(Remote Tech) How to calculate space between satellites depending on orbit.

[DCWarHound]

Hey all,

Playing with Remote tech a bit and i'm having a hard time figuring out how to evenly space my satellites so here are a few questions.

1.If my orbit altitude is 300km, is my orbit circumference 1884.96km ?

2.Does KER (Kerbal Engineer Redux) calculate direct distance to the selected target?

3.How can i find the direct distance my satellites need to be from eachother at 300km using KER? ( basically putting a square in a circle and then measuring the distance ?)(i've found the distance to be 424.26km , is this correct?)

4.if i have 2 objects in the same orbit (lets say 300km) and i increase the apoapsis of one object to 310km, how much closer would they get with each pass and how do i calculate this?

Thanks

DC

Edited October 27, 2015 by DCWarHound

[Van Disaster]

You forgot Kerbin's radius ( if you're at Kerbin ) - that's 600km, so add your 300km altitude to get the radius of your circular orbit ( the circumference is somewhat more than your number ). If you've raised your AP and left PE alone you have an elliptical orbit - the maths for elliptical orbits isn't that complicated, it's worth looking it up. The direct distance between your sats is a chord, also simple to calc if you have the angle around the orbit and the radius.

[LostOblivion]

What you want to find is the orbital altitude that will give equal spacing between your sats, given antenna range (direct distance, but you probably want some margin, say 1 %).

[Jovus]

KER will also give you orbital period; making this as close to 6h is your real goal. It doesn't matter if your orbit isn't perfectly circular.

[Snark]

KER will also give you orbital period; making this as close to 6h is your real goal. It doesn't matter if your orbit isn't perfectly circular.

Actually, as long as you don't have any big gaps in your satellite formation around Kerbin, it doesn't matter what their orbital period is, as long as they have exactly the same orbital period as each other. Whenever one of them rotates out of view of KSC, another one rotates into view.

[LostOblivion]

I found it. Previously, I used trigonometry to find the semi-major axis (orbital altitude essentially, for circular orbits) for a given range and number of evenly spaced satellites.

a=2*r*sin(180 degrees/n)

Where:

a is the semi-major axis, in km

r is the preferred range between the satellites, in direct line, in km

n is the number of satellites

a is given from the center of the parent body, so for Kerbin, you'll have to subtract by 600 km if you wish to use it to set Pe/Ap.

a=2*r*sin(180 degrees/n)-600

Example:

We want to place three evenly spaced satellites into orbit with an antenna range of 2500 km (the stock antenna). We probably want some margin, so we say 2490 km instead.

Let's solve it.

2*2490*sin(180 degrees/3)-600=3712.8 km

So we want our three sats to have their Pe and Ap of 3712.8 km above Kerbin.

Edited October 24, 2015 by LostOblivion

[GoSlash27]

DC,

Figuring out the distances between satellites doesn't really help solve the problem of how you're going to get the satellites in position.

What you really need to know is orbital period from orbital radius and vice- versa.

For Kerbin, period (in seconds) is

p= 2À√(r³/µ)

where

r= orbit radius from Kerbin's center in meters. Subtract Kerbin's radius (600km) to get altitude.

and

µ= Kerbin's mass x std gravitational constant; approx. 3.5316x10^12 in cubic meters per second squared.

algebraically flipping this around,

r=³Ã¢Ë†Å¡[µ(p/2À)²]

To get sats into precise positions, you play around with orbital periods rather than distances. For example, if your final orbit had a period of 1 hr, you might use an elliptical parking orbit with the Ap at the desired altitude and an orbital period of 40 minutes. Or you might pick a higher parking orbit with a Pe at the desired orbit and an orbital period of 1hr 15 minutes.

It's much easier to do if you launch all of the sats into a parking orbit in 1 bus assembly and kick off the individual sats 1 at a time.

If that's not an option, then the process gets a great deal more complicated. You have to figure out launch windows based on time, track when the sats pass overhead KSC, then correct their orbits as necessary.

Best,

-Slashy

[Alshain]

1. If you don't want to bother doing the math this little tool helps

https://ryohpops.github.io/kspRemoteTechPlanner/

2. Yes, but so does stock. It's just like rendezvous. Set it as a target, but instead of trying to get the closest possible approach, get a separation that is equal to the distance you need for the network. I usually check both satellites on either 'side'. Even then it doesn't have to be absolutely perfect unless you are in a really low orbit skirting the atmosphere with your signal.

3. You can't, use the tool I linked above.

4. Use the target tools in stock to find out. The closest approach arrows will tell you this. You can often plan this out in 2 maneuvers using the maneuver nodes. You only need KER to get your final orbital period once you are in position.

Sorry to Slashy but I have to contradict him. It's not easier to take one massive satellite launcher up (it might be cheaper though). I have done it that way, but you have to spin around the planet quite a bit, this gets worse the higher your orbit is, at synchronous you will be in time warp for a very long time. It's actually easier to launch each individual satellite into a 'somewhere in the ballpark' position and use the target tools to check their separation and adjust their orbits from there. For synchronous, obviously you will always end up in the same place on launch so you need to launch into a lower parking orbit and extend out in the direction you want.

Edited October 24, 2015 by Alshain

[GoSlash27]

Sorry to Slashy but I have to contradict him. It's not easier to take one massive satellite launcher up (it might be cheaper though). I have done it that way, but you have to spin around the planet quite a bit, this gets worse the higher your orbit is, at synchronous you will be in time warp for a very long time. It's actually easier to launch each individual satellite into a 'somewhere in the ballpark' position and use the target tools to check their separation and adjust their orbits from there.

Alshain,

That's cool. We just have different definitions of "easier".

What I mean by "easier" is getting the sats into position. If you do a single launch with a resonant orbit, you set up a machine that neatly drops the sats where they're supposed to be like intermeshing gears, no adjustments necessary. If you launch them one at a time, then you wind up with a ragged constellation that requires you to adjust each sat to get it where it's supposed to be.

But I suppose it's a different problem if you use mods to assist in the process.

Best,

-Slashy

[Alshain]

Alshain,

That's cool. We just have different definitions of "easier".

What I mean by "easier" is getting the sats into position. If you do a single launch with a resonant orbit, you set up a machine that neatly drops the sats where they're supposed to be like intermeshing gears, no adjustments necessary. If you launch them one at a time, then you wind up with a ragged constellation that requires you to adjust each sat to get it where it's supposed to be.

But I suppose it's a different problem if you use mods to assist in the process.

Best,

-Slashy

Yeah I know what you mean, but using that method you have to adjust them anyway unless you manage to get them the perfect distance in a single orbit of the 'machine'. I've found the target separation I want always ends up offset from the Periapsis so I have to constantly adjust the 'machines' orbit so I hit the target separation where it's orbit intersects the target orbit. This can be very difficult to do, especially for newer players, as it requires more than just a prograde/retrograde burn often since the machine is in an elitpitcal orbit.

If you are already in a near-matching circular orbit as you would be without the 'machine', you can start the adjustment anywhere it's orbit and use nothing but prograde/retrograde. Doing that you can easily ensure your target separation occurs near the point of intersect (Assuming you aren't attempting a Molniya network)

The only other mod I use is KER to get orbital period so they don't drift out of place, but that doesn't help me get them into position. I use that website to quickly calculate the distance I want but I could do that mathematically if I had the inclination to do so. So yeah, I don't really use other mods.

Edited October 24, 2015 by Alshain

[GoSlash27]

Yeah I know what you mean, but using that method you have to adjust them anyway unless you manage to get them the perfect distance in a single orbit of the 'machine'. I've found the target separation I want always ends up offset from the Periapsis so I have to constantly adjust the 'machines' orbit so I hit the target separation where it's orbit intersects the target orbit.

Alshain,

Yeah, I'd imagine that's why you're having trouble with it. When done properly, no adjustment is necessary. The parking orbit is computed to be perfectly resonant with the final orbit, so the correct spacing is assured.

The only other mod I use is KER to get orbital period so they don't drift out of place, but that doesn't help me get them into position. I use that website to quickly calculate the distance I want but I could do that mathematically if I had the inclination to do so. So yeah, I don't really use other mods.

What I mean to say is that if you're setting up a constellation without any mods, then you need to plan out the process mathematically. In that situation, it's much easier to do a one- up launch than it is to do a bunch of individual launches. I can well- imagine that if you are using mods, then the advantage of doing a one- up launch would be reduced or even eliminated.

Best,

-Slashy

Edited October 24, 2015 by GoSlash27

[Stone Blue]

As to the original question of determing distance and/or proper orbit altitude, I'll just drop this here:

http://wiki.kerbalspaceprogram.com/wiki/Tutorial:Satellite_Coverage

I also use and highly recommend the Visual Planner site that Alshain posted.

And, after trying to setup a 24 sat GPS constellation using timed, single launches, rather than motherships, because my GPS sats were oversized and over-purposed, and finding how difficult that is, I now TOTALLY prefer simplifying my sat designs to KISS, single-purpose sats, and using the single launch, "mothership" method GoSlash describes. Resonant orbits makes it so much easier to get the spacing very close...I detach the sat, then use RCS and an Ant engine to tweak and finalize all their orbits once all the sats are in position.

Although, if I DO have to use multiple, "timed" launches, I dont figure the exact timing, launch, and hope I get it close to where its supposed to be. (If you can do the math for this, then yeah, its probably easy...) But No way I could do that, so i use the rendevous-resonant orbit approach...I'll launch, rendezvous with the first sat, then go into the appropriate resonant orbit, then place the next sat...Rinse & Repeat for each following sat...Yes it uses more dV and time, but its the easiest way for a non-mathematical person to do it, I think...

[GoSlash27]

Although, if I DO have to use multiple, "timed" launches, I dont figure the exact timing, launch, and hope I get it close to where its supposed to be. (If you can do the math for this, then yeah, its probably easy...) But No way I could do that, so i use the rendevous-resonant orbit approach...I'll launch, rendezvous with the first sat, then go into the appropriate resonant orbit, then place the next sat...Rinse & Repeat for each following sat...Yes it uses more dV and time, but its the easiest way for a non-mathematical person to do it, I think...

Actually, even using the math it's still difficult. The problem lies in the fact that you don't know precisely how much time it will take to get to orbit from the pad and once you do know that, it's highly variable from one launch to the next.

What I do is first launch into a parking orbit with a fixed period (for example 11 orbits per day, which passes over KSC 10 times per day).

I can then track it and mark the time that it passes over KSC (using a guidance head on KSC grounds) and compute down to the second when to begin the transfer.

The math is a lot hairier than it is for the single launch resonant spacing setup.

Best,

-Slashy

[maltesh]

Alshain,

What I mean to say is that if you're setting up a constellation without any mods, then you need to plan out the process mathematically. In that situation, it's much easier to do a one- up launch than it is to do a bunch of individual launches. I can well- imagine that if you are using mods, then the advantage of doing a one- up launch would be reduced or even eliminated.

Best,

-Slashy

You can use the KSP UI to create your resonant phasing orbits /without/ calculating orbital periods and semimajor axes. You can use the close-approach UI, the fact that the close approach is /always/ shown after the last maneuver node in your future, and the ability to push maneuver nodes multiple orbits into the future with a click, to create a phasing orbit with no calculations necessary.

Wound up throwing together a tutorial here: Phasing Without Calculation: One Weird Trick to Spacing Satellites Using Just the KSP UI

[GoSlash27]

You can use the KSP UI to create your resonant phasing orbits /without/ calculating orbital periods and semimajor axes. You can use the close-approach UI, the fact that the close approach is /always/ shown after the last maneuver node in your future, and the ability to push maneuver nodes multiple orbits into the future with a click, to create a phasing orbit with no calculations necessary.

Wound up throwing together a tutorial here: Phasing Without Calculation: One Weird Trick to Spacing Satellites Using Just the KSP UI

maltesh,

Yeah, I had just looked that over. Pretty freakin' brilliant! That would simplify things immensely. I'll have to try it out.

Best,

-Slashy

[FancyMouse]

multiple launch might be ok for Kerbin orbits (and personally I don't think it's anywhere better than a single launch), but once go further - unless you're practicing interplanetary travel, otherwise it just becomes really tedious.

[Cairol]

The link of Alshains first post is my way to go.

It delivers everything you need to plan and insert your satellites precisely where you want them,

plus you can calculate your needed battery capacity for 100% uptime.

[DCWarHound]

Thanks for all the responses, everyone.I'm still going though the posts, math is not my strong point so i am still googling on how to do some of the calculations posted.

@Alshains

I'll try it out, thanks for the link.

[Stone Blue]

The link of Alshains first post is my way to go.

It delivers everything you need to plan and insert your satellites precisely where you want them,

plus you can calculate your needed battery capacity for 100% uptime.

PLUS I like that it gives you the necessary, corresponding resonant orbits, and it also shows a stable orbit, out to where your sats have 100% coverage (equatorially, at least...??)

[DCWarHound]

Thanks for the help guys,got 4 in perfect orbit.