Week 5: 7/31/2017 to 8/4/2017

Progress this week (more details below):
-DC motor issue research
-Battery charge no longer lasts as long on the bogie research
-Now able to control bogie with a Virtual Device uploaded with mobile application on Android Studio
-Soldered 2 pairs of hall effect sensors
-Tested new hall effect sensor brackets, no longer need to use glue




DC MOTOR

Issues with DC motor on 1/12th scale Bogie:
-Particles (dirt, metal shavings from gears, etc.) get stuck in the gear assembly.
-A few motors (so far 3 out of 6 tested) do not work at all
-Initially a total of 8 motors that cannot rotate
-Motor output is 4.6 volts using volt meter while battery output is 6.3 volts. This could be possibly due to the efficiency that brushed motors provide, a lot of energy lost due to heat.

Since the DC motor gear assembly is covered with a bracket, I am lead to believe that the gears are deteriorating or the bogie is too heavy and applying too much load on the gears causing extra wear.

Figure 1: Black bracket with two screws cover the gear assembly

Short Term Solution:
-Took out the metal gear shaft assembly and:
       - Sprayed Carburetor cleaner on the metal gear assembly to clean
       -Air compressor to dry gears and make sure all dirt is removed
       -Applied oil to all the gears in the assembly and rotated the shaft to allow oil to evenly spread


Summary:
We are currently using brushed dc motors and over time the metal brushes and commutator inside the armature wear out. Brushed dc motors have a small lifespan and even worse when they are continuously running. Brushed dc motors prove two advantages, that is, they can easily be wired directly to DC power and they are not expensive. The alternative to brushed motors would be brushless motors and these provide a great deal of advantages such as; longer lifespan, quiet running time, 90% efficiency and not effected by heat. But, they do come with an expensive price tag, and a motor driver or controller would need to be implemented.


BATTERY

We are currently using NiMH 6V 5 cell batteries that as of this week only are able to last 10 minutes or so running on the track with the bogie. These batteries are not expensive making them an easy buy, but they only last a few months and significantly decrease power output when exposed to heat.

In the 2016-2017 SSW 1/12th scale solar team's report, they mentioned how the 5 cell NiMH batteries only last 20 minutes running on the bogie and take 55 minutes to charge.

General Future/long term:
There are multiple ways to address the motor and battery issue and all depend on how we plan to present the 1/12th scale to people. For example, if we want to take the entire 1/12th scale to an event and plan to have the bogies continuously run while the audience can interact with the bogies or use it as a visual showpiece while running the mobile application and bogies a few times.

These two different scenarios are important to know because it will ultimately be the deciding factor on whether we should do a full motor and battery overhaul or use the parts we have and try to rebuild them instead of replacing them.

Either way the batteries do need to be replaced soon since the new rotary encoder design for bogie positioning uses two Arduino's (one Mega and one UNO).

Figure 2: Old hall effect sensor bracket, but we are using zip ties instead of glue.
                                          This will allow us to easily replace the brackets without damaging the hall effect sensor.

Figure 3: Soldering the wires onto the hall effect sensors.
         Using cardboard to avoid damaging sensor. 

Figure 4: Guide to use when soldering hall effect sensors. The shaded dot/circle indicates the label part is facing down.

Week 4: 7/24/17 to 7/28/17

I spent a lot of time the beginning of this week trying to document as much information about the 1/12 scale mobile app and testing process, unfortunately there is still more clarifications I hope to get situated on Monday. I am also working on a build procedure for the wiring portion of the bogie, trying my best to capture any points that may confuse future teams. 

Aside from the documentation, I implemented some new designs for the bogie that will allow for a longer life span. A big thanks to Victor Mercola for help with modeling/designing the new components in SolidWorks and getting them 3D printed! Victor has placed all new design changes in the proper folder in Spartan Superway archives google drive. 

Figure 1: Previous Servo switch design using tape and a bent paperclip 

Figure 2: New servo switch design using bolts and nuts 

Figure 3: Image on left is a 3D printed attachment for the servo.
                          Image on right is the original attachment from manufacturer.

The bogie housing for the Arduino and wires has two versions from previous teams; the oldest version, shown in Figure 4 on the right, and the latest version on the left. The issue with the older version arises when you have to screw on the cover to the cabin, takes a lot of time to unscrew just to make minor adjustments.  To avoid wasting 3D printing supplies, instead of printing a new cabin, a simple bracket will be placed on the inside, shown in Figure 5 and 6.

Figure 4: New version on the left and Old version on the right, both designed by previous teams.

                                         

Figure 5: No longer need to screw on cover.

Figure 6: New design of bracket.

We implemented a new design for the hall effect sensor bracket that allows for adjustments. Also, no longer need to glue hall effect sensor to bracket, shown in Figure 8. This is a good first prototype for this design, I plan to make a few more adjustments.

Figure 7: New hall effect sensor bracket allows for adjustments.

Figure 8: The hall effect sensor no longer needs to be glued.

By the end of the day on Friday I finished building 2 more bogies for a total of 5, I did not have time to run them on the track, I will do so on Monday. My goal is to have a 6th "indestructible" bogie, using the CNC aluminum parts left by the previous teams. The "indestructible" bogie would be used for testing new software changes where there is a risk of the bogie potentially falling off the track.  

Figure 9: Total of 5 bogies fully assembled, far right 6th bogie using aluminum.

Week 3: 7/17/17 to 7/20/17

Monday night was the Silicon Valley Automotive Open, so most of the day was spent preparing for the presentation as well as making sure the bogies run well around the track. A few things caught my eye that night and the biggest was how much the track changed from 1:30pm to 7:30pm. When we first ran the bogies in the afternoon they had no problems around the track, but later that night the bogies were constantly getting stuck at certain places along the track. With further investigation, I noticed some of the components that hold the track together are broken. I plan to work on the track more next week.

Most of my time this week has been spent trying to debug the phone application program, with the status code 300 issue. My process involved watching numerous videos of Java programming and using Java as a guide when trying to understand Kotlin programming language.

I also managed to get another bogie put together (making our total bogies assembled up to 4) since we had one more Arduino Mega left. The new hall effect sensors had arrived this week and I got one set soldered with the correct wires matching the rest. I attempted to test the bogie Thursday afternoon, but the hall effect sensor brackets were too wide and restricted the bogie from moving. I will continue to work on this next week.

week 2: 7/10/2017 to 7/14/2017

This week I spend a lot of time trying to get the track leveled, so the Bogies don't get their switch lever jammed as well as widening portions of the track to avoid the Bogies getting stuck.

Figure 4: The change in width of test track, which results in the bogie to get stuck. 

There was a problem with one of the bogies that resulted in the bogie to constantly go to station 1 (the inner loop) even when the outer loop program was installed. The issue was that the left hall effect sensor was not reading the magnet, reading the magnet allows the lever switch to activate. In this case, the hall effect sensor was more than the minimum distance away from the magnet. I added two washers to extend the hall effect sensor, this is a short term solution. I plan to redesign the hall effect sensor bracket to avoid constant breaking as well as come up with a way to adjust the distance from the magnet to the sensor.

Figure 1: At this distance the hall effect sensor was not reading the magnet.

Figure 2: Short term solution to bring hall effect sensor closer to magnet.

Figure 3: New distance where hall effect sensor reads magnet, after installing washers. 

I was able to get 3 bogies assembled and properly wired to run on the track. One thing to take note is that the pins on the wires easily come out of the slots on the shield connected to the Arduino. In cases as small as changing the battery, one should always double check to see if all the pins are in the slots. There were a few times when we would test the bogie again after a 45 minute lunch break and it won't work the same, this was due to some of the pins coming out.

                    

Figure 5,6,7: Working Bogies labeled 1,2,3

Week 1: 7/5/17 to 7/7/17

I am currently working on trying to implement the encoders on the DC motors to locate the position of the bogie.

The first step was to make a program in Arduino that runs the motors and outputs information of the encoders on the serial monitor. This was a big challenge for me, I spent a day of research and concluded that the code needed interrupts. At this point I reached out to Khang for assistance, he was able to get a program running to get useful information from the serial monitor by rotating the encoder.The next step is to correlate the values from the serial monitor, which are based on the rotations of the encoder, with distance traveled by the bogie on the test track. Khang and I will be running tests and measurements starting Monday.

My to-do list for week 2 consists of: finalizing the program for the encoders, redesigning the mounting bracket for the hall effect sensors, redesigning bogie components to allow for easy assembling, and designing a safety harness for the 1/12th scale track to avoid the bogie from falling to the ground during program errors.

About Me

Hello everyone, my name is Shane Fatehi and I will be working with the 1/12th Scale Team for the Spartan Superway! I recently finished my courses at Foothill Community College, the reason behind my late start, and will be attending SJSU in 2017 Fall Semester. My major is Mechanical Engineering and my goal is to emphasize on Mechatronics in grad school. Some of my responsibilities for this Summer internship includes; implementing encoders to detect bogie position, redesigning components on the bogie to allow for quick assembly, and assisting my team members with the Outer Loop Arduino program.