Pick and Place – Trade Off Study – Camera System

April 5, 2017/in 3D SMD/by Belinda Vivas

By: Kevin Ruelas (Electronics and Control)

Due to the use of a servo driver to control the existing and additional servos, a separate Arduino can be used to house an independent camera system. This Arduino will have its own code as well as utilize visual software for edge detection.

It was important to choose a camera that is lightweight, and small enough to be mountable on the current Z-Axis. The following are the cameras that I looked at that could possibly work.

	Price	Size	Weight	Resolution	Operating Voltage
Miniature TTL Serial JPEG Camera with NTSC Video	35.95	20mm x 28 mm	3g	640 x 480	3-5 V
TTL Serial JPEG Camera with NTSC Video	39.95	32 mm x 32 mm		640 x 480	5 V
Weatherproof TTL Serial JPEG Camera with NTSC video and IR LEDs	54.95	2 in x 2 in x 2.5 in	150g	640 x 480	5 V

The Miniature TTL Serial JPEG camera fits the needs of the project the best, as it is the smallest, lightweight and still maintains a decent resolution.

The Arduino UNO by itself does not have the capability for image processing so once the picture is taken, it will need to be stored externally and then sent back to the computer to a software called Processing.

https://processing.org/

Processing is an integrated development environment (IDE) written in Java. It is here where edge detection code will be written to detect the edge of the PCB, or whatever the camera is looking at. From there the coordinates of the edge will be extracted and fed back to Gremote.

Figure 1 – Miniature TTL Serial JPEG Camera with NTSC Video

Figure 2 – SD Card Breakout board

Figure 3 – Diagram

SD CARD PINS	ARDUINO UNO PINS
CS (SS)	Digital Pin 10
DI (MOSI)	Digital Pin 11
DO (MISO)	Digital Pin 12
CLK (SCK)	Digital Pin 13
3V	3.3 V
GND	GND

CAMERA PINS	ARDUINO UNO PINS
5 V	5 V
GND	GND
TX	Digital Pin 1
RX	Digital Pin 0

References

http://web.csulb.edu/~hill/ee400d/Project%20Folder/Camera/Camera%20Document.pdf

https://processing.org/

https://www.adafruit.com/products/1386

https://www.adafruit.com/products/254

Pick and Place-Preliminary Project Plan

March 1, 2017/in 3D SMD, Communites, Members and Communities/by Belinda Vivas

Belinda Vivas (Project Manager)

Chastin Realubit (Mission, Systems, and Test)

Kevin Ruelas (Electronics and Control)

Tyler Jones (Manufacturing)

Table of Contents

Work Breakdown Structure

By Belinda Vivas (Project Manager) and Chastin Realubit (Mission, Systems, and Test)

The Work Breakdown Structure (WBS) is a summarized visual representation of the divided work between the three divisions (Systems, Electronics, and Manufacturing) to successfully complete the second generation of the Pick and Place. The WBS dictates the overall project schedule for the duration of the project, 16 weeks.

Figure 1 – Work Breakdown Structure

Source Material:

http://web.csulb.edu/~hill/ee400d/Lectures/Week%2003%20Creativity/b_Requirements.pdf

http://web.csulb.edu/~hill/ee400d/Documentation%20Lecture%20Series/05%20Preliminary%20Project%20Plan.pdf

Project Schedule

By Belinda Vivas (Project Manager)

Following the defined Work Breakdown Structure, we can now present a detailed Project Schedule. This will give structure and clear direction to all divisions of the project, it presents a beginning and ending date, duration for each task, and how for certain tasks to begin the previous tasks have to be successfully completed.

Top Level Schedule

System/Subsystem Level Tasks

By Belinda Vivas (Project Manager)

The Top Level Schedule is a detailed visual representation of a detailed project schedule. If the team or a division engineer fall behind it will affect the remaining due dates of the tasks to follow, therefore the dates will have to be revised and shorten the period of future tasks to allow for the project to be finished on the desired due date. A longer time than what is needed was allocated for the Software and Assembly Phase for when a task is running behind schedule we can take days from those items. The Top Level Schedule further shows the System/Subsystem Level Tasks. The chart below shows the detailed Top Level Schedule for the second generation of the Pick and Place.

Figure 2 – Top Level Schedule from 01/23 to 04/02

Figure 3 – Top Level Schedule from 04/02 to 05/15

Burn Down/Project Percent Completion

By Chastin Realubit (Mission, Systems, and Test)

The Burn Down is graph representing the amount of work we have finished verses the amount of work we still need to do. We have currently finished more than 20% of the project.

Figure 4 – Product Burndown

System Resource Reports

By Chastin Realubit (Mission, Systems, and Test)

Mass Report


Vacuum System Components	Preliminary Mass (g)	Uncertainty (%)	Margin (±g)	Expected Mass (g)	Actual Mass (g)
Stepper Motor (A-Axis)	245.00	5%	12.25	245.00	247
Stepper Motor (Z-Axis)	245.00	5%	12.25	245.00	246
Vacuum Nozzle	40.00	5%	2	42.80	TBA
Z-Axis Actuator	292.00	5%	14.6	300	244.12
Detection Camera	300.00	5%	15	312.00	TBA
Project Allocation	Trade-Off Study will be obtained (Test to see when the Z-Axis can no longer pick up and item either due to vacuum strength or motor strength)
Total Expected Mass	1144.8 g
Total Margin	29.6 g
Total Actual Mass
Contingency

The purpose of the mass report is to ensure that the Z-Axis will be able to pick up any ICs or components that will we need to place on the 3dot board. Our design now is based on the first generation Pick and Place and will use mostly the same parts expect that we will be building on to it. We will add a camera in the Z-Axis so that the camera can be used for edge detection which will increase the accuracy of our machine. Next we will change the nozzle to one that has a smaller tip which will also increase accuracy. Using the first generation Pick and Place: Final Documentation, we were able to get the actual mass of each part except for the ones we plan on adding or changing.

Size Report

The size and weight of the Pick and Place will not need any report because instead of being placed in a small cabinet in ECS316, it will have a permanent place in ET111 We will discuss this matter with the instructor and the president. Because of this the only dimensions that we need will be the workable area that the nozzle can work in inside of the aluminum table. Those dimensions were given to us based on the work area of the XY plotter, which is 12.2” X 15.35”.

Power Report

Components	Expected Current Draw (A)	Uncertainty (%)	Margin (±A)	Measured Current Draw(A)
Stepper Motor (X-Axis)	.5	5%	.025
Stepper Motor (Y-Axis)	.5	5%	.025
Stepper Motor (Z-Axis)	.5	5%	.025
Stepper Motor (A-Axis)	.5	5%	.025
Detection Camera	.5	5%	.025
Display Screen	.5	5%	.025
Project Allocation	3 A
Total Expected Current	3A
Total Margin	.15 A
Contingency	.15 A

We will be using the same electrical components powered by CSULBs power lines coming from the outlet as the first generation Pick and Place but we will be removing the Microservo and adding cameras and a screen to let the user know which component the Pick and Place in currently working on. We assume that they will all get the same amount of current. Testing needs to be done to check if true.

Project Breakdown/Cost Estimate

Manufacturing

By Tyler Jones

The original pick and place was design and built using a reel feeder system that was proven to work with 0402 components. In the second generation of the pick and place it should have the ability to keep the reel feeder system as well as add ABS plastic trays for individual part strips to be added. This allows for the machine to still utilize the reel feeders if necessary, and to add a way to manufacture boards with many different parts using multiple ABS plastic trays for all types of IC packages and sizes. An example of the size and dimensions in inches of the plastic trays are shown below in Figure 5. Implementing two of these trays allows for 8 possible channels for the SMT part tape to be feed onto the table. This can then successfully populate the 3DoT board with enough parts and covers all possible part diversity used in the 3DoT board.

Figure 5 – Component Feeder

Another requirement to fulfill is to populate a 3DoT board which contains many different components including resistors, capacitors, and IC of different sizes. In order to successfully manufacture boards with all these components it is necessary to make the machine easy to setup and to use. It should also be able to be easily disassembled and moved. The user must be able to setup the pick and place machine and operate it in a timely manner of less than two hours. It must be able to populate a 3DoT board in a reasonable time frame of ten to sixty minutes.

In order to accomplish this task, the pick and place will incorporate organized easily configured and usable mechanical design features. A storage cabinet can be created using acrylic plastic sheet stock. This plastic can be shaped and fashioned to fit underneath seamlessly into the legs and support system. The cabinet can be mounted underneath, and will be able to house the SMT part reels, the vacuum pump, spare nozzles, SMT part strips, scissors, and solder paste, and stencils. Additionally, the legs and support system of the machine must be raised in order to meet this requirement. This includes lengthening the legs from a current height of 7 inches to a new height of 10 inches. This will allow the cabinet to be placed underneath and house the pump, and other components, without adding too much additional height, length, and width. The current pick and place is (27.6875 inches long x 15.9375 inches wide x 17.625 inches in height) (27.6875” x 15.9375” x 17.625”) It is also essential to make all housings and compartments easily and quickly able to be disassembled, and assembled. This means that the bolts must be properly tapped and dyed into the plastic material. Countersinking will need to be done as well in order to provide no stress fractures or bending of the plastic material. The plastic can be pre-ordered from Home Depot and its specifications are below in Figure 6. Purchasing two of these sheets will be necessary to create the cabinet. The used will not have to be purchased and would only total to around $15-$20 if they were to be purchased. This would bring the total to $121.76 for the machines new layout.

Figure 6 – Acrylic Sheet

In addition to the creation of the plastic housing, the machines dimensions were written down and noted for further design in Solid works and machining. The current wiring of the pick and place is lacking organization, and functional design. The Ethernet CAT 5 cables cables connecting to the Me Uno microcontroller are too short to fully re-program the machine to operate on every operable inch of the aluminum picking and placing surface. This would call for longer 5-foot CAT 5 RJ 25 port sized Ethernet cables. These can be found on Digi key in many lengths and styles, for roughly $5-9 The link for the cables can be found below.

http://www.digikey.com/products/en/cable-assemblies/modular

A wire mold similar to commercially available V5000 series wire mold can be manufactured using Solidworks STL files, and ABS plastic. These wire molds can be mounted in channels on the inside of the housing so that the wires are not stretched or damaged. Further Solidworks files will be generated.

A trade off study should be used to measure vibrations or disturbances at the base of machine using different methods of testing for shaking, vibrating, and shock. This is important because stability is vital in a precision robotic machine used to pick up very small components. Rubber shoes can be fitted onto the legs and tested as the machine operates for disturbances in the pick and place routine. Figure 7 shows a block of rubber that can be purchased and shaped into shoes that cover the legs of the machine. The rubber block can be purchased for $6.95 online.

Figure 7 – Rubber Bench Block

Another trade off study should be taken in positioning the reel feeders so that the tape can freely and easily be stripped of tape and moved the same consistent distance by the micro servos motors. This will affect the height and dimensions of the plastic cabinet.

A key facet in creating the pick and place to be more user friendly is implementing a LCD and LED screen that displays instructions and live camera feed of the machines operations. These components need to be large and accessible in order for the user to view and interact with the machine. The plastic housing underneath the machine will be able to house folding or telescoping style mounts for the LCD and LED screens. This allows the screens to be tucked away without taking up unnecessary space. A trade off study can be conducted in order to find whether or not the folding or telescope system is more durable, and stable.

Inside the plastic housing design and small slide out drawer can be easily fashioned and located underneath the tape that is pulled from either the stationary trays or the current reel feeders. This cleans up the mess as the machine functions and can easily be slid out and thrown away. A force analysis should be calculated to see how much force will push back on the tape. This is critical because if the force of the tape hitting the edge of the drawer is too great it will throw off the alignment.

A rough preliminary scale model below in Figure 8 shows the relative positioning of devices, and layouts represented in this article. Further rendering will be provided in Solidworks.

Figure 8 – Rough Preliminary Draft

MECHANICAL COST PROJECTION

Item	Projected Cost	Margin	Actual Cost
Anti-vibrational Rubber	$6.95	5%
Acrylic Plastic Sheets	$121.76	5%
5’ RJ25 cables	$7.50	5%
STL 3D prints	$15.00	5%
Spring Loaded nozzle	$10.00	5%
TOTAL	$162.11	5%

MECHANICAL SYSTEM WEIGHT

Total system weight added to current system is 14 pounds with a %5 margin. This estimate is contingent upon design and features.

Electronics

By Kevin Ruelas

Cost

2 Cameras + additional drivers ~$50 ea

LCD Screen ~$50

RJ25 Adapters/cables ~$15

All cost has a margin of 5%. When adding the cost for Manufacturing and Electronics it gives a total cost of $277.11 for the second generation of the Pick and Place.

Pick and Place – Preliminary Design Document

February 22, 2017/in 3D SMD/by Belinda Vivas

Belinda Vivas (Project Manager)

Amber Scardina (Mission, Systems, and Test)

Kevin Ruelas (Electronics and Control)

Tyler Jones (Manufacturing)

Chastin Realubit (Manufacturing)

Table of Contents

Program Objective/Mission Profile

By: Belinda Vivas (Project Manager)

Objective

The second generation of the Pick and Place will create a 3-Dot 454 PC Board. It will ensure precision through the addition of a camera system, addition of a variety of components, design upgrade, and a user friendly mechanism. A CSV file will be created for this generation, to implement a better interface between the user and the machine. A manual will be provided for this generation, as well as for the Madell Pick and Place for which extensive research is being done to implement the design of the camera (edge detection) system and software interface. A more detailed list of the customer’s needs can be found on the link provided.

http://web.csulb.edu/~hill/ee400d/S’17%20Project%20Objectives%20and%20Mission%20Profile.pdf

Requirements

Level 1 Program/Project Requirements

By: Belinda Vivas (Project Manager)

Through research on the Pick and Place first generation, the Madell Pick and Place, and the customer needs the following requirements will be implemented for the design of the second generation. This requirements will allow to define a clear structure for the design, engineering, and further requirements for the project.
The pick and place shall have an attached compartment to hold accessories for the pick and place.
The pick and place shall meet the EE 400D cabinet specifications for storage purposes.
The pick and place shall incorporate a camera for edge detection.
The pick and place shall have resources to aid the user of the pick and place in set-up and configuration of the pick and place.
The pick and place shall have a case to protect the pick and place from its surroundings.
The pick and place shall produce a 3Dot board in a specified time. This specified time is currently set for one hour, although after trade-off studies are conducted this time may change.
The user of the pick and place shall be able to set-up and configure the machine in a specified time. This specified time is currently set for two hours, although after trade-off studies are conducted this time may change.
The pick and place should have minimal movement from outside sources.
The pick and place shall cost no more than $500.
The pick and place shall be completed by Wednesday 17th, 2017.
The pick and place should have an emergency shut off button.

Level 2 System/Subsystem Requirements

Level 2 System Requirements of Pick and Place (1^st Generation) Analysis

By Amber Scardina (Mission, Systems, & Test)

Listed above are the Level 2 Requirements for the first generation pick and place. The Level 2 Requirements for the first generation demonstrates the functionality of the current pick and place machine. After analyzing these requirements, the following updates need to be made:

Size: The pick and place needs to be stored in the EE 400D cabinet.
More Components: Additional feeders and component trays need to be added to accommodate all components for a complete 3Dot EE 400D board. The reels on the first generation will be removed.

Given the updates listed above, the other level 2 requirements from the pick and place first generation should still hold true for the second generation, exhibiting the same functionality. After trade-off are conducted, more updates and level 2 requirements for the software and manufacturing subsystems may be required in order to successfully verify and validate the level 1 requirements. Additional Level 2 Requirements will be included in the second generation to make the pick and place user-friendly. In the next section below, the level 2 requirements for the pick and place second generation is listed.

Source Material:

Level 2 Subsystems Requirements of Pick and Place (1^st Generation) Analysis

By Amber Scardina (Mission, Systems, & Test)

The level 2 subsystems requirements for the pick and place 1^st generation can be found in the following link:

https://www.arxterra.com/spring-2016-smd-pick-and-place-machine-preliminary-design-document/

The subsystem requirements should still hold true for the second generation. The system should contain the original manufactured parts (except the reel mechanism), with the addition of new parts to verify and validate the level 2 system and level 1 program/project requirements. The level 2 subsystem requirements helped describe the functionality of the system that will be incorporated into training materials for the second generation.

Level 2 System/Subsystem Requirements of Pick and Place (2^nd Generation)

By Amber Scardina (Mission, Systems, & Test)

1. The pick and place shall have an attached compartment to hold accessories for the pick and place. Accessories for the pick and place may include: pump, paste, components, etc.

a. The dimensions of the pick and place and the attached compartment shall meet the EE 400D cabinet specifications.

b. The attached compartment shall not interfere with the functionality of the pick and place machine.

2. The pick and place shall meet the EE 400D cabinet specifications for storage purposes.

a. The legs of the pick and place should be raised to meet of the specifications of the attached compartment.

b. The Z/A axis should not be removed from the pick and place in order to be stored in the cabinet.

3. The pick and place shall incorporate a camera.

a. The camera of the pick and place shall be used to incorporate edge detection technology.

4. The pick and place shall have resources to aid the user of the pick and place in set-up and configuration of the pick and place.

a. The resources for the pick and place shall include a video tutorial on how to handle (set-up) and configure the machine.

b. The resources for the pick and place machine shall include a written manual to assist the user in set-up and configuration.

c. The pick and place shall include sample (test) files for the user to practice with the machine.

d. The resources for the pick and place shall include a section for troubleshooting the machine.

5. The pick and place shall have a case to protect the pick and place from its surroundings.

a. The case shall enclose the machine to meet the EE400D cabinet specifications.

6. The pick and place shall produce a 3Dot board in a specified time. This specified time is currently set for one hour, although after trade-off studies are conducted this time may change.

a. The pick and place shall be faster than human production time. Human production time for a 3Dot board is currently established as four hours.

7. The user of the pick and place shall be able to set-up and configure the machine in a specified time. This specified time is currently set for two hours, although after trade-off studies are conducted this time may change.

Source Material:

1. https://www.arxterra.com/spring-2016-smd-pick-and-place-machine-preliminary-design-document/

2. http://web.csulb.edu/~hill/ee400d/Lectures/Week%2004%20Modeling/b_L2%20Requirements.pdf

Design Innovations

Creative Solutions

By: Belinda Vivas

A creative exercise was executed by the members of the team to begin the process of brainstorming ideas to incorporate new innovative design ideas for the Second Generation of the Pick and Place. Focusing mainly on the problem of mounting the electronic components into the PC Board without them moving out of place. Also, how to implement a better overall design for a more user friendly generation.

https://drive.google.com/open?id=0B9iWYCBTJWEERHB6Y1BHX2owdFU

Systems/Subsystem Design

Product Breakdown Structure

By Amber Scardina (Mission, Systems, & Test) and Belinda Vivas (Project Manager)

The Product Breakdown Structure (PBS) demonstrates the updated system and subsystems for the Pick and Place 2^nd generation. The updates for the pick and place can be classified in three categories: Customer Interface, Software, and Manufacturing. These categories are described by the Level 1 Program/Project Requirements. The customer shall interface easily with the pick and place machine by giving the customer numerous resources for set-up and configuration. The software will be updated to accommodate customer interfacing, in addition to new edge detection software that will be written for the camera. A possible LCD may be incorporated to display the current electrical components that need to be loaded or placed. The pick and place will also undergo several manufacturing updates. The manufacturing updates on the pick and place includes a nozzle redesign to accommodate a camera to implement the edge detection software. The pick and place machine will also add more feeders and component trays to accommodate a 3Dot board with different type of components as well as an underneath compartment to store accessories for the machine. Accessories for the machine include: pump, paste, stencil, power cords, and electrical components. Other manufacturing additions include updating the specifications of the pick and place machine to accommodate the EE400D closet. After trade-off studies are conducted, the technical requirements for the pick and place 2^nd generation will be more clea defined.

Source Material:

1. https://www.arxterra.com/spring-2016-smd-pick-and-place-machine-preliminary-design-document/

2. http://web.csulb.edu/~hill/ee400d/Documentation%20Lecture%20Series/04%20Preliminary%20Design%20Document.pdf

Electronic System Design

By Kevin Ruelas (Electronics and Control Engineer)

Camera

A Cameras should be installed on the machine to incorporate edge detection software. Edge detection will function when the component is picked up and measured and placed onto the board. More research will need to conducted in order to develop the edge detection software.

Another design option: An additional camera can be used to detect whether a component is present on the tray or not and display an “error” message if the tray is empty. Cameras should be chosen so that it is able to capture a high resolution image of the smallest component the machine can pick up. (0402 size). They may also require flash LEDs in order to achieve a bright and crisp image.

LCD

A small screen should be installed and display the current component being picked up and placed onto the board. This display should also display the current status of the machine. Depending on its current operation, a user input pad will be considered. User should be able to input component coordinates and name and press a “Place” button to place the part. Size of the display and place of installation is still pending. Software for the LCD display will need to be developed.

Wires

The wiring for the pick and place should remain unchanged from the first generation.

Button

A kill switch should be implemented to isolate the machine from power in case of emergency or for maintenance. Software for the kill switch will need to be developed.

System Block Diagram

By Amber Scardina (Mission, System, & Test) and Belinda Vivas (Project Manager)

The system block diagram above describes the how information is sent in the pick and place system. The information is sent from the user using EagleCad, converted into Gcode, then sent to the microprocessor (Arduino Uno). From the Arduino Uno, the data is sent to a Me Orion board to control the motors for each axis. With the addition of a camera and edge detection technology, the system will allow for calibration of coordinates if necessary. Once the nozzle is confirmed to be at the correct coordinates, the data is sent to

Interface Definitions

By Amber Scardina (Mission, System, & Test) and Belinda Vivas (Project Manager)

The diagram shown above describes how the user will interface with the pick and place. Note: this method has remained unchanged from the first generation.

The link below shows the pin mapping of the Arduino Uno, which will be the same implementation for the second generation of the Pick and Place:

https://www.arxterra.com/spring-2016-smd-pick-and-place-machine-preliminary-design-document/#Design_and_Unique_Tasks

Mechanical Design

By Tyler Jones – Manufacturing

The basic overview of the current design for the pick and place is fully functional and can place 0402 components, as well as larger size components such as integrated circuits and 0603 sized components. The pick and place machine operates on a belt system in the X and Y axes. The Z axis contains the nozzle which can be lower and lifted using a linear actuator. The machine is controlled from an Arduino Uno microcontroller and Me Orion shield. Currently the software is being retested and calibrated for use.

The second generation requirements for the pick and place however will impact the current design. The second generation mechanical design will need to fulfill a size requirement of being able to fit in storage cabinets. Also a much more user friendly design must be incorporated so that the average user can easily manufacture boards without advanced technical knowledge. This will call for a smaller, more compact machine, and must be able to maintain the current precision and accuracy of the first generation. The Z- axis must be able to turn on a geared platform 90 degrees for setup and take down this eliminates the height problem.

The requirements also specify a speedy setup process so that boards may be manufactured quickly. This will call for an all in one style machine without any lose parts. Also the current design implements an automated reel feeder system. This will have to be re -designed however because the second generation will be designed to populate a 3DoT board, using a variety of more components. This renders the reel feeders to be useless because they can only provide thousands of only six types of parts. Additionally the current IC tray must be re-designed because it should also offer a more versatile range of ICs to be used.

In order to fulfill user friendly requirements the pick and place should incorporate a small emergency stop button on the side this will shut off the power to the machine if there are any errors or malfunctions being used. Each of these design elements; the new feeder system, the new sizing, new Z axis that can rotate, and the push button will be discussed further on in this article.

Second Generation Feeder Trays

The feeder trays can be made so that the grooves fit a universal tape reel size. This will allow multiple types of tape sizes to be fit into the grooves.

The trays can be 3D printed using ABS plastic and color coded according to part sizes.

Multiple trays can populate the table so that all parts can be utilized by the machine.

The groove size for the tape reels to fit snugly without stalling the tape is 0.317 inches.

The length of the groove should be enough so that at least 30 of each part can fit. This will safely cover the maximum number 3DoT board’s parts that are identical for each individual board design.

A tape peeling and reel feeding mechanism may have to be employed after exploring the software capabilities.

Downsizing Second Generation Pick and Place

The 2^nd Generation Pick and Place must be made smaller. This is so that the machine can be easily transported, and setup for ease of use. The image above represents a solid works draft of the machine. The legs can be lengthened by 4 inches providing a space below the pick and place so that the pump can be mounted.

This also makes it so that the pick and place can be stored in a cabinet in the upright position. It provides a convenient area to have the used part tape land, as well as an area for putting the spare components or component strips. The under tray can be made from aluminum or sheet metal so that the pick and place is still lightweight.

Getting rid of the reel feeder system provides more stability, because there is no longer an aluminum feeder and large reel hanging from the side of the machine. This also cuts down on height and width, as well as weight.

The absence of a reel feeder also creates more surface area to populate the picking surface with more IC trays and cameras.

Rotating the Z-Axis

The diagram above shows how the Z – Axis can be rotated by 90 degrees from vertical to horizontal.

Using a rack and pinion the whole Z – axis can be rotated and locked into position. This allows for more height clearance and ease of use. The Z axis stands 7.5 inches above the XY Axis. Allowing the Axis to be locked in 90 degrees from the Z axis eliminates all but about 1 inch to stand above the Axis. This creates enough room to store in a cabinet.

It is essential that the locking mechanism be very tight in order to prevent misalignment. It is also essential for the mechanism to be simplistic and small itself.

Emergency Button

The schematic above shows the sizing and dimensions for the emergency push button. This will be implemented as a way to stop the machine from processing any further components. The position of the button needs to be within reach of the operator, and on the same axis as where the parts are placed into the grooves.

Depending on the software development the emergency button may be implemented in power circuit as an override switch.

Design and Unique Tasks

By Amber Scardina (Mission, System, & Test) and Belinda Vivas (Project Manager)

Trade – Off Studies

Several will be conducted in the following weeks to confirm our level 2 requirements. The tradeoff studies conducted:

· Set Up time – The time to set-up and configure the machine by the user will be measured.

· Production Time – The time to pick and place an entire 3Dot board by the user will be measured.

· Material of Case – Possible materials of the case will be tested in order to support the weight of the pick and place machine.

· Material of Compartment – Possible materials for the compartment will be tested in order to set the weight specifications for the pick and place machine.

· Size of Component Trays – The “free space” on the pick and place machine will be measured to set the size specifications for the additional component trays.

· Size of the EE400D Cabinet – The EE 400D cabinet will be measured to set the size specifications of the pick and place machine.

Modeling

The modeling for the second generation pick and place should follow the software design of the first generation. Mechanical design will be upgraded for a more user friendlier generation and better calibration system.

https://www.arxterra.com/spring-2016-3d-smd-final-documentation/

Rapid Prototyping

The pick and place should be functioning as it is defined by the final documentation of first generation. The pick and place machine is currently in progress on setting up, running the software, and checking the wiring connectivity. Depending on how the tests run to implement the software we will decide on introducing a new software for the second generation (OpenPnP).

Spring 2016 3D SMD: Final Documentation

May 13, 2016/in 3D SMD/by BaoDoan

Table of Contents

Executive Summary

By Bao Loc Doan (Project Manager)

Program Objective

When humans manually pick up and place down surface mount components onto a printed circuit board (PCB), there are normally problems with efficiency. A surface mount component may be placed at the wrong spot or dropped. These mistakes cost the user money as well as time. An answer to those problems is a pick and place SMD machine. A pick and place surface mount device (SMD) is an automated device that can populate a PCB with surface mount components (resistors, capacitors, and IC chips) by referencing an EAGLE PCB file through the use of software. The pick and place SMD machine will be able to pick up the surface mount technology (SMT) components from 8mm reel feeders and an integrated circuit (IC) tray and place the components down at the correct location until the board is finished. The pick and place machine will replicate the error specifications of an industrial machine. The customer has expressed the desire to keep the budget of the project below $650 and finished before the end of Spring 2016.

Mission Profile

Once an EAGLE PCB file provided by any project from EE400D up until Spring 2016 is uploaded, the pick and place SMD machine shall begin populating SMT components from four 8mm reel feeder channels and one IC tray onto the PCB. The smallest SMT component that will be placed is component size 0603 (*the smallest possible component that the pick and place machine can pick up is 0402 component size*) and the heaviest component will be the ATmega32U4. One surface mount component will be rotated 45 degrees relative to starting position and be placed at a specified location. This process will repeat for 90 degrees, 135 degrees, and 180 degrees of rotation. The pick and place SMD machine will be modified from a Makeblock XY plotter and replicate the error specification of 0.05 mm (Madell Corporation Model DP2006-2).

Level 1 Requirements

These requirements provide traceability to the program objectives and mission profile. We created each requirement to be quantitative and be able to flow into system and subsystem requirements.

The SMD pick and place machine shall pick up and place down all SMT components provided by any EE400D PCB up until the end of Spring 2016.
The SMD pick and place machine shall be modified from an XY Plotter to have the same error specification of Madell Corporation Model DP2006-2 (n.d).
Software for the SMD pick and place machine shall accept all EAGLE PCB files of EE400D projects up until Spring 2016.
The SMD pick and place machine shall have four 8mm reel feeders and one IC tray.
SMT component size 0402 shall be the smallest component that the pick and place SMD machine can pick up.
ATmega32u4 chip shall be the heaviest component that the pick and place SMD machine can pick up.
Total cost of finished project must be under $650.
Deadline to complete the pick and place SMD machine shall be before the end of Spring 2016.

The Design

In order to have a clear understanding on how to proceed with the implementation stage, all of our systems and subsystems design needed to be complete. During this process, we had to create 3D models in Solidworks and decide what functions our machine should have in order to satisfy our requirements. The figure below will be the Solidworks model encompassing the systems and subsystems. There will also be an explanation for each subsystem and the project features that this pick and place machine has.

Figure 1 – Systems/Subsystems Model

The major systems and subsystems are the Z-axis actuator, automated reel feeders, vacuum nozzle system with A-axis stepper motor, aluminum surface, and XY-axis linear movement.

Z-Axis Actuator

This subsystem will control the “up and down” movement of our pick and place machine. A standard sized nema 17 Makeblock 42BYG stepper motor will control a threaded rod which has a bar attached to it was the base for our vacuum nozzle system. The original kit that was purchased is called Makeblock Thread Drive pack. We also had the height of the actuator cut due to the unnecessary large size of it. We found that our nozzle will shake when our Z-axis was in operation; therefore in order to eliminate approximately 80% of the shaking movement, we had to attach a rubber band to our nozzle. Due to this error, we decided to redesign our Z-Axis actuator. By using a linear slide actuator, we were able to completely eliminate all shaking of the nozzle caused by our previous thread drive kit. More information is provided in the mechanical design section.

Vacuum Nozzle System with A-axis Stepper Motor

A bracket was designed in order to hold a standard sized nema 17 Makeblock 42BYG stepper motor which will control the A-axis rotation of components. A 4mmx6.18mm rigid coupler will attach the shaft of the stepper motor to a T-connector (purchased from buildyourcnc.com) that has several nozzle fittings for 0402 sized components to IC chips. The vacuum will connect into a solenoid valve and then into the T-connector with aquarium tubing. The bracket design will be linked in our mechanical design section.

Aluminum Surface

The aluminum surface table allows the pick and place machine to have a static surface to operate on as well as the ability to attach our reel feeders, IC tray, the Makeblock XY Plotter, and origins. The aluminum surface blog will be linked in our mechanical design section.

Reel Feeders

The reel feeders is where all of our surface mount resistors and capacitors are docked. An aluminum bracket was designed in order to hold four continuous micro servos. These continuous micro servos will be for reeling and holding the plastic tape so the plastic tape does not go into the working area of the pick and place machine during operation. The reel feeders will be controlled by G-code and be automated by having our vacuum nozzle tip pull the cut tape forward before every picking of a component. The reel feeder solidworks model will be linked in our mechanical design section.

XY Axis

Originally, the Makeblock XY plotter came with their standard 42BYG stepper motors, but during our PDR and CDR demos, we came to the conclusion that the steppers were too slow and inaccurate. We decided to purchase the 42BYG 5:1 geared stepper motors from Makeblock in order to fix those problems. These new geared stepper motors will control our XY axis movement and will also be controlled by G-code. A trade-off study for the geared stepper motors will be linked in our experimental results section.

G-code Interpreter

Onboard our pick and place machine will be an Arduino Uno with a Makeblock shield attached to it. There will be C++ modules that will interpret G-code which will automate our machine. There will also be a java GUI interface that the user can use to send a G-Code file to, which will automatically send every line serially to our Arduino. A more in-depth explanation will be in our software design section.

System Design

By Christine Vu (Missions, Systems, and Test) and Henry Nguyen (Electronics and Control)

Figure 2 – Me UNO Shield Pinouts

For our machine, we used a Me Uno Shield attached to an Arduino Uno. The shield has 10 RJ25 ports. Each component for our machine is listed in the image above. All of our stepper motors are inserted into the red ports (1,2,9, and 10) which outputs 12V. More details on our Me Uno Shield can be found here:

https://www.arxterra.com/spring-2016-3d-smd-me-uno-shield-and-software/

Figure 3 – Update System Block Diagram from CDR

Based on these Me UNO Shield Pinouts, we were able to map out our system, which is shown in Figure 3. From the Me UNO Shield, there are 4 Me stepper drivers for the 4 stepper motors to control our axes and adapters to convert the cables to 1-2 pinouts. We used a total quantity of 6 RJ25 adapters to attach our axis limiting switches, reel feeder servos, and solenoid valve.

Experimental Results

By Henry Nguyen (Electronics and Control)

Origins and Scaling

By Henry Nguyen (Electronics and Control)

In order for us to accurately pick and place components on our pick and place machine, we must be able to identify our origins. We have decided that the bottom left corner of any PCB to be the origin. Once we determine our origins, our reel feeders and IC tray G-Code locations will be dependent on this. Details can be found in our blog post:

https://www.arxterra.com/spring-2016-3d-smd-origins-and-scaling/

Trade-off Studies

We had to perform several trade off studies in order to validate what products we can used based on our requirements. This will also allow us to avoid any future problems caused by choosing the wrong product.

Vacuum Pump

By Henry Nguyen (Electronics and Control)

For our vacuum pump, we needed to find a cheap but reliable vacuum pump for our machine. This pump and nozzle will need to be able to pick up components as small as 0402 and ICs as large as Atmega 32U4. In order to satisfy these requirements, we were able to change an aquarium pump into a vacuum pump. The details is linked below.

https://www.arxterra.com/spring-2016-3d-smd-vacuum-pump-trade-off-study-v2/

Solenoid Valve

By Henry Nguyen (Electronics and Control)

Our vacuum needs to be able to shut off and on based on the G-Code values we input. In order to do so, we needed a solenoid valve. We found that our solenoid valve heated up very quickly to high temperatures. In order to make sure the temperature did not exceed the melting point of our plastic tubing, we performed this trade-off this study.

https://www.arxterra.com/spring-2016-3d-smd-solenoid-valve-trade-off-study/

Geared Stepper Motors

By Loc Doan (Project Manager)

We upgraded our X and Y axis stepper motors to geared stepper motors from Makeblock. Geared stepper motors allowed us to achieve the precision we needed while half stepping verses our 16th stepping from our old stepper motors. Details can be found here:

https://www.arxterra.com/spring-2016-3d-smd-geared-stepper-motor-trade-off-study/

Rapid Prototyping

For our machine, we needed to perform rapid prototyping in order to ensure that our machine works the way we want it to before purchasing parts. We used extra brackets and pieces from our X-Y plotter in order to rapid prototype. Once we ensure that everything runs properly with our software we were able to create a design of parts that we needed to get manufactured.

Z-Axis

By Henry Nguyen (Electronics and Control)

For our Z-Axis, we purchased a Thread Drive Pack from Makeblock which is able to move our thread drive beam up and down with great precision. However, we wanted to control the thread drive using a stepper motor vs the given DC Motor. In order to do so needed to rapid prototype brackets and run tests using our X-Axis software and ports. Details can be found in the link below; however, due to accuracy errors, we redesigned our Z-Axis with a linear slide actuator which can be found in the mechanical design section.

https://www.arxterra.com/spring-2016-3d-smd-z-axis-rapid-prototype/

Subsystem Design

Interface Definition

By Christine Vu (Missions, Systems, and Test)

To help us obtain the correct pinouts for our system, we decided to map out the Me UNO Shield and decide which pins we needed. Port 5 was noted as unusable due to its use of bluetooth serial communication.

Figure 4 – Me UNO Shield Pinout

Each port has a general format and has 1-2 analog/digital pinouts. Figure 5 shows the general format for a Me UNO Shield port. On the left side is our microprocessor pinouts, ATmega328. The middle column indicates the pinouts for our microcontroller, which was the UNO. The right are the RJ25 connector ports for our Me UNO Shield. One port contains 6 pins, each for I2C communication, which we did not need to use, Vccm, ground, and 1-2 analog/digital pins.

Figure 5 – General Port for Me UNO Shield

Based on our system block diagram, the interface definition indicates which specific pins were used for the subsystems — stepper motors, limiting switches, reel feeders, and solenoid valves.

Figure 6 – Stepper Motor Controls

The Arduino Code variables in Figure 6 were originally based on the Makeblock X-Y Plotter Robot Kit code, but had to be modified due to switching from the Makeblock Orion to the Me UNO Shield.

Figure 7 – Reel Feeder, Limiting Switch, and Solenoid Valve control

As shown in Figure 7, most of the ports hold two different analog/digital pinouts, so the limiting switches, reel feeder servos, and solenoid valve controls were grouped. The two ports, Port 4 and Port 6, are shown with only one digital pin, so we decided to go with the solenoid valve and one of the limiting switches.

Hardware Design

The mechanical designs were generated based on the project’s subsystems, which were the aluminum surface, X,Y,Z,A – Axis and origin brackets, Z-axis thread beam, and z-axis linear slide actuator.

Aluminum Surface Table

By Henry Nguyen (Electronics and Control)

Our pick and place machine required an aluminum surface table in order to place our PCB, reel feeders, and IC tray. This surface table will allow us to have a flat surface in order to pick and place components. Details on the design of the aluminum surface table can be found here:

https://www.arxterra.com/spring-2016-3d-smd-aluminum-surface-table/

X, Y, Z, A -Axis and Origin Brackets

By Henry Nguyen (Electronics and Control)

Since our machine is first of its kind for EE400D, we had to create everything from scratch that was not purchased from Makeblock. Since we had to customized our machine, we needed new brackets with exact drilling in order to attach our X, Y, Z, and A-Axis stepper motors. We also had to manufacture a bracket to hold our PCB in place. Details can be found here:

https://www.arxterra.com/spring-2016-3d-smd-x-axis-y-axis-z-axis-a-axis-and-origin-brackets/

New Z-Axis Thread Beam

By Henry Nguyen (Electronics and Control)

We found that our current thread beam on our Z-Axis thread drive would cause a lot of accuracy error. This beam would wiggle a lot because the holes on it is slightly too large meaning our 4mm rod became too small. The thread beam will now tilt forward due to the weight of our A-Axis stepper motor. Now that the thread beam is tilting forward, it caused more friction on our 4mm rods forcing our stepper motor to stall. A new design of our thread beam can be found here:

https://www.arxterra.com/spring-2016-3d-smd-z-axis-beam/

Z-Axis Linear Slide Actuator

By Henry Nguyen (Electronics and Control)

After testing our Z-Axis thread beam, we found that although our accuracy greatly increased from before, we would still have some accuracy issue when our beams moves up and down. When moving down our nozzle would shift to the left and when moving up the nozzle will shift to the right. This was not a major problem because although the nozzle will shift to the left when moving down, it will always shift in the same position towards the left every time. Although we would still be able to pick up components, we wanted our machine to be accurate and have a smooth linear movement. We decided to purchase a linear slide actuator using a lead screw. Details can be found here:

https://www.arxterra.com/spring-2016-3d-smd-z-axis-linear-slide-actuator/

Creating an Automated Reel Feeder From Preliminary Design

by Nasser Alsharafi (Manufacturing)

The basic design included the fabricating of an aluminum base, and axle to dispense SMD parts. I tried to go into detail of the design by figuring out how to automate the reel feeder without adding too complex of a system design to the project. This included the addition of a servo platform, four servos, and channels for the tape to run through. The servos were chosen to rotate many times in order to collect the tape and move the tape at the same time. The servos were positioned on a custom fabricated platform at a 45-degree angle to optimize the tape peeling off and disposed of into four spools attached to the reels. This method worked much more efficiently and simply than the other designs, and fully automated the reel feeder after programming.

https://www.arxterra.com/spring-2016-3d-smd-reel-feeder-bracket/

Designing a Universal IC Tray

by Nasser Alsharafi (Manufacturing)

The developments of the project lead to the need to create an IC tray with universal chip, and part sizes. I designed an IC tray from ABS plastic material by exporting a Solidworks file to a 3D printer. It is critical that the Tray can successfully contain the main ATMega32U4 IC, as well as other widely used IC sizes. The tray features 21 compartments described in detail in the link below. The IC tray needed to be designed to be small enough to save real estate on the machine table. It also needs to be less than 0.3 in height from the table so that the machine only lifts a maximum amount of 0.3 inches on the Z – axis. This I was able to configure in order to make the machine speed more efficient.

https://www.arxterra.com/spring-2016-3d-smd-ic-tray/

Software Design

The initiation of our machine is when one line of G-Code is sent serially to the Arduino. If there is data, then the process will begin deciphering which type of control we would like to the machine to output, such as checking whether to turn on the solenoid valve, activating a rotation for the reel feeder servo, or moving to a coordinate. Figure 8 is the general process for our software.

Figure 8 – Software Block Diagram

Arduino Coding

By Christine Vu (Missions, Systems, and Tests) and Henry Nguyen (Electronics and Control)

Our arduino software has 4 modules: G-CodeParser.ino, process_string.ino, stepper_control.ino, and move_reel.ino. G-Codeparser.ino, process_string.ino, and stepper_control.ino are modules obtained from Makeblock, which is an open source company. Since we used their X-Y Plotter Robot Kit as our baseline of our project, we used their software. In order to adjust the software to our needs, we wrote codes several modules. We included an off and on switch for our solenoid valve, four reel feeder servos, and two new stepper motors. More information on how our software was modified can be found here. It includes flowcharts and explanations for each module we modified and created:

https://www.arxterra.com/spring-2016-3d-smd-software/

Java

By Henry Nguyen (Electronics and Control)

For our machine, we had to tap into the Java code for our GRemote.gui in order to initialize our X, Y, and Z-Axis stepper motors and limiting switches. Details can be found here:

https://www.arxterra.com/spring-2016-3d-smd-me-uno-shield-and-software/

Conversion of Eagle file to G-Code

By Christine Vu (Missions, Systems, and Test) and Henry Nguyen (Electronics and Control)

One of the main requirements for our machine is to be able to receive an Eagle board file and convert it into G-Code in order to know the proper locations of where we need to place our components. We found that we can convert Eagle to DXF and from DXF to G-Code. Details can be found here:

https://www.arxterra.com/spring-2016-3d-smd-conversion-of-eagle-file-to-G-Code-initial-process/

Verification & Validation

By Christine Vu (Missions, Systems, and Test)

The Validation matrix created was based on the program objectives and the goals that the customer wanted the team to achieve. Below is a list of functions our machine performed for the customer to evaluate.

Table 1 – Validation Matrix

Verification tests were conducted to reflect the requirements on the physical design of the SMD pick and place machine. A list of procedures were presented in the verification test plans. Details and results are found here:

https://www.arxterra.com/spring-2016-3d-smd-verification-test-plan/

Project Update

Work Breakdown Structure

By Bao Loc Doan (Project Manager)

A work breakdown structure was essential to the completion of the project. The WBS allowed each engineer to understand their roles and what tasks they needed to focus on completing. Our WBS had several iterations as the semester went on and each iteration there would be something new added or revised. This was necessary because there were some things that were missing and some new design ideas that were implemented. The final pick and place SMD machine WBS will be shown in the figure below.

Figure 9 – Work Breakdown Structure

Resource Reports

Mass Report

By Christine Vu (Missions, Systems, and Test)

The purpose for our mass report is to assure that the weight on the Z-Axis would not possibly obstruct the progress of picking and placing. Since the Z-Axis was on the X-Axis rods, we considered the weight on the X-Axis. The project allocation of 2000g is from a rough test we conducted by placing a heavy object, which was approximately 2000g, and determined that this the motors almost began to stall. Table 2 indicated that our actual mass was 767.47g, which is two times less than our allocated mass.

Table 2 – Finalized Mass Report

Power Report

By Christine Vu (Missions, Systems, and Test)

Our resource for power is limited by the CSULB’s power lines and our power adapter. It is important to note that the actual current draw, which is shown at the bottom, is the maximum amount that would have been drawn at one time. The Total Expected Current Draw will not be met because that would be assuming each component/subsystem is operating.

Table 3 – Finalized Power Report

Cost Report

By Bao Loc Doan (Project Manager)

Our original allocated project budget was $650. The total cost of the project ended up being $756.94. We asked for permission from the customer each time we went over budget and got approval. A good bulk of the cost of this project was due to the need of hardware that was not readily accessible. It was tough to find any local hardware shops that could supply the parts that we needed at the price we wanted. We looked on McMaster Carr but their prices were much too high for our budget (~$37 dollars per coupler). Anything we bought from Makeblock or Pololo shipped from China, and this resulted in high shipping costs. The cost report will be linked below.

Table 4 – Finalized Cost Report

Project Schedule

By Bao Loc Doan (Project Manager)

We were able to successfully complete our project schedule by the end of the semester. Our project schedule and Gantt charts were created in ProjectLibre. We were able to export our ProjectLibre schedule file later to Microsoft Project to produce burndown graphs and the project overview. The project schedule was extremely helpful because by CDR we were behind schedule but since we knew what tasks we were behind on, those tasks were focused on and completed before we moved on.

Figure 10 – Gantt Chart from ProjectLibre

Figure 11 – Burndown Plot

Figure 12 – Project Overview of Completion

Concluding Statement

By Bao Loc Doan (Project Manager)

Although we completed all of our requirements for this semester, I feel like there are some things that can be added to improve the functionality of the device. If we had more time and resources, there are two features that we would like to implement onto this pick and place machine. First, we would implement a second nozzle onto the z-axis actuator that could extrude solder paste. Applying solder paste before setting the component down onto the PCB is a critical part of the success of this pick and place machine. The solder paste will allow the component to be set down properly rather than possibly vibrate out of position when the solenoid valve cuts off the flow rate from the vacuum. Second, we would implement a camera self-calibration system using an inexpensive usb endoscope. The self-calibration system will save the user a substantial amount of time as it will automatically determine the origins and error rate and self correct for the entire pick and place machine.

There are some lessons that I took away from this semester. When we first started this project, the main concern was the software. Although the software was indeed the largest issue of this project, the hardware was nearly equally as important. Understanding what parts were needed to maintain our accuracy was critical. At the end of the semester, we had the software portion of this project finished but due to janky hardware, we didn’t have the right accuracy. We went over 5 iterations of a simple z-axis beam until our accuracy was satisfactory. Another important lesson is that this project will require many iterations. Do not get discouraged by how many iterations that will eventually be redesigned and reimplemented. This was the first semester of the project so there was no backbone for us to go off on except for research we did on the internet. Sometimes what we did find couldn’t work for our needs because of the limited resources we had. Another lesson we learned the hard way was issue of miscommunication. Communication was critical in this project. During this semester, three months were spent on what was supposed to be a simple reel feeder bracket design. This was due to a communication issue between me and our manufacturing division. Once we got the issue straightened out and we received help, the bracket was finished in a weekend. Other than that, we really did enjoy doing this project. It was great seeing something come alive from the very beginning of the process to the end.

Resources

Project Video – https://youtu.be/KDmNSbDamOA

CDR – https://drive.google.com/folderview?id=0B1CO1_x-4KFOd19ESlhYejJXZHc&usp=drive_web

PDR – https://drive.google.com/folderview?id=0B1CO1_x-4KFOa1g5Nk1NMnQ0Mkk&usp=drive_web

ProjectLibre File – https://drive.google.com/folderview?id=0B1CO1_x-4KFOWTdaeXBhRXlOLVU&usp=drive_web

Verification and Validation Documents – https://drive.google.com/open?id=0B1CO1_x-4KFOWjlDZXRuS3RVSkk

Solidworks Files – https://drive.google.com/open?id=0B9oA00YQwNGMT1VzMWZpSDdjak0

EagleCad Files – https://drive.google.com/folderview?id=0B1CO1_x-4KFOSzBwZHVNY05SUzA&usp=drive_web

Arduino and/or C++ Code

*** Link was added to this blog post: http://arxterra.com/spring-2016-3d-smd-software/ ***

https://github.com/cvuchristine/MakeblockPlotter.git

Other Applications: G-Code Files –

https://drive.google.com/folderview?id=0B1CO1_x-4KFOMGpFTC11dXV1UzQ&usp=drive_web

Complete BOM – https://drive.google.com/folderview?id=0B1CO1_x-4KFONjBGMVNLMl9sMG8&usp=drive_web

Spring 2016 3D SMD: Verification Test — Working Area

May 12, 2016/in 3D SMD/by BaoDoan

By Christine Vu (Missions, Systems, and Test) and Henry Nguyen (Electronics and Control)

Working Area Test

Working area is defined by the area that includes PCB to assemble, reel feeders, and IC tray. To verify that our design meets the size criteria for working area, a test will be conducted to calculate its parameters.

Requirements

Section 1 EE400D Assembly

The SMD pick and place machine shall pick up and place down all SMT components provided by any EE400D PCB up until the end of Spring 2016.

Section 1.1

Working area must be within 12.2”x15.35”(310mmx390mm) based on the Makeblock X-Y

Plotter Robot Kit.

Section 1.1.2

Surface to hold PCB shall be smooth with dimensions tolerances to be ±.001″, parallelism to .001″, and flatness to .001″ (TCI Precision, 2005).

Section 1.1.3

All pick and place processes shall begin at the bottom left corner of the PCB.

Section 1.1.4

All moving cables using RJ25 connectors shall have a minimum bend radius of 10x its diameter (Telecommunications Industry Association, 2001).

Section 1.2

Pick and place SMD machine shall change the orientation of each SMT component before placement.

Applicable Tools

Equipment Type	Name (Brand)	Tolerance
Mitutoyo S/N 12519090	Caliper	0.001”
Measuring Tape	Stanley 33-425 Powerlock 25-Foot by 1-Inch Measuring Tape	+/- 3%

Procedure

Attach Z-axis and all other subsystems.
Using X- and y-axis limit switches, move the X-axis and Y-axis at the maximum and measured the allocated work area.
Generate an EAGLE file by placing capacitors on each corner of the design.
Convert to GCode and verify the dimensions of the PCB Outline.
Using another Gcode file, test the rotation movement of an SMT component by rotating about 180 degrees in 45 degree increments.

Results

The following table indicates the parameters for our working area. The orientation of the EAGLE file parameters are 4” along the X-Axis and 3.2” along the Y-Axis. It is important to note the minimum reel feeder distance from the X-Axis

Working Area Parameters	Distance (in.)
X-Axis	12.6
Minimum Reel Feeder Distance from X-Axis rods	7.034
EAGLE Outline	4.001 x 3.2002

Rotation movement for procedure 5 can be done using G-Code programming operations. Since our rotation movement is using a stepper motor, we decided to half-step the orientation for 45 degrees.

Rotation Coordinate	Degrees
B5	45
B10	90
B15	135
B20	180

Conclusion

Our working area measurements indicate that each subsystem is able to fit in our X-Y Plotter. The rotation of each SMT component can be done using G-Code software.

Spring 2016 3D SMD: Verification Test — Reel Feeder

May 12, 2016/in 3D SMD/by BaoDoan

By Christine Vu (Missions, Systems, and Test) and Henry Nguyen (Electronics and Control)

Reel Feeder Design

To verify the reel feeder design has the correct function, a test will be conducted to determine its parameters and design.

Requirements

Section 4

The SMD pick and place machine shall have four 8mm reel feeders and one IC tray.

Section 4.1

All SMT resistors and capacitors shall remain in cut-tape of the reel feeders until the vacuum nozzle is ready to pick up the component.

Section 4.1.1

All reel feeders shall be installed on the working area, 12.2”x15.35”(310mmx390mm).

Section 4.1.2

Bracket to hold cut-tape of the reel feeders shall be higher than 0.04″.

Section 4.1.3

All motors used to peel off cut-tape of reel feeders must rotate 360°.

Section 4.2

IC tray shall store all IC chips required for one PCB assembly.

Section 4.2.1

IC tray shall be installed within working area, 12.2”x15.35” (310mmx390mm).

Applicable Tools

Hardware:

Reel Feeder

Micro Servo FS-90

Equipment Type	Name (Brand)	Tolerance
Mitutoyo S/N 12519090	Caliper	0.001”

Procedure

Install reel feeder to the aluminum surface. Attach reel with SMT components.
Install the IC tray to the working area.
Install micro servo to reel feeder bracket. Servo wheel shaft shall be placed at an angle of approximately 45°. Record Servo wheel angle.
Attach Z-axis and move the X-Y-axis motors to a component. Measure the clearance around the Z-axis. Record the clearance tolerance, which includes the distance around both the reel feeder and IC tray. Measure the height of the IC tray.
Operate the micro servo by turning 360°. This requires timing of the speed of the micro servo and must be determined in the Arduino Code. Record this time.
Determine initial measurement of the reel on the reel feeder. Record this measurement
With the cut-tape attached to the micro servo wheel shaft, reel in the cut-tape.
Observe the elevation of the reel feeder. Record the height of the reel feeder.

Results

	Parameter	Can the vacuum nozzle reach? [Y/N]
Microservo rotation	>360°	Y
Distance between Nozzle and X-Axis Limiting switch (in.)	3.011	Y
Distance between Nozzle and cut-tape (in.)	0.515	Y
Distance between Reel Feeder component and X-Axis Limiting Switch (in.)	4.161	Y
Bracket for microservo height (in.)	4.214	Y
Distance between microservo and end of cut-tape (in.)	2.975	Y
Distance between vacuum nozzle and IC tray (in.)	4.021″	Y

The angle of the cut-tape:

It was observed that increasing the angle at which the cut-tape was reeled required more force due to the way we changed the angle. If we changed the angle to approximately 60 degrees, it required us using the back bracket attached to the reel feeder and more force to pull the cut-tape.

Placement tolerance between the edge of reel feeder closest to the X-Axis and the nozzle:

4.161” – 3.011” = 1.15”.

Conclusion

Based on these parameters, we were able to conclude that our placement of the reel feeder system will be reached by the vacuum nozzle. Maximum rotation is more than 360°; however, the cut-tape would not stay in place of the servo if the wheel is holding too much of the cut-tape. Increasing the angle at which the cut-tape is allowed to turn resulted in ripping the cut-tape.

Spring 2016 3D SMD: IC Tray

May 10, 2016/in 3D SMD/by BaoDoan

By Nasser Alsharafi (Manufacturing)

The Custom 3D Printed IC Tray

The custom 3D printed IC tray is designed to house 21 evenly spaced different sized common IC components. One of the largest and most desired ICs we will be using is the ATMega32U4 IC. This IC is the standard for which we will calculate our pressure and rotation of the A-Axis. The different sizes are placed in order or square mm from largest to smallest respectively. The size of the IC must be less than 0.3 inches from the aluminum table. This is due to the optimization of the machines speed. Therefore the total height of the IC is 0.25 inches. The well sizes of the 21 part compartments are 0.1 inches deep. The entire IC tray is 4.45 inches long by 3.75 inches wide. The IC tray has four holes that allow the IC tray to be attached to the aluminum table.

The compartments fit components the following components:

SO8 x3
SO14 x3
SO16 x3
ATMega32U4 x3
Standard 14 mm x 14 mm square ICs TQFP x3
SO28 x3
Standard 12mm x 12 mm square ICs TQFP x3

Fig 1 ATMega32U4 Dimensions

Figure 2. Above shows the dimensions of the ATMega32U4 IC. The D/E dimension shows the size in mm of the IC. The IC is approximately 12mm x 12mm from pin to pin.

Spring 2016 3D SMD: Reel Feeder Bracket

May 10, 2016/in 3D SMD/by BaoDoan

By Nasser Alsharafi (Manufacturing)

Table of Contents

The Reel Feeder

The Reel Feeder shown above is a major component of the pick and place machine. This component is fully automated, and has custom fabricated aluminum parts. It is made up an aluminum base, four SMD part reels, four servos, four legs, and one axle, 4 spacer drums.

The base has four cut channels that are 8mm wide. There are four reels that are positioned on an axle that spins freely. This forces the component tape in the reels to be channeled through the 8mm grooves. There are additional legs that are bolted to the base that hold the axle and the SMD part reels.

The Base

The base is (10.0 inches long x 4.0 inches wide) the base has four channeled grooves spaced by 8mm across. The base has a tape guard that spans the entire width this is to ensure that the tape stays in the grooved channel.

The Reels

The four reels made of plastic that are positioned on an axle that spins freely. The reels spin and dispense the parts onto the aluminum table. The part reels are 5 inches in diameter. Two 5-inch long legs bolt into the base hold up the reels and axle. The reels hold SMD component tape. A plastic cover seals the tape. The size of the components in the SMD tape are very small at 0402.

The Custom Servo Platform

The four servos are mounted onto an adjustable platform by 8 screws. There are four spools attached to the servo, which contain the tape as it wraps up the plastic. The platform is set at 45 degrees from the base. This allows the plastic cover to be pulled off the SMD part tape, and advances the tape simultaneously. The device has legs for the custom bracket that are mounted by 2 screws to the base. The custom servo platform is fabricated from a rectangular piece of aluminum. The servos placed in a 0.9 inch cut out section of the aluminum are evenly spaced out the entire length of the 4.0-inch platform.

Spring 2016 3D SMD: Z-Axis Linear Slide Actuator

May 9, 2016/in 3D SMD/by BaoDoan

By Henry Nguyen (Electronics and Control)

Introduction

For our Z-Axis actuator, we found that our thread drive will constantly cause our nozzle to shift towards the left when going down and towards the right when going up. This was caused by the two 4mm rods and the thread screw we were using. A slight solution was to attach a rubberband to our nozzle to prevent it from shifting. The following Youtube video shows the problem our nozzle has when going up and down:

https://www.youtube.com/watch?v=Rw_LVb8K-G4

We found that in order to prevent this type of error, we had to have a completely new design for our Z-Axis actuator.

Linear Slide Actuator

We found a cheap linear slide actuator as shown in the title picture on eBay. After purchasing this sliding unit, we found that the slide will be sturdy enough to attach our A-Axis stepper motor as well as keeping the nozzle in a linear fashion. I needed to design a new bracket that will screw into the sliding unit as well as attaching to our red nut on the lead screw.

Figure 1. Linear Slide Bracket Sketch

For my bracket design, I would attach the linear sliding unit to the top of my bracket with the four holes. The red nut will go through the large circle in the middle. The two screws on the bottom will allow us to attach our A-Axis stepper motor bracket. The last screw hole on top of the A-Axis stepper motor bracket will allow us to screw into the red nut. This will keep the red nut in place and allow us to move the hole unit up and down.

Figure 2. Linear Slide Actuator Bracket

For the bracket, I had to apply some silicone to the red nut in order for it to stay in place. The screw did help to keep it in place; however, I wanted to make sure the red nut is completely stationary.

Figure 3. Full View of Linear Slide Actuator

Figure 4. Side View of Linear Slide Actuator

I uploaded a Youtube video showing the linear slide actuator in action.

https://www.youtube.com/watch?v=UqnxF44cEr8

Conclusion

Overall, this linear slide actuator is a success. It completely eliminated all wobbling caused from our old thread drive kit from Makeblock. We are able to go up and down smoothly as shown in the Youtube link above. This will allow us to rerun all of our accuracy tests to ensure that we are as accurate as possible when picking up surface mount components.

Spring 2016 3D SMD: Verification Test — Cable Fatigue

May 8, 2016/in 3D SMD/by BaoDoan

By: Christine Vu (Missions, Systems, and Test)

Table of Contents

Cable Test

A test will be conducted to determine that the bending radius will have a minimum bend radius of 10x its cable diameter.

Requirements

Section 1 EE400D Assembly

The SMD pick and place machine shall pick up and place down all SMT components provided by any EE400D PCB up until the end of Spring 2016.

Section 1.1.4

All moving cables using RJ25 connectors shall have a minimum bend radius of 10x its diameter (Telecommunications Industry Association, 2001).

Applicable Tools

Equipment Type	Name (Brand)	Tolerance
Mitutoyo S/N 12519090	Caliper	0.001”
Measuring Tape	Stanley 33-425 Powerlock 25-Foot by 1-Inch Measuring Tape	+/- 3%

Background

According to previous studies (Weaver & Gutierrez, 2015), wire fatigue may occur during repetitive movement. The Makeblock X-Y Plotter Robot Kit moves in a confined area with multiple repeated values of X-coordinates and Y-coordinates. Therefore, our project may be subject to repetitive movement that may cause damage to the cables.

The Makeblock company overall lacks a large amount of datasheets for their parts, so it was difficult to determine the standards of their RJ25 cables and connectors. It was noted that it was a 6P6C connector, so we decided to reference Digi-Key’s supplier, Assman WSW Company, whose cable diameter size is 26 AWG. The datasheet is shown in Fig. 1.

Figure 1. Reference datasheet for AT-S-26-6/6/B-xx-R.

In addition, we referred to commercial telephone standards (Telecommunications Standard Association, 2001), which noted that wire bend radius should have a minimum of 10 times its cable diameter.

Due to the funding of this project, we will not imitate the rotary fatigue tests as done by Weaver & Gutierrez (2015). As a replacement of this test, we will measure the cable bend radii at each important area to assure compliance.

Diameter of entire cable = 6.8 ± 0.2 mm

Worst-case scenario = 6.6 mm

6.6 * 10 = 66 ± 0.2 mm = 2.59 ± 0.007 in.

Therefore, minimum bend radius shall be 2.59 ± 0.007 in.

Procedure

Connect the wires attached to the X-Axis and Y-Axis.
Control the Y-Axis and move to the farthest end of the plotter with respect to the PCB until the limiting switch is hit. Record the length of the cable.
Control the Y-Axis to the other side of the limiting switch. Record the bend radius as shown as the photo below.The bend radius will be measured at the curved cable touching the aluminum surface.

Figure 2. Obtained from: https://en.wikipedia.org/wiki/Bend_radius#/media/File:Bendradius.svg

Results

Locations

The locations of each measured area is shown below.

Figure 3. Cable #’s of the RJ25 cables.

Cable #	Y-Limiting Switch (in.)	X-Limiting Switch (in.)	Requirement Met? (Y/N)
1	N/A	3.843”	Y
2	N/A	3.605”	Y
4	4.587”	N/A	Y
5	4.655”	N/A	Y

Table 1. Cable Bend radii measurements.

Conclusion

Cables using RJ25 connectors can have a minimum bend radius of 2.59″ +/- 0.007″. The smallest bend radius was 3.605″, which is larger than the required amount. Therefore, this requirement is passed.

References

Assmann WSW Components. (2013). “Modular cable – Straight 6P6C 50u”, 26 AWG, Black, 5-50 feet”. Datasheet.

Telecommunications Industry Association. (2001). TIA/EIA-568-B.1: Commercial Building

Telecommunications Cabling Standard.

Weaver, J. , & Gutierrez, E. (2015). Comparing rotary bend wire fatigue test methods at different test speeds.Journal of Materials Engineering and Performance, 24(12), 4966-4974.