New to rovers, and very confused

[Klapaucius]

Hello,

So I am pretty new to KSP, and I have watched a few tutorials on rovers which I am keen to start using. While learning KSP I have been alternating between building my own stuff and using some of the prebuilt rockets and aircraft.  I've landed and returned from the Mun and Minmus and stranded a Kerbal on Moho,  so now I am keen to explore more.  But one thing I do not at all understand:  how do you take a rover, whether built by me or one of the stock ones, and do anything with it?  Once we get into multiple vehicles I am lost.  

 

1. How do I attach it to a rocket?  Do I need to make it a sub assembly?  Do I always need some sort of decoupler?

2. Just to see what I could do, I was trying to get a stock Prospector rover into the cargo bay of the stock Mallard. I cannot for the life of me figure out how to do this.  I opened the Rover, saved it as a subassembly and then opened the Mallard.  I could not place the Rover in the Mallard in any way that did involve fusing it to the plane. Can't you just roll the thing up the ramp and secure it like you would a real jeep in a C5A?  I've pasted some pictures below.

3. Testing the same rover just on the Kerbal runway, I get a message that it has no remote control or manned command modules. But it has two seats for Kerbals, and yet I cannot fill these.  What is required that I am missing?

4. I was also playing with the stock Rover and Skycrane. Once I detach the rover from the skycrane I no longer have any control over the skycrane. Is there any way to retain that?  What if I want to safely land the skycrane or send it back up to a mother ship? 

5. Totally unrelated to the posted topic, but is there anyway to toggle Kerbal Engineer off or just hide the window when you are not using it? It's an amazing mod, but occasionally I'd like to be able to turn it off.  You can see in the pictures how it is kind of in the way at the moment.

Thanks for helping out a newbie 

 

[AccidentsHappen]

What I've found to do if you want to possibly reattach is to put a docking port on your craft that is to become a sub-assembly, and then a random part on top of the docking port. Use the Root tool to make the random part the root vessel and then drag the craft by the docking port into the sub-assembly area. Go into your transport craft and put a docking port in the appropriate spot and orientation and grab your sub-assembly from the menu. You can hold down Alt to make sure it only attaches to nodes and then click it on the the port on the transport craft! You can also do this with decouplers, and if you are sometimes lazy like me you can just stick a bare sub-assembly onto a radial decoupler that is already on your transport craft. As for fitting rovers to rockets, you can use the same process, but when it comes to fairings, it can be a mix of trial and error fitting it under the shell. I would recommend using trusses and beams to orient it without craft-destroying clipping. And remember to use struts! If the sub-assembly is wobbling around, slap some of that space duct-tape on to fix the problem. Also, if you don't want the strut endpoints dirtying your brand new shiny rover, place them on the transport craft. When you decouple or detach, the beams will disappear and the endpoint will stay on the transport craft end.

To answer your 3rd question, if they are external command seats, you need kerbals from another pod or craft to EVA, walk over and occupy them. If you don't want it to be manned, you can always add a probe core. This also answers your 4th question. The stock sky-crane is only controllable whilst you have a probe core attached, and the only one on the craft is in the rover! If you want to control the sky-crane after you detach it, you will need to put a probe core on there. Make sure you bring sufficient power though; probe cores and SAS need power to operate.

As with KE, you can just click the icon in the lower-right hand of your screen to toggle it on-screen.

Hope this helped, and if this didn't make sense, I'll add pictures!

Edited March 26, 2018 by AccidentsHappen

[bewing]

What AccidentsHappen said is correct, but I'll reiterate it again.

1) Yes, you always use a decoupler -- because that's what allows you to detach the rover from the rocket. Typically you attach the decoupler to the base (or interstage node) of a fairing, so the fairing will help the rover get through the atmosphere when you launch. You don't need to make it a sub-assembly. It's easier to build a rover in the SPH, and then use the Merge function in the "open" menu, I think. But there's a trick to doing that, involving choosing the correct root part for the rover.

2) Putting a functional rover into a cargo bay is one of the hardest things you can possibly do. To be able to detach it, and then drive in and out of the bay ... extremely tricky stuff, in need of a thread all of its own.

3) As said above, you have to launch the kerbals separately to get them into seats. You cannot prefill seats. Or use a trick to launch them with the rover. The kerbals can only board the seats when they are on EVA. But that rover won't rove until it has a kerbal in a seat.

4) In the stock version of the game, you can only control one vessel at a time. When the rover separates from the skycrane, you have to choose one or the other.

Edited March 26, 2018 by bewing

[Klapaucius]

1 hour ago, AccidentsHappen said:

The stock sky-crane is only controllable whilst you have a probe core attached, and the only one on the craft is in the rover! If you want to control the sky-crane after you detach it, you will need to put a probe core on there. Make sure you bring sufficient power though; probe cores and SAS need power to operate.

 

1 hour ago, bewing said:

4) In the stock version of the game, you can only control one vessel at a time. When the rover separates from the skycrane, you have to choose one or the other.

Okay, but I can choose, as long as I have a probe core (or create a skycrane with a pilot)?  And I assume I change control via the tracking station or is there a shortcut for that?

[Cpt Kerbalkrunch]

42 minutes ago, Klapaucius said:

 

Okay, but I can choose, as long as I have a probe core (or create a skycrane with a pilot)?  And I assume I change control via the tracking station or is there a shortcut for that?

You can switch ships on the Map screen as well but, when you're within Physics range, you can switch focus quickly on the Flight screen by hitting the bracket keys [ ]. This will let you jump back and forth quickly; which is especially useful when dropping something off with a skycrane.

You already have good answers above; I would just add, when building rovers, I always have a Rovemate probecore onboard (the one that looks like a little white box) and a docking port or Klaw on one or both ends. If you right-click the docking port you want to be the "front" and hit "Control from here", the Rover will move as expected and be much easier to control. I generally right-click the rear wheels and disable steering on them as well. This will make it operate more like a car. I find it easier to control this way. And you can perform tests easily at KSC to see how well it works and, especially, to see if docking ports line up correctly. This will be hugely important if you're going to use your Rover as a fuel truck or to put modules together to build a base.

Edited March 26, 2018 by Cpt Kerbalkrunch

[Zhetaan]

6 hours ago, Cpt Kerbalkrunch said:

[...] especially, to see if docking ports line up correctly [...]

Rover docking port alignment is tricky at best.  Many people choose to use claws for this purpose, instead.  (Never mind the difficulties in suspension of disbelief:  while you certainly can fill a fuel tank after ripping into it with a hydraulic grappler in real life, keeping it full is decidedly more difficult.)

However, you can use a docking port by itself as a sort of decoupler.  This gives you the control possibilities that were mentioned before, but without requiring even more additional parts.  Simply attach the rover by the node on the docking port's mating face, and after delivery, you can undock even though you're not mated to a matching docking port.  Of course, you can't dock there again, for for delivery purposes, it's unnecessary.

For delivery of stock rovers via stock rocket (if you were also asking about that) I typically use three different sorts of delivery system, but the choice, as always, depends on the mission.

If you're sending a tiny Sojourner-type rover, you can make it so that it fits into a 2.5 metre service bay.  Usually, I attach these to the ceiling of the bay with a separator and contrive to have the bay at the base of the rocket (this necessitates a skycrane, radial-engine design, or a parachuting atmospheric lander).  However, I don't usually have enough room for a chair, so these are remote-controlled only.

For more car-sized Curiosity-type rovers, the mass is large enough to affect the balance of the rocket.  For this reason, I usually take two rovers mounted radially.  They may have skycranes for landing or the rocket may have a different system:  I sometimes mount them low, near the landing legs, but have the rover mounted lower still on the decoupler.  That way, decoupling imparts a torque that can turn them to land upright but they don't fall too far.  Always have an engineer to repair the tyres if you try this; 'too far' is entirely subjective.

For giant, lumbering, no-real-analogue-exists-yet resource harvester rovers with the giant tyres and where the Kerbals can pilot from a cab rather than a seat, I usually make the rover into the rocket and incorporate belly-landing engines (that may or may not be mounted on decouplers so they may be shed after landing).  Mind your profile; usually, you'll want to launch these on a lifter stage one size (or more) larger than the nominal diameter of the rover:  a 2.5 metre rover is sent up on a 3.75 metre lifter, for example.  That lets you use a fairing that keeps the rocket relatively uniform in size.  It's also important to remember to balance the rover:  small ore tanks on decouplers can provide useful ballast (and without the decouplers, ore storage if it's a resource harvester).

Edited to add:  Kerbal Engineer's torque values are useful for balancing, but because they only give magnitude, the usefulness is limited:  it can be difficult to tell how to modify the payload mass distribution in order to minimise that torque.  You'll have to move things around to see what you can do.  I don't want to load you down with mods, so I will tell you that RCS Build Aid exists without linking; what is more useful, I think, is for you to ensure that you have as much mass concentrated into compact central structures as possible.  Keep heavy and dense parts to the core of the rocket.  Use the centre-of-thrust indicator and see whether its arrow is in line with the centre of mass.

Mind the effects of staging; your first-stage lifter may have a great deal of torque (because the rocket is longer and the thrust is greater) or not (because the angle between thrust and mass centres is smaller) and yet also be able to overcome that torque because of the power and gimballing of the first-stage engines, but the second-stage engine may not be able to overcome the subsequent torque changes.  There are further tricks that you can use with reaction wheel tweakables and the like, but they are no substitute for good design.

Edited March 26, 2018 by Zhetaan

[Dark Lion]

Honestly, a lot of folks here could tell you better than I could about rovers. I mean, I'm really great at making them crash, but practicality and missions aren't always a part of my game. x] I can tell you about some Kerbal Engineer though. If it's in your way, you can just grab the interface and move it while building. There's even an option to make it minimalist. In flight though, you can customize it for pretty much anything you'd like it to tell you. There's a button in-flight that looks just like the one you turn it on/off with in the SPH/VAB... but there's options there for customizing in-flight telemetry displays. You don't have to use the given groups either. You can change the size, where to put the display, give it a background, and tons of information to use in your custom group(s). I created a custom group to mess with editing it to my personal preferences by just launching a random craft to sit in. Also, uh... when building your crafts, don't just trust Kerbal Engineer's staging or TWR calculations until you've tested them in flight. The torque offset has been invaluably accurate, however. Hope that's helpful.
You may or may not need a 3-star engineer aboard to access the in-flight data available from KER. This doesn't apply to Sandbox games.

Edited March 26, 2018 by Dark Lion

[Klapaucius]

Thanks everyone. You all answered everything!  As a test of principle, I tried again, and success!  I built a VERY basic test lander, launched it on the runway and parked it just off to the side on the grass. Then I launched the Mallard and opened the back door. Voila, I was able to drive the lander right up into it.  Of course, the first time the lander was not held down by anything and got thrown around the cargo bay.

So I added a docking port to the lander, and I built a bulkhead inside the Mallard with a docking port as well.  The lander did fine, despite my terrible landing (you can see the tail is now missing from the plane).  The only issue is I cannot get the docking port to detach now.

 

Lander with docking port

Bulkhead in Mallard

Driving in.

Attached to docking port

 

My landings have room for improvement.  Though I do consider the lack of a fireball to be a step forward.

 

Interior: I cannot get the port to uncouple.

Edited March 26, 2018 by Klapaucius

[HvP]

10 hours ago, Klapaucius said:

Interior: I cannot get the port to uncouple.

After you re-dock only one of the two docked ports will have the option to un-dock. If you are having trouble clicking the right one you might try decoupling by setting up an action group for that function instead.

[Klapaucius]

9 minutes ago, HvP said:

After you re-dock only one of the two docked ports will have the option to un-dock. If you are having trouble clicking the right one you might try decoupling by setting up an action group for that function instead.

I tried an action group after I posted this, but no luck.

[Aeroboi]

5 hours ago, Klapaucius said:

I tried an action group after I posted this, but no luck.

You can set the "decouple" and "undock" action group on both the docking ports also so it is always the right one that activates the un-docking mechanism.

Also, according to the picture it seems the docking port is clipped in the radiator panel. But I cannot entirely make up whether this is true or whether that white ribbon square object on the last picture is a radiator panel in the first place. It kinda looks like it but maybe my eyes are playing tricks on me.
Clipping often bugs the decoupling and parts with attachment nodes front to back which are close to one another like the Jr docking port often attach on the wrong side of the node.

Detach the docking port and then make sure it is attached on the rearest node as you can attach the docking port on the front and on the backside of it even when it faces in the same direction. A pair of docking ports should only barely touch one another. If it fits beyond the outer edge of the hatch by facing into one another then it is the rear node of the docking port (the backside of it) that gets attached to the other one in the pair. In that case you have no docking mechanism at all, just 2 docking ports welded together.

[Dark Lion]

52 minutes ago, Aeroboi said:

But I cannot entirely make up whether this is true or whether that white ribbon square object on the last picture is a radiator panel in the first place. It kinda looks like it but maybe my eyes are playing tricks on me.

That's because it's not a radiator. That's actually a mid-sized structural panel from Making History.

 

6 hours ago, Klapaucius said:

I tried an action group after I posted this, but no luck.

Be sure you've set up said action group to both "undock" and "decouple" -- both, with the same action group, and on both docking ports an see if that helps.

[HvP]

@Klapaucius as @Aeroboi said, be sure that when you assign the action group that you add both docking ports and both "decouple" and "undock" functions to the action group. That should cover all possible combinations to undock the rover.

This assumes that you have been trying to undock the rover after driving it into the cargo bay from a separate launch. If you are only having the problem when launching the craft with both of them already connected then I'd guess that some nodes on the structural panels have been accidentally attached together in the hangar instead of the docking ports themselves.

[Klapaucius]

4 hours ago, Dark Lion said:

Be sure you've set up said action group to both "undock" and "decouple" -- both, with the same action group, and on both docking ports an see if that helps.

I did not realize that I needed both decouple and undock--what is the difference? 

Anyway, that worked as far as decoupling it, but it also blew the rover up in the process.

Edited March 27, 2018 by Klapaucius

[Dark Lion]

1 hour ago, Klapaucius said:

what is the difference?

Decoupling can be used with any sized docking ports when you surface-attach or node-attach them to other parts in order to have them separate via staging, but only once and done.

Undocking is specific to using the various docking ports as paired with their respective twins on the two (or more) crafts you wish to re-dock an indefinite amount of times.
 Or put more simply, decoupling breaks the attachment. Undocking readies the ports for next docking.

1 hour ago, Klapaucius said:

but it also blew the rover up in the process.

I'd follow the advice above and be doubly sure I've attached my docking ports correctly. The docking kraken is an elusive and tempermental beast, but can be won over by keeping your distance . Also, after said explosions, push F3 to see the flight log. Check the damage report on the left side each time and it may or may not reveal the source of the problem. Don't give up d00d, the explosions are a clear sign of engineering improvement. 

Afterthought:
Maybe try to avoid using a bulkhead and attach the Jr docking port directly to the inner, rearmost vertical plane (under the door inside the cargo bay) of the Mk3 Passenger? This would further simplify the design to help isolate the "boom troubles."

Edited March 27, 2018 by Dark Lion
Afterthought

[HvP]

2 hours ago, Klapaucius said:

Anyway, that worked as far as decoupling it, but it also blew the rover up in the process.

You may want to raise the cargo bay's docking port slightly higher if you can.

When you first test the height of the two docking ports inside the building hanger the wheels on the rover will be in their "compressed" state, which makes them a tiny bit shorter. But this won't be the right height for the wheels once you start driving it. After undocking the springs in the wheels extend a little bit. But, when the craft are docked together it will ignore any parts that clip through the same craft, which means that your wheels could end up slightly below the floor of the cargo bay. But as soon as you undock the game will read the collision of the wheels sticking through the floor and explosions are likely.

If it's possible to raise the cargo bay docking port enough and still be able to drive up to connect with it then that should prevent that from happening.

Edited March 27, 2018 by HvP