Relay Networks in 1.2

[Geschosskopf]

INTRODUCTION:
The stock communications system in 1.2 is like a cross between RemoteTech and AntennaRange.  It has (optional) control issues similar to RT but the network itself links up pretty much like AR. If you've never played with either of those mods, then hopefully this tutorial will demystify the whole communications network thing.  It's really not that complicated.  If you're used to playing with AR, then you'll find stock pretty much the same except for the new antennae and the different performances of the old ones.  If you're used to playing with RT, then you'll probably welcome the vast reduction (to nearly zero) of the micromanagement, but might have to learn new habits depending on how you did RT.

GENERAL STOCK NETWORK FACTOIDS:
Stuff to understand before you get started.

1.  No Perfect Networks
There is no such thing as a network that allows 100% uninterrupted coverage to every square inch of every planetary surface.  So don't lose any sleep over that.  No matter what you do, something will always get in the way periodically.  The best you can do is make something that works everywhere most of the time, and then time your activities to happen when you've got a link, waiting the minutes to hours when you sometimes don't.

2.  You Don't Need Perfection
You really only need the network in 2 situations:  1) when you're trying to control a ship/probe without a Kerbal pilot aboard (assuming you selected that option); and 2) when you're transmitting science.  The rest of the time, the network doesn't matter.  Thus, you only need to design the network to handle these situations.  This means you need a network to allow landing on the far side of Mun/Minmus, to cover the mid-course tweaks of interplanetary trips, and to land on other planets.  Once you're on the ground, you can wait a bit if necessary to transmit your science and take off again.  That's all you really need, so don't sweat trying to do more unless you just want to.

Also keep in mind the duration of the mission.  Are you ever going back to this planet again, or is your main focus elsewhere and you're just here for some quick science?  There's no point in getting too elaborate for a 1-and-done mission.

3.  Getting Acceptable Coverage
To do this, you have to minimize the impact of the 2 things that break links:  1) celestial bodies in the way; and 2) distance in excess of antenna range.  You can pretty much eliminate the 1st problem with the network design described below.  The distance thing is another matter.  Right now, the biggest stock relay antennae barely work between Kerbin and Jool when they're on the same side of the sun, and only hit Eeloo when it's at or inside Jool's orbit.  You can reduce this problem somewhat by using intermediate relays at Eve, Duna and/or Dres, although certain planetary alignments over time will still be problematic.  And that's just with the stock solar system.  If you use something like the Outer Planets Mod, where the closest new planet is twice as far from the sun as Jool (as in, you get just as bad a link from Jool to Sarnus as from Kerbin to Jool and the distances between the rest get worse from there on out,, the stock communications system just flat won't work much if at all  (EDIT without jacking up the DSN range a LOT).  So for that you'll need a mod.  Surely somebody will make a bigger mod relay antenna to work with OPM, or you can use a ModuleManager patch to change the stats of the stock antennae for longer range.  Thus, this tutorial will just worry about the stock solar system, where distance isn't THAT much of a problem (except for Eeloo) most of the time.  Then it's just a matter of avoiding blocking by celestial bodies, which is relatively easy.

4.  Forget Geostationary Networks
Kerbin now has a number of communications centers scattered all around its surface and your network can link back to any one of them, not just KSC.  IOW, as in AR, your distant ships essentially talk to Kerbin as a whole, not just 1 spot on its surface.  This also means that Kerbin by itself covers the near sides of both Mun and Minmus, plus anything within Kerbin's SOI not blocked by Mun or Minmus.  The only reason anybody ever built a geostationary network before was because in RT, you could only talk to KSC itself, so you needed the geostationary constellation to handle the 50% of the time when KSC was facing away from your ship, which happened even in LKO.  Because in the stock system you can talk to Kerbin regardless of which way KSC is facing, you don't need a geostationary network.  You can build one if you want, but it's really just a waste of time and money.  And FWIW, you can't build a geostationary network at most other planets and moons anyway because the required altitude is beyond their SOIs.  So just ignore geostationary networks as the archaic "back in the day" things they are and move on.

5.  Redundancy is Good
It's a good idea to provide multiple possible link paths between any 2 points.  That way, if one path is blocked, hopefully another will be open.  It's easy to go overboard with this, but it can be done with a fairly minimalist approach.  This tutorial will show how to do that.

 

AVOIDING BLOCKING BY CELESTIAL BODIES:
There are 3 types of blocking by celestial bodies.

1.  Blocking by the Planet/Moon You're At
This happens whether you're in orbit or landed.  This is the most troublesome type of blocking because it will happen the most often (like 30-50% of the time), and you can't do much about it without going to a lot of trouble and expense.  Often, this is more trouble than it's worth because, as mentioned above, it's actually pretty easy to make a simple network that provides coverage when you actually need it.  Normally, there will be some time (minutes to hours, perhaps a few days even at very tiny moonlets) when you have to wait for a satellite to pass over before you have a link, but the wait shouldn't ever be long  so long as to cause a real problem.  Just warp through it.  Going beyond a minimalist approach doesn't really gain you anything that's actually useful in real game terms, although you might want to for purely role-playing purposes.

2.  Blocking by an Intervening Planet/Moon:
Statistically, this happens quite infrequently.  It's also the easiest type of blocking to overcome, to the point where it'll never be a factor.  Just put your relays in highly eccentric polar orbits so they talk across interplanetary distances well above or below Kerbin's ecliptic and you're golden.  That's the method recommended here (detailed below).

3.  Blocking by the Sun:
This is only a problem on interplanetary missions, but it is certain to happen during them at some point, usually between arrival and waiting for the return window.  In purely game terms (ignoring  the role-playing aspect of needing to phone home every day), this is only a problem if you're still wandering around in the system doing science when the sun gets in the way.  And that's usually only an issue at Jool because the system is so large it takes months to explore it all.  Problem is, the sun is too big to talk over/under with your highly elliptical relays, so you have to talk around it.  You do that by creating relays at all planets except Moho.

Eve is especially important because it's far enough from the sun usually to have an LOS from Kerbin to Eve to wherever, and it move faster than Kerbin so will provide these links more often.  When Kerbin is opposite the sun from wherever your ship is, Eve is often placed to bounce a signal around the sun.  This is especially important for communications with Jool .  Anyway, as mentioned above, you need redundancy and multiple possible paths, and Eve is often a key step in that alternate path.  Moho, not so much because it's so close to the sun that most of the time it can't provide a bank shot around it.  Also, it's way expensive to get to Moho compared to Eve.

 

NETWORK DESIGN:
OK, so that's the underlying principles and realistic objectives.  Now, how do you accomplish all this?  Finally, we get into the nuts and bolts of the network.

1.  Relay System:
The basic unit of the system is a set of 2 long-range relay satellites in highly elliptical polar orbits around the central planet.  One goes up, the other goes down, and they're 180^ out of phase with each other, so that when one of them is at Pe, the other is at Ap.  Because their orbits are highly elliptical (like 2000m/s worth above low orbit), the relay satellites will spend the bulk of their time high enough above or below the ecliptic to be able to talk over or under any intervening body except the sun.  Putting a pair like this at each planet (except Moho) will then provide a path around the sun to any other planet ALMOST all the time.

These relays also cover nearly all the surface of the central planet most of the time.  One will get most of the northern hemisphere, the other most of the southern.  There will be a blind spot centered on the equator on the opposite side from where the satellites are (both orbit in the same direction so will both be on the same side of the polar axis most of the time), but central planets all rotate fast enough that a ship on the ground in the blind area will rotate into coverage soon enough.  Every week or 2, there will be a few minutes when both satellites are over the same pole, one at Ap and the other at Pe, so that the whole opposite hemisphere is blind, but this won't last long because the one at Pe will be moving so fast, so don't sweat it.

For most of the time, these relays will also cover much of any moons of the central planet.   There will be a gap on the far side, however, wider than on the central planet, which you fill with other component of the network, the moonsat.

The relay satellites should always have the biggest relay antennae you have.  This is so they can provide alternate paths around the sun in as many ways as possible.  As you unlock the bigger antennae, you might have to replace some of your early relays.  Kerbin should definitely always have the biggest available.

2.  Moonsats
For each moon of the central planet, you'll need something in equatorial orbit.  The purpose of a moonsat is to cover the gap on the far side of the moon left by the relay sats.  The "moonsat" MAY be a dedicated commsat if you're doing something long-term on the far side, but for a 1-and-done mission, you can do things Apollo-style so that the ship left in orbit serves fills this role.

The moonsat should orbit at a conveniently low orbit so that it passes over a ship landed on the far side often enough that blackouts aren't a bother, and for long enough that it can land while still in contact.  But it also must be high enough that the relays can see over/under the moon to it most of the time.  If being high enough for this makes the moonsat's orbital period (and thus blackouts for farside landers) longer than desired, you can use multiple moonsats spaced more or less evenly around the moon's equator.  But most of the time you'll only need 1 for gameplay purposes.

Munsats need an antenna big enough to reach the local relays at their Aps.

3.  Modular System
So, putting this together, for each planet, you have 2 relays satellites in highly elliptical polar orbits and for each moon you care about, you have (at least) 1 satellite in a relatively low equatorial orbit.  It looks like this:

 

CREATING THE NETWORK
Setting up this sort of network is pretty simple.  At Kerbin, you can launch each relay on a separate rocket but everywhere else, it's best to send them both out on the same carrier vehicle.  If you want multiple moonsats somewhere, they should always go out on 1 ship.  Using 1 ship to carry multiple satellites facilitates spacing them out evenly.  Thus, this description will always use single ships even at Kerbin.  The mechanics go like this:

1.  The Kerbin Relays
These have to be set up first because they're the essential link from other planets and the farside of Mun/Minus.  Design a probe with the biggest relay antenna you have, enough juice to run it, and about 2000m/s dV in its own little tank.  Save this as a subassembly.  Then start the carrier vehicle.  Make it so you can mount 2 of these relay probes under the same fairing and then add the subassembly probes.  Launch into a polar orbit of about 150km.  This will be the Pe altitude of the relays, high enough to avoid stuff in 80-100km parking orbits but low enough to whip by Pe in no time.  Detach one of the probes and switch to it.  Plot a maneuver node for when the probe is directly over one of Kerbin's poles, burning all or most of the probe's fuel to reach an Ap out near the edge of Kerbin's SOI, then do this burn.  You should have a few hundred m/s left in the tank to tweak the orbit later if needed and to deorbit the relay when it needs replacing.

Then you wait about a week for the 1st relay to reach its Ap.  When it's starting to get close, switch back tot the carrier vehicle, detach the 2nd relay, and switch to it.  Plot a similar burn for it, but located above the other pole so this relay will go in the opposite direction as the 1st.  Adjust the number of orbits into the future when you actually do this burn as needed until it's as close as possible to when the 1st relay reaches Ap.  It doesn't have to be exact, just close enough.  Do this burn, then switch back to the carrier vehicle and de-orbit it.

NOTE:  It could be that the 1st relay won't have its own link back to Kerbin's surface when it's time for it to burn.  This could require putting a "0th" relay in equatorial orbit first, if nothing you already have in Kerbin orbit can do this.

2.  Munsats and Minmussats
Until you have the Kerbin relays up, you'll need at least 2, unevenly spaced (or 3 evenly spaced) commsats  in Munar orbit to talk to the farside.  Minmus isn't tidelocked but rotates slowly, and orbital velocity is pretty slow, too, so you might want the same there to avoid long blackouts.  By unevenly spaced, I mean 120^ apart, like a formation of 3 with a missing satellite.  Having 2 spaced like this will give farside coverage 2/3 of the time---3 will give full-time coverage.  But remember, you'll be putting relays up, at which point you only need 1 moonsat to give farside coverage 50% of the time, so don't spend money you don't need to.  The basic technique is the same whether you drop 2 or 3 moonsats off the same carrier vehicle.  The basic method for spacing multiple moonsats is as follows:

First, decide on the altitude you want the moonsat to be so it can usually talk over/under the moon to one of the relays.  Use math to determine what the orbital period is for a circular orbit at that altitude at that moon.  Send the carrier vehicle to the moon and capture into an elliptical orbit with this altitude for the Pe.  As for its Ap, again use the math to determine what altitude would give the carrier vehicle in an elliptical orbit an orbital period 1.33 (or some whole multiple of that number) times as long as for the circular orbit at Pe altitude.  The idea is, every time the carrier vehicle reaches Pe, it will be 1/3 of a circular orbit behind any moonsat previously released.  The carrier vehicle thus makes 1 elliptical orbit for every moonsat carried.  Each time it reaches Ap, it releases a probe.  You switch to the probe and plot a burn to circularize its orbit at its Pe.  Tweak the probe's orbit as needed with RCS to fine-tune its orbital period to be within 1 second on the calculated circular period.  Once all the moonsats are in position, de-orbit the carrier vehicle.

NOTE:  Generally, you'll need a mod like KER or MJ that will give you a display of your orbital period.  As mentioned, using a tiny amount of RCS prograde or retrograde as required will allow for very fine control of your orbital period.

3.  Expanding the Network
Repeat the above process with the relays for every other planet.  Send at least 1 moonsat to every moon you expect to visit regularly, but those can wait until you actually send out such expeditions.  The relays are the important thing, to be able to talk around the sun and to give a better connection out to Jool, and should be set up as soon as you have the antenna to make them worthwhile.

NOTE:  Relays require a lot of EC/sec, but only when you're actually using them.  Solar power is practically useless at Jool for any purpose other than slowly recharging batteries between widely separated uses.  RTGs aren't any more powerful and are quite expensive, while fuel cells run out of fuel.  Thus, relays for Jool, Eeloo, and even Dres should have rather huge battery capacity to cover their periods of infrequent activity.  It might also be a good idea to have a fuel cell for emergency instances of prolonged or frequent use, remembering to turn them off as soon as possible afterwards.

Edited September 9, 2019 by Geschosskopf

[JohnnyPanzer]

Great guide! My only suggestion would be to add a note that when playing with the extra DSN stations off (I believe more players than expected go for this option), three equatorial relay sats around Kerbin will be needed to avoid annoying blackouts at the very core of the network. They only need large enough antennas to communicate with the two polar sats.

Edited October 5, 2016 by JohnnyPanzer

[JAFO]

Great guide, and very helpful.. one minor point, however..

 

On 9/30/2016 at 5:07 PM, Geschosskopf said:

Regardless of where you come from, however, you can't avoid a network now

 

This is just plain wrong. Of course you can avoid networks. All it takes is one click in the settings, and you won't need to bother with them at all.

Edited October 10, 2016 by JAFO

[Kerbart]

You don't need elliptical orbits. Two high altitude polar orbits at 90° will do the trick. As long as they're at 90° angles, the laws of geometry dictate that if one of them is blocked, the other one is visible. With all the tracking stations on Kerbin you don't even need the equatorial one.

On a side note, you don't geostationary satellites in RT either. A ring of three of four equatorial satellites (evenly spaced) will do the trick. One of them will always be over the horizon of the KSC.

Aside from that, thank you for writing the manual. Well done!

[SchwinnTropius]

This is good for a start of a network. I haven't tinkered yet with the networks side, but this will come in handy.

[John Cochran]

Nice guide, however you may wish to mention that when you're using a single carrier to insert multiple relays equally spaced, it's perfectly valid to have the carrier craft with an orbital period of (1-1/N) time the desired period of the relays or a period of (1+1/N). The (1-1/N) is for when the carrier takes a faster, lower orbit and inserts each relay at apoapsis. The (1+1/N) period is for inserting the relays at periapsis instead. Nice thing is the the circularization burn the relays have to do is lower if they're dropped off at periapsis with the carrier taking the higher orbit. For instance, dropping off three relays into Munar orbit at 250km altitude can be done with the carrier having an orbit with a periapsis/apoapsis of 36,829km/250km with insertion happening at apoapsis and a prograde burn of 64.5 m/s to circularize. It can also be done with a periapsis/apoapsis of 250km/440.272km with insertion happening at periapsis and a retrograde burn of 31.9 m/s to circularize. In pretty much every case you can imagine, it's better to insert at periapsis to both reduce the circularization burn deltaV and to prevent inadvertent lithobraking. About the only case I can think of where insertion at apoapsis is when the desired orbit is close to the SOI of the body being orbited. In that single case, periapsis insertion would result in the carrier vehicle going outside the SOI so it couldn't perform its job, so apoapsis insertion is the only feasible choice.

 

 

[Blaarkies]

Anyone tried a ground based link up to North Pole yet? 
I disabled the ground stations and the occlusion percentages(so that relays links cannot pass through half the planet), and I found that dropping a simple probe on the mountains just east of KSC helps out a lot. LKO craft get signal about 30 seconds earlier before they switch over to a direct KSC connection.

I we could build a surface relay up to the North Pole, and place the biggest baddest multi dish comm array there, we should have the core/Kerbin connections sorted. Even the Mun will only get in the way once every 100 orbits...to be honest, I just want to build a Square Kilometre Array on ice 

[Geschosskopf]

On 10/5/2016 at 4:12 AM, JohnnyPanzer said:

Great guide! My only suggestion would be to add a note that when playing with the extra DSN stations off (I believe more players than expected go for this option), three equatorial relay sats around Kerbin will be needed to avoid annoying blackouts at the very core of the network. They only need large enough antennas to communicate with the two polar sats.

Well, sure, but that begs the question....   If Kerbals can colonize Laythe, why can't they built a ground relay on the far side of Kerbin?

 

On 10/9/2016 at 8:19 PM, JAFO said:

This is just plain wrong. Of course you can avoid networks. All it takes is one click in the settings, and you won't need to bother with them at all.

Which is a great thing.

 

On 10/9/2016 at 9:30 PM, Kerbart said:

You don't need elliptical orbits. Two high altitude polar orbits at 90° will do the trick. As long as they're at 90° angles, the laws of geometry dictate that if one of them is blocked, the other one is visible. With all the tracking stations on Kerbin you don't even need the equatorial one.

True, but the elliptical orbits mean your relays spend way more time not blocked than blocked, and their periods are so long that being off  a bit on matching them up makes a lot less difference to the longevity of the system.

 

On 10/13/2016 at 8:01 PM, John Cochran said:

Nice guide, however you may wish to mention that when you're using a single carrier to insert multiple relays equally spaced, it's perfectly valid to have the carrier craft with an orbital period of (1-1/N) time the desired period of the relays or a period of (1+1/N). The (1-1/N) is for when the carrier takes a faster, lower orbit and inserts each relay at apoapsis. The (1+1/N) period is for inserting the relays at periapsis instead. Nice thing is the the circularization burn the relays have to do is lower if they're dropped off at periapsis with the carrier taking the higher orbit. For instance, dropping off three relays into Munar orbit at 250km altitude can be done with the carrier having an orbit with a periapsis/apoapsis of 36,829km/250km with insertion happening at apoapsis and a prograde burn of 64.5 m/s to circularize. It can also be done with a periapsis/apoapsis of 250km/440.272km with insertion happening at periapsis and a retrograde burn of 31.9 m/s to circularize. In pretty much every case you can imagine, it's better to insert at periapsis to both reduce the circularization burn deltaV and to prevent inadvertent lithobraking. About the only case I can think of where insertion at apoapsis is when the desired orbit is close to the SOI of the body being orbited. In that single case, periapsis insertion would result in the carrier vehicle going outside the SOI so it couldn't perform its job, so apoapsis insertion is the only feasible choice.

Chacun à son goût.  Me, I figure that it costs more money to have the carrier circularize than the probes, because the carrier has so much more mass when it enters the SOI of the target planet.  It's got all the probes and their fuel, plus its own fuel aboard.  The lower the Ap the carrier captures into, the more fuel it needs for that burn, which means the initial lifter has to be bigger and pricier.  At the same time, simple probes usually have way more dV than you need and are still cheap.  So I'd rather skimp on the carrier's fuel, giving it just enough to capture into an elliptical orbit without bringing its Ap way down, then use the cheap fuel of the probes to circularize them.

Edited October 20, 2016 by Geschosskopf

[The Spaceanator123]

Hey could you help me? My "Display Relay" button and all that stuff don't appear and I don't know how to fix it. (I've tried reinstalling)

[xendelaar]

Very useful tutorial! I've got just one noobish perhaps slightly off topic question...

What's the difference between a direct antenna a relay antenna? 

 

[Wraith977]

13 minutes ago, xendelaar said:

What's the difference between a direct antenna a relay antenna? 

 

Direct antennas are for transmitting data from a craft back to the KSC and will not relay data from other direct antennas.

relay antennas cannot directly transmit data but form a path for direct antennas to transmit across.

Eg. Direct antenna -> Relay antenna -> Relay antenna -> KSC

[Vorg]

Maybe wrong, haven't gotten far enough to check, but I put up a sat with 2 HG-5's and 2 16's using I think it was the HECS (first core after the ball) and once in orbit, The game marks it as a relay sat. Are you sure you need relay antennas to make a control link sat? Or just more then 1 antenna and the right core?

[storm_soldier2377]

Still don't know what the difference between the antennas is, what their ranges are, or when to use which.

[Tarheel1999]

On 10/30/2016 at 5:19 PM, Vorg said:

Maybe wrong, haven't gotten far enough to check, but I put up a sat with 2 HG-5's and 2 16's using I think it was the HECS (first core after the ball) and once in orbit, The game marks it as a relay sat. Are you sure you need relay antennas to make a control link sat? Or just more then 1 antenna and the right core?

Isn't the HG-5 a relay? I believe the tool tip says it's both a direct and relay. 

[JacktoseIntolerant]

Aha, this is the first logical guide I've found! Thank you @Geschosskopf!

Wouldn't orbiting the polar sats in opposite directions improve their coverage gaps?

[Geschosskopf]

7 hours ago, JacktoseIntolerant said:

Aha, this is the first logical guide I've found! Thank you @Geschosskopf!

Wouldn't orbiting the polar sats in opposite directions improve their coverage gaps?

Thanks.

If you put the polar relays orbiting opposite directions, then you have more trouble getting them there.  Either you have to send them out on separate rockets, which increases the management burden of the expedition flotilla, or you have to give their single carrier enough dV to change directions, which makes it big and expensive.

Also, if the relays are going in opposite directions, you end up with 2 blind spots on the central planet instead of 1, so you'd have more frequent blackouts in the equatorial region than if they were both going the same direction.

[Vinhero100]

On 11/5/2016 at 8:06 AM, Tarheel1999 said:

Isn't the HG-5 a relay? I believe the tool tip says it's both a direct and relay. 

the HG-55 is only direct. I'm looking at it right now

 

 

[Wraith977]

6 minutes ago, Vinhero100 said:

the HG-55 is only direct. I'm looking at it right now

From the HG-5 cfg and in game - "description = A short range dual purpose communications antenna that can handle either direct communications or short range relays."

I have also tested in game and it seems to do both...

[Nicias]

On 10/20/2016 at 0:19 AM, Geschosskopf said:

Chacun à son goût.  Me, I figure that it costs more money to have the carrier circularize than the probes, because the carrier has so much more mass when it enters the SOI of the target planet.  It's got all the probes and their fuel, plus its own fuel aboard.  The lower the Ap the carrier captures into, the more fuel it needs for that burn, which means the initial lifter has to be bigger and pricier.  At the same time, simple probes usually have way more dV than you need and are still cheap.  So I'd rather skimp on the carrier's fuel, giving it just enough to capture into an elliptical orbit without bringing its Ap way down, then use the cheap fuel of the probes to circularize them.

You never have to circularize the carrier. This is what I've been doing, If I want to put my satelites in say a 250km orbit, I enter the SOI, adjust my Pa to be be 250, then adjust the Ap to 440km (or whatever the exact value is), and fix my inclination, then drop and circularize the probes on three laps. If I have more probes to drop off from the carrier, I leave without ever circularizing. 

[Geschosskopf]

1 hour ago, Nicias said:

You never have to circularize the carrier. This is what I've been doing, If I want to put my satelites in say a 250km orbit, I enter the SOI, adjust my Pa to be be 250, then adjust the Ap to 440km (or whatever the exact value is), and fix my inclination, then drop and circularize the probes on three laps. If I have more probes to drop off from the carrier, I leave without ever circularizing. 

Yeah, I never circularize the carrier, either.  In fact, your description above is my preferred method, leaving the carrier's Ap above the desired altitude of the moonsats.  This is because when the carrier arrives at the planet, it has no Ap at all until it captures, then the Ap starts out at the edge of the SOI and comes down lower the longer you burn at Pe.  Thus, leaving the Ap above the moonsat altitude requires less of a burn than bringing it down below the moonsat altitude.  And you want to minimize carrier burns because it's big and heavy so needs more fuel for a given dV burn, which makes it even bigger, which makes the initial lifter that much bigger, too.  I find this preferable to the method proposed by @John Cochran, to which I was replying.  Both work, but leaving the carrier higher is cheaper overall.

Edited November 8, 2016 by Geschosskopf

[Tarheel1999]

19 hours ago, Vinhero100 said:

the HG-55 is only direct. I'm looking at it right now

 

 

The HG-5 is not the same thing as the HG-55. 

[Vinhero100]

20 hours ago, Wraith977 said:

From the HG-5 cfg and in game - "description = A short range dual purpose communications antenna that can handle either direct communications or short range relays."

I have also tested in game and it seems to do both...

ohh right I was looking at the HG-55 the antenna dish that was in ksp before 1.2

[Blaarkies]

On 10/30/2016 at 3:36 PM, Wraith977 said:

relay antennas cannot directly transmit data but form a path for direct antennas to transmit across.

I am not sure about data, but they can definitely connect your craft directly to the KSC. A small probe with only a relay dish does connect with KSC without any hops, so I believe that transmission would work as well...but i have not tested that, so i wont argue about it yet

[Wraith977]

38 minutes ago, Blaarkies said:

I am not sure about data, but they can definitely connect your craft directly to the KSC. A small probe with only a relay dish does connect with KSC without any hops, so I believe that transmission would work as well...but i have not tested that, so i wont argue about it yet

I just checked and it seems as though you can actually transmit science data via a relay dish . That being said it's hardly the best practice, the relay dishes weigh 3-6 times as much as their direct antenna equivalents and have quite a bit less bandwidth (which raises the question of whether the bandwidth is capped by the antenna at the origin point or the relay dish along the path with the lowest bandwidth...)

[Geschosskopf]

On 11/14/2016 at 11:34 AM, Wraith977 said:

I just checked and it seems as though you can actually transmit science data via a relay dish . That being said it's hardly the best practice, the relay dishes weigh 3-6 times as much as their direct antenna equivalents and have quite a bit less bandwidth (which raises the question of whether the bandwidth is capped by the antenna at the origin point or the relay dish along the path with the lowest bandwidth...)

All antennae do both things:  provide control and transmit science.  Relay antennae add a buff to transmitted science (provided the transmitted value is less than 100%).  However, I'm not sure how that works.  Does the ship transmitting the science need to have the relay antenna, or can the relay antenna be anywhere along the line between the ship and Kerbin?

I think it more likely that the ship initiating the transmission needs the relay antenna, and that this has to have a direct link to Kerbin.  If that's true, there's a reason to use relay antenna on your 1-way science probes.  

Edited November 16, 2016 by Geschosskopf