Design notes, 8888Z sack attack

Our teams don’t usually do reveals, however, I wanted to post about a recent design that was built by a freshmen team as it’s a good example of how a competition robot can be relatively simple yet still be competitive.

This design was created with the following goals.

  1. Drive system using mecanum wheels. No other reason here other than that we have them and wanted to try them out.

  2. Height under 15 inches so as to be able to drive under trough.

  3. Scoring in trough only, no attempt to score in the high goal. This was to simplify the design.

  4. As light as possible.

  5. Finish building with enough time to practice (wishful thinking).

  6. Try and use as many uncut standard structure pieces as possible.

  7. Modular design that separated the three major sub-systems (drive, lift and intake) and allowed them to be worked on independently.

So with this in mind the following design was created.

Drive system

The drive is based around a standard 25x25 steel chassis. Aluminum could have been used but we have a number of these that came with classroom kits and were available. Version 1 of the drive constructed the chassis in the standard way, after testing it was decided to push the front wheels forward to create a larger wheel base and help keep the center of gravity behind the front axles, more on that later, this meant replacing the front angle with a C channel behind the wheels. Power for the drive comes from four 393 motors geared for torque directly driving the four mecanum wheels.

[ATTACH]6912[/ATTACH]

Lift System

The lift system is a simple (2-bar ?) arm that uses two small turntables as the final pivot points for the arm. Another of our teams had tried to use the turntables last year but were unsuccessful, this time the idea was to try and put the arms outside the chassis so as to theoretically be able to drive them a full 360 degrees. I posted some details of the compound gearing used in the design this thread;

https://vexforum.com/t/compound-gearing/22130/1

power for the lift comes from two 393 motors and we stuck with the 8.33:1 gear ratio that was initially chosen. The lift has a potentiometer on both sides for positional feedback to the software.

[ATTACH]6908[/ATTACH]

The lift attaches to the drive using 5x10 aluminum (or steel) plates, this allows some amount of forward/backward and height adjustment. Some additional bracing was added after the final position was determined.

Arm

The arm is very simple, 1x2x1x35 Aluminum C channels, scoop driven by two 269 motors using HS chain with a 4:1 ratio to provide more torque to rotate the scoop. Several different scoops were tried before the final simple design was arrived at. The scoop has a quad encoder so the software can “electronically gear” the scoop movement as the arm raises.

[ATTACH]6909[/ATTACH]

The three sub systems come together as follows.

[ATTACH]6911[/ATTACH]

[ATTACH]6913[/ATTACH]

The 1x5x1x25 C channel at the back is ballast needed to counteract the weight of the sacks.

Part 2 will look at the center of gravity and force needed to lift the arm.





2 Likes

Part 2

Before building we used the CAD to determine the estimated robot weight and center of gravity.

This shows COG with four sacks and the arm in the worst case position.

[ATTACH]6919[/ATTACH]

COG is still behind the front axle so the robot (when not accelerating/decelerating) should not tip.

The CAD shows the weight of the arm as 1.6lb, this, along with the weight of 4 sacks, was used to calculate the force needed to raise the arm.

[ATTACH]6920[/ATTACH]

The robot can actually lift 10 sacks without problem, however, in competition it’s usually limited to 4 due to the current scoop design.

We also used CAD to check scoring geometry.

[ATTACH]6921[/ATTACH]

Also behind

[ATTACH]6922[/ATTACH]

Ok, finally a picture of the finished robot. In part 3 I will give a summary of the programming.

[ATTACH]6923[/ATTACH]





1 Like

Part 3

Programming

The 8888 teams have an informal structure with members contributing to more than one robot. We currently have one member who is the primary programmer for all of the designs, obviously some of the programming uses my contributions from vexforums.

Programming is done using ROBOTC, the drive runs as an independent task as does the arm and intake system. A potentiometer on the left arm is used as part of a PID control loop to maintain arm position, only two presets are used, floor and trough. A joystick can be used to manually override the presets when, for example, trying to de-score.

The scoop/intake uses a second PID control loop to maintain the geometry of the scoop to the arm, buttons are used to add an offset to this which allows sack pickup and scoring. The scoop is initially folded under the front wheels and unfolds during the autonomous period, this could use improvement as it depends on the initial setup of the robot (and therefore the initial encoder position) which is a variable.

The drive system has a gyro available for “field centric control” but the current freshman driver prefers not to use this. The plan was to add IMEs to the drive system also but this has not yet happened.

There is currently a very simple, time based autonomous implemented which is about 90% reliable and usually scores just two or three sacks (no preload).

BOM/Cost

As an exercise I took a look at the BOM required to build this robot.

Description                           Qty   Part No     cost     ext cost
Steel chassis Kit                       1   275-1033    21.35    21.35
Steel C channel 1x5x1x25 (4 pack)       1   275-1138    17.99    17.99
Aluminum C channel 1x2x1x35 (6 pack)    1   276-2289    49.99    49.99
Aluminum C channel 1x2x1x25 (6 pack)    1   276-2288    39.99    39.99
Aluminium plate 5x25 (6 pack)           1   276-2311    34.99    34.99
393 Motors                              6   276-2177    19.99   119.94
269 Motor                               2   276-2181    12.99    25.98
MC29                                    8   276-2193     9.99    79.92
Mecanum Wheels                          1   276-1447    59.95    59.95
Turntable pack                          2   276-1810    19.99    39.98
HS Gear Kit                             1   276-2250    29.99    29.99
HS sprocket & chain kit                 1   276-2252    39.99    39.99
Potentiometer (2 pack)                  1   276-2216    12.99    12.99
Quad encoder (2 pack)                   1   276-2156    19.99    19.99
                    
VEXnet system Bundle                    1   276-1604   399.99   399.99
7.2 V batteries                         2   276-1491    29.99    59.98
Power expander                          1   276-2271    49.99    49.99
                    
Screws/bearing flats/standoffs etc.     1                       150.00
                    
                    
                    
                                            Structure           593.04
                                            Control/power       509.96
                                            Misc parts          150.00
                    
                                            Total              1253.00

Not 100% accurate but total cost, including cortex etc. about $1250. Spare batteries/chargers etc. are not included.

A couple more pics of the finished product.

1 Like

This is an absolutely amazing job especially for a rookie team. They did fantastic. Gratz on a job well done.

And thanks for the reveal. It is enlightening to see what other programs can do.

1 Like

They are a rookie team but the programmer is a high school senior, this is the third year that the school has been involved with VEX so there is some experience. They were supervised quite closely, last year we tried something different that did not work as well.

1 Like

I would recommend putting standoffs connecting the sprockets on the intake. Otherwise, the axle will most likely get twisted from the weight of the sacks.

looks like a well built robot for a rookie team. James, were you at the Chula Vista tournament last weekend? I saw that your teams were there but I did not see you around.

Yes, we have talked about this, the issue is that currently the intake can rotate 360 deg, adding the extra standoffs would limit that, I left it to the students to decide what to do. So far it has not been a problem and, as the sacks tend to be very close to the axle, the torque created is not that great.

Yes, I was at Chula Vista, I met some members from one of the 569 teams. The driver for this robot is also the cousin of one of your students.

I should also mention that 8888Z was Tournament Champion along with their alliance partner 2150H from Chadwick robotics. They had a good run of luck during the qualifiers and finished as the #1 seed, quite unexpected. It may be the only time this robot is ever run, we plan to run a different one in Granada Hills this weekend but may try and enter it if there are any no-show teams.

1 Like

Looking forward to seeing you guys at Granada, i think its going a pretty competitive tournament. Also i really like how you guys used CAD to test the center of gravity (with sacks). Really impressed with the design process, etc, good luck at Granada.

Wow you guys are impresive you analize every single little detail hahahaha with us is you try it out if it works great if it doesnt well back to the drawing board.

That’s impressive, especially for a freshman team. Do you find that you’re able to descore using your wrist? We were considering doing something similar for one of our other teams.