Team 1493 will be documenting our build season for 2020
Who is 1493?- This will be team 1493’s 18th season in FRC. Like many teams we spent way too many years building over complicated robots that did poorly. We were getting burnt out, frustrated, and spending way too many hours in meetings. In 2018 we decided we needed a drastic change to our approach.
We decided to apply the KISS principle (I know groundbreaking idea), but this time we would actually stick to it and not get sucked into the temptation every kickoff to throw that out the window, and over extend ourselves because the game didn’t seem “that hard”.
The decision we made that forced us to stick to this principle was to meet 50% less than we currently were. We currently meet 2 days a week after school. Mondays from 2:50-5:30 and Thursdays from 2:50- 7:30*. This constraint forces us to keep our designs simple, but effective decisions based on our game analysis and strategy selection.
*students are allowed to come to shop on free periods to work and we have small groups of drive team and select students meet at practice facility we share with local TVR teams outside of normal meeting hours for drive practice/fine tuning.
Since we made these changes we have experienced our greatest success on the field and the stress level for mentors and students has dropped significantly.
In is our hope that we can provide some level of assistance to teams who have similar struggles.
Some things we plan on providing are
How to best utilize RI3D + MCC
Game Analysis, Strategy, and our design choices
More info about 1493 our students, mentors, and resources
I highly recommend anyone who wants to build a simple robot pay attention to this blog. 1493 has absolutely nailed the MCC/MCC+ robot the past two years, meeting us in TVR finals in both 2018 and 2019 (and beating us in 2018). Thank you Brent and team for sharing this resource, I’m super excited to see what you come up with
Can you post up more photos of your robots from '18 and '19? I got to watch your matches both seasons and was super impressed with the improvements made from prior seasons. Definitely excited to see how your design develops this season.
Team 1493 starts today off with a parent/alumni meeting breakfast- we started this last year and its a great way reconnect with old mentors and give parents a better idea of our program.
After kickoff we spend the rest of our day looking at all of the rules and details of the game- No talk about what the robot might look like, what we should do, etc. (Once again nothing groundbreaking here from what other people have been preaching for years)
@Brian_Maher posted a great resource to help you work through this process.
We will continue this discussion on our team Discord all weekend.
On Monday we take the team on a field trip to our local practice field during the school day. We make up a mock up version of the field with tape/cardboard/and whatever we can find to mimic game pieces. You can identify scoring strategies, congestion points, defense effectiveness, and scoring/cycle times.
This process has been super helpful in framing our strategy for building a MCC+ robot.
If your team is thinking about going with this approach I cant stress enough how amazing the guide @AllenGregoryIV and Spectrum posted is- we will be reviewing this before we start our strategy discussion on Monday.
The afternoon on Monday is spend determining our strategy for the season.
Have fun today everyone!!!
Like most teams we spent Sat reviewing all the rules and making lists of all possible actions a robot could complete during the game.
Today we went to our local practice field and reviewed the rules, checked if we missed any important actions and role played the game with students on office chairs with wheels.
Thoughts from this-
scoring 49 balls will be hard
The balls on the field to start a match will be at a premium in auto and start of match, if you can get those and score them quick you have a chance to start a ball overflow for your opponents and disrupt their game plan.
Some matches will look like a youth soccer game with 3/4 robots all fighting over a ball
communication and planning for cycles before a match is very important the field gets cramped in the middle and timing who will score and when.
climbing like many have said is very important
someone needs to come up with a term for other teams getting slammed into the Shield Generator aka Steel Cage for the WWE fans
With this we came to some decisions as to what our robot would be capable of this year.
1- drive
2- gather balls from ground
3- score low
4- hang
Our next discussion is what would a robot look like that could do these tasks and be fast, reliable, and efficient at the task. And how could be make this MCC and add some + to it.
We unanimously decided ( which never has happened) on a ball low scoring system similar to the Everybot from last year that will have a over the bumper intake to increase our collection area. Our hopes is to shave some seconds off cycle times. We also believe that this system can be build rather quickly which allows us time to make our climber great and not so/so.
The UIUC RI3D has been working on a design which is very similar to what we thought would be ideal for our team. They have been doing a great job documenting their work.
We will continue to watch them and draw ideas for prototyping and looking for areas we could innovate. When you are a team that meets limited amounts of time and wants to build a robot that you can get plenty of practice time with to work out any bugs it is important to make use of the numerous resources available and select the ones that best for your team and the strategy you think best for your team.
For our intake we really like the designs team 3476 and 2363 posted from 2016
We plan on using 8 WD WCD with 6” pneumatic wheels- we have used them the last few seasons and really like them. Last years robot had a perimeter of around 100” 26” wide x24” long and we don’t plan on being much bigger (maybe a perfect square this year).
The +’s we hope to add to this MCC are
+1 - multi ball auto which with the intake and scoring on opposite sides should be much easier.
+2 if we manage to pull off all our goals we will add a device to spin the wheel- most likely for the 10 point bonus- don’t see 49 balls happening in many matches and especially in eliminations when defense will increase.
Thanks guys! Let us know if you have any questions for us about our prototyping or MCC design! You can join our discord to ask questions (or just use chief) https://discordapp.com/invite/eUXqYV3
After taking a deep dive into some of our design choices we decided that we will no longer be a low bot.
Reasons we made the switch-
we want to climb 100% of matches and no matter what height the bar is- this is can be easier when tall
just bc you are low does not mean you will have clear run of the trench. You may have other low bots on your alliance and high goal shooting bots will likely want to use this area to shoot from
We felt a little validation in our choices when we saw this morning that 118 came to the same decisions and analysis for their Everybot design this year.
At tonight’s meeting students worked on CAD of drivetrain, assembled drive gearboxes, worked on writing code and tested some prototypes for an over the bumper intake, belt ball system vs. roller system.
One thing we learned is teams will want their bumpers very low to the ground as it is really easy to drive up on top of the balls.
One of the hardest things this time of the year is second guessing your decisions to keep designs simple. It’s so tempting to see all these RI3D designs revealing high shooters and low climbers and want to change your approach and try to do it all. No bag this year makes this temptation even harder. Has this happened to your team already?
We made a small error last night- We ordered 6 Falcon 500’s seconds after they went on sale from WCP and received them Weds.
We have run VEX Single Speed Double reduction gearboxes for the past 5 years- they have been great for us.
When we mounted the Falcons to the gearboxes we were unable to turn the drive shaft at all.
Took them apart and tried to figure out what we did wrong. Turns out after re-reading the instructions that we installed the 10-32 screws much to far and that causes motors not to rotate. The 10-32 screws can not go more than 1/4” into motors
Ordered some shorter 10-32 screws and hopefully will have things back together on Monday.
Been awhile since we updated this- Learned a few things this week and made some big progress.
Falcon motor shafts rub against one of the ribs inside the VEX Single Speed double reduction gear boxes- we needed to file some of it down to get the gear boxes to fit and a axle to rotate
stupid moment of the week- using an odd number of 35 chain links in our center to center calculation on our drive train design. (Wont make that stupid mistake again)
Keeping a blog of your work on CD is also hard, we do not come even close to providing some of the info that other teams do. Big props to those teams that constantly update and provide valuable/timely information.
Our original plan for scoring balls had been to build something like the 2019 Everybot/ UIUC RI3D team. Some of us were split on that design because it would be forcing our drivers to have to drive in 2 directions constantly to intake/score. Everybot 2020 swayed the vote back to a front intake/scoring over the bumper design. We will be looking at ways to add our own spin on that design and be a great low scoring robot. We know that trading 1 point scores vs 2-3 point shots will be a big hole to dig out of. Hopefully we can do it at a slightly faster rate than opponents and maybe pick up the 10/20 wheel points in some matches.
To hang we plan on being tall and using a single stage elevator. We ordered some of the new elevator kits from Competition Robot Parts-
We really like them and are super impressed with how easy they are to use right out of the box. We still have not settled on hook/balance design but we usually get those details done much later in the season.
I am so glad we decided to build a “tall” robot this year as it is making the most important task of hanging so much easier. Our goal is to climb 100% of the time no matter the height of the bar. Our reach goal is to have a method of balancing after the hang.
We did end our week with some success as we have our drive train assembled and sent off to be powder coated today- this season has had our highest participation rate- 28 students have attended all the meetings so far and all are being productive. We have also had students coming to the lab on free periods to knock jobs off the checklist. That combined with only building one robot has us slightly ahead of our usual pace.
One features we added last year that we really liked was laying a 1x2 versatube flat along the length of the frame as you can see in images below. We then are able to mount our game piece mechanisms thru the holes in the tube. Our arm last year was mounted this way.
Next week is a testing week at our school and the teachers/mentors were not give proctoring duties for 2 days- this allows us 2 full days to meet. Our plan is to CAD and build our version 1.0 of intakes/low scoring and elevator climb.
I’m Andre Castagna, programming and elctrical mentor for team 1493. I been with the team for 8 years and am a math teacher at Albany High School. It’s been an exciting year so far, and we have been really impressed at how motivated and many of our students have been. We have a core of about 10 students working regularly on the programming and electric systems of the robot. We are trying to up our game this year and finally make good use of vision guidance, motion profiling in autonomous, and enhaced control of mechanical systems.
We had a group of programming student today working on a variety of projects. The drive train programmers have been writing and testing (on last year’s robot) code to control the drive motors. They were driving it around the shop today. They also completed code for a custom joystick ramp rate. We like the idea of the Talon ramp rate function, but want it only on acceleration, not deceleration. The custom ramp code works great. Their next tasks are to work on code to raise and lower control the intake mechanism using position control with a feedforward term to compensate for gravity.
Other students worked on vision targeting for the loading station. They were able to get limelight camera images from a target mockup, and worked on the image processing end. They tried image processing with the limelight software and also with Grip. Their best results were from Grip processed images - very clean contours from a variety of distances and angles. They imported the grip pipeline into a robot project, and are working extracting tartgeting data.
We also are setting up sonar sensors - a student exploring using multiple sonar sensors to provide alignment guidance with walls.
We have also completed mapping this year’s field layout onto our motion profile path planning software, which we will use for plotting autonomous mode paths. Code is being fine-tuned for controlling controlling the robot in both autonomous and teleop mode. We are using a combination of Talon motion profile mode, magic motion mode, and custom PID methods.
Finally, we made progress in using the Rev color sensor to detect colors. Detecting colors, now working on code to link a wheel-spinning motor to the color detecting algorithm.
Looking forward to begin wiring the competition robot later this week.
