where would you put your rcs ports on this?

[putnamto]

im trying to put rcs on this with the least amount of rotation as possible, my brain just hurts trying to figure this out. im going to ponder this all night at work tonight.

[bigcalm]

It looks rather asymmetric - how are you planning to push it to its intended destination in the first place?  But perhaps that's really your answer - RCS ports should really be spaced an equal distance away from the centre of mass - so that's the way you should go and you should be able to do this using this existing design plus some cubic octagonal struts and RCS ports on the end.  However, my own preference would be to make it symmetric even though it'll be heavier - and that way putting RCS on it would be a lot easier - any reason you can't replicate the Mk1 crew cabin and fuel tank on the other side too?  Could you do this and just add a decoupler later to get rid of the relevant bits when you no longer need them to keep the centre of mass in a sane place?

 

[linuxgurugamer]

40 minutes ago, putnamto said:

im trying to put rcs on this with the least amount of rotation as possible, my brain just hurts trying to figure this out. im going to ponder this all night at work tonight.

Try using RCSBuildAid, will make it ridicously easy

[Zhetaan]

@putnamto:

The immediate answer is that you can put RCS anywhere, because without monopropellant, they won't work and thus will contribute no torque ... but I'll assume that that's not the issue.

Since it appears that you already have RCS Build Aid installed (I believe it is the icon at the leftmost point on your toolbar), I'm going to assume that your issue is one of actual placement rather than one of finding the centre of mass, which is good, because that makes this a lot easier.

Since the 4-way RCS block's thrusters all lie in a plane, you need at least two to guarantee three dimensions of motion but you need three to get the counterbalance you need to prevent torquing.  Four would be better, but because that will force issues with placement and available space, start with three and work from there.  It is certainly possible to position thrusters three-dimensionally about an asymmetric, irregular mass such that you get zero torque, but the solution space for that is ... not fun, to say the least, and can result in solutions that have you thrusting in odd directions--a problem, definitely, if those landing legs are not purely decorative.  By far, the ideal approach is to find a solution in which the thrusters all lie in a plane and are aligned with the axis of the vessel.

I do not presently have access to KSP or a good modelling program, I'm going to rely on a Mk. I eyeball and guess at the placement.  I believe that the mobile lab has the greatest share of mass, so I will consider that the primary part--meaning that the best solution places most of the thrusters on the lab.

Practically, one thruster will be on the longer radial part; the way this works is to imagine a circle drawn about the centre of mass in the plane horizontal to the ground and to place thrusters at the points where that circle intersects the skin of the vessel.  The centre of mass for the whole structure is below the centre of the lab, naturally, and pulled somewhat towards the radially-attached parts, so that circle (its size is adjustable so long as it intersects the skin of the craft in three places) will erupt and submerge at two places on the far side of the lab and a third somewhere on the larger radial part (and possibly a fourth and fifth on the shorter parts).  So long as the thrusters are correctly spaced on that circle (and the circle is on a horizontal plane), they are all equally spaced from the centre of mass, the torques cancel out, and you won't have to concern yourself about torque from up-and-down translation.

The centre of mass is easily determined, so I won't go into further detail about that, but the remaining issue is lateral translation; your solution needs to be rather precise.  To solve it, figure out a circle that has the thrusters closest to equally-spaced around it.  In other words, the three (or four) intersections need to be at as close to one hundred and twenty degrees (or ninety degrees) as you can make them.  You are stuck with the one on the radial part; the most you'll be able to do is move it in or out from the centre of mass unless you add a cubic octagonal strut or the like--but it should be on the inside of the right angle.  The finesse comes from placement of the thrusters on the far side of the lab from the long radial tube, since those thrusters can be moved from side to side.

For thruster placement, forget symmetry; you have to place them one at a time.  That ought to make sense:  symmetry is only effective on symmetrical craft.  If the craft lacks symmetry, so does your propulsion solution.  The trick is that the solution you want is symmetrical about the centre of mass, even though the vessel is not.  Ideally, you would be able to calculate the correct thrust and limit the thruster nozzles accordingly, but there is no way to throttle individual nozzles.  Single-direction RCS ports provide more flexibility but also potentially more headache.

When you move this vessel, don't forget to activate caps lock while translating, assuming you're going to do it in space--on the ground presents different challenges.  That puts the thrusters into precision mode, which works to throttle asymmetric output a bit to correct for minor deficiencies in placement and further reduces or eliminates problems with torque.

After all that, note that there probably is no in-game way to totally eliminate torque on that vessel while constraining the major translation directions to coincide with the principal axes of the craft, meaning parallel to the axes of the lab (for the landing legs) and the radial tubes (for the docking ports), but it likely is quite possible to reduce it to something manageable.

To conclude, I would place my ports slightly higher than the axis on the long radial tube, between the blue line and the hatch on the lab (at about the bottom of the door seam), and on the other side of the hatch a bit farther away from it than the second thruster is.  Then I'd work from there.  If need be, you can put one thruster on each of the three radial tubes and a fourth on the lab; however, the increased radius both allows you greater precision to reduce torque in placement and extracts greater punishment in the form of a longer lever arm if you err in that placement.  You may be best served by using advanced tweakables to shut off thrusters for rotational adjustments (use reaction wheels for that) and keeping RCS solely for translation, in which case you would want your thrusters as close to the centre of mass as you can get them.

Edited April 26, 2018 by Zhetaan
There's another short section hiding on the other side that I didn't notice at first.

[putnamto]

for those that asked, this is how im getting it to orbit, with a test flight i was able to reach a perfectly circular orbit of 400k ap and 70k pe.



the reason its assymetrical and it has landing legs is because im planning to build a surface station in orbit of the mun, then land it their. this is one of the four cardinal modules that will plug into the station core before landing.

and thank you very much @Zhetaan you have given me plenty of food for thought.

Edited April 26, 2018 by putnamto

[Tyko]

As others have said, it's a LOT easier if your craft is symmetrical. If you have to RCS balance an asymmetrical design here are a couple of tips:

1. Make sure any consumables are as close to CoM as possible. You already have a challenge on your hands, no reason to make it worse by having to worry about consumables levels.
2. Don't use identical thrusters on both sides. RCS ports placed further from the CoM need less thrust to give the same turning force. You can either use tweakscale or right-click and adjust thrust limiting to control how powerful each thruster group is.

Good luck!

[Superfluous J]

Wherever you put them, disabling yaw/pitch/roll on ALL of them will make your life much easier.