P16601: Glass Cutting Machine:Guide Rollers
/public/

Preliminary Detailed Design

Table of Contents

Key Questions for Preliminary Detailed Design

Our goal for Phase 4 was to answer the following questions:

  1. Will we model off of the DS264 or develop a unique design?
  2. How many rollers will we use and what is the approximate size? Two Rollers - 420 mm long with an outer diameter of 320 mm
  3. Where is the most power being used? During the cutting process but the inertial power is not negligible
  4. How powerful do the motors need to be? Approximately 7.5 hp
  5. What material will the rollers be made of? 7075 Aluminum

Mechanical System Design

Design Selection Pugh Charts

public/Reviews/MSD1 P4 Preliminary Detailed Design/Roller Choice.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Roller Choice Total.JPG

The live document can be found here

public/Reviews/MSD1 P4 Preliminary Detailed Design/Roller Manuf Choice.JPG

The live document can be found here

Concept Design

public/Reviews/MSD1 P4 Preliminary Detailed Design/Design Sketch.jpg

Installation of Guide Roller

Installation of Guide Roller

3D Model

Full Assembly

Full Assembly

Front Insert Assembly

Front Insert Assembly

Motor Side Insert Assembly

Motor Side Insert Assembly

Subsystem Design

Roller - Led by Evan N.

Full Guide Roller

Full Guide Roller

Exploded Uncoated Guide Roller

Exploded Uncoated Guide Roller

Adapter Plates - Led by Noah S.

Front Adapter Plate

Front Adapter Plate

Motor Side Adapter Plate

Motor Side Adapter Plate

Bearings - Led by Ken W.

SKF Tapered Roller Bearing

SKF Tapered Roller Bearing

Frame - Led by Samantha D.

Framing

Framing

Insert Mounting Plate

Insert Mounting Plate

Material Selection

7075 Aluminum

6061 Aluminum

Additional materials to be determined

Electrical System Design

Power Source for Machine

The smaller machine will still consume a fair amount of power probably at least more than 20kW therefore 3 Phase power at 480 VAC would be better in terms of wire sizing, flexibility (more options available in terms of motors, drives, etc.), and because motors at 5 to 7.5 hp at 230 VAC are hard to come by. At 20kW, the required current at 230 VAC would be about 87 amps (3 AWG wire at 125% overload) while at 480 VAC the amperage would be about 42 amps (8 AWG at 125% overload). It would most likely even be cheaper to build a 3 phase 480 VAC machine. In the industry this machine would be designed for 480 VAC 3 phase and this was also recommended by David Cicero, our Rockwell Design Expert.

Regenerative Analysis

The DS264 currently has an Active Front End (AFE) power module that powers all the major motor AC drives. This module produces the 750 V DC Bus that provides power to the AC Drives powering the guide roller motors as well, the spool motors, the work piece feed motor, and others. When the motors are "motoring" power is supplied from the AFE module to the drives to the motors. When the machine is decelerating, the DC Bus becomes regenerated from the kinetic energy and can even be converted to 480 VAC back on the main lines. The current AFE Module is a 110 kW module, so is fairly large. The smallest AFE modules offered by Rockwell are about 38.4 kW, which just might be too large for what is necessary for our smaller machine.

A simple regenerative analysis was done to see how much energy could potentially be harvested from the DS264. The major components were the guide rollers and their motors and the spools with their motors. Ignoring friction, a good amount of energy could be potentially harvested.

Regenerative Analysis Ignoring Friction

Regenerative Analysis Ignoring Friction

The potential for some decent energy savings exist if the machine is really run frequently. However, running the machine frequently would also cost more as well. However, there is a large problem with the guide roller system when friction is added in. The equation below shows that at high enough friction the deceleration will actually be much greater than 2 m/s^2 and will therefore require the motors to still run.

Deceleration Due to Friction

Deceleration Due to Friction

When adding friction into the picture, the DS264 should theoretically produce no regenerative energy from the guide rollers when cutting the glass or even when there is just bearing friction. On the other hand, the smaller machine would offer approximately 335 W if bearing friction is considered. At a smaller number of cuts the bearing friction would be smaller so at absolute most, the rollers would require 500 W of heat dissipation if not cutting.

Regeneration with Friction

Regeneration with Friction

Therefore, the savings calculated earlier would cut in half for the DS264 and the savings for the smaller machine would cut in half again. Therefore, at most 100 to 150 $ would be saved per year possibly. This would not offset the cost of investing in an AFE Regenerative Module even after 10 years of operation at 90% up time. Therefore, the AC drives will simply have Dynamic Braking Resistors or Braking Modules in order to dissipate the extra energy produced.

The live document can be found here

Motor Selection

Here is where our search for motors took us at first before hearing from Dave Cicero. The reason these won't work too well for our application is that they are induction motors which do not have great control performance. We started looking into Rockwell Servo motors since they use different controllers (much faster refresh rate than Powerflex)and are permanent magnet motors with absolute encoders perfect for our application.

Rockwell AC Motor Research

Rockwell AC Motor Research

Then after a meeting with David Cicero from Rockwell, we used Rockwell's motion analyzer software to analyze our system and recommend possible solutions. The following table represents some possible solutions we encountered, though is far from an exhaustive list. We have narrowed our motor search to three different motor models: Medium Inertia Motors (MPM), Low Inertia Motors (MPL), and Rotary Direct Drive Motors (RDB). A lot more work needs to be done in order to specify the exact solution during the next phase.

Rockwell Servo Motor Research

Rockwell Servo Motor Research

The software provides detailed analysis of each solution such as shown below for the RDB most cost effective solution:

RDB Cost Effective Solution

RDB Cost Effective Solution

Below is a snapshot of the Motion Analyzer Software of the profile customization that can be done.
Profile of Guide Motor Operation

Profile of Guide Motor Operation

In addition, anyone can download the Motion Analyzer tool from Rockwell for free. Ryan Cavanaugh produced a basic guide to using the software that can be found here

Driver Selection

The Rockwell Kinetix series drives will most likely be the drives that will be used to power the guide roller motors as well as the other motors in the machine. They are designed to be used with Rockwell Servo motors so are the perfect match. The refresh rate of these drives is much faster than regular AC motor drives such the Powerflex series, which we were thinking of using the previous phase. These drives can be installed side-by-side and special connectors can be purchased to power all the drives off of each other instead of wiring every single one. If a particular model such as the Kinetix 6500 drives are used, all the drives in the system can share the same DC Bus to share any regenerated energy from deceleration.

PLC Selection

The PLC controller has not been specified exactly either, though the 5730 Series Controllers have been seriously contemplated. There are easily enough I/O points for the entire system and it should integrate with the Kinetix Drives. The table below was obtained from a Rockwell 5730 Automation Controller Brochure.
Compactlogix Controller 5730 Specifications

Compactlogix Controller 5730 Specifications

Sensors

Temperature Monitoring

Our temperature sensors need to be simple yet provide accurate temperature readings. The four bearings will simply have an over-temperature reading and if necessary the motors could also have extra temperature monitoring. Motors generally have an over temperature sensor but once that alarm goes off it is not good for the motors, therefore an external temperature thermocouple could monitor motor temperatures. Since the planned PLC module is the compactlogix 1769-L30XX controller, a special temperature module can be purchased to interface with this PLC called the 1769-IT6 Thermocouple/mV Input Module.

This module can support up to 6 thermocouple's which potentially is as much as we would need (4 bearings and 2 motors). This module costs approximately 1080 $ as listed on the Rockwell website. The K type thermocouple could be obtained to measure temperatures in the range of approximately -50 to 900 degrees Fahrenheit and they would not be too costly either ranging anywhere from 10 to 100 $ each after a quick search on Amazon. In the final phase we will have exact thermocouple models, manufacturers, and prices.

Another option is to use a regular analog I/O module and program it to interface with the thermocouple. Anolog I/O Modules are not cheap either, but if we have extra analog I/O it might be the cheaper option requiring more coding. Below is a chart obtained from the 1769-IT6 module manual for what temperatures the module can support from different thermocouples.

1769-IT6 Supported Thermocouple Temperature Ranges

1769-IT6 Supported Thermocouple Temperature Ranges

Door Safety Switches

Our machine also requires safety monitoring specifically for access doors to the machine. This can be done very easily via simple normally open (NO) switches and the digital inputs on the PLC. 24 VDC would go to one end of the contactor and the other end would go to the PLC input. If the switch is closed the PLC would register 24 VDC and if the door is open or there was a problem with the contactor, the PLC would register 0 VDC and would not allow the machine to run. The exact model numbers/manufacturers will be determined this next phase. These safety switches would not be very expensive.

Encoders for Motor Synchronization

An incremental encoder monitors only relative position which means that it can have the same amount of accuracy but if the machine powers off and back on it will have no idea what position the rollers are in. Absolute encoders solve that problem where each of 8192 bits are unique and therefore the controller would know exactly what position the motor ended up in. There are also single-turn and multi-turn encoders. The multi-turn encoder monitors position at 12 bits for about 4096 revolutions while a single-turn encoder would monitor exact position for only one revolution. Currently the DS264 has an ECN-113 absolute single-turn encoder on each of the guide roller motors with a resolution of 13 bits (8192 pulses per revolution). That's once every .044 degrees! The servo motors we have looked at MPL, MPM, and RDD all come with encoders on them. The RDD motors we are currently looking at can have either absolute single-turn or multi-turn encoders so we will just make sure we go with a 13 bit absolute single-turn encoder as available.

The live document can be found here

Motor Synchronization/Wire Break Analysis

The synchronization of the rollers plays a crucial role in wire tension maintenance. Using Young's Modulus and the Wire Spooling Teams wire analysis, the force required for the wire to break for even a minimum of 30 wires was found to be quite large. Thanks to the Wire Spooling team, the wire breakage point was assumed to be 3% and the Young's Modulus of Elasticity assumed to be 80 GPa at a minimum and 180 GPa at a maximum. Below is the equation that was derived to determine the torque required per wire to brake it starting from about 25 N.

Torque Required to Break One Wire Loop

Torque Required to Break One Wire Loop

Here is some general data obtained from doing this analysis:

Guide Roller Wire Break Chart

Guide Roller Wire Break Chart

The motors would not be able to muster up enough enough force even if they spun in opposite directions. The one thing not understood about the system is if tension will accumulate, however from the basic analysis so far, it does not seem like wire tension would accumulate even if the rollers had a slight torque differential. With such high resolution from the motor encoders and such a large force to brake the wire, the risk for wire break from un-synchronized rollers seems to be diminished.

The detailed document can be found here

Drawings, Schematics, Diagrams

The common understanding now with all the electrical engineers on all three glass cutting teams is that this machine will require a professionally designed and installed enclosure with proper safety ratings. David Cicero has stated that he will contact a company called Agile in order to get the enclosure system design talks started. This enclosure may just be the longest lead-time "part" in the electrical systems since it will take time to design it and then manufacture it. Therefore it makes no sense for us to draw up schematics since we are very inexperienced with industrial enclosure design. However, some sketchy diagrams have been provided to get a rough idea of some of the aspects of the enclosure.

Rough System Network Concept

Rough System Network Concept

Rough System Power Diagram Concept

Rough System Power Diagram Concept

Again, these diagrams are very rough ideas but do show general concepts that will be implemented in the panel design! For example, the DC Bus can be reconfigured in multiple ways as well as all the other wiring.

Sources and other Reference Documents

A document with most of the sources we visited and were able to document in our research can be found Here

A presentation we prepared for David Cicero can be found Here

Bill Of Materials

public/Reviews/MSD1 P4 Preliminary Detailed Design/BOM.jpg

The live document can be found here

Build/Assembly/Debug Plan

Build and Assembly

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 1.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 2.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 3.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 4.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 5.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 6.JPG

public/Reviews/MSD1 P4 Preliminary Detailed Design/Build Assembly Plan 7.JPG

The live document can be found here

Electrical Debug

Motor problems

Motors do not move:

1. Check starter coils in ladder logic to see whether or not the starters of the motors are getting energized.

2. Check ladder interlocks for motor to see if something is keeping it from starting.

3. Check drives for power.

4. Check drives programming for functionality.

Motors are not synchronized:

1. Check logic for programming mistakes.

2. Check inputs of encoders for accuracy.

3. Check ladder logic during run time for energizing of rungs for proper functionality.

4. Determine time that the error is occurring (if it can be localized)

Motors are overheating:

1. Check temperature sensors for functionality.

2. Check fans/cooling system for power/functionality.

Cabinet Problems (PLC, DRIVES, POWER)

No Power in the System:

1. Make sure 480 VAC breaker panel was not locked out due to work on enclosure wiring or if it is just in the OFF Position (DO NOT TURN ON!).

2. Check main disconnect on enclosure to see if it is off (DO NOT TURN ON!).

3. Call technician/electrician to check cabinet/wiring and safely turn on the system.

No Power to a Single Drive:

1. If other drives work, fuses and wiring to non-working drive will need to be checked- call technician/electrician.

No Power to PLC:

1. Make sure cabinet has power.

2. Check LED status of 24 VDC power supply. If 24 VDC is non-existent, call technician/electrician as power supply will need to be checked (fuses, wiring).

3. If 24 VDC is present and working check power wires to PLC with a multimeter.

4. If no voltage present at PLC power input check 24 VDC fuses. If blown make sure problem that caused blown fuse is rectified and replace fuse.

5. If voltage present, PLC might be damaged/destroyed.

The live document can be found here

Test Plan

public/Reviews/MSD1 P4 Preliminary Detailed Design/Test Plan.JPG

The live document can be found here

Risk List

public/Reviews/MSD1 P4 Preliminary Detailed Design/Phase 4 Risk Assesment.jpg

The live document can be found here

Project Plan

Phase 4 Gantt Chart

public/Reviews/MSD1 P4 Preliminary Detailed Design/Gantt Chart.png

Phase 5 Plan

public/Reviews/MSD1 P4 Preliminary Detailed Design/Project Plan.jpg

The live document can be found here

Deliverable Completion Plan

public/Reviews/MSD1 P4 Preliminary Detailed Design/Deliverable Completion Plan.JPG

The live document can be found here

Individual Contributions

public/Reviews/MSD1 P4 Preliminary Detailed Design/Individual Contributions.jpg

The live document can be found here


Home | Imagine RIT | Final Technical Paper | Final Poster

Problem Definition | Systems Design | Subsystem Design | Preliminary Detailed Design | Detailed Design | Gate Review

Build & Test Prep | Subsystem Build & Test | Integrated System Build & Test | Integrated System Build & Test with Customer Demo | Customer Handoff & Final Project Documentation | MSD2 Gate Review