raptop changed the topic of #kspacademia to: https://gist.github.com/pdn4kd/164b9b85435d87afbec0c3a7e69d3e6d | Dogs are cats. Spiders are cat interferometers. | Космизм сегодня! | Document well, for tomorrow you may get mauled by a ネコバス. | <UmbralRaptor> egg|nomz|egg: generally if your eyes are dewing over, that's not the weather. | <ferram4> I shall beat my problems to death with an engineer. | We can haz pdf
e_14159 has quit [Ping timeout: 190 seconds]
e_14159 has joined #kspacademia
egg|cell|egg has joined #kspacademia
egg|cell|egg has quit [Ping timeout: 180 seconds]
egg|cell|egg has joined #kspacademia
<_whitenotifier-3d18>
[Principia] pleroy closed issue #2107: Journal replaying is no longer deterministic - https://git.io/fjJyD
<_whitenotifier-3d18>
[Principia] pleroy closed pull request #2141: Synchronize the prognostication when recording and replaying the journal - https://git.io/fj39e
<_whitenotifier-3d18>
[Principia] pleroy pushed 11 commits to master [+9/-2/±57] https://git.io/fjGe1
<_whitenotifier-3d18>
[Principia] pleroy b3863b3 - Merge pull request #2145 from pleroy/Linux Fix problems found on Linux
egg|cell|egg has joined #kspacademia
egg|cell|egg has quit [Ping timeout: 190 seconds]
egg|laptop|egg has quit [Ping timeout: 183 seconds]
egg|laptop|egg has joined #kspacademia
<egg|laptop|egg>
UmbralRaptor: whitequark: the number of moons of Jupiter is no longer Nice: "On 17 July 2018, the International Astronomical Union confirmed that Sheppard's team discovered ten more moons around Jupiter, bringing the total number to 79."
<whitequark>
aw
<UmbralRaptor>
oh well
<_whitenotifier-3d18>
[Principia] pleroy opened pull request #2146: Next release is Fatou - https://git.io/fjGTH
<egg|laptop|egg>
ferram4: assuming you are interested mostly in hypersonic/reentry behaviour, how few parameters can you give a model for aerodynamic forces and still have something useful
<ferram4>
Sorta around
<ferram4>
Depends on what part of reentry you're interested in
<egg|laptop|egg>
ferram4: context is someone asking whether trajectories could interact with principia (this sounds like a mess) and me wondering whether we can predict where something not overly weird will land
<ferram4>
If you're only interested in forces, you really only need the forward facing shape of the body, the angle of attack of the vehicle and the density
<egg|laptop|egg>
(STS is overly weird for our purposes)
<egg|laptop|egg>
trajectories the mod, not the concept of trajectories
<ferram4>
Assuming that it's stable in that orientation
<ferram4>
So, as always, assuming you're dealing with something sane
<egg|laptop|egg>
ferram4: I suppose what I could do also is ask you to compute forces and interpolate from that
<ferram4>
But for the level of reentry forces that FAR or any simple aero model would work with, very easily
<ferram4>
There's already an API that you can use for that
<ferram4>
I'm not sure how possible it is to figure out the proper orientation for that, but you can grab the forces from FAR for that already
<ferram4>
I believe I added that to interface with Trajectories awhile back
<egg|laptop|egg>
ferram4: well
<egg|laptop|egg>
I can't call your force computation from the middle of my integration
<ferram4>
That's true
<ferram4>
You could call the API outside of the integration and store a couple values to interpolate between
<ferram4>
It'll be close enough
<ferram4>
I mean, the actual accuracy will be trash
<ferram4>
But we're talking adding more chaos and inaccuracy to a system filled with chaos
<egg|laptop|egg>
:D
<ferram4>
"There was a point where we needed to stop and we have clearly passed it, but let's keep going and see what happens"
<egg|laptop|egg>
ferram4: yeah, but interpolation on many variables is a right mess, so if I interpolate I want to choose my variables wisely
<egg|laptop|egg>
e.g. can I say things depend on Q instead of ρ and u, or things like that
<egg|laptop|egg>
I have very little intuition of how these things work
<ferram4>
Alright, how much do you care about stock vs. RSS?
<egg|laptop|egg>
I sort of care about both >_>
<ferram4>
Alright, so here's the absolute minimum
<ferram4>
Q
<ferram4>
Angle of attack, sideslip, roll
<ferram4>
You know, orientation, figure out how you'd like to store that and vary that
<ferram4>
That should be good for RSS. Stock you might need Mach number, but I'd judge that on a case-by-case basis
<ferram4>
Like for Jool, you certainly don't need Mach
<ferram4>
General rule for hypersonics is that above Mach ~5, you can kind of assume that the Mach number is infinity and run with it. Don't actually enter infinity into the calcs though
<egg|laptop|egg>
:-p
<egg|laptop|egg>
well, it can be fun,
<ferram4>
At least on the level I'm working with
<ferram4>
For more accurate models with shock-boundary layer interactions that won't work
<ferram4>
But I don't do that
<ferram4>
Because that's not really doable in real time
<ferram4>
Nor do I have experience with that
<ferram4>
Forces should vary linearly with Q, so I don't think you actually need more than 2 or 3 points for that.
<egg|laptop|egg>
ok, that's interesting
<ferram4>
You'll need to figure out what maximum Q you should look for though, it'll vary body by body
<ferram4>
Remember, the point of calculating aerodynamic forces as [area] * [dyn pres] * [coeff] is to make scaling stuff really simple
<egg|laptop|egg>
so basically if I assume a fixed orientation (if you tumble it's not my problem, and if you have a plane fly the damn thing yourself), it's theoretically easy
<ferram4>
Yes
<ferram4>
Now
<egg|laptop|egg>
is there a common API to get your forces or stock's in that context?
<ferram4>
I don't think there's a common one for both FAR and stock
<ferram4>
I'm actually not 100% sure how to get stock's
<egg|laptop|egg>
(I also want to avoid explicitly calling FAR because licensing madnesss >_<)
<ferram4>
Understandable
<egg|laptop|egg>
I really like the thing we managed to do where your real-time forces look like engines and it Just Works with Principia
<egg|laptop|egg>
(debugging that one was fun)
<egg|laptop|egg>
(something something 300 m/s surface winds at one point)
<ferram4>
Yeah, I don't think there's an easy way to do this without calling things directly
<ferram4>
It's not exactly like stock is designed for nice, sane ways of getting predictions either
<egg|laptop|egg>
yeah
<egg|laptop|egg>
ferram4: by the way, how is your FAR-building computer doing?
<ferram4>
Currently still shut off
<ferram4>
And in storage halfway across the country
<egg|laptop|egg>
ow
<ferram4>
GOt a job offer, I'm in training, can't really work on it if I wanted
<egg|laptop|egg>
oh, congrats on the job
<egg|laptop|egg>
ferram4: https://forum.kerbalspaceprogram.com/index.php?/topic/179445-14-17-ferram-aerospace-research-continued-v015101-lundgren-170419/ is a thing so it's possible to use FAR with recent KSPs at least
<ferram4>
Yes, I've seen
<ferram4>
If I've got time when I'm past most of the hard training and the probationary period I'll see if I can start getting back into things
<egg|laptop|egg>
yay
<egg|laptop|egg>
what's the job if I may ask?
<ferram4>
But for now, job is more important
<ferram4>
Air traffic control
<egg|laptop|egg>
oh nice
<ferram4>
So, yeah. They'
<ferram4>
They're not going easy on any of us newbies
<egg|laptop|egg>
somewhat understandable
<egg|laptop|egg>
ferram4: btw, the next version of Principia will have text entry for the flight plan, in addition to the sliders
<egg|laptop|egg>
(after years of users clamoring for that)
<egg|laptop|egg>
clamouring?
<ferram4>
[insert glare]
<ferram4>
FINALLY
<egg|laptop|egg>
:-p
<egg|laptop|egg>
ferram4: look it's hard to get it right
<egg|laptop|egg>
even now there are a bunch of cases where the UI will revert your edit without telling you why
<ferram4>
Yeah, it's always the small things
<egg|laptop|egg>
at least it highlights when you have a syntax error so that's nice
<ferram4>
I have ranted about how FAR is more geometry than aero?
<ferram4>
I feel like I have
<egg|laptop|egg>
probably
<egg|laptop|egg>
ferram4: btw not sure I told you, we have geopotential modeling beyond J2 now
<egg|laptop|egg>
the Moon has mascons
<egg|laptop|egg>
muahahaha
<ferram4>
How many users have complained that their orbit crashed into a mountain?
<egg|laptop|egg>
none
<egg|laptop|egg>
my users seem mostly happy about the mod/grateful for its eggsistence by and large; where they get grumpy is mostly UI questions
<egg|laptop|egg>
there is also the odd user who complains that their solar system is unstable and that this should not be a problem, but mostly they're OK
<egg|laptop|egg>
we have a ton of Chinese users amusingly
<egg|laptop|egg>
I found a page in Chinese that explains what principia is, gives some guidance to use it, etc.
<ferram4>
*shrug*
<egg|laptop|egg>
ferram4: it says that you can use a lunar calendar to find out when the next version comes out :D
<ferram4>
Well, I'd argue that supersonic Mach, but still less than 5 (where the actual Mach number is significant in determining the forces) is still in the range of "high" Mach numbers, but the gist is right at least
<egg|laptop|egg>
ferram4: it's low Mach number in eggland :-p
<ferram4>
Also, the landing site will be unpredicatable
<ferram4>
Period
<egg|laptop|egg>
yeah but the size of the uncertainty matters
<ferram4>
It'll be somewhere along that trajectory, but downrange won't be predictable
<ferram4>
It depends on the shape
<ferram4>
Unfortunately
<ferram4>
I can't even give you ballpark numbers
<egg|laptop|egg>
long ellipse then, the usual
<egg|laptop|egg>
ferram4: we could conceivably run multiple integrations to try to bracket things
<egg|laptop|egg>
and draw an ellipse
<ferram4>
Reentry capsule going through supersonic->transonic->subsonic will be very different than reentering debris on the same course
<ferram4>
If you care that much, you can try and account for Mach number
<ferram4>
Because that will be the big thing. Assuming you somehow maintain the same orientation (you won't, AoA will change with mach number)
<ferram4>
Of course, if it's a reentry capsule with a lifting reentry, 10 degrees or so of AoA change won't be too bad
<egg|laptop|egg>
better simulation is good, but pessimistic estimate of uncertainty might be easier and usable
<ferram4>
egg|laptop|egg, I was originally thinking you were only concerned with aerobraking accuracy when I said Q alone would be good enough. If you want landing accuracy, you probably want to account for Mach as well
<ferram4>
So make sure to get the speed of sound as a function of altitude for each of these bodies
<egg|laptop|egg>
yeah iirc that's a spline
<egg|laptop|egg>
ferram4: 2-parameter interpolation is infinitely more annoying tbh
<ferram4>
I know
<egg|laptop|egg>
ferram4: is it mostly-linear in Q for all Mach numbers?
<ferram4>
Yes
<egg|laptop|egg>
ah good
<ferram4>
That [coeff] thing we calculate?
<ferram4>
We calculate it as a coefficient so that we can throw all the nasty ugly shit in there
<ferram4>
Including variations in Mach number
<egg|laptop|egg>
lol
<ferram4>
The only reason it won't be perfectly linear is because we're ignoring Reynolds number effects, which should be minor
<egg|laptop|egg>
ferram4: yeah if you're landing in an atmosphere of tar it's not my problem
<ferram4>
There will be some funkiness at the very edge of the atmosphere
<ferram4>
Because non-continuum flow results in slightly higher drag coefficients there
<ferram4>
With, like, barely any Q to apply the forces though
* B787_300
wanders in
<ferram4>
But still, you might see some underprediction of drag so...
<B787_300>
what is the discussion?
<B787_300>
looks like a discussion on how to estimate reentry drag?
<egg|laptop|egg>
ferram4: at some point it can start being fudge factors on my side instead of more stuff to interpolate I guess
<ferram4>
Minimum number of variables that can be used for predicting hypersonic aerodynamics assuming we're keeping things down at the level of Newtonian flow
<ferram4>
And how to make it work
<_whitenotifier-3d18>
[Principia] eggrobin commented on issue #2137: Integration of Trajectories aerodynamic prediction mod - https://git.io/fjGLb
<ferram4>
Mostly so that Principia can predict where things will through an atmosphere
<egg|laptop|egg>
ferram4: mind you there is the related question of making Principia toss your LEO sats to the ground so that's also fun
<ferram4>
Ah, right
<ferram4>
that is a thing
<B787_300>
how high does the atmo go in Principa?
<egg|laptop|egg>
B787_300: right now it doesn't exist :-p
<egg|laptop|egg>
ferram4 is an engine
<egg|laptop|egg>
literally, ferram4's force go through the same code path as main engine thrust
<B787_300>
if it is an engine why woudnt you be able to have exosphere-like atmosphere where the thrust is nicronewtons
<_whitenotifier-3d18>
[Principia] eggrobin edited a comment on issue #2137: Integration of Trajectories aerodynamic prediction mod - https://git.io/fjGLb
<egg|laptop|egg>
B787_300: because somebody would have to compute that thrust
<egg|laptop|egg>
we just do what ferram4 says should be done
<B787_300>
also it would probably make timewarp a pain
<egg|laptop|egg>
yeah I'm in no hurry to add that given that stationkeeping is not a solved problem
<egg|laptop|egg>
the fancy geopotential and third-body effect already make GEO unusable :-p
<egg|laptop|egg>
(well, you can stay at the two stable longitudes, but you still have third-body effects)
<B787_300>
average the effects of stationkeeping to stay in a userdefinable box at geo and just decrement fuel every year or so?
<egg|laptop|egg>
possibly
<egg|laptop|egg>
that requires thought and design and is very different to most things Principia does, we'll have to ponder
<B787_300>
because IIRC even at the super unstable points the dv needed is like 5 m/s E-W and 15 m/s N-S to stop the moon effects
<egg|laptop|egg>
B787_300: there are plenty of orbits that need stationkeeping, so you need a way to define what property of your orbit you care about
<egg|laptop|egg>
do you want to keep mean solar time of ascending node constant (for a sun-synch orbit)? do you want to keep your ground track within a few km of taiwan? do you want to stay in the same plane as your friends in the same constellation? etc.
<B787_300>
how about keep the kelperian elements within X % of desired?
<egg|laptop|egg>
the keplerian elements as defined using which perturbation theory,
<B787_300>
...
<B787_300>
good question
<egg|laptop|egg>
all of these are hard problems
<egg|laptop|egg>
B787_300: the first thing I'm trying to do is an analyser that doesn't try to put you in an orbit or help you get to one, just tells you things about your orbit
<egg|laptop|egg>
does it have a nearly-recurrent ground track, what are its nodal/apsidal/sidereal periods, what are its precession rates, etc.
<egg|laptop|egg>
also huh, stock engines are starting to look decent O_o
<egg|laptop|egg>
I see no tankbutt on the new poodle
<kmath>
<sigfig> you can reduce dimension of a variety of models in hypersonic flight by assuming the atmosphere is a rigid body fre… https://t.co/ICWvDWMVij
<egg|laptop|egg>
mofh: have you looked at the IGS mirrors
<egg|laptop|egg>
IVS?
<galois>
IVS: International VLBI Service for geodesy & astrometry
<egg|laptop|egg>
how many of those services are there
<egg|laptop|egg>
!acr -add:PASTEL PAssager SPOT de TÉlécommunication Laser
<galois>
Definition added!
<egg|laptop|egg>
HRVIR?
<egg|laptop|egg>
!acr -add:HRVIR Haute Résolution Visible et InfraRouge
<galois>
Definition added!
<UmbralRaptor>
Dear NASA: linking to ApJ in general instead of to the actual paper is not useful. (Technically the paper hasn't been published yet. Fortunately, arχiv exists)
<egg|laptop|egg>
hm, so for eggsample orbits with precision ephemerides, the IDS should have SPOT which covers heliosynch, IGS orbits for GNSS sats and the IDS orbits for TOPEX/Poseidon & the Jasons provide a variety of non-heliosynch repeat ground track orbits, and the ILRS orbits for ЭТАЛОН and LAGEOS -1 and -2 for less special orbits
<egg|laptop|egg>
not much in the way of eggscentric orbits though
egg|cell|egg has quit [Ping timeout: 180 seconds]
egg|cell|egg has joined #kspacademia
egg|laptop|egg has quit [Ping timeout: 202 seconds]
egg|laptop|egg has joined #kspacademia
egg|cell|egg has quit [Remote host closed the connection]
egg|cell|egg has joined #kspacademia
<egg|laptop|egg>
!seen mofh
<galois>
egg|laptop|egg: I last saw mofh at 2019-04-26 - 12:48:20 in here, saying egg|cell|egg: damnit I need to teleport to Paris
<egg|laptop|egg>
;8ball does mofh eggsist
<kmath>
egg|laptop|egg: My sources say no
<egg|laptop|egg>
hm
<UmbralRaptor>
!8 does mofh eggsist
<galois>
UmbralRaptor: yes
<egg|laptop|egg>
;8ball did mofh just spring into eggsistence
<kmath>
egg|laptop|egg: Concentrate and ask again
<egg|laptop|egg>
;8ball did mofh just spring into eggsistence