Over our three seasons as a team, we’ve learned (the hard way, as many things are learned) that superior build quality results in a superior performing robot. It’s not just that good building practices result in a better *looking* robot, they result in a longer-lasting, more durable robot. And that translates directly into how your team does in match play.

Here are some tips from my own experience—as well as collected from the VEX Forum—that will move your team toward a better robot.

Rule #1: Think Ahead

Really, this can’t be overstated in building a quality robot. Thinking ahead results in better planning in robot design, and far fewer after-the-fact fixes and add-ons, which generally don’t contribute to a well-built and well-functioning robot.

Think about the step that happens after the step you’re working on now . . . and the one after that (at least).

General Building Tips & Overall Design

  1. Avoid half-assed work and “good enough” fixes (to quote a VEX Forum member)—they will fail in short order and end up wasting more time to redo and fix than if you had done it right the first time.
  2. Design your robot to be no more than 17.5″ in any dimension (17.5″ also happens to be the length of the long c-channel). You will have big headaches later if your robot is designed too close to the 18″ limit (ask me how I know this). Building to 17.5″ gives you leeway later if you need to make adjustments to the structure.
  3. Plan ahead for where your sensors will be placed; sensors take up space—some more than others—and need to have that space built into the structure of the subsystems where they will be used. Attaching the sensors *during* the build process also saves tons of disassembling time compared to adding them later.
  4. Make sure you can access your motors without disassembling anything on the robot.
  5. Make sure you can easily access the cortex (especially the power switch), and that you can see the cortex from where drive team members will be standing; being able to see the cortex lights while waiting for a match to start is a must.
    1. Use a USB extender cable if needed to get your VEXnet key into a good location that will minimize the possibility of disconnection.
  6. Plan ahead for where you will place your batteries; they’re heavy & take up considerable space. Figure out where that weight will be useful on your robot and make sure that you can and change them easily. 
    1. Use battery extension cables to situate them where you want.
    2. Battery extension cables are also important from a reliability standpoint: they save wear-and-tear on the cortex and power expander ports, which in turn reduces the likelihood of battery disconnection when hitting into field elements or other robots.
  7. Document everything in your engineering notebook, including things that failed, and the reasoning behind your design decisions.

Friction

Friction is not your friend, it’s a vampire, sucking off power that could otherwise go toward the operation of your robot. Minimize friction in every way possible.

  1. Use bearing flats wherever an axle goes through a piece of metal.
  2. Support shafts in 2-3 places (but not 1 place or 4+ places).
  3. Fill up empty space on shafts with spacers, but don’t make the spacing *too* tight. Leave about 1-2 washer-widths free so that spacers can spin easily; if things are too tight then the spacers will prevent the axle from turning at full speed.
  4. Make sure there is a plastic/teflon washer in between any 2 moving metal parts.
  5. Use lock (nylock) nuts on components that are supposed to move at a hinge point, such as on a lifting arm; you want the arm to swing freely without the nut falling off.
  6. Make sure gears are perfectly aligned (instead of only meshing, say, 90% of their width).
  7. Pay attention to weight; a heavy robot will not drive as fast as a light one. Use aluminum if your team’s finances allow for it; otherwise, think through which pieces are absolutely necessary and which are not.
  8. Use round idler inserts for gears that don’t need to turn an axle; using square inserts means that some of the energy will be given to turning the axle, instead of making its way through the gear train.

Structure

  1. Everything that SHOULD be symmetric between left & right sides MUST be symmetric between left & right sides, all the way down to the number and placement of screws. (Ditto for front/back.) Put a schematic of bearing flats and spacers used on each axle into the engineering notebook so that you can build the second side (or put stuff back together) without a lot of hassle. Be sure to update the schematic when you modify the robot.
  2. Stuff needs to be straight; pieces that are slightly crooked will not produce a reliable, consistent robot, and can translate that crookedness into, say, bent axles, which makes everything much much worse.
  3. Tighten all nuts with a nut driver. For connections that have the possibility of having shear (side-to-side or twisting) force on them, use several screws on that connection, and make sure that the nuts are super-tight. (See our guest post on nuts & bolts for more information on this topic.)
  4. Every piece should be connected to 2 other pieces.
  5. All superstructure pieces should be connected to 3 other pieces; this makes night-and-day difference for the sturdiness of your robot.
  6. Try not to cantilever anything important.
  7. Superstructure components should be evenly balanced over the center of the robot.
  8. Triangular supports add strength to square angles.
  9. File down sharp edges; you will likely fail inspection with a lot of sharp edges (filing them down also keeps you from scratching yourself every time you work on the robot).
  10. Pop-rivets seem like they should be useful, but they don’t last (handy in quick & dirty prototyping, though).
  11. Attach a lifting arm c-channel to a large gear with an axle running through it, instead of using a lockbar to connect the c-channel directly to the shaft.
  12. Think about the robot’s center of gravity both when folded up and when fully extended—before you get too far into the build process (see Rule #1).
  13. Use high-strength gears (and high-strength chain) as your default; use low-strength gears only when (a) your system is under very low stress, or (b) the real thing won’t fit (though there’s less excuse for this one; see Rule #1). Use low-strength chain . . . never.
  14. Use omni wheels; they are superior to all other VEX wheel options.
    1. The one downside to them is that other robots can push you from the side rather easily.
    2. To remedy this drawback, some teams use 2 omni wheels and 2 regular/friction wheels.
  15. Take a moment to figure out where to place shaft collars on each axle to minimize side-to-side motion of the axle; some locations are better than others.
  16. Once the robot (or a subsystem) is built, follow the “You drop it, you find it” rule. When modifying or improving parts of the robot, if you drop a screw, you have to find *that screw*; you can’t go and get another one. This is one way to ensure that all of the screws and spacers that started out on the mechanism make their way back to the mechanism in the same configuration. 

Wiring

  1. Label wires and motors—it’s a fussy job, but will save you tons of time throughout the course of the build process.
  2. Label wires at many points between motor/sensor and cortex, including on each side of every linkage (motor controller or extension wire), so that when things come apart or are taken apart, they can easily and quickly be put back together the correct way.
  3. Employ good wire management practices—keep things neat, and be mindful of where wires pass near a moving part that could pinch or sever them.

Testing

How do you know if you have succeeded in creating a durable robot? How do you know if you have good build quality?

  1. Do the shake test. Pick it up, tip it to one side and shake it around; does stuff fall off? Tip to the other side, repeat; hold upside down, repeat.
  2. Drive it. A lot. You can’t finish the robot the night before a tournament and conclude that you have a well-built robot; you need to include time in your schedule to drive it and confirm that it is built to last.
    1. Corollary #1: Clean up the field before you drive around, and see if screws or other parts fall off.
    2. Corollary #1a: If stuff does fall off, fix it in such a way that it won’t happen again. (“Blue” Locktite can come in handy when the design becomes finalized.)
  3. Drive it under tournament conditions. Hold a scrimmage with other teams in your area if you can. Otherwise, be sure to run the 15-second autonomous followed by a 1:45 driver-control period (or run your 60-second driver skills pattern over and over and over). Use the VRC Hub app to help simulate tournament conditions.
    1. Keep track of each practice run in your engineering notebook, including mechanical failures; it will be easy to look back through your records and notice whether you have to stop and repair the robot every 4 or 5 runs—or not.
    2. Simulating competition conditions is the only way to know if you will have motor overload/stall (i.e., PTC tripping) issues.
    3. Students generally drive the robot harder in competition (more abrupt changes in direction, harder acceleration, etc.) than during in-the-lab driving. Don’t be unpleasantly surprised by stalling in a tournament—test this out ahead of time, over and over and over.

♦         ♦         ♦

The recently announced changes to the playoff structure for VEX tournaments place a new emphasis on a durable, consistent robot. The tips above can help your team make it through the “sudden death” elimination rounds. (See my previous post for a longer discussion of the new changes.)

Share this post: