STEM Robotics 4 All

STEM Resources for all

Cone Grabbers

We designed various cone grabbers at the beginning of the FTC 2022-2023 Season (First Energize), using the main idea of entering the cone from the top, with hooks which then deploy. This general principle is one of the two main ones for the cone manipulation, the other one being from the outside, with a pinching mechanism. At the top of the FTC cone there is a cylinder attached on the inside. Thus, the inner part is not conic, but cylindrical. The cone can thus be grabbed from the inside with hooks which would support it from the lower rim of the inner cylinder. We wanted to explore the hook-inside-from-top idea in depth, since it had the potential to minimize the need for servo and for very precise positioning, by making use of passive-latching and possibly passive release.

Our first working device was a spring-loaded grabber, with two main states: 1. its three hooks-tucked-in, allowing thus to enter the cone, or to release the cone, and 2. the hooks-deployed, when holding onto the cone, for transportation. The grabber has a main platform and several moving parts: the three hooks, pivoting around screws fixing them onto the main core-platform; a skirt-ring with compression springs around pillars, moving up and down; a central spring-loaded piston, to be pushed by a servo, then pressing onto the back of the hooks, to tuck them in; a latch fixed horizontally under the platform, and spring-loaded with a horizontal extension spring. Once the piston is down, the latch prevents it from springing back up. The skirt-ring has a wedge which can press onto the latch when the device is being set inside the cone, to release the piston and thus allow the hooks to deploy. Moreover, when the cone is grabbed, the skirt-ring also presses onto the top rim of the cone, since it is spring-loaded. This helps securing the cone during transport. At the bottom of the page you can find a link to download the SolidWorks files for it.

By clicking on Cone1 you can download a zip archive with the SolidWorks files and the overall assembly.

Notes: when examining the inside of the top of the cone, we noticed the eight ridges attaching the central cylinder to the outside cone. As these ridges start right at the inner rim of the central inside cylinder, they can prevent the hooks from grabbing the rim, if they land in these spots. However, as the number three is not a divisor of the number eight, and as all (hooks and rims) are equally distributed around the circle, at most one hook can ever end up prevented from grabbing, by being blocked by the ridge. Finally, as the three hooks move independently, the other two can still grab, and this is a necessary characteristic of any design with hooks which can be blocked in that way. It is also possible to have hooks or support levers which rotate right below the ridge, since these ridges extend away from the cylinder.

Subsequent variants on this idea moved the hook fulcrum at the bottom and removed some of the numerous springs and screws, making the operation smoother. We then thought that a good improvement could also be to not need a servo at all. Indeed, it is possible, for the main operation of picking up a cone from the ground and placing it onto one of the junction assembly poles, to have a completely passive mechanism. The hooks would be extended when inserted, and then released when touching and slightly pressing against the top of the pole, which is capped with a rubber stopper.

Our first functional such device had a central compression-spring-loaded piston which this time would move the hooks using a double-collar — an outer ring and an inner ring. The inner ring would push the hooks outwards as the piston is lowered, and the outer ring would tuck them back in, when the piston would be raised, from touching the top of the pole. This way the cone would be gravitationally released, simply falling down onto the pole. In this design the hooks do not have springs anymore. To hold the piston up, with the spring compressed, small scoop-like levers would lodge themselves in a niche attached onto the piston, and would release it when the grabber is lowered into the cone. A textile elastic can be threaded through the scoops and the fixed pillars, to have the scoops come back when the piston is pushed up. The compression spring for the piston is to be inserted at the top, right below the rounded head. A small plate holds it, and two small horizontal “pins” hold the plate against the head.

Each hook-lever and each scoop-like lever was 3D-printed separately. We have also 3D-printed a compression-ring for each of them, to affix it on the notch of the shaft of each lever. However, the central piston and the main platform were 3D-printed together. To do that, we first created a Solidworks assembly out of these two parts, carefully aligned them using the ‘mate’ facility (where we can specify that two cylinders would be concentric, that two faces would be some distance apart, etc), and then simply saved the assemble as one (new) part, and saved it further as STL. When slicing it, we inserted the necessary supports, and, after 3D-printing them, we gently removed them. This way we obtained the two parts “together”, sliding one with respect to the other, without needing to put supplementary screws or fixing plates.

By clicking on Cone2 you can download a zip archive with the SolidWorks files and the overall assembly.

A better design would have less moving parts and be sturdier. It turns out we can indeed simplify this much further, while keeping it passive. Instead of having long lever-hooks with fulcrum at the bottom, we can have wedge-hooks rotating about their top corner, and blocked to not go below horizontal. This way the cone can rest onto the hooks, which would be maintained in the horizontal position by individual springs. We can notice that this is actually a simple idea already being widely used in day-to-day life, for doors which can close without having to turn the handle: the sliding block which locks the door has exactly a wedge shape on one side, to allow it to be pushed in when the door is closing. Here, the wedge-hooks are lifted when the device is lowered into the cone, allowing it to slide inside. As it reaches past the rim of the inner cylinder, the wedge-hooks are pulled back by their springs, individually, and start supporting the cone. The release happens as before, with a central piston, which now has only one outer ring, sliding between the device and the inner cylinder, to push the wedges up.

%d bloggers like this: