User:Andyha0933/portfolio/enes-100/Project 3

Task
Identify next steps and prerequisites

Doing
Met with Mark Edelen. All prerequisite research and information collected by Mark is on his user page.

Key redesign requirements

 * Add on/off switch
 * Make snap-fit cover and housing for battery
 * Redesign device to allow easy fit of all components
 * Easy assembly for possibility of multiple productions

Necessary prerequisite skills

 * Knowledge of CAD program, preferably Inventor (done)
 * Soldering (done)
 * 3D printing/Makerbot certification

Current working model is being used by Anatomy & Physiology courses. Will collect feedback and adjust redesigns accordingly. Staff member that is using the device is Asst. Professor Ann Repka.

Next steps

 * Get certifications for soldering and makerbot
 * Introduce self to Ann Repka
 * Begin drafting up new designs after collecting feedback

Task
Test the circuitry on a breadboard

Wiring

 * Followed Mark's circuit diagram:

Troubleshooting
The circuit did not work. Several reasons were brainstormed by me, that were resolved:


 * Faulty breadboard: Tried to use different breadboard and circuit was still completed.
 * Problems uploading code to ProMicro: Used "blinkies" test code, and confirmed to work
 * Wiring error: On the wiring diagram, two components were labelled incorrectly: SCL was labeled SDL, and SDA was labelled SCA. Tried switching wiring as SCL to 2 and SDA to 3. No change in functionality.

Mark thinks that it may be weak connections, considering none of the connections were soldered. Trying to use an Arduino Uno, where wires can be inserted similar to a breadboard, did not make a change. Will test again after soldering LEDs to LED board.

Next Steps

 * Solder connections
 * Compare wiring to working model

Task

 * Solder pin headers to LED Backpack
 * Add switch to circuit

Soldering
I determined that the problem of the LEDs not lighting was due to the connection between the pin headers and the LED backpack circuitboard by angling the circuitboard in such a way that the terminals made contact with the pins and observing that the LEDs lit up. After soldering the pins to the board, the LEDs lit up. After that, I loaded up Mark's program to the ProMicro, and the circuit now functions as it should: the more pressure applied to the FSR, the more LEDs are lit.

Demonstration

Adding the switch
Added a simple two-phase switch on the connection from "POWER: +" to "ProMicro: RAW". Power supply to circuit is now controllable.

Diagram Demonstration

Design flaw in current model
In the current working model, the FSR positioning is flawed. The user must hold the grip with the bottom of the handle in the middle of their palm, reducing ergonomics and providing an inaccurate reading. Furthermore, the FSR shifts around.

Other Work Done

 * Re-taped FSR to handle to secure it. The FSR was shifting around and causing reading troubles.
 * Scraped off left-over dried glue on the area between the handle and housing to improve stability of device.
 * Tightly strapped circuit housing to handle to reduce amount of shifting and further improve stability of device.



Next Steps

 * Do field testing to collect user feedback data
 * Draft new designs that implement a switch, FSR improvements such as positioning and security, feedback data, and ergonomic & aesthetic improvements such as longer handle and adequate sized circuit housing

Task
Work on improving the design of the model. My specific section was to improve the handle part.

Repairs
A connection in the working model was disconnected, causing none of the components to power on. Some troubleshooting I did: I re-soldered the connection and the device is operational again.
 * Connected a USB cable to the ProMicro to ensure that the ProMicro was not dead. The ProMicro's LED lit up, showing that it still got power. Furthermore, the entire device worked, showing that the problem was a power connection.
 * Dissected the device. The connection between the switch and "ProMicro: RAW" came loose.

Recording dimensions
One of the main complaints of the grip was the that the grip was too small to be comfortably squeezed.
 * Recorded dimensions of current model

Scaling up design

 * Using Makerware, scaled the length of the grip up to 4.2in(106.68mm)

3.76 inches to 4.2 inches is a total change of +11.7%.
 * Percent change

Printing prototype

 * Printed scaled up prototype

Prototype cracked under squeezing pressure due to low quality print settings causing the prototype to be structurally weak.

The larger prototype proved to be more comfortable, from feedback given by 10 randomly selected people. Key feedback included: easier to squeezeand feels overall better in hand
 * Feedback

Soldering circuitry

 * Soldered the new components together

The circuit didn't work. I used a multimeter to gauge where the circuit was having issues, and found that it was the connection where "ProMicro:VCC", "LED:VCC", and FSR meet. Otherwise, the circuit works fine.
 * Troubleshooting

Other work done

 * Fixed the Makerbot: The old generation Makerbot was having issues printing, where the extruder was intermittently extruding an insufficient amount of PLA. All diagnostics pointed to either the PLA or the extruder. I decided to change the filament first. Changing from the yellow filament to green filament solved the issue, and 3D printing could resume.


 * Concept ideas for the handle: As it currently stands, the way the grip is designed (one open end) and the positioning of the FSR (at the bottom) causes the readings to be inaccurate. I hypothesize that moving the handle towards the middle and fixing the open end will fix this. When a user squeezes the handle, a majority of the force is being applied to the center of the handle, rather than the bottom. Moving the FSR to the middle will solve this. To further increase accuracy, fixing both ends to that the middle is the part that is bending would prove valuable because the two sections of the handle will meet at the center while evenly distributing the force being applied.

Next Steps

 * Implement design changes
 * Fix circuit

Task

 * Adjust team member's designs to prepare for printing
 * Print final parts

Housing Adjustments
To account for scaling up the pieces in Makerware, I had to make the windows for the switch and LED bars smaller. The LED window also had to be shifted towards the top end slightly because it was in the way of the switch.

All dimensions shown in inches

These dimensions will be the correct sizes when scaled up by a factor of 11.7%

Rail Adjustments
Using Marcelo's initial rail concepts, I made a rail that would connect the housing and the handle.

Measurements Height of rail: .226 inches from bottom of rail to bottom of housing Gap between rail and housing: .125 inches Rail thickness: .125 inches Total length: 2.052 inches Total width: 1.8 inches

Dimensions will be scaled up by a factor of 11.7%

Snap-Fit Cover Adjustments
Adjusted Tyler's snap-fit cover design for a more exact fit and more sturdy connection.

Printing
All measurements were scaled up by 11.7% in Makerware.

Next Steps
Finish printing all components and assemble device, adjusting any mis-measured components.

Task
Assemble printed parts: handle and housing

Overall thoughts and observations

 * Grip is too large to fit into rails
 * Switch and LED windows are perfect size to allow for snap-in fitting. Gluing is not needed.
 * Gap on the handle is too large for an accurate reading on the FSR

Handle Adjustments and Rail Demonstration
The handle was too wide and thick to fit into the rails of the housing. I sanded the edges of the grip to the point where it tightly fit, and then greased the rails to ensure a smooth slide on and off.

Demonstration



Switch Fit and Soldering
The switch snaps perfectly into place, so there is no need to glue it on. Soldering it onto the circuit required me to feed the wires through the window to solder it in, effectively making the circuit permanent inside the housing.

Demonstration

Operation and FSR Issues
After insulating an exposed wire, the device operates. However, the gap in the handle is slightly too large to get an accurate reading. To fix this, I used thick tape to close the gap. After this, the reading is more accurate, but more fine tuning has to be done with the programming to perfect the accuracy.

Next Steps

 * Print battery snap-fit cover and attach
 * Adjust software to account for new handle