The main thing to remember with these exotic mounts is to always account for the shift in CG...
Short version: Anything that moves the CG FORWARD will increase stability... anything that moves the CG REARWARD will decrease stability or cause the rocket to go unstable... therefore, when installing these aft-of-the-rocket-looking-forward boom mounted cameras, one MUST re-balance the rocket to account not only for the weight of the camera, but also the distance that weight is mounted aft of the rocket... the moment-arms MUST be the same.
What I mean by that is, say you're mounting a 20 gram camera 50 centimeters behind the rocket's "unmodified" but loaded-in-flight-condition CG... If you can add 20 grams of nose-weight 50 centimeters AHEAD of the same CG point, the CG point will not change with the addition of the camera and the noseweight-- their moment-arms will be equal, and therefore COUNTERBALANCE. If the rocket is short and you can only add noseweight 25 centimeters ahead of the unmodified (no camera booms) CG, then you will have to add TWICE the weight to make up for the HALVING of the distance from the CG, in order to counterbalance properly... that would make 40 grams of noseweight 25 centimeters ahead of the CG to counterbalance a camera weighing 20 grams located 50 centimeters behind the unmodified loaded CG... to keep the CG in the present location (basically, multiply the weight of the camera and boom by the distance behind the present CG location of the rocket, loaded and prepped for flight (less ignitor) and balanced on a ruler to find the loaded, unmodified CG... in this example, 20 gram camera times 50 centimeters equals 1000... so whatever location you add the noseweight to, in this instance, 25 centimeters ahead of the loaded unmodified CG, would require 1000/25= 40 grams of noseweight... )
When adding cameras and counterweights to the nose end of the rocket, the shift in mass results in the CG moving forward, which should increase stability. Adding weight to the AFT end of the rocket moves the CG rearward, which is INHERENTLY DEstabilizing... That's why doing these AFT mounts is much more difficult than doing even elaborate boom mountings on the front... and it's also the reason why MOST cameras are mounted near the FRONT of the body tube, as far forward as possible or practical. Mounting the camera at the CG will have the least effect on movement of the CG point.
ANY time you add a camera AFT of the unmodified (no camera, boom, additional noseweight, etc) CG point, you WILL shift the CG rearward, which is destabilizing. The further aft you mount a given weight, the more destabilizing it will be, simple as that, just the opposite of mounting the weight as far forward as possible will move the CG forward as much as possible.
One should also not neglect the aero-effects of mounting cameras and things out on booms "sticking out in the wind"... While the aero-effects of small diameter dowels, skewers, CF or metallic rods, etc. should be negligible and well within the abilities of the rocket's fins to overcome, flat wooden strips or other materials can act like very long, thin fins, and can cause problems from lift they generate... Remember than the SPAN of a fin generates much more force than the CHORD of a fin-- that is to say, a "fin" (or anything acting like a fin, like a long, thin, flat boom sticking out from the side of the rocket) will produce more force the further out it sticks from the side of the rocket; lengthening the fin along the long-axis of the rocket will generate less increase in (hopefully stabilizing) force than increasing its span, or how far out it sticks from the tube... that's why long, thin, "skinny" fins are more effective than short, long, "fat" fins, strakes, or other such things. Remember too that the forces generated by any such long flat booms will act on the rocket according to not only how strong the aerodynamic force it generates, but it's "leverage arm" or location from the CG... the "moment arm of inertia" as it's called in Rocksim...
While round booms should have virtually no aerodynamic effect from a stability standpoint (other than drag, which can add up obviously, and too much additional drag too far forward CAN act like "shuttlecock" or "plate" stability (drag stability) and reduce the effectiveness or even overpower the stabilizing forces generated by the fins, one should consider the shape and orientation of the camera as well... flat, "plate like" cameras put on booms out away from the body tube can 'catch a lot of wind' and act like a deflected control surface aerodynamically... sort of like a landing flap or aileron, and out on the end of a boom, any force it generates will be multiplied by its leverage arm (the longer the boom, the more the force multiplication!) SO, be sure that the counterweight shape is aerodynamically the same, if at all possible, to minimize any unbalanced aerodynamic forces.
One can minimize nose weight requirements by putting the noseweight into a ball or teardrop shape on the end of a boom itself sticking out the front of the nosecone... this would increase the noseweight's moment arm of inertia, therefore making it more effective at counterbalancing the mass of the camera and boom behind the rocket's unmodified loaded CG. Going back to the original example, the 20 gram camera mounted 50 centimeters behind the unmodified loaded CG (ULCG)... (20x50=1000)... If we put the noseweight on a boom 100 centimeters ahead of the ULCG, we could reduce the mass of the noseweight to 10 grams and still have the same effect on CG (10x100=1000), counterbalancing the camera with half the additional noseweight.
What this shows is that long, skinny rockets with large fins are best suited to these "aft mounted facing forward" rear boom cameras trailing behind the rocket. They have a higher margin of stability to start with, and therefore we don't have to be AS careful about keeping the CG in the same spot without the camera or with it. It also makes the addition of noseweight to the rocket to offset the aft mass of the camera and boom much easier to accomplish with less additional noseweight needed. Basically, don't try this with a short, fat rocket!
One other thing to remember-- adding counterweights aft of the rocket is a VERY bad idea... MORE weight aft of the ULCG is a BAD, BAD idea... keep the mass of the boom and camera to a bare minimum, mount the camera NO FURTHER AFT than ABSOLUTELY NECESSARY, and try to set the camera up so that it will have minimal aerodynamic effect... turn flat, plate-shaped cameras (like the keyfob cams) so that they are flat to the airstream when angled toward the motor nozzle/rocket... do not mount them "top side up" or whatever where they will catch more air and act like a deflected elevator and induce pitch into the rocket-- remember, ANY aerodynamic effect the camera has WILL be MULTIPLIED into the airframe due to the VERY LONG moment arm of inertia behind the CG!!! Even more than the fins, which are mounted much closer to the CG, since they're IN FRONT of the camera!
Mounting the camera closest the CG will minimize any forces generated by aero-effects due to the short moment arm of inertia between the force and the CG... remember that fins and aero effects caused by them act just like levers lifting a weight... the longer the leverage, the less force is required to move the load. The shorter the lever, the more force must be applied to move the load. Remember too that aero-effects can include cameras acting like DRAG BRAKES deployed out away from the rocket, on the end of a LONG arm, giving them more leverage, and therefore multiplying their aerodynamic force. That's why its important to keep the stack as SYMMETRICAL AS POSSIBLE, with the boom and counterweight approximately equal in "wetted area" and ORIENTATION of that counterweight to the airflow around the rocket... so that the forces are balanced. Due to the additional mass of a counterweight FAR BEHIND THE CG, I WOULD NOT advise the use of a counterweight on the opposite side of the rocket on aft-boom-mounted-looking-forward type camera setups... this only COMPOUNDS the need for additional noseweight to offset the aft location of the camera AND its counterweight! Plus, all the extra weight makes the fins that much less effective, because it makes the rocket MUCH heavier, therefore MORE force must be generated by the fins to get that extra tail boom weight, camera, and noseweight moving to correct the flight path of the rocket... Basically, the fins act like they're too small due to all the extra mass they have to push around, or its leverage on the rocket itself (moment arm of inertia from the CG).
This is an interesting topic and provides REALLY cool video, BUT, one has to REALLY stop and think about the effects of the camera setup one is contemplating on the rocket and its CG/CP relationship, and on the mass distribution (CG) of the rocket compared to the "baseline" unmodified loaded CG, and on the aero-effects of the camera and boom on the rocket (which is harder to visualize...)
Later and keep it safe! OL JR