Posts Tagged ‘ cad & design

Day 10: Prototyping, Manufacturing & Design Progress

Prototyping

Today, a prototyping group met to finish the roller prototype that was almost complete yesterday.  Round belt pulleys were fabricated and round belts were added so that one motor could be used to power both wheels simultaneously.  The prototype was made mobile and we tested picking up tubes off of the floor and spitting them onto pegs.  It works very well but has issues picking up the corners of the triangles.

Manufacturing

Today, the wheels were finished.  The first operations were performed on the chassis siderails which should be completed tomorrow.

Completed Wheels

Design

The design is coming along great.  The baseplate is almost complete and the spool gearbox is nearing completion as well.  We are on track to be finishing the robot design within a week.

Spool Gearbox

Day 9: Rollers Prototyped, Wheels Complete & Preliminary Electronics Layout

Rollers Prototyped

Today, our prototyping group finished putting together a jig for testing tube intake rollers.  The prototype was very successful and successfully picks up and holds tubes.  Furthermore, we were able to rotate tubes within its grasp by rotating the two rollers at different rates.

The design still needs to be tweaked for optimum tube grabbing performance.

The initial roller prototype. Driven by drills, the prototype is very effective at sucking tubes in and holds them quite well once they are in its grasp.

Manufacturing

After several days of hard work, the drivetrain wheels are finally complete.  During machining, there was an issue with the original CNC code where the spokes were not being milled deep enough.  This was resolved and the problem wheels were fixed.

Completed Wheels

The wheels look great and are visibly much smaller than previous wheels, as shown in the image below.  The wheels will be anodized by sponsor Pacific Coast Metal.

The Evolution of Cheesy Poof Wheels. From left to right: 2006 6" Wheels, 2008 4" Wheels, 2010 4.5" Wheels, 2011 3.5" Wheels.

The DXF CAD drawing for the gearbox sideplates was created and sent to Mike D of Team 233 for Manufacture.

Design

Today, the design team worked on the electronics layout, the spool gearbox and preliminary design for the minibot.  The electronics board is coming along great.  We are happy with the positioning of most of the elecronics but are still not clear on where we want our compressor.

Preliminary Electronics Layout

A second team worked on CADing the spool gearbox.  An issue arose where one of the needed gears was not available, so we had to modify the gear ratios slightly.  The design for the gearbox is coming along and should be finished tomorrow.

A third team started to analyze the parts available for the minibot and started to develop preliminary ideas for how to efficiently lift the small robot to the top of the pole.  Work will continue into the next week.

Day 8: Design & Manufacturing Progress

Design

Today, we worked on space claim analysis and tried to figure out how we could even fit a roller claw onto our elevator and still have room for a minibot while building a nice looking robot.

Space Claim Analysis Sketch for the robot. The minibot is housed on top of the horizontal elevator support and can slide out backwards to attach to the pole.

Next, we started to analyze what our tube grabber could be shaped like.  We knew that one of the major issues would be getting the tube grabber over the bumpers to pick up tubes on the floor.  To solve this problem, we thought about creating an arm that would come down from the elevator carriage and have a second roller below it.

A second issue we were worrying about was depositing the tubes on the rack.  Our experience with the 2007 FRC game as well as the 2011 VEX game led us to believe that simply spitting the tubes out of the rollers could result in unpredictable scoring.  To eliminate this issue, we thought about having the two rollers separate with a pneumatic piston so that the tube could be gracefully released and deposited on the rack in one fluid motion.

The tube grabber in pickup position.

The Raised Tube Grabber. The grabber would be able to raise to perfectly vertical so that the tubes can be carried entirely within the volume of the robot.

We also talked some about the speeds for the elevator.  We decided that we could use one BaneBots 775 Motor and one Fisher Price Motor on the spool gearbox to still leave enough motors for the rest of the robot.  We decided to gear for a faster cable speed than either of the elevators we have built (for comparison, the 2007 robot’s elevator moved at 3.6 ft/sec and the 2011 will move at 5 ft/sec).

The gear ratios for the 2011 elevator. The first stage will use 32DP gears and the other two stages will use 20DP Gears.

Prototyping and Testing

We decided to build some prototypes to test how much force would be needed to keep the tube grabber closed.  We built a jig to hold two wheels at a specific angle and allow us to apply force between them.  It was found that with the tubes, once the wheels had passed the diameter of the tube, the frictional force was so great that the tubes could not be freed from the grabber.

Testing using two wheels as a tube grabber. With static wheels, once the wheels pass the diameter of the tube, the tube cannot be freed from the grabber by pulling.

Another issue we addressed was the tendency for the tubes to rotate in the grabber.  By putting a metal tube behind the tubes, it improved tube centering on most tubes.  Problems, however, arose when testing on the corners of the square and especially triangular tubes.

On the corners of the triangle, the wheels cannot get past the major diameter of the tube, causing more difficult tube retention.

To allow the tubes to still be held on the corners, we tried applying various amounts of force to the top of the lever.  We found that after about 12lbs of force, it was extremely difficult to dislodge the tube from the grabber.

Manufacturing

The machining of the wheels is coming along very well.  The first operation is complete on almost all of the wheels.  We hope to complete the wheels tomorrow.

The wheels with the first operation completed.

Last year, we had an issue where the hole in the wheels created by the rotary hex broach was not big enough for the hex shaft. This year, we have purchased a new hex broach to fix this issue. The wheel fits very nicely on its shaft.

Aluminum Chips after Milling the Wheels.

Day 6: Prototyping, Programming, Design & Manufacturing Progress

Prototyping

Today, we had a large group of students working to prototype mechanisms for subsystem 2 (tube grabber.  One group was working to construct a jig for testing roller prototypes.  The other group worked to think up alternate solutions for tube grabbing.

Several Team Members working on a prototype.

Ryan and Erik "Prototyping"

Programming

The programming team met today to refine the control of the robot during teleoperated mode.  Progress is being made.

Design

Today, several team members were working on the CAD Design of a part for the elevator carriage.  Furthermore, later in the evening, we continued working on the mounting interaction between the elevator and the drivebase.

Manufacturing

The machining of the bearing housings is finally complete.  After five long days of machining, all of the parts are done with machining.  After they are deburred and polished, they will be ready for anodizing.

Completed Inner Bearing Housings

Completed Outer Bearing Housings

Day 5: Design & Programming Progress

Design

At the lab today, a number of students met to work on robot design.  We spent the entire day working on Subsystem 1: Arm/Elevator.  We began the day by reviewing our game objectives before proceeding to create a weighted objective table.

Elevator 1 Joint Arm Scissor Lift Conveyor 2 Joint Arm Telescoping Arm
Designability (x3) 3 4 3 1 2 2
Manufacturability (x3) 4 5 3 3 3 3
Speed (x4) 5 4 3 2 4 4
Accuracy (x5) 5 4 1 3 3 4
Weight (x3) 4 5 3 2 3 3
Weighted Total 85 78 44 41 55 60

Using the table, we were able to rate all of the different potential superstructure designs on factors such as designability, manufacturability, weight, speed and accuracy.  In the end, the Elevator and 1-Jointed arm came out way above all of the other designs so we decided to focus on those two designs.

Benefits of Elevator

  • Easier to Drive
  • Easier to Score
  • Faster
  • More Compact
  • Center of Mass Can’t Leave Robot Footprint
  • More room for Minibot
  • We are more familiar with Elevator Design & Construction
Benefits of Arm

  • Easier to Design
  • Easier to Manufacture
  • Potentially Lighter

While discussing some of the pros and cons of each design, we focused especially on the designability of each design.  We concluded that an arm would be much easier to design.  However, because our team has built elevators in the past, we are familiar with their design and will be able to modify some of our old CAD models to work in the 2011 competition.  Because of this, we decided that it would be feasable to design and build an elevator and that it was the best design for our team.

After deciding on an elevator, the team moved forward with design, looking at old robots and old CAD models.  After some analysis, we started to build the new elevator assemblies, using some of the concepts used on previous robots.  At the end of the day, we have completed drafts of the first and second stages of the elevator; only the carriage still needs to be completed.

The robot design as of today. Many parts are still missing including the rear supports for the elevator.

The bottom of the second stage of the elevator. The second stage requires no welding.

Programming

The programming team was working hard today to perfect both the simulator and robot code.

Day 3: Field Construction, Programming, Design & Manufacturing

Field Construction

The field construction team was hard at work today working on the two elements of the game field that we are building this week.  We have constructed a shopping list and hope to buy the remaining materials tomorrow.

Programming

The programming team was working hard today on teleoperated and motor control code.  They have designed an interface to separated the control system (joysticks, arrow keys, etc) from the code so that the robot can be easily modified to be driven by a variety of controllers.  Furthermore, they performed experiments to measure how linear the Victor Speed Controllers are in their outputs.  Using MatLab, they will be able to analyze this data and derive a function to linearize the Victor outputs.

Design

The design team was hard at work today.  The first order of business was the drive gearbox.  After the gear ratios were finalized last night, we were able to move forward with the drive gearbox CAD which his now completed.  When doing the design, we were able to develop a new way of retaining the shifter shaft between the bearings to ensure that we can shift reliably.  Without motors, the wheel and bearing housing, the gearbox is less than two pounds.

Drive gearbox as of tonight.

In addition to working on the gearbox, a group of students worked to design and CAD the drivebase frame.  After careful review of the bumper rules, we decided on a design that would support the bumpers all along the back while still being extremely lightweight.  We added a heavily pocketed baseplate and put the whole thing together for a design that weighs about 30 lbs.

The drivebase design as of today.

Manufacturing

Production continued on the first operations of the bearing housings.  We hope to finish the bearing housings by Wednesday.

Lab Improvements

Today, the wired network between the lab computers was completed so that they can all share printers.

Day 2: Drivetrain Design & Manufacturing

Today, at the lab, a group of team members met to continue designing the robot drivetrain and continue manufacturing bearing blocks.

Networking & Lab Improvements

We started by networking all of the upstairs computers at the lab together.  This will allow us to share printers before the computers.  We were able to connect all of these using a switch generously donated by our sponsor, Vivid-Hosting.  The networking will be completed tomorrow.

We also procured two power strips which were wired under the floor to provide power to two upstairs tables for laptops.  Previously, the lab did not have a good place for laptops to work.

Manufacturing

Manufacturing on the bearing blocks continued today.  Progress is being made and the manufacturing will continue into the next few days.

Design

Today, we discussed and decided on several key aspects of the drivetrain.

The first topic of discussion was wheel size.  Several team members brought up the idea that with such a flat game field, we could potentially go to wheels even smaller than 4″.  Although we quickly ruled out 3″ wheels due to lack of ground clearance (~1/4″ clearance), we were interested by the prospect of 3.5″ wheels which would lower our center of gravity slightly and could decrease drivetrain weight by more than one pound.  Furthermore, because less gear reduction is needed with smaller wheels, the weight of the spinning components could be decreased by up to 20%, leading to a potential significant increase in robot acceleration.  However, some team members had reservations due to the decreased ground clearance in the move from 4″ to 3.5″ wheels.

To come to a consensus, we created a weighted objective table to weight some of the pros and cons of each design.  After much deliberation, the table resulted in almost a tie so we continued to discuss.  In the end, it was decided that any problems caused by the decreased ground clearance (1/4″ lower) of the 3.5″ wheel system could be avoided so we decided to move forward with the lighter 3.5″ wheel design.  This caused the outer bearing blocks to need edits so that they would not interfere with the newer, smaller wheels.

Benefits of 3.5″ Wheels

  • Lower Center of Mass
  • Less Weight
  • Increased Acceleration
  • Smaller Gear Reduction Needed
  • Less material means less cost
Benefits of 4″ Wheels

  • More familiarity with the design
  • More ground clearance
  • Less Tread Wear

The second topic of discussion was robot speed.  After yesterday’s strategy session, the whole team was in agreement that having a fast and maneuverable robot would be key to successful performance in Logomotion.  After deciding on small, 3.5″ diameter wheels, we also knew that these smaller wheels would not require as large of a gear reduction as has been present in past robots.

With the ability for a smaller gear reduction, we started with our gear ratios from last year and worked with them to both speed up the robot and minimize the use of large gears to decrease robot mass and increase acceleration.  In the end, we found a set of ratios that we liked that will both allow for the fastest 254 robot ever built and for smaller and lighter gears while still maintaining a slower low gear.

Finalized gear ratios for the 2011 robot. Robot speeds are estimated for 3.5" wheels in ft/sec at 100% and 80% efficiency. The first line represents the common reduction. The second and third lines correspond to the secondary reductions for High and Low Gear.

Finally, today was our deadline to finalize the gearbox shafts to be sent out for manufacturing by sponsor Pacific Precision.  All four of the shafts were updated and finalized by a team of students.  All drawings were checked by team leaders and mentors and were sent to manufacturing.

FRC Season Day #1

Today was an extremely productive day at the lab.  We built a significant part of the field, tested the game elements, began design, started manufacturing and continued to work on programming.

Field Construction

As soon as we got to the lab, a large group of team members got started building a replica of the most essential field elements – a rack and a tower.  We intend to build enough of a low-cost field for testing and the NASA Robotics Alliance Project will purchase a complete field for the west coast in the coming weeks.  The field will stay at our lab during the build season but will travel around the state for various off-season competitions afterwards.

Several team members cut parts for the field.

The structure to hold the rack starts to come together.

One of the vertical poles of the rack is assembled.

Game Element Testing

After we had a basic understanding of what the rack was going to look like, we moved forward with game piece testing.  We experimented with how far the tubes could be thrown and discovered that a well-trained human player could likely throw tubes across the entire field.

Ryan tests a game piece.

Programming

The programming team was working hard all day.  They began the day by installing the new versioins of the programming software and then spent the day getting Onslaught to drive under driver control.  The simulator group was working on data logging, 3d modeling and collision detection.

The programming team working hard.

Design

After dinner, a group of team members met to discuss robot design.  We immediately decided on a wheeled robot and decided that at least 6 wheels would be best for maneuverability.  To save weight, we all agreed that it was not practical to consider building a robot with more than 8 wheels.  We also decided that we would like a two-speed drivetrain with a high gear to quickly traverse the field and a low gear to push other robots out of the way if necessary.

When looking at the game field, we were very worried by a 1/4″ tall plate under the carpet surrounding each tower.  The concept of a 6 wheel drive or 8 wheel drive drivertrain with dropped center wheels relies on the fact that not all of the wheels will be on the ground at all times to improve turning.  We were worried that when in proximity to the plate under the field, more wheels would be in contact with the carpet which could inhibit turning.  We thought that this might be a greater issue with the six wheel drive robots than with eight wheel drive robots, but decided to test to make sure.

We put down a metal plate under carpet and tested it with both 8 wheel drive and 6 wheel drive robots.  It was determined that the 8 wheel drive robots drove considerably better when in proximity with the plate when in high gear.  However, in low gear, both robots had great performance as if the small bump wasn’t even there.

Testing robot maneuverability on a bump covered by carpet.

We took the data from our tests and moved on to discuss how important turning near the towers would be in high gear.  It was determined that the only time turning near towers would be very necessary is when lining up to deploy the Minibot.  We decided that this is only for about 5 seconds of the match and that there is no reason to not just use low gear during this time.

After we decided that it was not important to be able to turn in high gear near the towers, we all decided to move forward with a six-wheel drive robot.  Everyone in attendance voted that the benefits of saving up to 3 pounds outweighed the drawbacks of potential limited turning in high gear when in close proximity to the towers and agreed that a six-wheel robot would be best.

The next topic that was discussed was traction.  Although with ideal physics, contact area with the ground does not affect traction, it was determined through observation that due to the tread material interlocking with the carpet fibers, more contact area does indeed result in greater traction.  Because of this, we discussed moving to slightly wider wheels in order to increase the contact area with the ground.  We were worried that wider wheels could negatively affect turning so we decided that the robot should be designed with thin wheels (similar to what we have used before) but should be compatible with wider wheels if necessary.

Finally, we discussed speed.  The consensus was that we want to be able to drive fast.  However, no decision was made on exact robot gear ratios or speed.  Both will be discussed and finalized tomorrow.

Manufacturing

As soon as it was finalized that we would be building a wheeled robot, manufacturing began on the bearing blocks to support the wheels.  All of the stock for the bearing blocks was cut and the first operation began on the outside bearing blocks.  We expect to complete the first operations on all of the bearing blocks tomorrow.

Deburring the stock for the bearing blocks.

Bearing blocks being milled.

FRC Build Update

Friday the 10th marked our third and final CAD Friday and our final FRC Friday of the semester.  We had a great time and started to explore some of the more advanced assembly and sketch features of SolidWorks.  Although it would have been nice to have more time so that we could have gone into more depth and gone over more features of SolidWorks, we had a great opportunity to give many students a broad overview of the program.  Overall, I think that the FRC Friday program this semester was very successful and I hope we try something similar again in future years.

Today, we had our second Machine Tools Training of the semester with two more students getting trained on how to use the machine tools at the lab.  For reference, I’ve put up a list of the students who are trained here.  If you would like to sign up for machine tools training in the coming weeks, sign ups are here.

As the semester wraps to a close, we have very little time left to get everything ready for the FRC season.  We have scheduled Open Lab time for December 20-23.  Please sign up for this time here and come in if you are interested in working on Programming, CAD or just helping get the lab ready for build.  There are no space limitations but if nobody signs up, we will likely cancel the session and the lab won’t be open 🙁  There are still numerous items on the lab to-do list that need to get done and we need as much help as we can get.  Let me know if you have any questions for this.

Remember, students interested in CAD and programming: Coming to the workshops alone is not enough to become a CAD or Programming master.  To gain all the skills neccesary for the robot work, you will have to put in some extra time outside of the workshops.  Coming to the open lab days (mentioned above) is an excellent opportunity to get the much-needed experience with CAD or Programming.

26 Days Until The 2011 FRC Season Begins!
Happy Holidays and Go Poofs!
Nick Eyre

First CAD Friday

Today was Team 254’s first of three CAD Fridays. After a fun team lunch, we had over twenty team members come together at the NASA lab to learn CAD skills.

We started out with an introduction to SolidWorks and went into some of the skills used in basic extruded features and sketches. We discussed sketch relations and dimensioning and made several simple parts.

Later, Travis gave a presentation on the Fundamentals of Graphic Communication, which can be found here. The presentation was long, but went very well and was extremely informative. All who attended were able to learn about expressing three dimensional objects in two dimensions through engineering drawings as well as other valuable graphic communcation skills. Travis even showed the team one of his parts from his work and showed how many drawing views needed to be used to effectively communicate the part’s design.

After the presentation and dinner, we worked to develop our CAD skills by analyzing a pre-made engineering drawings and creating 3D CAD models of the parts based on the drawings. It was a good excercise for using the graphic communication skills we learned and applying them to actual design.

After the CAD was wrapped up, several team members stayed late at the lab and built shelves upstairs that can be used by our team and by team 1868 for storing and displaying our trophies and awards.

At the next CAD Friday (12/3), we will split into two teams and work on a design project, using SolidWorks to model our different solutions to the design challenge which will be announced. It should be a great way to continue to learn CAD skills through practice and problem solving.