For the prototype, I bashed a Mosquito kit; I narrowed, shortened, and reswept the fins, and added a ringtail at the tips (made, in the case of the prototype, from a paper-towel roll, though my original design calls out BT-60), as well as a pair of wire retaining hooks intended to keep the motor in the rocket, some 40 mm to the rear of its launch position. This movement of the spent motor at ejection would render the rocket completely unstable, allowing it to continue the tumble imparted by venting the ejection charge through the side of the body tube just behind the nose cone.
Part of the theory here is that a tumbler, as opposed to featherweight recovery as used in the Mosquito, will have less tendency to be thrown far off trajectory by the motor ejection, and the rocket will be easier to track as it falls to earth due to the light glinting from the body as it turns. Even better, it'll be less likely to simply disappear in the grass, as Mosquitos often do after streamlining in.
This rocket weighs an estimated 3 grams without motor; that would give a recovery weight of 6 grams with a spent mini-motor casing in place, and a launch weight of 10.6 grams with an A3-4T installed -- a motor which, according to Digitrak, would toss this tiny rocket to over 700 feet (and eject still rising a bit). For the development period, I intend to stick with 1/2A3-2T and possibly 1/2A3-4T motors, which will give altitudes around 400 to 450 feet, and thus enhance the chances of recovery.
There is a possibility of this rocket appearing as an inexpensive kit once development is completed.
The original prototype was lost on its fourth flight on a 1/2A3-2T motor. After ejection, everyone present appears to have lost sight of the rocket, though at least a half dozen people saw something land in the grass a few yards from the launch control table. After almost ten minutes of combing the landing area with a crew of six to eight people, I was forced to conclude that it was the ejected motor casing (the wire retainers never did work right) that was seen landing, and that the rocket had gone elsewhere, unobserved.
I do have three more Mosquito kits awaiting surgery for conversion to Pixels, though; with a small modification to the motor retainer, and an increase in the size of the vent in the body tube, I hope to make the tumble recovery work.