Selecting a core

[MPDerksen]

Multiple options to select a core, HECs, OKTO etc.  I have the tree mostly unlocked.  I'm not clear on the difference between them, and how to select between.

Also, even with battery power, I sometimes cannot maneuver without engaging thrust.  Is that simply because I don't have any RCS?  I had a probe with an Ant engine, and couldn't rotate even the little probe.  I reverted and replace it with a spark, then I could orient just fine.  I think I'm missing some of the "rules" for making a probe.  Do I need a reaction wheel?  RCS?  

Last sub-question:  If I want to put 3 relays in orbit, how close do I have to be for them to stay separated?  Example, if I want to put a relay network around Duna, I'm thinking about 200K orbits.  Can this be 198-202K?  Or should everything be tighter than that?  I'm not modded, but would add one if that's the way to be successful.  I ask this part assuming I want to send unmanned probes in the future, but assume I need CommNet active to control the landings on the backside.  (Unless I am wrong about that part).

[FloppyRocket]

The different cores vary in their SAS capability level and amount of built-in battery storage and amount of torque.

When you run out of battery power, the ship can't be maneuvered unless you have RCS or start up an engine that has gimbal capability. If the engine also has electricity generating capacity, that will also recharge the core battery very quickly, which then makes its SAS torque capability work again.

Edited April 2, 2019 by FloppyRocket

[MPDerksen]

2 minutes ago, FloppyRocket said:

The different cores vary in their SAS capability level and amount of built-in battery storage and amount of torque.

When you run out of battery power, the ship can't be maneuvered unless you have RCS or start up an engine that has gimbal capability. If the engine also has electricity generating capacity, that will also recharge the core battery very quickly, which then makes its SAS torque capability work again.

So more to do with my probe design (forgetting batteries) than needing a reaction wheel?  That's not what the wheels are for?

My KSC tracking station is fully upgraded, so a probe should only need a simple "16" antenna from Duna, right?  And when Duna gets between my probe and Kerbin, I just lose all control?  If so, that would be why I want to establish my relay network first.  

Any suggestions on the orbits (and quantity of probes) for my relays would be welcome.  I can visualize just 2, in synchronous orbit in positions that cover the "backside", but 3 in rotation seems actually easier.  Provided they stay separated.

[bewing]

With the default settings for the game, when you lose communications, you lose the ability to steer your craft with WASD. (Only SAS modes can be used to steer.)

No, the Communotron 16 will not reach KSC from Duna with a T3 tracking station. It'll barely reach to Minmus.

For setting up a spaced relay network, don't look at the altitudes --  they don't matter. Look at the total periods of the orbits -- you need to get them very close.

You can land without a CommNet signal. I prefer to have an OKTO2 to do it, so I have "radial out" SAS mode available. But your engine is either Full or Zero thrust without connectivity, so you need to plan for that if your engine is overpowered (reduce your engine thrust while you have signal).

 

[MPDerksen]

9 minutes ago, bewing said:

With the default settings for the game, when you lose communications, you lose the ability to steer your craft with WASD. (Only SAS modes can be used to steer.)

You've mentioned this to me a few times in the past, and I'm not clear if I altered any settings (PC).  The rest of your comments are helpful, as always.

How do I check my orbital period?

Michael

Looking back, it's quite amazing what I've already learned to do:  EVA between ships and/or a station, docking, biome-hopping on Minmus.  Each was overwhelming at first, but this forum (Bewing, Snark, Floppy and others) have all been very helpful and patient.  I appreciate it.

[bewing]

The important setting is in your Advanced tab in your Difficulty Options: "Require Signal for Control". If it's off, then you can always steer using SAS modes (provided you have some electricity). If it's on, then your craft becomes a brick once your CommNet signal strength goes to zero.

One easy way to check your orbital period is to create a maneuver node. Write down the time to the node. Then right click the node  and click the "+1 orbit" button. Then subtract the two times.

Or, take the time to your Ap, and the time to your Pe, subtract, and then multiply by two.

Edited April 2, 2019 by bewing

[MPDerksen]

2 minutes ago, bewing said:

 

One easy way to check your orbital period is to create a maneuver node. Write down the time to the node. Then right click the node  and click the "+1 orbit" button. Then subtract the two times.

Clever. Thanks again

[FloppyRocket]

4 hours ago, MPDerksen said:

So more to do with my probe design (forgetting batteries) than needing a reaction wheel?  That's not what the wheels are for?

My KSC tracking station is fully upgraded, so a probe should only need a simple "16" antenna from Duna, right?  And when Duna gets between my probe and Kerbin, I just lose all control?  If so, that would be why I want to establish my relay network first.  

Any suggestions on the orbits (and quantity of probes) for my relays would be welcome.  I can visualize just 2, in synchronous orbit in positions that cover the "backside", but 3 in rotation seems actually easier.  Provided they stay separated.

The HECS and OCTO core series (and their '2' variants) provide varying amounts of battery capacity and torque - but not as much battery and torque as the pods.

You can add reaction wheels and batteries if more torque or power are needed. 

I usually refer to the handy table in the Wiki to compare different parts.

https://wiki.kerbalspaceprogram.com/wiki/Parts#Pods

I have the commnet features disabled, so don't have any opinions about that.

 

[Capt. Hunt]

10 hours ago, MPDerksen said:

Multiple options to select a core, HECs, OKTO etc.  I have the tree mostly unlocked.  I'm not clear on the difference between them, and how to select between.

Also, even with battery power, I sometimes cannot maneuver without engaging thrust.  Is that simply because I don't have any RCS?  I had a probe with an Ant engine, and couldn't rotate even the little probe.  I reverted and replace it with a spark, then I could orient just fine.  I think I'm missing some of the "rules" for making a probe.  Do I need a reaction wheel?  RCS?  

Last sub-question:  If I want to put 3 relays in orbit, how close do I have to be for them to stay separated?  Example, if I want to put a relay network around Duna, I'm thinking about 200K orbits.  Can this be 198-202K?  Or should everything be tighter than that?  I'm not modded, but would add one if that's the way to be successful.  I ask this part assuming I want to send unmanned probes in the future, but assume I need CommNet active to control the landings on the backside.  (Unless I am wrong about that part).

1. There are slight differences between the probe cores, generally, the higher up the tree they are and the larger they are, the better, but for most applications it will probably come down to which style fits best with the rest of your probe's design.  Some shapes work better in tight spaces, some offer more space to attach things, and some fit best with certain cross-sections.
2. Unless you're planning on doing precision maneuvers, such as docking, you don't really need to use RCS.  Most, but not all, of the probe cores do have some reaction wheel torque, and of course adding more wheels adds more torque.  The reason the Ant didn't work is that not all engines have gimbaling nozzles, the spark could turn you under thrust because it does have one, although gimbals have a limited range of motion, so it's not going to be much help.
3. That depends, if you want them to stay in the same place relative to one another, they should be as close to the same orbit as possible, I'm not talking Kilometers, the orbits should be within meters of each other, and as close to the same inclination as possible too.  Their orbital periods should match as close as possible, or else they will drift out of alignment with each other.   I know there are people who will disagree with me, but the easiest way to ensure precision like that is to use something like Mechjeb or KER.  That said, for mission comm coverage, that's not as important as just making sure at least one relay has line of sight on your lander when you need it.  There are different techniques to achieve this, using as few as a single relay or a whole constellation of them.  On Mars for example, NASA routes all of their probes signals through the Mars Reconnaissance Orbiter, a single satellite, which is in a sun-synchronous polar orbit, the downside of course is that it is not overhead all the time, but that's not a problem for NASA, since the rovers are mostly autonomous, and any instructions can be sent during the comm window, the orbit was mostly chosen for MRO's own mission.  One or two relays in Dunastationary orbit over your planned landing sight could also work, they will always stay in the sky over your landing site, and be high enough that line of sight to Kerbin is only an occasional problem when the planets are aligned just right.  You could probably even cover a whole hemisphere with one relay.  I usually use a swarm approach.  Put enough relays in orbit and it won't matter where they are as long as something has line of sight more often then not.

Edited April 3, 2019 by Capt. Hunt

[Kryxal]

For keeping satellite spacing, the orbital period is MORE important than the exact apoapsis and periapsis, you can afford the intervals to shift a bit over an orbit as long as you come back to where you started.  Also, two satellites about 120 degrees off give very good coverage as long as they're high enough for them to relay to each other, since the CB you're orbiting won't block both at the same time.  It's not 100%, but it's a cheap 80+% probably.

[MPDerksen]

11 hours ago, Capt. Hunt said:

1. One or two relays in Dunastationary orbit over your planned landing sight could also work, they will always stay in the sky over your landing site, and be high enough that line of sight to Kerbin is only an occasional problem when the planets are aligned just right.  You could probably even cover a whole hemisphere with one relay.  

Very thoughtful reply.  Thank you.  One note, however, is that Ike is in the way of positioning Dunastationary satellites, from what I have read.  Also, after bewing's comment about the required antenna,  I checked the Wiki CommNet page and realize I not only need an actual RELAY option, but the -15 if I want to be sure to connect.  This has downstream effects of needing a larger amount of solar, and a bigger base to attach everything to.  

If I use a rocket with 3 satellites on it to drop them off at the requested orbit, how much dV is likely required to circularize?  The one below has 779 dV Vacuum.  Last, do I need the battery on here?  Since the HECs 2 has 1,000 battery, will the solar panels recharge that?  My intention is to plop 3 of these around Duna, more for RP than an specific mission.

11 hours ago, Capt. Hunt said:

1. I know there are people who will disagree with me, but the easiest way to ensure precision like that is to use something like Mechjeb or KER.  

I think it's time to get Mechjeb.....

[MPDerksen]

Electrical:  RA-15 requires 68.6/sec, and the HECs is 3/sec.  2, 1x6 solar arrays generate a mere 3.28/sec, and around Duna on 0.431 of that.  (see me trying to read, understand and learn BEFORE asking?).

Please explain the gap in my math.

[bewing]

The relay antennas only require EC when they are actively transmitting a science report that the current craft generated. Otherwise, they require zero. A HECS2 is 3.0 per minute, so you are a factor of 60 too high on that one. Unless you hibernate it, and then it falls by another factor of 500. And yes, the solar panels recharge all the batteries on the craft, including the probe core. So really, a HECS2 is probably way more probe core than you really need for these little relays -- unless you plan to use them to look for anomalies (then they are pretty good).

 

Edited April 3, 2019 by bewing

[Capt. Hunt]

10 hours ago, Kryxal said:

For keeping satellite spacing, the orbital period is MORE important than the exact apoapsis and periapsis, you can afford the intervals to shift a bit over an orbit as long as you come back to where you started.  Also, two satellites about 120 degrees off give very good coverage as long as they're high enough for them to relay to each other, since the CB you're orbiting won't block both at the same time.  It's not 100%, but it's a cheap 80+% probably.

I know, but matching the orbit is probably the easiest way to match the period of the orbit.  Bottom line, the larger the difference between either the period or the radius, the more they will drift.

7 hours ago, MPDerksen said:

Very thoughtful reply.  Thank you.  One note, however, is that Ike is in the way of positioning Dunastationary satellites, from what I have read.  Also, after bewing's comment about the required antenna,  I checked the Wiki CommNet page and realize I not only need an actual RELAY option, but the -15 if I want to be sure to connect.  This has downstream effects of needing a larger amount of solar, and a bigger base to attach everything to.  

If I use a rocket with 3 satellites on it to drop them off at the requested orbit, how much dV is likely required to circularize?  The one below has 779 dV Vacuum.  Last, do I need the battery on here?  Since the HECs 2 has 1,000 battery, will the solar panels recharge that?  My intention is to plop 3 of these around Duna, more for RP than an specific mission.

I think it's time to get Mechjeb.....

Ike does like to gobble up unsuspecting spacecraft, one of the reasons I like the swarm approach.  I use Scansat, so usually I have a few sats in polar orbits anyway, I sometimes also throw small relays on spent boosters and payload busses to help.  Once I even released a cubesat (literally a QBE core with an RA-2 and a small solar panel) during a Trans-Duna injection burn, so it would settle into a Kerbiocentric orbit between Kerbin and Duna and act as an intermediate relay at least until the payload reached Duna..

In a pinch, the RA-2 will also work at Duna, as long as Kerbin and Duna are on the same side of Kerbol.  If you time your mission right you can make it before the planet gets out of range.

Fortunately, there is no penalty for having a weak signal, as long as it does connect.

 

BTW, a side benefit of mechjeb is it does not require a signal to operate any of the auto-pilot functions, as long as you're comfortable with being hands off.  Think of them as pre-programmed maneuvers.

Edited April 4, 2019 by Capt. Hunt

[MPDerksen]

So the table below, from the Wiki, is what I'm basing my probes on.  I also went into the VAB to confirm the presence/absence of the reaction wheel.  Without one, I need an engine with a gimble.  Only the bottom 4 have that.  I really don't need even Prograde/Retro control, just basic SAS to hold while I burn to circularize.  I really just need a core, a few small solar panels, minimal fuel and engine (an Ant would likely even be just fine) as a base for the RA-15.  

I want to rack up 3-4 of these for delivery. But that part is fairly clear.  I'd love to see a few ideas of what parts you would use to make these.  When I look at KerbalX, it's hard to find basic stuff.  Everyone wants to show off their Tie Fighter, but I rarely find simple, helpful examples.

[Snark]

2 hours ago, MPDerksen said:

Only the bottom 4 have that.

Well, not quite.  The HECS and OKTO do in fact have reaction wheels in them-- they're very low-torque, but it's sufficient for little probes.

[MPDerksen]

Just now, Snark said:

Well, not quite.  The HECS and OKTO do in fact have reaction wheels in them-- they're very low-torque, but it's sufficient for little probes.

So what is the Maneuver column for?  I looked in the Wiki before asking.  Can I stick an RA-15 on these "tiny" cores and not have any issues?

[bewing]

1 minute ago, MPDerksen said:

So what is the Maneuver column for?  I looked in the Wiki before asking.  Can I stick an RA-15 on these "tiny" cores and not have any issues?

One of the SAS modes is called "Maneuver Hold". That column is if the core has that mode. Yes, you can easily stick an RA-15 on one of those tiny cores with no issues.

[Jestersage]

Updated the page to reflect which probe has reaction wheel

[MPDerksen]

14 minutes ago, bewing said:

One of the SAS modes is called "Maneuver Hold". That column is if the core has that mode. Yes, you can easily stick an RA-15 on one of those tiny cores with no issues.

THAT makes much more sense.  Thank you

9 minutes ago, Jestersage said:

Updated the page to reflect which probe has reaction wheel

I see that.  Also see in the table below it that the force of 0.5 and 0.3 for the HECS and OKTO is quite small, as bewing said.

About as cheap, and simple as I could imagine.  Hopefully the dish won't shadow the panels, right?  HECS, Oscar tank and an Ant.  1150dV in Vacuum.  Next I'll figure out how to fit a few in a scaffold for Duna.

[juanml82]

I'm playing with comm net disabled, but the best way to do relays (if ground antennas are on) isn't to put three relays in equatorial orbit. It's to put two relays in very elliptical polar orbits, one going north, the other going south. That way, there is almost always a relay within line of sight, and both relays will have line of sight to the ground stations.

So, you make a probe carrying two relays. In your example, you send it to Duna. You put the probe in a low circular polar orbit and detach both relays. Once a relay is at the North pole, you burn prograde to raise your Ap to the border of the planet's SOI. Then you set an alarm (use Kerbal Alarm Clock) to the time of that relay's Ap. Your second relay remains in a circular orbit. Near the time your first relay will reach Apoiapsis, you set up a burn for the second relay at the South pole, also raising the Ap as high as you can and execute.

That way, both relays will zoom over the planet very quickly, so they'll be occluded by the planet for a very short while, and most of the time both will be in line of sight of the entire Kerbol system.

[Superfluous J]

Here's how I choose a core:

• Q: Are you making a rover and have the Rovemate unlocked?
  • Yes:  Use the Rovemate
  • No: Continue
• Do you have the OKTO2 unlocked?
  • Yes: Use it.
  • No: Continue.
• Do you have the HECS unlocked?
  • Yes: Use it.
  • No: Go do some more science.

IMO the OKTO2 is the best core because it's light, does all the directional holds (maneuver's nice and all but I never really *need* it), and can scan for biomes. HECS is second only because it's the first one you unlock after the essentially useless Stayputnik. Rovemate's got a niche that it fills well.

Apparently the big cores allow a pilot to control a nearby probe. I've never bothered with that. Seems like a lot of work for a small benefit when you could just put a better antenna on the ship with a pilot.

Edited April 4, 2019 by 5thHorseman