Week 6: Race to The Wall

This week was the week we did the first important challenge for our robot, it would determine our place on the leader board and who we would be fighting first in the tournament. But we still had a lot of work to do this week as the bot wasn’t built yet, we wanted to start building the previous week but half the team were sick and unable to come into the workshop on the Wednesday so we left it until the next Wednesday as that was the only time we had access to the workshop. Before going into the workshop we didn’t know what materials and equipment was available so we were limited to designing the bot beforehand. When we got to the workshop we seen that we were fairly limited to the equipment, there was hacksaws, hand drills and a guillotine so we had to make do with what we got.
So for constructing the bot, David and Nathan took charge as they had the most experience in metalworking, and me and Sean would throw in ideas and help were ever we could. When building the bot the whole design was centered around the twin gearbox motors we had as they took up nearly the whole 10cm by 10cm box restrictions, we found the motors to be a whole load of problems like barely having enough room to fit in the front wheels. After spending a few hours in the workshop and trying a few different ideas we finally had a chassis that we could mount the breadboard and batteries onto, David took up the responsibility of taking the bot home to make any improvements to the chassis and we would finish the rest of the bot the next day.
David came up with a new design for the chassis which allowed us more room to build around after a lot of problems with the old design (we don’t have any drawings of designs as they were changing with every new problem we encountered). So on the Thursday we went about finishing the whole bot and testing how it ran. This took a good few hours to finish the housing of the bot which David took charge of and when he was finished that it was me turn to work with the circuit, soldering the sensors and cutting the wires to the right lengths. After I did this we tried to test the bot and it didn’t work, one wheel spun slowly and the other not at all. I hooked the motors directly to the battery and both motors spun very fast so we weren’t sure what was wrong, I went about testing voltages on the circuit and all the voltage on the MSP430 were right but the voltages on the power chip(SN75441ONE) wasn’t right when the motors were attached, there should’ve been around 6V on the motors but was only getting around 2.5V on the high side and 1V on the other resulting in only 1.5V actually on the motors which wasn’t enough to power them efficiently, when there was no load the power chip was giving out very close to the battery voltage. So we thought that the power chip might’ve been broken, it was about 7:30 at that time and we couldn’t replace it until the next day, so to get around the only one motor going problem I connected the two motors in parallel on one side of the power chip which pulled the voltage lower but both motors ran.
So we tried to test the bot to see if the program worked and to just get a time with the slow motors. We noticed that the front axle would lock up because we used a threaded bar for the axle and they would lock on the nuts holding the wheels on so we would file down the threads where they weren’t needed but we didn’t have a file so we left that til Friday morning. We also noticed that the bot had to hit the wall straight on to click the switch but if it did the bot would do what it was suppose to, we got a time of 24 seconds which wasn’t great but we hoped with the new power chip it would be faster and with that we went home and would finish the rest on Friday morning.
So Friday morning we switched the chip and tested the motors but nothing changed, same voltage and speed coming out, while I went about trying to find the problem Nathan went to file down the axle to stop the locking up. I called over our lecturer Dave Doren to help us as everything I tried didn’t make a difference, the only thing we found to make a difference was to add extra wires form the 6V power rail to the supply of the power chip which would let more current through(Martin Tully gave us this idea) which increased the voltage by 0.4V each wire but resulted in both those wires and chip to heat up so we left just two wires in. Me and David sent a good hour talking to Dave the lecturer about our problem which he couldn’t think of a way around as he wasn’t sure what was causing it either, we tried changing the supply from 6V to 12V with no difference, after the hour of trying to find something to speed up the motors Dave told us to just race the robot and get a time so if we change too much and it doesn’t work we at least had a time. We got a time of about 20-21 seconds(Timer stops after 20 seconds) which was awful compared to some bots getting around 3.7 seconds better still better then some, after that was done Dave got Ted Burke,another lecturer, to come and help us find our problem. When Ted got there we found another problem which was the motors only sometimes started and if they didn’t start we would here a very high pitch “scream” from the motors. So again we spent about an hour trying to fix the problems with Ted and found a few things to improve the motors, first was the the ground pin of the MSP430 was connected to the 6V rail which would experience a lot noise from the motors which could cause the chip to reset, to help with the noise we add a capacitor to the 6V rail as well because our motors made a lot of noise on the lines. The final thing which we were thinking could be our problem was the motors were too difficult to run meaning they pulled a lot of current which the power chip just couldn’t supply, the only way to get around this problem was to change the gear ratio of the gearbox or get a new motor and both of these options were undo able because we would have to take apart the whole robot to do both and we didn’t have the gears to change the gear ratio, so we suck with the slow time and are going to start from scratch with a new robot with new motors.
And with that the race to the wall challenge was over and we learned that we should actually test our new motors with the circuit and not just with the battery next time and get motors with a lower gear ratio that we can actually power but it was a good learning experience and it was good that we had these problems now rather than in the real tournament. That bot is now going to be taken apart for pieces so that a new, better bot can be born.
Here are pictures of the bot before it was taken apart:
IMAG0011[1]IMAG0009IMAG0010

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s