Hey guys i’m new to CD so I’m still trying to figure things out, please click the link to see the pictures
Hello people of chief delphi!
I am Caleb Hayes and I have been working on a little project that I want to share with you. I have been creating a WCD for my team. I wanted to do this cause our team has used the kit bot for years and I think its time to move on. It is meant to be an off season bot and to be really small, this is why there are no bumper support beams between the wheels. It currently has 6 in pneumatic wheels geared at a ratio of 5:1. I understand this really aggressive but on a less than 60 pound bot I think it will manage. I want to hear what you guys think of my project and how to improve it. Also forgive if it doesn’t look cool or really good, I’m pretty new to CADing, but I don’t think it’s to bad for my first real CAD project.
Fun WCD attempt by: caleb7hayes
This is my first attempt at a WCD ever and I would like for you to give your thoughts.
Hello people of chief delphi!
I am Caleb Hayes and I have been working on a little project that I want to share with you. I have been creating a WCD for my team. I wanted to do this cause our team has used the kit bot for years and I think its time to move on. It is meant to be an off season bot and to be really small, this is why there are no bumper support beams between the wheels. It currently has 6 in pneumatic wheels geared at a ratio of 5:1. I understand this really aggressive but on a less than 60 pound bot I think it will manage. I want to hear what you guys think of my project and how to improve it. Also forgive if it doesn’t look cool or really good, I’m pretty new to CADing, but I don’t think it’s to bad for my first real CAD project.
Looks like solid CAD work, so I’m just going to comment on sizing.
The JVN calculator tool I’m using here is posted to CD here https://www.chiefdelphi.com/media/papers/3188
I kept all of the standard assumptions and just tweaked the weight, wheel size, and gear train to match what you posted. http://imgur.com/MTbNNXP http://imgur.com/MTbNNXP
TL;DR 60A = sketchy. Run 1:7~1:8 for no brownouts & more fun
Without advanced controls, you will get brownouts. Keep the “pushing” current draw to 35-45A for a better user experience, or add advanced controls, or both. You don’t need to be in a pushing match to get those high torques - just turning or changing direction on a skidsteer robot is enough!
We ran close to these outputs this year (20 FPS / 55-60A pushing on 70lbs skidsteer robot) and had to run TalonSRX speed controllers to get current limiting to 40A per motor.
You’ll have trouble getting enough torque (without brownouts) to turn those nice sticky inflatable wheels via skid steer/tank drive. Can’t tell what your center drop is but I’d guess you want at least 1/8" - I wouldn’t want to go any smaller without extensive testing.
From a ‘driver experience’ perspective - we’ve also found that this speed range is so fast it’s hard to control. It’s frankly terrifying to drive our robot this year, because it just leaps around - IMO it’s more fun to learn on a more forgiving speed ratio and be able to pass the controls around to inexperienced drivers on the team. EDIT: This can also be accomplished by putting an artificial software limit on the control inputs at like 70% during training, then ratcheting it up as folks get more comfortable.
Bottom line I recommend going to a higher gear ratio to prevent brownouts and add fun. It’ll “just work” when you build it.
Hey I appreciate you give me some feed back from experience. I’ll go ahead and increase the gear ratio to make it more reasonable. I haven’t ever used that calculator so thanks for linking it to me. That should help me be more precise and not have to guess the correct ratio. I will update the design to better suit the motor curve.
For pneumatic tires, you may want to bump up the drop center to 1/4". If you keep it at 1/8", you’ll want to be checking the air pressure on the center wheels very regularly to keep the total weight off the corner wheels to reduce scrub forces in a turn. For a single speed gearbox when you aren’t worried about ramps, it is often a good thing to design for “time to a desired range” which will be larger in years such as this one where you had to carry a gear from one end of the field to another, and shorter when the field is more broken up or there are no long distance runs anticipated.
One very minor point, it looks like you are using a cRIO vs a roboRIO. Is that true?
Its an interesting part layout. We typically put the transmissions in the center and attach the belly pan the the bottom of the frame. In your design it looks like the belly pan is higher. While it give you an open space to place the electronic components, it also raises the center of gravity of the robot (CG).
Only by specifying your design goals can you determine which trade-off is better.
Same goes for the reason to use 6" wheels on a small robot.
If you didn’t really have a need for the 6" wheels and larger gearbox, you could do something like this using 4" omni’s and single reduction clamping gearboxes driven by two mini-cim’s. It’s next to an average sized swerve for comparison.
Looks good for a first CAD project, very good. JVN Calc is absolutely amazing too.
A pushing current of 60 amps is absolutely fine. Browning out isn’t the end of the world as long as you avoid getting into pushing matches in the first place. Given that this would be an offseason deal, that’s fine, but anything above 20fps will start to get uncontrollable for drivers not used to it.
If you try and gear for very low pushing currents on typical FRC robots weighing between 100-150lbs fully loaded, you’re not going to be moving very fast at all.
+1 on using 4" wheels; less gear reduction is needed. If this is your first WCD, try using COTS gearboxes as well.
Also, for future reference, you can post pictures directly to CD instead of using a paper.
OP’s pushing 70A on the original design, I had the wrong wheel size input. And that’s after assuming OP hits the 60lb weight target, all in - miss to 70lbs, pull 80A… better sizing will reduce the sensitivity.
We got repeated brownouts at SF with 60A on the calculator. It sucked, took mentors to diagnose, and we couldn’t actually fix it until our second regional. It “might be okay” for an offseason demo bot, but when it’s someone’s first project and destined to be the template for a competition frame, I tend to go in a “momma-bear” direction on sizing advice.
Browning out isn’t the end of the world as long as you avoid getting into pushing matches in the first place.
We thought so too… until we found out the 35 second field reconnect time was enough to collect >100 points in penalties…
…COTS gearbox…
+1 if OP’s goal is “accomplish the act of driving ASAP” - which is an important team capability (see: “drivetrain week 1” thread) but might not be the only goal of this project
Shutdowns/reboots/etc. only occur under very high voltage drops. A regular run-of-the-mill brownout will cause stuttering due to output being shut on an doff quickly but it won’t reboot your robot unless the current draw keeps up or other things happen.
1072 ran with an 80A+ pushing draw all season without issues (geared for ~16.5 fps free speed). We had one reboot after getting high from both front and back at the same time at SAC, but I’m not sure whether that was a router reboot or a RIO reboot. 115 ran considerably higher in their high gear (22fps+ free speed) without autoshifting code and only suffered a single robot reboot the whole year. Any robot in a typical FRC game that weighs 100lbs+ fully loaded is going to draw a lot more than 60 amps in a pushing match if they’re geared reasonably fast (say, 16fps free speed). IMO, there are three options in an FRC robot: gear fast and risk brownouts, gear slow and have reduced performance, or have a shifter. Swerve drives and holonomic drives do other weird things.
All of the above assumes 4 CIMs in the DT. 6 CIMs will brown out the robot much faster at the same current draws per motor.
You mentioned you had a skid-steer robot; does that mean you didn’t run a center drop?
At the same current draws per motor, a 6 CIM drive will certainly brown out faster (because it draws 150% more current). However, if you keep the same gearing (i.e. same free speed) and just add an extra motor, you only increase the total current draw by 2*free current (~5A for CIMs). You’ll get more power (better acceleration) and lower per motor current draws (less tripped breakers). There are obviously downsides to this (heavier, not improving top speed), but for some strategies it may be valuable. Likely the best solutions for going to 6 CIMs would be to bump up the top speed a bit, but not to match current draws per motor where you’ll start browning out.
Hm. We were able to replicate our radio reboots by stalling a 4-CIM drivetrain with same high gearing in the lab.
1072 ran with an 80A+ pushing draw all season without issues (geared for ~16.5 fps free speed). We had one reboot after getting high from both front and back at the same time at SAC, but I’m not sure whether that was a router reboot or a RIO reboot. 115 ran considerably higher in their high gear (22fps+ free speed) without autoshifting code and only suffered a single robot reboot the whole year. Any robot in a typical FRC game that weighs 100lbs+ fully loaded is going to draw a lot more than 60 amps in a pushing match if they’re geared reasonably fast (say, 16fps free speed). IMO, there are three options in an FRC robot: gear fast and risk brownouts, gear slow and have reduced performance, or have a shifter. Swerve drives and holonomic drives do other weird things.
All of the above assumes 4 CIMs in the DT. 6 CIMs will brown out the robot much faster at the same current draws per motor.
You mentioned you had a skid-steer robot; does that mean you didn’t run a center drop?
Interesting data points.
4 CIM, ~23fps free speed/19 adjusted, “center drop” accomplished by running a 2.85" wheel at the front and 3" wheels in the middle and back. Double checking on the calculator, I think we were closer to 70A than 60A. Admittedly, there was also a convolution of the brownouts/sizing issue with at least one loose wire at SF. At SJ we had both gotten the talons implemented & had gone on a diet - down to 70lbs.
Somewhat true. With the right gearing, a sharp and short collision will pull your battery voltage low enough to reboot the radio. The RoboRIO goes in to “blackout mode” (turns off) at 4.5V, and I don’t have an exact figure on the radio but I believe it’s slightly under that.
Makes sense, with the gearing you’ve described this is one of the above scenarios, IMO.
6:1 doesn’t tell us the full story. How serious are you about staying in the 60lb weight budget? How fast will you go at free speed? What stall current will you draw?
Have you considered going to a COTS gearbox? Even if you keep fabrication in house, look at the VEX and West Coast Products gearboxes, they do some clever things that you can put in your design, even if you’re fabricating on a manual mill instead of a CNC…
