Remotetech GPS satellite placement made easy

[John FX]

Using the remotetech microsat I have found it is easy to set up a network that will provide very good coverage without many gaps and will not take long to put into orbit.

Say for example you have launched a manned craft into LKO with 8 microsats and you have put yourself into an orbit of about 1000Km and drop all 8 of the microsats. One by one burn them retrograde by 1m/s,2m/s, then 3,5,7,11,13, and 15m/s respectively and they should naturally fall into non resonant orbits which will mean there should always be a sat in the sky to link to.

It`s not the most optimum but it is really easy to set up and it works.

[Kimberly]

That's an interesting method...most people just use constellations relatively fixed. I guess it might be worth trying for someone new to RemoteTech. (They're not GPS satellites, by the way. That's a whole different ballgame.)

[John FX]

That's an interesting method...most people just use constellations relatively fixed. I guess it might be worth trying for someone new to RemoteTech. (They're not GPS satellites, by the way. That's a whole different ballgame.)

Yeah I know but I don`t know what to call the network of satellites you put up for global coverage from KSC. I put up a FSM Spray of these then a pair of uplink stations with huge dishes on that are to be targetted by the rest of the system. I had trouble getting the fixed orbits to be fixed enough so I devised a way of making them work because they are unfixed so to speak.

As I say, not optimum but really easy, really quick and works fine after the first orbit.

[John FX]

After a bit of experimentation I would say that my method is good for the very first comms sats you put up (microsats) as these have a good 1000Dv for orbital maneuvers so you can set up a basic working network with no effort except launch them one after the other and let them wander around the orbit then refine it down to 3-5 sats in the right places later when you want things to work *and* be tidy...

Currently I have 4 sats around kerbin to provide coverage that are geosync and two uplink stations for the rest of the system. I put two on each moon and planet for uplink and three for surface coverage.

[M3Man03]

V.21 made RemoteTech obsolete. (or I think it was .21). There's now a slew of remote control pods that don't require that super hard to use RemoteTech network.

I fought with that damn thing forever and always would lose signal behind a planet at a crucial time or something. Almost made me give up Kerbal.

Get rid of RemoteTech - it's junk.

[Firenexus13]

uhhhhh...... how did 0.21 make it pointless?

[daffy]

I really enjoy RemoteTech, and don't think it's obsolete, at all, on .21. I've tried a couple of methods, and found the most efficient to be a set of 5 launches, parking the first three in geosynchronous orbit around the planet with the 50Mm Sat Dishes. Then the last 2 SATS in a polar orbit (opposite each other, crossing each pole simultaneously, or as close as possible). You get a huge amount of coverage in this set up, and good range. I did manage to finally match two SATS up with the MUN, on Kerbans orbit, one leading the Mun and the second trailing it about 500km each, ahead and behind. I placed the giant 900Gm dish on each of those...(might have that size wrong. The large folding orange dish). Those will get you a signal out pretty far. I guess I would like to start building relay stations, but know it's going to be a project. I like the idea of it though. I recommend RemoteTech.

[DAVIDESCOTT]

Strictly speaking there is a resonance period, its just really long. Its on the order of 2*3*5*7*11*13 orbits.

[PatriotBob]

My go-to method of satellites networks (All math for Kerbin):

Take a ship up with 4+n satellites, so lets say 6.

Circularize your orbit at GSO altitude (2868.750km) @1009.018 m/s. At this point you release the first satellite.

Then burn retrograde and lower your periapsis to 2074.746km. (If my math is correct) This should give you an orbital period of 5 hours. (GSO orbital period is 6 hours. 6 satellites should be 1 hour apart.)

Do a full orbit and circularize at apoapsis to 2868.750km and set velocity to 1009.018 m/s. You should be 1 hour behind your previous satellite.

Drop another satellite and repeat.

You can do this with any number of satellites you're just going to change the transition orbital period/periapsis to adjust.

4 satellites == 4 1/2 hours == 1658.031km

5 satellites == 4 hours 48 min == 1909.807km

372 satellites == 5 hours 59 min 1.93 seconds == 2856.313km

[DAVIDESCOTT]

(If my math is correct)

It is. This is a stolen modified copy of the Celestial Body Parameters spreadsheet. https://docs.google.com/spreadsheet/ccc?key=0AthzlH8eMa0DdFZJSjlMLTA0ZlBtTnhMQWVlc2hDNmc&usp=sharing

On the Physical Parameters sheet columns AC/AD/AE are new put in your phase angle to synchronous orbit in AC2 and the Pe you want will be calculated. In AD put your Pe and AE will spit out your phase angle.

[PatriotBob]

Well I mean that's a mighty fine spreadsheet and all. I just find it a bit overkill. The orbital adjustments can be done in a matter of seconds with any calculator. About the only thing I that makes it difficult is knowing the gravitational parameter for all the celestial bodies.

I find I benefit more by knowing a few equations and using those on the fly than switching to a spreadsheet. (Probably influenced by my game's hatred of Alt+Tab) Being able to work out orbital adjustments, convert deltaV to burn durations and a few other things is a huge benefit if I need to do something in the next 30 seconds.

But maybe I'm just old fashioned.

[legion1202]

whats the math for all this. I got 3 remotetech sats up and they provide perfect coverage. The mun is a different story..

[little square dot]

V.21 made RemoteTech obsolete. (or I think it was .21). There's now a slew of remote control pods that don't require that super hard to use RemoteTech network.

I fought with that damn thing forever and always would lose signal behind a planet at a crucial time or something. Almost made me give up Kerbal.

Get rid of RemoteTech - it's junk.

hahaha... oh the pain that you've caused yourself.

[John FX]

Strictly speaking there is a resonance period, its just really long. Its on the order of 2*3*5*7*11*13 orbits.

It would be a bit longer as I only adjust the orbits by 1-13Dv instead of a large percentage of an orbit (your maths assumes a resonance of half, third etc). It takes 3-6 orbits just for them to spread out enough for full coverage, closer to ten orbits by the time one passes another depending on orbit height.

[DAVIDESCOTT]

"on the order" I'm not disputing it is effective as there is an enormous multiplier in front of that, but it is a resonance. That big multiplier should be something like the time it takes for two orbits with a 1Dv difference to match up (which would depend on the various orbital parameters).

[John FX]

"on the order" I'm not disputing it is effective as there is an enormous multiplier in front of that, but it is a resonance. That big multiplier should be something like the time it takes for two orbits with a 1Dv difference to match up (which would depend on the various orbital parameters).

I`ve just done some back of the pad maths and if I have it right, it would be something like 1000 in-game years for them to align again. Not a precise alignment but enough to stop coverage on the other side of the planet. I think we may have another patch before then that breaks the save...

[Ralathon]

Or just say "screw it, I wanted to practice circularizing and aligning orbits anyway!"

[John FX]

Like I say, this is only a technique for people making their very first Remotetech network so they can get full coverage from a single launch and figure out the niceties later. This is my current setup around Kerbin. I have a single Geosat uplink over KSC, three more to provide ground coverage and two high power interplanetary uplins for the rest of the system. Total planetary and system coverage in just 6 satellites. The high power uplinks are modular and can be upgraded later if I get a lot more interplanetary craft.

[VaporTrail]

I use a set of six microsats as my first low orbit network, and a set of three keosync dished sats as my high orbit network for the Kerbin planetary system.

Note, I use MechJeb, but it is possible to set up these orbits without using the autopilot options, it just takes more manual control.

My Microsat mission design: (this has changed since then... but premise is still valid).

Modified Microsat:

It uses the cubic octagonal truss to connect the microsat to the docking ports, and the Mechjeb unit is placed so it goes along for the ride, so I've got all the MechJeb data for an accurate final orbit, and the autopilot if I want. Since the mass of the MechJeb unit is so small, and the truss is physics-less, the design works without flipping out (literally). You've just got to be sure to decouple from the docking port, rather than the sat.

Basically you get your insertion package into your final desired orbit. (Technically not necessary, more on that later).

- I usually use a 202.5km x 202.5km, 0deg inclination orbit as my basis, or as close as I can get. This is a 40min (2,400sec) orbit of Kerbin.

This ensures contact with the 800km (not the listed 8,000 km) antenna capability of the Microsat. ((Radius of Kerbin + ASL) * 3.14159 <= 800km)

- Then I lower my periapsis until my orbital period is 33 m 20s. (2,400 sec - (2,400 sec/6) = 2,000 sec = 33m 20s)

You could do the same by raising your apoapsis by the same amount, but since you're returning the Kerbal to Kerbin doing so just wastes fuel.

-Then I warp through orbits (checking at 120 seconds before apoapsis) until:

-- Commlink with KSC is established.

-- KSC is ahead of me in my orbit.

-Then I begin deployment of the sats.

Checklist:

-- Deployment package attitude to prograde and stable.

-- Deployed sat antenna.

-- Activated sat engine.

-- Barbecue Deployment package (slow rotation about longitudinal axis).

-- Manually decouple (docking port) activated sat. (Wait till the right time, you want the vector imparted by the spin to be in the same orbital plane as the delivery package).

-- Switch to sat

-- Deploy Panels

-- Circularize orbit (@ apoapsis), making sure orbital period finishes at 40m 0s, and inclination is 0 degrees.

-- Switch back to delivery package and prepare next sat.

-- Warp through an orbit and restart checklist.

After all sats are deployed, return deployment package to Kerbin and prepare to send High orbit Sats. The high orbit sats need at least one dish on the delivery package, and dipole antennae on both the sats and the delivery package. The folding dish (the only one I've felt the need to use, so far) should have enough range to pick up the low orbit sats from keo-sync, and the dipoles prevent annoying loss of contact during deployment.

Deployment of the geosync sats is pretty much the same, except the phasing number (6 in the case of the low orbit sats) is 3.

Formula: Desired orbital period +/- (desired orbital period / phasing number) = insertion orbital period.

Now about that bit about not needing to get into your final orbit. If you know your insertion orbit, you can set that orbit up direct from launch, and save a decent amount of deltaV from not having to circularize after launch. Most of the time though, it's not mission crippling if you circularize first.

My current network (and plans for the rest of the Kerbin, Mun, Minmus system) involve a total of six microsat relays and three geosync relays in kerbin orbit, and six microsat relays each in Munar and Minmal high polar orbit.

That's total Kerbin planetary system link, space and surface to surface, in twenty-one sats and approx 4 launches.

Kerbin

Low orbit through microsat relays out to about 1000 km.

High orbit through Geosync sats (1000 km to interplanetary) with specific dish assignment for missions not going to Mun or Minmus.

Munar orbit/surface

Through Geosync sats -> Munar Polar relays.

Minmal orbit/surface

Through Geosync sats -> Minmal Polar relays.

Interplanetary is going to be interesting. Thinking about putting two comms stations in high Kerbin Polar orbit. Use the really big fixed dishes on stuff really far away, and the big auto-track dishes for anything they can reach... but that's for later.

Edited October 6, 2013 by VaporTrail

[DAVIDESCOTT]

Why not just use module manager to add mechjeb to the parts you want. These days there is a MM patch that allows you to make wildcard additions to all control pods.

[Stone Blue]

I originally had a 4 comsat geosync constellation that worked well. I had added 2 in polar orbits, phased roughly 180°...Never really seemed to need those, though. But then my use was strictly below 1Mm...After playing a couple times, for some reason, the two sats in polar orbit just disappeared out of the game...lol

Anyway...I have since started using Figaro with a fresh game install. I have 13 GPS-com sats in a semi-sync constellation...My question is, is a separate geosync comms constellation redundant if all my GPS-com sats have plenty of comms antennas?

[John FX]

Why not just use module manager to add mechjeb to the parts you want. These days there is a MM patch that allows you to make wildcard additions to all control pods.

There is a mod called remotetech and mechjeb for all (IIRC) which adds RT and MJ to standard pods without changing the originals. Very handy, just put a probe core on and you are set.

Edited October 8, 2013 by John FX

[John FX]

I have noticed another reason to use my method. One that applies to users of all abilities.

I set up a perfect sat system after reading the comments in this thread, all orbits perfect to the resolution of the readout. After doing a Jool mission and a Duna mission I noticed the sats were all out of place by about 70 degrees, causing some dead zones, yet their orbit timings were still perfect to less than a second. Using the method I describe above, they would still not have moved into places where there would be dead zones of no signal.

Robustness is the reason to use this method. Set up your satellites as perfect as you like, warp forward a few years and see how `perfect` their positions remain or set them up `messy` like this and see how perfect your signal coverage remains...