As mentioned in my first post, we are using the RD02 Drive kits on our robot along with 3-7/8″ Banebot Wheels. We’re using these to power our robot from the back – rear wheel drive – and are using a pair of large castor wheels at the front.
After mounting our wheels onto the motor’s shaft, we measured the distance between the base and the surface it was on, which came to about 80mm. We wanted the base of our robot to be level and parallel with the surface, but could not find a suitable ball transfer unit. We had the dilemma of the bolt attached to the ball transfer unit not being long enough on the small units, but the weight being too high for the large units. We settled for a ball transfer unit of the size 24.5mm, which was weighty and the bolt was not long enough, but we figured we would find a way around this.
After brainstorming ideas to make up for the lack of height of the ball transfer units, the best solution we could come up with was to use M8 bolt in conjunction with the ball transfer unit, with a spacer joining the two. We hoped that, once the spacers have been threaded, they would provide a tight fit and should solve the height issue of the ball transfer unit. With this in mind, we ordered some M8 spacers, and also went out and bought a couple of M8 bolts and nuts.
The deliver took a few days, with the castors arriving before the spacers, but when the spacers arrived, we noticed that the shaft diameter was too big, and the bolts slid straight through. It was only then we realised that in order to make the M8 thread on a spacer, we would need a spacer with a 7.5mm diameter. As a temporary fix, we glued the nuts onto either side of the spacers using multipurpose adhesive, and screwed on the ball transfer unit on one end, and the bolt on the other end, with the base in between it and the nut. We hoped that this would be more of a temporary fix, but after a few runs, the spacer would give in, and the would separate from the nuts.
We are going to purchase more spacers, ones that fit the bolts better, so that we can start testing without anymore issues. But for the time being, we’re using the ball transfer units without the spacers, which is causing a slight tilt to the robot. The tilt has affected our compass calibration, but as this is just a temporary issue, we’re not looking to re-calibrate it. In order to re-calibrate the compass, we would need to account for the tilt, for which we would need an accelerometer, which we don’t have.
I had, also, mentioned that we were having issues with our Banebot Wheels’ hub screws in post #3. The replacements for the screws had arrived and were a perfect fit, but after a few runs, the wheels would constantly fall off. On close inspection of the robot, we could see that the weight of the robot, and mainly the battery, has warped the base – both length ways and width ways.
Due to this, the brackets and, by extension, the motors and wheels were angled inwards by about 10 degrees. We believed that this may be causing the wheels to gradually slip off, and might be the source of the inaccuracies we experienced during testing. When performing the compass turns, we noticed that the first 3 turns are close to 90 degrees, but the final 90 degree turn (when trying to complete a full 360 degree rotation) was more than 90 degrees by about 5 to 10 degrees. We thought that this may have been the addition of inaccuracies built up in the previous 3 turns, which may have been caused by the tilt of the wheels.
Due to my boyfriend recently getting a temporary job, we’ve had to reduce the amount of time we can spend on this project, and therefore I won’t be posting updates as often. Apologies in advance but do keep an eye out for a post at the end of this week.