scale images

worklog/alex/README.md → worklog/alex/index.md

+# Alex Worklog
+
+[[_TOC_]]
+
 # 2021-02-05 - Discussion with Professor Schuh
 
 Professor Schuh gave us the idea to use piezo resistors or Velostat to perform the weight/pressure sensing that we need. Additionally, the tilt sensing probably won’t be necessary - we can correspond the rear left wheel with the front left wheel and rear right wheel with the front right wheel, as those pairs of wheels should be spinning at roughly the same velocity during turning. This simplifies our design.
@@ -9,8 +13,7 @@ Here are some examples:
 
 ![](esc1.png)
 
-[link](https://flipsky.net/collections/electronic-products/products/dual-fsesc6-6-based-upon-vesc6-with-
-aluminum-heatsink)
+[link](https://flipsky.net/collections/electronic-products/products/dual-fsesc6-6-based-upon-vesc6-with-aluminum-heatsink)
 
 This one has dual MCUs, with a physical switch to include or disclude the CAN connection
 between both of the VESCs.
@@ -42,3 +45,39 @@ For our project, I would prefer a 12s battery for adequate performance. The pack
 This will be plenty, for up to 10 miles of range. I have never worked on batteries before. When researching how to do so, I found this over-an-hour-long video [here]( https://www.youtube.com/watch?v=7QjO90LG67g), that clearly depicts and describes everything required to safely build a reliable battery pack.
 
 I will need access to a spot welder to complete this - we’ve reached out to Illini Solar Car, as they have claimed the spot welder on campus.
+
+# 2021-02-18 - Battery Update
+
+Received the pair of 6s4p batteries, along with battery building supplies such as a Smart BMS, nickel strips, fishpaper, 10 gauge wire, and more. Upon closer inspection, splitting the 6s4p pack into a 12s2p pack is difficult, as there is no leverage to remove the existing H-shaped nickel strips.
+
+The pair of 6s4p packs take up a lot less space than I expected. It would be easier to make a 12s4p pack out of them instead, which will provide us the same power as I required from earlier at nearly 500 Whr. Additionally, I have gotten in contact with a coworker who has a battery spot welder I can use.
+
+Once I have planned out the build and assembled the materials, we should be good to go to complete the battery.
+
+# 2021-02-24 - Investigation acquiring RPM data from VESCs
+
+I have done some investigating of the VESC hardware and software and discovered that you can query them for RPM telemetry data. This removes the need for us to develop RPM sensing
+around the motorized wheels.
+
+To accomplish this however, we may need to talk to each VESC individually. Having a dual VESC is convenient, but this may interfere with our ability to query data from them properly. Some dual VESCs have dual MCUs, other more recent designs have single MCUs as shown earlier in this journal.
+
+# 2021-03-01 - Parts update
+
+From further discussions about the issue of communicating to both VESCs individually or to a ‘master’ VESC to both, we have decided to go with a pair of single VESCs. For example, we could attach a CANBUS cable between the pair of VESCs in order to allow them to communicate with each other, or simply split two PPM signals to pass throttle information individually, among various other choices. In other words, the VESCs can work in tandem with one another or can be independent from each other.
+
+I have assembled a significant portion of the board at this point, as seen here in this picture below:
+
+![](parts_update.png)
+
+Some issues:
+
+- The motor pulley is scraping on the motor mount. This is an issue of tolerances, as the
+motor mount I am using is 6mm wide and the Motor shaft has a tolerance of 6.2mm +-
+0.5mm. In theory, we should have 0.2mm of space, but the margin of error was not in our
+favor.
+  ![](motor_pulley_scraping.png)
+- The threads on the ends of the trucks are not deep enough. We need bearing spacers to
+push the wheel out.
+  ![](bearing_spacer.png)
+  
+