As part of my design for a VEX competition I have a series of gears set up for a decent amount of torque. The gears finally, after several drive gears in the middle, drive a metal bar that on another side of the structure holds an arm that goes up and down based on motor movement. This torque is absolutely critical in order to lift the arm’s weight and any weight that the arm lifts.
However, after a competition with flawless design performance, during improvements and repairs before the next competition I realized that between the drive gear and the arm the bar twisted on its axis to look like a drill bit. I was wondering if there is any way to strengthen the system so it uses only VEX parts (so the design is legal for VEX competitions) yet does not twist like it did. (as a small note, the bar was cut and grinded for aesthetic reasons but only on one end, so under common sense I would think that the ends would possibly be weakened but not the dead center of the bar)
I would post this in the general forum because i think a little creativity is needed to solve this problem. I had a similar problem lifting a heavy arm. I was just using the long metal bar i got from home depot and cut it to length. My solution was tempering. I used a torch to heat the axle until it was red hot then dumped it in water. The axle did not bend and never broke but because of the properties of tempering if it did break it wouldnt bend, like the normal one, it would shatter or break clean through.
That must be a heavy arm to bend your axle. It might work to use the standard worm gear setup from the advanced gear pack for a physically smaller torque gearing, and a shorter axle would probably less likely to bend than a long one. If this doesn’t work, use something such as the Vex rubber tubing to take some of the arm load off of the axle
For our elevation robot we’ve been very conscious of avoiding this particular problem so we have thought about this:
To avoid axle twisting, counterbalance the other end of your arm in some way. We used Vex Latex Tubing, but you could also just put a few wheels or a bunch of metal on the other end.
To avoid axle bending, make the space in between whatever your axle is rotating on closer together (in most cases, move the two ends of your tower closer together). We instead put another support in the middle to reduce strain since moving the tower wasn’t possible.
These are very “within-elevation-rules” methods. You could just use stronger axle.
Another way to take the load off of your axle, if the arm is really heavy, would be to use a liner slide and rack gears from the advanced gear pack to make a liner drive. Basically, the electomechanical equivalant of a hydraulic or pneumatic cylinder.
Of course you would have to re-design your arm, but it would take the load of the axle by replacing your existing gear train. The axle would just become a pivot point. The draw back would be a decrease in the range of motion - maybe only 120 - 150 degrees depending on how your arm is designed.
I am working on an arm for my elevation robot and if I use the idea I just gave you, the arm can go from the floor all the way up to the high goal with no problems. It all depends on your design and where you mount the drive. If your arm is really heavy, you can always use 2 of these liner drives - one on each side.
I wish I had pictures that I could post, but I hope you get the general idea. Just look at construction equipment.
Another way to give more torque to the axle while balancing the force is to put motors on both sides of the axle. You would have to program them to be activated simultaneously but driven in opposite directions. Also, cutting the axle to exactly the right length is important.
*]Can you bolt the arm directly onto your final drive gear? This will make it so NO torque is transfered through the final shaft, it will all be transferred directly from the gear to the arm.
*]Can you use some VEX Latex Tubing or Rubber Bands to reduce the load applied on the geartrain? You mentioned how important it is that the robot lift a certain load, but with a little “passive assistance” the motor (and geartrain) won’t need to do all the lifting by themselves.
A superior list of suggestions – we’ve used them all over the last three years.
A well-designed arm uses tension assist devices (“rubber bands”) to balance the weight of the arm. In a perfect world, the arm would point up when left to freely rotate. The motor would pull the arm down to pick up a load, and then only have to lift the load and not the arm.
If you must dead-lift a heavy weight, a pair of linear sliders combined with rack and pinion gears is the hot ticket. We’ve found this to be a lot more fault-tolerant than a gear train.
The advice that John V-Neun gave about bolting the gear straight to the arm is another good one. We’ve done a bunch of lifting bots and have never twisted an axle shaft by doing this. We are – however – the owners of a large collection of savaged 12-tooth gears collected in the period before we figured out how to design around this weakness.
Good luck. If I get a chance I’ll post a picture of our newest rack-and-pinion lifter.