Jump to content

RCS on large, asymmetric structures


Recommended Posts

I'm planning a new space station. It will consist of at least six individually launched parts that get docked together in orbit. Each of these parts will come with RCS installed, and the station is ultimately supposed to use RCS for turning and stationkeeping (what little of it you need to do).

The question is: does that even work? Can you stick together random parts that all have enough RCS thrusters for themselves, and then have them work together as a whole? Or is any RCS equipped station destined to be an uncontrollable mess unless you pre-assemble all of it in the VAB and then never change its weight distribution (making it useless for refueling)?

I could of course stick reaction wheels on everything but I kind of want to avoid using magic this time around...

Edited by Streetwind
Link to comment
Share on other sites

The question is: does that even work? Can you stick together random parts that all have enough RCS thrusters for themselves, and then have them work together as a whole? Or is any RCS equipped station destined to be an uncontrollable mess unless you pre-assemble all of it in the VAB and then never change its weight distribution (making it useless for refueling)?

This works just fine. I do it all the time.

RCS has 2 functions: translating and rotating. Translation is for docking, rotation is for changing orientation. Thus, in such a station, balanced RCS for rotation-free translation is only necessary for each module individually as they are docking to the assembly. Once assembled, the station itself isn't going to dock with anything but just rotate as needed by solar panels and pointing docking ports towards incoming ships. You don't need balanced RCS for rotation, so for the station as a whole, it doesn't matter how the individual modules are arranged.

Of course, the station probably won't necessarily require RCS to rotate anyway. Each module will have at least a probe core, probably also an SAS unit, and the juice to power them, for its trip to the station. Thus, the assembly will have scads of torque and the power to run it all automatically, and RCS will merely speed the rotation up.

As for station-keeping, I've never found a need to do that. The thing will stay close enough to where you park it, provided you don't ram it with another ship. But should this become necessary, hit Caps Lock to get into fine controls. This turns on a stock RCS balancer, with which you can translate without (much) rotation, even with wonky asymmetrical stations.

So really, the place to worry about RCS balance is in the design of the individual modules in the configuration they'll be in during the docking operation. So build just that much of the module and then use the RCS Build Aid mod to get it balanced out. The current version of the mod shows the CoM with and without fuel aboard. It's thus quite possible to build the modules so their CoMs don't move at all regardless of fuel (including mono usage). Then you can put the RCS thrusters on to get perfect, rotation-free translation and know that will always be the case.

Link to comment
Share on other sites

(...) You don't need balanced RCS for rotation, so for the station as a whole, it doesn't matter how the individual modules are arranged.

Of course, the station probably won't necessarily require RCS to rotate anyway. Each module will have at least a probe core, probably also an SAS unit, and the juice to power them, for its trip to the station. Thus, the assembly will have scads of torque and the power to run it all automatically, and RCS will merely speed the rotation up.

As for station-keeping, I've never found a need to do that. The thing will stay close enough to where you park it, provided you don't ram it with another ship. But should this become necessary, hit Caps Lock to get into fine controls. This turns on a stock RCS balancer, with which you can translate without (much) rotation, even with wonky asymmetrical stations. (...)

Good news, that.

I tend to decouple and deorbit the probe cores used to transfer the parts. And I worry about stationkeeping because of experiences with a station in Minmus orbit, which naturally comes with a very low orbit and a very low orbital energy/velocity. I noticed that the mere act of the docking collars sucking two crafts towards each other in the final coupling moment was enough to modify the opposite apsis by a two-digit amount of meters (depending on the size of the docked craft). And in an orbit of around 10,000m, that number may have been small but not insignificant over time and continued use. I suppose a station screaming around Kerbin at 2500 m/s would be a lot less affected.

Now, extra questions:

- Will rotating a highly asymmetric object induce translation, in the same way an unbalanced translation induces rotation?

- Am I understanding you right, that switching to precision control will actually make the thrusters dynamically throttle themselves as needed in regards to their position, whereas in normal control mode they always fire 100%?

Link to comment
Share on other sites

...I worry about stationkeeping because of experiences with a station in Minmus orbit....

Oh yeah, that might make a difference. But good news there is, you can actually escape Minmus with RCS so RCS should be able to tweak you back into the desired orbit :).

Now, extra questions:

- Will rotating a highly asymmetric object induce translation, in the same way an unbalanced translation induces rotation?

- Am I understanding you right, that switching to precision control will actually make the thrusters dynamically throttle themselves as needed in regards to their position, whereas in normal control mode they always fire 100%?

In general, rotation will not introduce translation because the object, no matter how asymmetric, rotates about its CoM. Pure rotation doesn't move the CoM, so no translation. HOWEVER, if you've got modules with RCS thrusters pointing every which way, then using RCS to rotate could conceivably result in some net translation forces. So when rotating stations, use torque or balanced (fine control) RCS.

Now, due to KSP's wonky physics, you can created FREE translation from rotation by violating Newton's 3rd Law. The classic example is to put fuel tanks on opposite ends of a long central structure. One tank is full, the other empty. Start the assembly rotating. It will begin rotating around the CoM, which is off towards the end with the full tank. Now, as the assembly rotates, pump fuel from the full tank to the empty tank. This moves the CoM to the other end of the ship, and the ship will now rotate about that point. Because you have moved the CoM, you have moved the ship as a whole. It is therefore possible to move the ship through space by walking the CoM from the heavy to the light end of the ship at the same point in each rotation. Pumping fuel costs no energy and creates no opposing force, so apart from the energy spent to start the rotation originally, this is all free translation :).

And you're correct about the effect of fine controls on RCS. With normal controls, the thrusters all fire 100%. With fine controls, they throttle themselves in an attempt to eliminate asymmetrical thrust. The system is not totally effective but it's pretty good. So, if station-keeping and rotating the station with RCS, I'd always use fine controls, just to be safe. Of course, you're getting way less than 100% thrust this way, so it will be a slow process.

Link to comment
Share on other sites

I've made a number of stations in LKO as well as around the Mun and currently have a long term station in orbit over Minmus. I never put RCS on non mobile stations. Once the station is brought into orbit and parked where I want it the delivery tugs normally detach taking their RCS with them. When new modules are attached to the station, the RCS system used to deliver them stays with the delivery tug which is de-orbited once the module is attached.

While the stations usually have a reaction wheel on the core module, I don't tend to have them on any modules added later. (they leave with the tugs)

I tend to think of engines and RCS systems as a disaster waiting to happen on a station. If they're not there they can't cause problems when they fire unexpectedly.

While I can rotate a station, it's very slow and I don't do it. When I approach to dock, if the port I want is on the far side I drift past the station and kill all my relative velocity then approach the port I want.

(you can park a car with out rotating the parking lot, right? This is similar)

In the unlikely event that I wanted to move the station to a new orbit, I'd dock a tug and drag it. It's easier to pull an unbalanced load than to push it.

I've never noticed any movement in my stations. Perhaps this is because the station mass is far higher than anything I dock with it. My current Minmus operation for instance has a 7 ton lander docking with a 45 ton mobile station. (this is fairly light as far as my stations go.)

Even if undocking gave the lander a push of 1 m/s it would only give the station a push of about .17 m/s. The push from undocking is actually much slower.

Since I don't rotate my station for docking that little micropush is in a random direction depending on where the station is in it's orbit so it's unlikely that any cumulative effect will build up over repeated undockings.

Link to comment
Share on other sites

I noticed that the mere act of the docking collars sucking two crafts towards each other in the final coupling moment was enough to modify the opposite apsis by a two-digit amount of meters

I am of the opinion you could almost sneeze yourself out of Minmus' gravity well.

(you can park a car with out rotating the parking lot, right?)

The things I learn here. NO WONDER everyone is so upset with my driving!!!!

Link to comment
Share on other sites

While I can rotate a station, it's very slow and I don't do it. When I approach to dock, if the port I want is on the far side I drift past the station and kill all my relative velocity then approach the port I want.

(you can park a car with out rotating the parking lot, right? This is similar)

I want my station to be rotate-able so that the command cupola can be made to face Kerbin at any given time. Is it necessary? Absolutely not! Do the Kerbals want to gaze at their homeworld anyway? Absolutely so! ;)

Link to comment
Share on other sites

And you're correct about the effect of fine controls on RCS. With normal controls, the thrusters all fire 100%. With fine controls, they throttle themselves in an attempt to eliminate asymmetrical thrust. The system is not totally effective but it's pretty good.

YOU CAN DO THIS?!??

That's up there with learning that staging can be locked and that manoeuvre nodes can be clicked and dragged.

Link to comment
Share on other sites

I noticed that the mere act of the docking collars sucking two crafts towards each other in the final coupling moment was enough to modify the opposite apsis by a two-digit amount of meters (depending on the size of the docked craft).

I think this is because your altitude is measured in regards to the part you are controlling the ship from, but the orbit itself being calculated from the COM or something like that - but I am leaning very far out of the window here!

- Am I understanding you right, that switching to precision control will actually make the thrusters dynamically throttle themselves as needed in regards to their position, whereas in normal control mode they always fire 100%?

Do they? I always assumed they just started really low and got to 100% if you pressed the key long enough - but never heard of automatically balancing, that would be kind of the holy grail of RCS building! :)

Link to comment
Share on other sites

Do they? I always assumed they just started really low and got to 100% if you pressed the key long enough - but never heard of automatically balancing, that would be kind of the holy grail of RCS building! :)

Give it a try. Make a simple test right with deliberately unbalanced RCS, HyperEdit to space, and translate it with and without fine controls turned on. Don't get too wild with the imbalance, though. This isn't designed to be a cure-all, it's just intended to help when your construction only approximated balance. Be advised that the more imbalanced the thrusters, the more throttling down they do, so the less thrust you get in total.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...