The meat of the satellite cluster. Notice the total cost (166.7k). The bottom stage, which provides the lion's share of the dV for orbital insertion, consists of a large SRB (1800) and 4x medium SRBs (2800 for the set). Hilariously, the fins and RCS thrusters cost almost as much as the rest of the bottom stage, costing 600 and 400 each respectively. The middle stage completes the insertion and costs roughly 3000 or so and I've used damn near every drop of fuel in it the past few launches in my sandbox game (After crashing two out of three in my career save and burning a third of my funds in the process, I decided to tweak the design and practice).
It might seem odd that I'm using RCS where one would typically put control surfaces, but even outside the fact that I lack a joystick and so can't do precise maneuvers with control surfaces, control surfaces become damn near useless past 30k elevation, and that's when I try to swing the rocket towards the horizon for the last 20ish seconds of burn. On top of that, they provide *too* much power at the lower elevations, meaning that the rocket constantly seems like it's in danger of moving out of the prograde cone and flipping out. Control surfaces also had a tendency to induce a substantial amount of rolling. So, yea, RCS definitely proved to be the better choice
Also, the SRBs are setup to burn out at roughly the same time, so there's only the need for a single decoupler at the top of the large SRB. Have I said I'm a cheap bastard yet? Heh.
Anyways, enough of that. Time for a selection of my failures when I remembered to grab a screenshot.
This was my second failure on my career save, which was back when I was still using control surfaces. The launch actually went reasonably well, but I got greedy and pushed my luck at the end of the burn and ended up doing a couple flips before the SRBs burned out. I managed to regain control as I was just about out of the atmosphere at this point, but even if I had started burning with all the ions in addition to the 909 immediately, I'm not sure I would've been able to save the cluster as I still had a Pe of -110km when I reentered the atmosphere (Ap was barely above 80k; I didn't exactly have a lot of room to work with). Also notice that I burned all the LFO on the little probe lander as well.
The result of being too aggressive with my gravity turn. I'm actually surprised nothing broke or burned up, although it's not out of the realm of possibility that that result is from some bugs in the mods.
The RCS was useless in saving this launch. It's a great way to add some quasi-thrust vectoring to the build, but once you're more than five degrees off, you're fucked.
Glamor shot of the above. You can see why I'm surprised nothing broke, though I did get the DRE sound effects that typically occur right before failure, so I might've just pushed the rocket to the very edge of its limits.
A change of pace: Here's a version I put together a couple weeks ago as a proof of concept.
The end result with connections turned off. Kerbin's cluttered to all hell as this is the sandbox game that I use to experiment with different builds. Mun's nice and neat, though, heh.
The equatorial orbit has two short and one long relay at 750km (Current version uses an additional long relay and orbits at 400km), high polar has two short relays and the high resolution ScanSat mapper at 750km (The high resolution works best between 750km and 800km), and the low polar has the low resolution and biome mapper at a 250km orbit.