As it's approaching the dog days of summer, this can only mean one thing: summer break is over. It's time to start building robots again.
Like a Lemming being driven to the sea, today I had a strong urge to build a robot. Not just any robot, but one which is new and innovative. Naturally, I wanted to set the stakes higher than before. For your dining pleasure tonight, here's a custom Vex drivetrain concept using a beefier motor and a worm gear.
Since this bad boy likes to suck over an amp of current, I cannot power it solely off the PWM ports - I needed to power it using a speed controller. In this case, it was a spare Victor 883 that I had lying around my house.
I don't know it I'll finish this robot, as all I really wanted to do was test the concept of seamlessly integrating non-Vex-legal parts into Vex robots. This is just a teaser for my next project which I am working on right now, which takes a Vex/non-Vex hybrid idea to the next level.
(Note: These are not VexLABS parts; these were parts that I was able to find in middle-school-age Tech Ed Teacher curriculum catalogs.)
Wowā¦ Iāve never seen anything like this before. A worm gear driving the 32-tooth (?) gear right to the wheel. Hmā¦ how well does it work? Does the worm gear ever slip from the Vex gear?
This was mostly just an idea concept of how to simply and efficiently use a worm-gear drive in a small compact space, although I do have enough motors, motor mounting clips, and worm gears to make a 4WD robot using the worm gears.
The motor here has a fairly high amount of torque for something itās size. Under a moderate load it runs at about 2,500 rpm, which is geared down through the worm gear 36:1 to about 69.4 rpm. Just to see how powerful it was, I geared it down an additional stage (using two 60t and two 12t gears, so as to split the load between them) using a 5:1 reduction. It was practically impossible to stop the output shaft of the stump-pulling second reduction.
At the moment I am working on a much more complicated design that mechanically will calculate the difference between two input shafts. For example if both shafts were rotating CCW at the same RPM, the difference between the two would be zero. But if the two input shafts were moving at different speeds the output shaft would rotate in one direction (CW if shaft A is faster, and CCW is shaft B is faster) at a speed proportional to the difference of the two shafts.
Nice work on the worm gear. Now that it is a few months later, how did it go?
Were you able to make a āSouth Pointing Chariotā?
Was it to make your holonomic drive ādriver centricā?
Links from here might give you some ideas on differentials.
I thought the screwdrive one has some promise due to no requirement for bevel or planetary gears. You might also consider just doing the integration computationally, based on optical wheel encoders on the drive wheels or drag wheels.
Somewhat. I got about 70% done with it, but at that point the number of gears required to mechanically keep it āpointing Southā was taking a toll on the motors and causing the robot to drift badly. This was just meant as a kind of interesting experiment in gearing, and nothing more really.
Was it to make your holonomic drive ādriver centricā?
Thatās still a work in progress; I have the a KoP gyro at my house, and Iām waiting on an order from Futaba for keyed connectors to make a jumper cable to go from FRC-style female leads on their sensors to the Vex-style male leads. Even once I get them, it probably wouldnāt be until a vacation break, like Thanksgiving, when I would actually have enough time to start programming an absolute holonomic drive.
You might also consider just doing the integration computationally, based on optical wheel encoders on the drive wheels or drag wheels.
I could of did that, but I just wanted to play around with differential gearing to create a complex mechanical-computer type of thing, just for fun. If I was serious about finding the exact location of the robot, I would have picked something much simpler.
