Automated Threaded Insert Installer (WIP)

Thanks for checking out my project! This will be a "Capstone" project for the completion of my associates degree in Electromechanical Technology at Thaddeus Stevens College of Technology. The Capstone is designed to be a cumulation of the knowledge I have gained during my time at Stevens as well as my industry experience. The goal of this project is to automate the process of inserting threaded metal inserts into plastic or 3D printed objects. If all goes according to plan, my machine will do the following:

• Recognize the object inserted into the machine using a vison system to detect shape or by a QR code
• Drill holes in designated locations dictated by the vison system
• Use a process known as "heat staking" to install the threaded metal inserts into the holes using and auto-feed system
• Involve as much automation as possible, eliminating human involvement
• Not burn down the house or initiate Judgement Day
• Provide me with even more knowledge about mechatronics

My original plan was to use the schools ABB robotic arm or a FANUC at my employer RG Group. It was my intention to design and build at least two custom end effectors and a system that would allow the robot to change out its own tooling. However, due to limitations from COVID19, it was unfeasible to use either arm.

So I switched gears, but maintained the original idea of multiple tooling to accomplish a single task. The idea behind that was to purposely add complication and complexity. I want the challenge of a multi-step process that will involve as much automation as possible.

During my time as an Engineering Technician at RG Group, one task was to occasionally help out in our light manufacturing department. The department has a number of customers in the medical industry and most products sold to them involve tubing and fittings. The one job was hand threading fittings into molded plastic reservoirs. The job was tedious and surprisingly hard on your hands. Not only was the repetition grueling, the rough edges took a toll on you your fingers. My inspiration was formed during 2 weeks of doing this job. I knew there had to be a way to automate this task. While my insert machine won't insert those fittings, it will take on a different role and prove a concept.

• Introduced the idea of a Cartesian style robot to drill, tap and insert threaded insert into a plastic block with both instructors Mr. Dagen and Mr. Rodgers
• Mr. Dagen liked the idea and suggested I look into a company called Penn Engineering as they make a number of different types of inserts
• Mr. Rodgers expressed concern with the difficulties of tapping a hole due to speed and control
• I spoke to a machinist friend of mine, Jody Gontero about the difficulties. He explained drill speed and rate of insertion. Said it would be fairly easy with plastic, but not without issue
• Discussed using a QR code reader/vison system to determine hole location with Mr. Rodgers

• Discussed QR codes more with Mr. Rodgers
• Talked about the possibility of using an air motor with Mr. Dagen
• Talked about getting in touch with Walker Gross at RG to source an air motor
• Realized how cheap and easy to use a small 24v DC CNC router motor is and scrapped the air motor idea

This was a really neat X/Y axis idea I found on YouTube. I thought the way it moved was brilliant in its simplicity and ease of build. I planned to use 2 servo/stepper motors on each side and adapt a 3rd to operate the Z axis.

• Discussed the idea with Mr. Rodgers. He seemed into the idea and suggested bike chains and gears which I have access to at the bike shop I work at part-time
• We discussed PLC's vs Raspberry Pi's for my application
• More discussion on webcam vison systems and programming with Python

• Came to the realization that building an X/Y/Z axis would be very expensive and would take up all of my budget
• Discussed a reduction in scope with both Mr. Dagen and Mr. Rodgers and decided to just make a machine with less moving parts.

• Bothered Mr. Rogers Saturday night to get an opinion on a cheap 24v drill motor and issues with drawing too many amps

• While researching what I could use for a heat staking device I found out some soldering irons have replaceable tips
• Some tips are specially designed to push things into plastic
• Researched soldering iron options
• More discussion about drawing amps on a low voltage soldering iron to use as a heat staking tool

• Spent the morning confirming the parts I definitely wanted to purchase
• Filled out and submitted the required PO paperwork
• I purchased 4 items from Amazon.com totaling $85.23 leaving me a balance of $214.77
  • 1 24v DC CNC Drill Motor with ER11-A Collet for $19.99 Chosen for its compact design, simple collet attachment and price
  • 1 set of 3 Heat Staking Tips for a Weller Soldering Iron for $11.65 Chosen for variety of common metric sizes in case I needed to use different sizes
  • 1 300pc Female Threaded Metal Inserts sizes M3,M4,M5 for $14.99 Chosen for best value
  • 1 Weller WLC100 40w Soldering Iron Station for $38.60 Chosen for ability to accept heat staking tips, controllable temperature and brand reputation

• Seretta and I worked through some technical details on the classroom whiteboard We bounced ideas o each other and drew up a few possibilities on methods to feed the threaded metal inserts automatically
• This was a great way to get o the single-minded track I was on for one feed method and look at other options
• I had planned on using a very small pneumatic cylinder of some sort of solenoid to push inserts out into a chute that would hold them using friction. A vertical tube would be used to resupply the mechanism
• We discussed a revolving cylinder to rotate and position an empty hole for the drill bit to pass through and then be held with an insert for the heat staking device to push into the newly drilled hole

• My original plan was to create a Cartesian style robot to perform multiple functions
• Budgeting didn't allow for the creation of the X/Y/Z axis and all the required components, so I had started to plan out machine that was much more simple
• While searching the lab and storage areas of the shop, I kept looking at a linear rail/smart motor setup Mr. Dagen had previously used as a robotic drawing device
• I didn't think I would be allowed to use it so never asked
• Figured I would ask, so I did and he gave me permission to use it
• MAJOR hurdle overcome. I now had a solid foundation to build my Cartesian robot on

• I started to work on the configuration of the axis's
• The modular design allowed multiple layouts
• Some layouts were better than others in that they minimized cantilevered load on bearings and some configurations allowed the slides to move but not the entire rail system
• I preferred the motors to remain stationary and the slides to move. Seemed like it would stress the motors less and allow more precise movements and more control

• Research on threaded insert tolerances and hole depth recommendations
• Straightened up my garage and workbench in preparation of frame/enclosure fabrication

• I fabricated the base out of 1/2" thick HDPE (High-density polyethylene)
  • 24" across and 28" deep
  • 1/2" thick HDPE base with 2" double rail aluminum T-slot structural framing

• Tools Used:
  • Drill Press
  • Table Saw
  • 4" level
  • 5/8" Forstner bit
  • 1/4" drill bit
  • 5/32" Allen wrenches
  • Sharpie Marker

• Materials Used:
  • 100" of double rail T-slot framing
  • 12 black-oxide 1/4-20" buttonhead hex drive screws
  • 12 5/8" at washers 12 single nut 1/4-20" T-slot fasteners

• Frame/enclosure fabrication begins Dimensions: 24"x28"x24" (excluding 1/2" HDPE base)

• The following is a list of considerations I made during design and construction of the frame:
  • Basic design: With as little redundancy as possible
  • Sturdiness: To support multiple heavy components
  • Rigidity: Key to accuracy, precision and repeatability
  • Self-sustainability: The frame also needs to be able to support all of the components within its structure
  • Safety: Design will allow for plexiglass panels to be inserted in the railing grooves later Ascetics: Form and function were goals from the beginning that I wanted to achieve

• Tools used:
  • Miter saw with 10" carbide tipped blade for cutting aluminum
  • Tape measure
  • Digital calipers
  • 4' and 6" level
  • Combination square
  • Roofing square
  • Sharpie Marker
  • 5/32" Allen wrench
  • 1/4-20" Tap/handle

• Materials Used:
  • 246" of T-slot framing
  • 120 black-oxide 1/4-20" buttonhead hex drive screws
  • 56 double nut 1/4-20" T-slot fasteners
  • 4 90 degree corner braces
  • 22 45 degree angle braces 4 2' double rail covers
  • 4 black-oxide 10-24 1/2" sockethead screws
  • 4 single nut 10/24" T-slot fasteners

• I decided on a method of feeding the metal inserts one at a time. Early on, Mr. Dagen mentioned a device that would be able to help me with this
• Through research I found this concept is generally called "Feed Escapement"
• Concept is a method of allowing a set number of product to move on. Commonly used in industry for timing and separation
• I choose a 2 finger design that should allow my inserts to advance one at a time and also allow for no insert to be in place.
• The unit I chose is made by SMC. Part #MIW8-8D1
• I found and ordered a NIB unit on eBay for $149. Approx. $50 less than buying directly through SMC
• This portion of the project may end up being one of the more difficult aspects from a troubleshooting/fabrication viewpoint. I feel there will be some difficulty in getting the escapement/inserts to line up with the heat staking tool and the material

• I reached out to my boss Jay Danks @ RG Group today seeking donations. Specifically, I was looking for pneumatic rotary cylinder. I want to use it to move the tooling (drill/heat staker) out of the way so they do not hit the material when not in use
• We are working together to see what RG may have that will work for my needs
• Jay put me in touch with RG's "Order Specialist", an employee that works with engineering staff to develop BOMs and order parts

• Today I started to prototype a method to mounting the soldering tip to the carrier of the linear actuator
• After disassembling the soldering gun and measuring the base of the soldering iron tip, I designed a housing assembly for it in Fusion 360 and then 3D printed it
• The design needs a little work, but I will wait on that until I finalize the method of the tool changing. For now this was mostly a way to re-familiarize myself with 3D modeling and to further test my new 3D printer

• Today I submitted a PO for materials to be used to fabricate the insert magazine
  • 1 Aluminum U-channel 4' long, /5/16"x7/16"
  • 1 Aluminum U-channel 4' long, 1/4"x9/16"
• The 5/16"x7/16" piece has an ID of .31 which will allow a nice fit for the 7mm wide inserts
• The OD of this piece is 7/16" which is the same as the ID of the 1/4"x9/16"
• I plan on using the larger width section of u-channel as cover on top of the smaller section
• This will create a rectangular tube that will be spring loaded
• Will have to source a long spring with minimal force
• I need to fabricate a follower for the spring, most likely will 3d print this

• Submitted a PO for 3 items on eBay
  • 2 used SMC MXS6-10 Pneumatic Linear Guided Air Cylinders for $15 each
    • Hopefully they are in as good of condition as they look, these can be quite expensive new
    • Different than a traditional pneumatic cylinder in that they combine a linear actuator for precision and control. I also like the mounting options
    • If I am able to source a rotary actuator, these will be used as clamps to hold the part that is being drilled/staked in place
    • If I am not able to source a rotary actuator, I will mount the pair vertically on the Y-axis carrier and and they will provide a method lifting the tooling out of the way
  • 1 NIB 2 finger escapement cylinder by SMC mentioned previously

• Spent most of the day researching and purchasing components needed for the programming, computing and control portion of the project
• Decided on a Raspberry Pi and a 4" touchscreen to control as many functions as possible
• From Amazon I purchased the following out of pocket and plan to keep after the completion of the project
  • Raspberry Pi 4 4GB Starter PRO Kit ($99.98)
  • Miuzei Raspberry Pi 4" Touchscreen w/fan and case ($34.99)
  • KNACRO 5v 8-Channel 10a PC USB Control Switch Relay ($16.99)

• My Raspberry Pi 4 came today
• Very nice packaging, easy to assemble
• I wish I went with a larger screen
  • I may upgrade to 5" or 7" screen and keep this 4" for another project down the road

• Designed a 4"x4" "Test Block" in Inventor Pro
• This test block will serve as a template to start testing the movement of the axis
• Test bed for different infill percentages and layouts to see what works best for the inserts

"Save Early, Save Often"

I am religious about saving documents often, but now I need to start backing them up. Today I somehow deleted my entire Instructable and tech support could not retrieve it. Luckily, the browser on my phone was not refreshed and I was able to grab screenshots of everything. Mr. Dagen was also able to save it as a PDF, something I was not able to do on my end. I spent a few hours re-entering all of my information. It was a tedious lesson in doubling up that I don't want to relive. Back up and running, waiting on parts!

• Today I started a Bill of Materials
  • I will include every fastener, piece of raw material and component
  • I will research and document part numbers and manufacturers/suppliers
  • I will include price and quantity
  • I will record the source that funded each part
• The BOM will replicate industry standards in order to provide an accurate cost to physically build my machine
• This BOM will be a continuous work in progress until the projects completion
• I have already started to see the true cost in just materials, not including labor. I am actually glad I did not start recording how many hours I have included. However, that would be a very interesting facet of this project. A true cost analyst would include labor and reflect how expensive projects like this can be and serve as a teaching tool in the importance of proper planning, sourcing, time management and the value of experience

• Today I reached out to my good friend Don Dresser, who works for an insurance company. I asked if he could assist me in gaining insight on the background savings when automating a process. I would like to know what savings there would be from an insurance standpoint. Automating should reduce not only operating cost by limiting human involvement, but also lowering the potential workers compensation claims with the reduction of injury due to repetitive movements being eliminated.

• Completed the BOM for the structural enclosure. The total is $544.37
  • This is higher than I expected, but not by much
  • I estimated in the $400 range
  • I was surprised most by the cost of the HDPE sheet
  • The other surprising element was the cost of the T-slot gussets and braces. They were easily 3x what I thought they would be and it is shocking how fast these parts add up

• I used AutoCad Inventor to design a handle to attach to a linear slide
• The slide will have a platform for the work piece to securely rest on while the drilling and staking will be performed
• It has a slotted section designed to allow for adjustment when mated to the linear rail

• Sometime overnight while printing, a failure occurred
  • I am not sure if it was a failed heatbreak or a jam in elsewhere
  • I cleaned everything up, put a new nozzle in to see what happens
  • I also played around with some Cura settings and revised the design a little
  • Additionally, I ordered an all metal hotend conversion kit on Amazon

• Spent a majority of the day troubleshooting and calibrating my 3D printer
  • I was having a number of issues with bed plate adhesion and extruder blockages
  • After consulting several websites and my friend Jody, I think we are back in business

• Used Inventor Pro to work on a few parts
  • I am now satisfied with the 4th version of what I am calling a "Slide Handle"
    • This part will act as a handle attached to a linear guide rail
    • It will be the method used to load and unload the work pieces
    • When closed, it will seal off the enclosure, ensuring no hands can be inside while operating
    • The variations mostly dealt with dimension changes and attachment style simplification
  • I am pleased with the look of the 3rd version of the part I modeled called "Work Piece Cradle"
    • This part will be mounted on the linear guide rail in which the Slide Handle is attached
    • It will serve as a base in which the work piece will sit
    • I modified it from the original design to allow the operator remove the finished piece easier
    • The newer version also will allow for pneumatic clamps on the sides if I do decide to add them
    • The latest version incorporated a simpler attachment design as well, one that allows for an easier method that also takes adjustments in consideration

• Today was a productive snow day. After finally figuring out the 3D printing gremlins I was able to get a successful print of the "Slide Handle" and "Workpiece Cradle" then assemble the slide
• I am particularly happy with the outcome of this sliding mechanism for multiple reasons
  • It works better than I expected. Everything is aligned and has a smooth action. I looks good and serves multiple crucial functions
  • First and foremost, this portion of the machine is designed to hold the workpiece in place while the drilling and staking operation is being performed.
  • The second function is this sliding mechanism and door act as an engineering control for safety purposes.
    • This mechanism makes it very difficult for the operator to place their hands anywhere near the moving or high temperature parts of the machine as it places a physical barrier between each other
    • Originally I had just planned on a door that the operator would open, but this method should allow for a higher level of safety. Also, it looks cooler

• I also took some time to redesign the Workpiece
  • First version fit a little tight
  • I also wanted to include a place to affix a QR code sticker for the optical reader in the future

• I made a mount for a pneumatic rotary cylinder
  • This cylinder is one I found in the back room of our schools lab
  • Its rotary action is compound
    • Cycles linear
    • Rotates 90" as it moves linear
  • This cylinder will rotate a chassis that holds both the drill and soldering tip
• I designed this in AutoCad Inventor
• I then 3D printed it
  • The print quality isn't very good
  • I need to play with some settings to get more precise results/tolerances
• Overall, I like the design, but it needs some work
  • I focused to hard on certain aspects, such as keeping it small and light
  • Need to strengthen the areas where the bolt heads are
    • Cracked PLA
    • Distortion of material when tightening
  • I will redesign this to clamp stronger
    • Spins under load
    • Can't tighten more due to cracking
  • Will redesign to use the clamp more as a stabilization
  • Will use the 4 bolt holes on the back of the cylinder to provide a firm mounting point

• Today was a very busy day for 3D modeling in Inventor
  • I modeled a what is my most complex design to date
  • It involved a lot of planning, measuring, calculating and spatial awareness
  • Things I had to consider:
    • Balance of the components on a shaft as they rotated
    • Their positioning in both static and dynamic states
    • Proximity to other parts of the machine and the workpiece
    • Cable management
    • Heat from the soldering tip
    • Cooling for the drill motor
    • Mounting hole clearance
    • Weight vs strength ratios
  • The hardest part was the spatial awareness
    • I had tried and failed to mostly draw the shape mostly as one unit and extrude it in sections
    • This failed and took longer than it should have
    • I remedied this by starting with one section and then adding on others by drawing on multiple planes
    • The main issue I was having was a getting certain sections inverted and having the entire thing mirrored
    • Ultimately, the 5th version was the best and came out nicely
    • I thoroughly enjoyed modeling this and am very pleased with the outcome
      • After drawing this, all other models became faster and easier with less iterations
      • I know I could have designed this as 3 different pieces and attached them together but I wanted to challenge myself to creating a 1 piece design that works well
  • I also learned more about 3D slicer settings
    • Resolution matters quite a bit, the one photo demonstrates this with the octagonal vs round hole for the drill motor
    • I may also consider redesigning the tool cradle to align the drill bit and soldering tip to accommodate the offset caused by the compound cylinders action. This would eliminate the need to move both the X and Z axis, theoretically this will be more accurate.

• I redesigned the cylinder mount after it failed to securely hold the cylinder
• The new design utilizes the 4-bolt base to prohibit the cylinder from turning under weight
• I also increased the wall thickness around the holes to prevent cracking

• The 2nd version of my cylinder mount proved to be a good example of overthinking and overengineering an idea.
  • It was however another way to get more 3D modeling experience
• I did not like the way the cylinder fit and did not like the limited adjustment
• So I went back to the drawing board and used T-slot to overcome both the adjustment issues and strength/stability/rigidness

• Today I designed and fabricated my favorite part of the project so far
  • I was so pleased with how this turned out due to its simplicity and the multifunctional aspects
• I designed simple clamps in AutoCad Inventor to be 3D printed then mounted to the two SMC MXS6-10 Pneumatic Slide Tables I purchased through eBay
  • The slide tables are mounted to T-slot railing allowing for adjustment
• The workpiece clamps are designed to hold the workpiece in place and for the workpiece cradle to be locked into place so the slide handle cannot be pulled out while in operation
  • This system is an interlock known as an "engineering control"
  • It is generally considered the best way to prevent injury as it places a physical barrier between the operator and the dangerous operation
  • It does not rely on a light curtain, sensors or other electronic components that could fail or be worked around
  • If time permits, I do plan on adding proximity sensors to the slide as well

• Today I placed another Amazon order for some more components
  • Pneumatic Solenoid Valve (2 position 5 way) ($39.89)
    • 4 of these valve are attached to a common manifold with a single inlet and dual muffler system
      • This was more economical, simplistic and compact than individual valves
      • The ability to mount the manifold rather than each individual valve is helpful too
      • I will likely not need the 4th valve, however wanted to have it in case I need to add another part
        • I would like to have the workpiece slide operate with a pneumatic cylinder, but will likely run out of time and resources
  • ER-11 Collet Set ($14.99)
    • The collet that came with my CNC drill motor will only fit a 1/8" bit
      • I need it to fit a 7mm drill bit for the inserts
      • It was more economical to buy a set rather than an individual collet
      • While my project is based around the 7mm OD inserts, I thought it would be a good idea to have the other sizes available for future modifications/projects

• I designed and 3D printed "finger couplers" for the escapement cylinder
  • These are meant to slide over the fingers which will hold a 1/8" rod that will be used to to separate and hold the metal inserts while they are being staked
  • The first version was a straight design, but I choose to go with an offset design instead
    • The offset would allow for the cylinder to be mounted higher and out of the way of the workpiece
  • The holes on the sides are for set screws to hold the finger couplers to the cylinder fingers
  • In addition to being offset about an inch lower, both fingers are slightly offset towards each other so the 1/8" rods would be approximately 8mm apart to accommodate for the 7mm wide inserts

• I modeled and 3D printed parts for the insert magazine
  • The magazine will be spring loaded
  • I choose aluminum C-channel stock to act as the body of the magazine
    • The channel has an ID slightly bigger than the OD of the metal inserts
    • I purchased a second aluminum channel with an ID that matches the OD of the other rail
      • This piece will act as a cover and keep the spring from escaping the magazine
  • The short simple looking rectangle with a cylinder on one end is a magazine follower
    • The tube slides into the spring pushes against the inserts
  • The longer rectangle with a cylinder and 2 countersunk holes was my first version of a stop to slide into the other end of the spring
  • The complex looking piece shown in the 3D model and with the heat staked inserts is essentially that piece attached to a custom bracket to be mounted to aluminum T-slot
    • It is a much more secure design and doubles as a mount
    • I liked this design a lot as it acts as 2 things at once and keeps the overall design compact

• Today I mounted the insert magazine as a proof of concept
• The overall idea has some flaws and some good ideas (on in my head)
  • One issue is the bottom of the magazine needs to be located just above the spot the inserts need to be pressed into
  • This creates a problem as the magazine will hit the workpiece clamps
    • I could eliminate the clamps
    • I could print the workpiece so the spots are higher
  • The magazine moves with the tooling but not up and down
    • This allows it and the soldering tip to always line up
    • The drill motor is out of the way and won't interfere
• I pointed out how the metal inserts had more friction than I expected in the C-channel
  • I suggested I could wax it to reduce friction
  • Mr. Dagen said it would be possible to use emery cloth to smooth the edges off
  • He also said if I was ordering things like this in bulk I could possibly have the manufacture tumble them and it would deburr the edges
    • I consulted my former boss Jeff Kauffman about his brass tumbler
      • Jeff is a certified marksman in competitive shooting knows a lot about deburring and brass preparation
• I ended up scrapping the entire idea of automating the insert magazine
  • It was too time consuming and complicated
  • I am convinced this can be done but it will be a challenge for anyone attempting it
  • My time was better spent in other aspects of the project

• I started out by designing a simple bracket that would allow me to mount the manifold to T-Slot
  • The holes on the manifold nearly lined up with the dual slot railing
  • Original design was to simply mount the manifold but no plans where
    • The modular design was nice, but a more well thought out dual purpose design was what I decided on
• My final design incorporates a manifold mount and a tube management system
  • I spent some time laying out the wire and tube management to allow space and flow of the components
    • I like to consider maintance and access when when managing cables
    • Aesthetics are also important
    • In this application, all of the above was considered as well as overall presentation, I didn't want to hide too many components

• I used AutoCad Inventor to design a simple bracket to mount the pneumatic regulator
• Pretty straightforward design
  • After doing so many other complicated things, designing and printing things like this is a snap

• I designed this custom bracket to mount 2 pneumatic T-fittings
• It slides in the groves of T-slot and provides a place to hold the fittings out of the way
  • More secure than zip ties
  • Looks rad

• One of my favorite parts of the build
• This is simply a junction/control box for the Weller soldering iron I am using as a heat-staking tool
• I removed all of the internal wiring and external controls from the soldering iron station and designed a box to house them
  • Designing this gave me an appreciation for what is involved in something as simple as a box to house simple wiring setups
  • Though unnecessary, I wanted it to mimic something more involved that would require cooling, hence the ventilation holes (which double as cable management) and the vents on top
  • This portion of the project gave me inspiration to design a much more complex enclosure for the HMI screen

• I used 2 of the large 24v power supply's for the MOOG smart motors
  • One is dedicated to the motors themselves
  • One is for pneumatic solenoids and the drill motor
• The power supply's are secured to the frame via a T-Slot frame
• The relay board is mounted to the power supply frame with a pair of adjustable stand-offs I designed in AutoCad Inventor and 3D printed

• I used soldering skills gained through my internship at RG Group to attach wires to the soldering iron tip I am using as a heat staking tool
• I also used these skills to clean up and properly secure the daisychain smart motor cable harness
  • The harness was previously wirenutted and used electrical tape
  • While technically functional, it was not pretty and seemed likely to fail
• In addition to soldering, I used heat shrink and wire look to manage the cables better

• I routed the pneumatic tubing and other wiring in such a way to allow it to move with the toolhead and each of the robotic axis's
  • This took multible attempts to find an optimal layout
  • I am still not happy with it
  • In the future I would like to see a better loom setup to manage everything better

• In the interest of adding components later, troubleshooting, wire management and replacing faulty components, I opted to use 2 sets of terminal blocks
• One set is mounted under the relay board and adjacent to the power supply
• One set is further away on the machine and ties in the 120v components

• One of my favorite aspects of building this project was 3D design
• Designing a part from scratch and using it is highly rewarding
• It is also highly time consuming, I found myself sitting for hours in the evenings or on a rainy Sunday afternoon designing a part
• I used my digital calipers constantly
• This was a great exercise in spatial awareness and thinking a part through
• Many considerations had to be taken into place
  • Mounting hole locations and accessibility
  • Door fit
  • Screen clearance
  • Cooling
  • Cable management
  • Power switch securement
  • Hinge alignment
  • A method to keep the door closed
    • I went with magnets that were to be glued into holes countersunk into the design

• Perhaps the biggest lesson learned when undertaking a large project is not spending time on ancillary features when you have not worked out the major parts
• I knew I was spending too much time on this
  • I worked on it anyway, because I enjoyed this aspect so much and detest programming
  • Basically, I was avoiding the inevitable and treating myself to a fun challenge when I was stressed out
• I overlooked the hinge alignment
  • I made the door hinges protrude as far down as the the case hinges protruded up
  • I needed to have them meet and somehow overlooked this crucial detail
• I knew printing this would be a challenge
  • The print time was very long so I changed too many settings to bring it down
  • I had everything dialed in nicely before that and should have let it go
  • In my haste to see the finished product, quality suffered heavily
  • I printed the door face down when I shouldn't have
    • Supports would have taken a lot longer, but the face down print required the first layer to be perfect and it wasn't
    • I also forgot to have the G-code print supports for the case and quality suffered yet again

• Programming is not my strong suit, it is by far my weakest
• I want to start out by thanking Bryce Hanaway and Mr. Rodgers for helping me work through the many issues with this portion of the project
• From the beginning there were many issues with this particular Raspberry Pi's touch screen
  • Intermittent glitches and bugs with it syncing to the screen
  • This was eventually figured out (sorta) when I found that I had to have the screen unplugged from the Pi, then powering it up, then plugging the screen into it
    • This has to be done every time I power it up or it will not work
  • Other challenging aspects were looking up drivers required for this specific Pi
  • Connecting to it remotely
  • Sensitivity of the screen
  • A persistent issue where you need to unplug the relay board and plug it back in occasionally
• My Python program is fairly straightforward
  • 6 buttons acting as basic on and off switches to control the drill motor and 2 pneumatic solenoids
  • I had wanted to control the heat staking tool with it, but for the sake of time, opted not to
• Diagram

• I designed a wiring diagram in AutoCad
• I wanted to develop a schematic that one would see in industry
  • I have a fair amount of schematic reading ability now and felt as though it would be pretty east
  • It was not as easy as I thought
  • Drawing a diagram like this is pretty straight forward and most anyone could pick it up and run with it
  • While that will help the next person, its not a very accurate representation as I wanted

Perhaps my biggest failure was purposely postponing programming the robotic arms. Admittingly, I do not like this aspect of the project. It is something I want to continue learning about and want to pursue further. I also knew I wouldn't be able to fully test this portion of the project until other steps were taken. Looking back, I should have at least figured out how to get the program in a basic pattern early on, then worked on the actual drilling aspect later.

I was able to use both the SMI software and the AniCNC software to connect to the motors and control them from my laptop. This gives me hope that I can figure it out and would have if I faced my fears of programming. If I had to do any one thing over, it would have been to make this high priority. I should have learned how to get a basic pattern down first.

I have included on this slide some notes of encoder locations for the 4 corners based on my attempts of programming a 4 hole rectangular bolt pattern circle.

Overall, I was pleased with the outcome of my capstone project. The challenges presented by the unique circumstances of the past did not allow me to achieve all of my my goals, but I was able to take on a lot and create something really interesting. My biggest failures were apprehension when it comes to programming and difficulty staying motivated as other opportunities came along. My greatest strengths were planning and 3D modeling. I enjoyed the challenge of creating a project that is rather complicated, yet fits into a mobile unit. I knew I may never achieve my goals but wanted to create a capstone that could be used by future ELME students.

There is a lot of potential in this project for others to take on. I would love to see a vision system to detect the workpiece to determine hole location. Additional safety features like e-stops, light curtains, sensors and guarding should be added. PLC driven logic would be a great addition as well. The current configuration has good bones, but there will be challenges to correct accuracy and when adding additional automation. A fully automated system with the inclusion of as many industrial components is the proper way to keep this project moving forward. I feel as though this could be a great teaching tool for many more classes. In that regard, I feel my project was a huge success!