would be useful for your project.  Specifically the drag information would be good up to the transsonic regime (where shock waves start to form) even with the 1996 PC CFD software I had, as would the stability calculations.

Now, your rocket will surely be deeply supersonic by burnout, if it's to coast to ~21 km, but I'd think this CFD capability has advanced some in the past twenty years.  Again, the computing power required, even to cover the transsonic and supersonic regimes, is much less than that available in a top end smart phone.  I'd try contacting some folks involved with either NAR (National Association of Rocketry), the governing body for "model" rocketry (but which also certifies fliers for high power, up to about 4" diameter motors 3-4 feet long), or Tripoli Rocket Association, who deal primarily in "high power" rocketry and also dabble in amateur rockets, twice the size of the NAR limits if not even larger.  Not only can they (mainly Tripoli) help with the motor work, they'll know a lot about the construction of a rocket that's to hold together for a 50-100 G launch and Mach 3+ burnout speed.

[Rybikson]

Hi,

The physics part is not so easy, but quite doable if u are familiar with derivative calculations.

U just need to make the spreadsheet for each stage including:

Velocity - initial from previous stage, final calculated (separate for x,y, tangent for easier calculation)

same for the acceleration and the distance

then use the basic accelerated movement equations and "orbit movement" equations (i mean the movement on the circular path, just dunno the EN words)

make equations for final velocity etc, use derivative to find the maximums according to the TWR

if you feel confident enough with those calculations ... add the drag factor (at this point I stopped with the mindblown coz i havent been using math for a couple of years and at this point you need to be careful about many sh#tty thing like the atmosphere density which decreases non-linear, the shape drag factor, and the drag force which is also non-linear dependent from the tangent velocity)

Finally for the game just keep some basic rules of TWR at different stages:

[disclaimer: this is my personal experience with some googled knowledge, the ascending rules are probably straight written and explained better, just google "KSP ascending path"]

- u start with TWR slightly over 1.2

- u start the gravity turn at 1600m and the speed boave 120ms/s (u need to start moving horizontal to add the centrifugal force to help you with gravity and 1600m seems to be the best moment to start, 70m/s is the first stable speed for small rockets to turn and 120m/s is quite enough for even heavier ones) u just drop the nose by 10-20 degrees

- u continue the gravity turn by dropping nose to aout 45degrees and should still be at your first stage

- at 8-16km u should decouple the first stage and continue with 0.8 TWR stage
- the final stage TWR at 0.2 is enough to maneouvre in space if u dont need to bypass any atmospheric influences and only need to make some fly-by's or land at low gravity objects.