Mechanism design end to end
Take an intake or gripper from an envelope sketch to a checked assembly, and iterate it in CAD instead of in aluminium.
By the end you can
- Design a mechanism from its required motion outwards rather than from parts inwards
- Verify travel, game-element clearance, and expansion limits in the assembly
- Use Onshape versions and branches to iterate without losing a working design
Everything in this track comes together here: sketches that hold, parts that relate, mates that describe the real joint, and drawings someone can build from.
Start from the motion, not the parts
The instinct is to open a Part Studio and start drawing a plate. Resist it. Start with what the mechanism has to do:
-
Sketch the game element at the position it is picked up and the position it is scored.
-
Sketch the path between them. This is the motion the mechanism must produce, and it is the actual requirement.
-
Sketch the envelope — the sizing cube at rest, and the expansion limit extended.
-
Only now choose a mechanism type. Four-bar, linear slide, turret. The motion you sketched rules most options out for you.
-
Place the pivot points that produce that motion, then build parts around them.
This sketch is the design
Everything after it is implementation. Teams that skip it end up with a beautifully modelled mechanism that reaches 30 mm short, and no cheap way to find out which dimension to change.
Verify before you cut
Before anything is manufactured, the assembly should answer all of these:
- Does it reach both positions, with the game element modelled in place?
- Is it clear through the whole travel, not only at the ends?
- Does it stay inside the expansion limit at every point?
- Can it be assembled, and can the motor be replaced without removing the mechanism?
- Does the bill of materials match what the team can actually order?
Model the game element properly
Size it from the official field drawings, not from a photo. A gripper designed around a sample that is 4 mm too small in CAD is a gripper that does not close on the real one.
Iterate with versions and branches
This is where Onshape's version control stops being a footnote.
- Version
- A named snapshot you can always return to. Make one whenever something works — before you try the risky change, not after.
- Branch
- A parallel line of development. Two people can try two intake concepts from the same starting point without either destroying the other's work.
- Merge
- Bringing a branch back into the main line once it has won.
The workflow that makes redesign cheap
Version the working design. Branch. Try the aggressive idea. If it works, merge it; if it does not, delete the branch and lose nothing. Teams that work this way try more ideas, because trying one stops being risky.
Check yourself
You are starting the CAD for a new intake. What should the first sketch contain?
Where this goes next
You now have the full path: sketch, part, assembly, drawing, mechanism. The fastest way to get better from here is to CAD a mechanism your team has already built in metal, then compare. The places the model disagrees with the robot are exactly the habits worth fixing.
Finished this lesson?
Mark it complete to track your progress.