We’ve finally got the CAD cleaned up to a point we feel comfortable sharing our 2023 Robot VIPER. This is the robot as it was at the World Championship, upgrades are already underway for the offseason. This is an export from solidworks, so somethings might be missing or appear incorrectly.
Let us know if you have any questions, we will do our best to answer them.
-Parker
It cooked.
-Ronnie
It did indeed…
(It was an honor to be on one of this incredible robot’s meals)
Can you change the share setting to allow copies of the document to be made?
I’m not super familiar with the Onshape interface. But I believe I have made it allow copies. Let me know if you are able to or still unable too now.
All good now. Thanks Parker. Great season, and great robot!! I didn’t realize you guys did a pair of angled inserts to move the end effector to the middle of the robot, that was a pretty neat detail.
Thank you, we are looking forward to many offseason competitions with even more robot improvements! Including two more drive motors…
Did you guys end up going with 3 motor swerve for IRI and Rainbow Rumble because watching you guys just flew on the field during those comps and you guys were already super fast at Champs or was it some other swerve modification.
Hey, I was curious what material you printed your end effector gears out of?
Between champs and IRI we added two additional motors to our swerve kits; bringing us up to a 10 motor swerve drive. Only two of our modules had three motors, leaving the remaining two modules with only two motors. This was done to as an attempt prevent brown-outs from occurring during a match, as reported by some teams that are running 12 motor swerve. This change also bumped our speed to 24.4 fps.
How did you choose which modules to add the additional motors too? Was it arbitrary, or did you evaluate a need for the extra power on one side?
We spoke a bit about your drive upgrades at Rainbow Rumble. I’m hoping that you get some time this summer to benchmark the improvements to your robot sprint times between 4 drive and 6 drive motors, adjusting the current limits as needed for each. I would love to see some real world data!
Thanks for being great pit neighbours this weekend, and congratulations on the win!
We added the additional motors to modules adjacent from one another. As for which pair of modules we decided to update, we picked the two most accessible ones in case we were to run into any issues with them. We also just didn’t feel like messing with the module directly under the arm’s gearbox.
All gears and pulleys on out claw are printed out of Onyx filament from a Markforge printer. This was done to keep the swinging inertia of the arm as minimal as possible.
We are working diligently to put together a comprehensive breakdown of the cost to benefit of 10 motor swerve with different chassis weights, robot architectures (shooter games), and cycle distances. We are doing this to prove to ourselves that the extra motors will benefit us if we chose to use them during a real competition season. Some of this information we will release, but most of the theory, conclusions and controls of will more than likely stay internal.
So far this offseason we have been loving the 10 drive motors, we’ve ran two events totaling 26 matches, and only experiencing two brown out cases. This was exciting news for us as the control scheme for those two events was making the second drive falcon a follower of the first one, and having a manual switch on the drive station to disable the extra motors if we were getting close to a brownout state.
To explain our reasoning behind the 10 motors compared to the “more traditional” 12 motors. We thought 12 was too many and that 10 sounded cooler. In all seriousness there was some math and simulations that went into it but there’s already threads discussing all that magic pixies stuff. More on this if you want to compare a 10 motor to a 12 motor directly watch this match from IRI where on red 2910 is running 12 falcon motors with FoC, and their dry robot weight is 95 lbs, and on blue we are running 10 falcon motors with FoC, and our dry robot weight is 96lbs. There is very little difference between the two and the eye test maybe favors the 10 motor being slightly faster than the 12 motor at this weight, which seems counter intuitive but also tracks with the simulations we ran after the season for a robot in the 85-100lbs range.
If there are any questions about any specify things we haven’t covered yet let us know and we will do our best to answer them.
-Parker
I assume this might be in the documentation you mentioned but would you be willing to talk about how you geared a 10 motor system? Mostly, was the gear ratio between all 4 modules the same and if so what was it?
For the 3 motor module we used L1 gears pared with a 24t pinion on the drive Falcons. And for the 2 motor modules we used L3 gearing with 18t motor pinions. This calculates to somewhere in the 24fps range (free speed.) Oh the gearing is also slightly off by .011:1 between the two we didn’t have any issues with it but ended up slowing down the 2 motor module in code to even them out. The reason why we didn’t make the gearing the exact same was because we didn’t want to make x4 top plates for the SDS module. To fit the 18t pinion we used a relocating plate. See 2910 CAD for basically the exact thing we made.
relocating plate? is that something thats needed to be machined?
I’m curious as to how you manufactured part 5460-FRC23-201-D07 - the part that connects the vertical tube for the arm. What material is it made of? Also, the vertical tube - your BOM says it is carbon fibre - where did you purchase it?
Yes this was machined out of 3/8 aluminum plate on our Omio X8. It shifts the C-C further away from the 50t gear allowing the 18t pinion to fit. View the 2910 cad to see the plate we made its on their drive motor for their MK4i’s.
This part was originally split into two pieces and printed out of Onyx. The two pieces where connected with 1/8in dowel pins and epoxy, this worked great until it didn’t. To clarify the connection never failed, the stress on the print would split the layer lines which pushed us to switched it to Laser Sintered Nylon 12, this part has yet to fail due to being completely solid as well as not having the delamination characteristic of FDM.
The Carbon Fiber tubes were purchased from Clearwater Composites This company was great to work with and shipped our tubes next day. As well as they provided us with a educational discount on our purchase. We reached out to many companies when searching for this tubing and Clearwater was the only one to provide such a discount.