4237 pre-season WCD review

Team Lance-A-Bot presents our 2019 Preseason Training and Prototype WCD Drivetrain. Special thanks to the 973 RAMP, 3506 YETI, and Triple Helix videos!


Summary:
Chassis

  • West Coast Drive principal
  • 1”x2”x.100” Aluminum VersaFrame
  • Eight 4” Colson Wheels
  • .09" Drop on 4 center wheels
  • 25H Chain- sequential drive
  • Wide grid Belly-Pan
  • Quick change Bumper mounting

Gearbox

  • WCP 3 Flipped Motor Gearbox; 6:1 total reduction
  • 3x Rev NEO
  • Estimated real speed 13.4fps
  • 18T output sprockets

I think we’ve overbuilt this, but want a solid example to start from. We’d like to reduce the Rail and Belly Pan thickness for weight, but need to learn where not to, or how to compensate. The gearbox standoffs would be shortened in a competition chassis.
I’m also curious about if a ~140lb robot will be traction limited, under powered, or overpowered.
The Bumpers are quite tight and somewhat complex in order to minimize direct impact on the quick release clips.
We’re still waiting for a few parts before we can actually test drive the unit.
Please share any advice ,thoughts, or questions. STEP files are available at this GrabCAD link

Hi, can you post a photo with the bumper hidden so it is easier to look at the rest of the WCD? It looks cool!

Good point. We wanted the bumper mounting included in our planning. The battery probably isn’t needed either. Hopefully this shows more of what you are interested in.

hole

Out of curiosity, what is the hole used for? If its not used for anything, I would suggest removing them for more structural integrity. Overall it looks solid in my opinion!

On a follow up to that block, is that block necessary as can’t you just put the latch-bushing on the frame and not extend it upwards?

The hole was only to reduce weight. The strut was to support the upper corner of the short bumper from a high impact. I’ve seen those short sections break off too often. But I think you are both correct if the corner bumper joint is solid. Thanks for the input.

1 Like

In response to being traction limited, under powered, or over powered. My team had a similar drivetrain with 6 Colson’s instead of 8. You should be fine on power with your current gearing. As for traction you definitely are going to run into issues, especially with 8 wheels. We had to current limit ours and we put Omni wheels on the 2 front wheels. Up to you for your goals with the drivetrain but don’t do what we did, test your drivetrain before the first comp lol.

Unintuitively colsons have nearly the same traction in 6 wheel and 8 wheel configurations, because traction is defined by normal force and CoF mainly. For treaded wheels contact area does make an impact due to the teeth on the tread interacting with the carpet.
Current limiting your drive train is typically a good idea and (I think) it’s standard practice for the comp robot.

Design feedback:
6 neo is commonly seen as unnecessary/excessive, although there is some noticable benefit when getting up to higher speeds like 16.5 FPS free as you have
Those gearboxes are, uh, quite large by todays standards due to being designed for CIMs, and 3 motors just don’t package well when flipped. You might look at some non flipped 3 motor gearboxes, or even custom ones.
I’d be worried about the bumper rail support brackets, they don’t have a lot of meat on them and might buckle.
I agree with removing the 4 bumper bracket mounts with the big holes, just put the latches straight onto the chassis.
Those slide latches are liable to come off in match and aren’t exceptionally robust but many teams use them without issue
Make sure you think through the electronics mounting.

Actually responding to OPs questions:
Almost all FRC robots are traction limited, under and over powered is related to sprint distance but 4 neos aren’t underpowered for a full weight robot usually.
In a pushing match you’re effectively optimizing for top speed vs current draw when pushing. 49A is pretty safe to run through a 40A breaker for a decent amount of time but 294A(49x6) through the 120A main breaker is not.
Rail thickness is a big trade-off, personally I think DT rails should always be 1/8" or .100" and alu belly pans .9"-.125" gussets can easily be reduced in thickness to .9" but some teams have run 1/16" wall

Side note: free speeds are most often used for comparing drive trains because the method used to calculate “real” speeds are inconsistent and usually pretty inaccurate

Disclaimer: I’m a student but this is all good information from what other teams have done in the past to the best of my knowledge

1 Like

Great Chassis! I see that you are using 25H chain. If you use 25 chain you can put the chain in the tube with double sprockets and save some room.(https://www.wcproducts.com/25-sprockets or FIRST choice)

We used 25H in tube. What restriction have you found trying to do this?

How are the 8 bumper mounts attached to the frame? Sheet metal screws through the small holes in the top surface? Are the 4 in the corners attached to the frame with only 1 fastener?

It appears that the battery is meant to be lifted vertically to get it out. It may be easier if you create a wide “indent” in the strips along each side of the battery so your fingers can wrap around the bottom corners of the battery.

The real question is “Are all the electronic going to be zip tied to the belly pan or is there another plan for that?”

Very nice job! It’s great to see a design approach where the bumper interfaces are so thoughtfully considered.

I wanted to quickly comment about a design tradeoff inherent in how this corner is constructed:
image

First of all:

  • Plenty of teams do it the way you have currently designed it, and it’s perfectly valid to proceed like this!
  • What’s great about your current design is that it’s very easy to support the bumper, because you are free to extend the ladder bar all the way to the frame perimeter.

If you look at the drive rail as a free body, there is a force couple on it [a torque equal to (half the weight of the robot) * (the wheel cantilever distance) ] that is totally reacted by both in- and out-of-plane loads on the gusset plates.

Alternatively, if you were to use construction method {B}, then this couple could be reacted by bearing between the faces of the tubes as well as by the gusset plates. Team 610 has some great historical examples of doing perpendicular tube joinery using this principle.

7 Likes

Is there any actual benefit of doing part B. Team 2046 uses A because we actually extend our front bars to the width of our wheels. We have never had any sort of issues and we’ve even jumped off hab 3 last year with no issues.

1 Like

If you end up overloading your chassis to the point of catastrophic failure, then construction method B will not fail quite as catastrophically as construction method A.

Both methods are strong enough for FRC drivetrains. This concept may be more applicable to mechanism construction, if you are deciding which tube to extend to the full length of the joint. If one of the tubes will see a large twisting force, then you want the other tube to butt up against it.

Thank you all for the thoughtful discussion. I’ll try to catch up…

We originally didn’t consider Chain in Tube because we didn’t want to reach too far. If we had seen the Triple Helix video and other discussions here on CD, I would have been less apprehensive. I will look for examples and discuss at our competitions this year and propose it for off season next year with any other major changes.

I chose an 8 wheel design because I expect it to be easier to adapt to a 6 wheel rather than visa-versa. We want to let the game strategy decide 6 or 8, drop wheels or none with Omni’s, and also if we use traction or Colson wheels. We have a lot to learn on all these options, so I really appreciate the comments like the one about how the traction corners affect the traction limited concerns.

We chose these gearboxes because of the flexibility for testing (if we have time). The gear ratios have many options that can be changed quite easily, removing the 3rd motor, and even a pneumatic shift if ever desired. Compared to what we’ve done previously, the space on the belly pan looks wide open, so that is very appealing. While most beneficial for CIMs, I am curious why the ‘Flipped’ option is not “packaged well for 3 motors”. Initially, I expected to reuse them, but agree a simpler option should exist for competition.

Considering that the NEOs have a built in encoder, is it worthwhile to add an encoder directly on a wheel axle?

Are you referring to the latches, angle bracket, or posts? Yes, I am worried about the whole system. The Posts are held on with a 10 Buttonhead, the clips using 2 pop rivets, and the angle brackets with wood screws. I’m afraid we will not get the bumpers done for testing.

Well thought out and well supported post. (and I’m not saying for “for a student”)

The 8 Bumper blocks/struts are attached with 1/4x20 screws, and the 4 corner ones only have 1 because I did not start the hole pattern correctly. Planning to fix this in the next revision.
Love the finger slots idea.

I expected more electrical and pneumaticc features in the BP, but others preferred separate boards, feeling that they are too hard to access (under mechanisms) if mounted to the BP. I’d be interested in how others feel about electrical and pneumatic access when components are mounted to the BP. This is compared to our practice of 1 or 2 separate boards mounted to various framing.

Extended Drive Rail -vs- Extended Ladder Rail @Nate_Laverdure, @NustinJewton, @s_forbes
Thank you for the insight; I was concerned on this without understanding the factors. In the end I couldn’t think of an easy method for Opt B to support the side Bumper Rail. But if we find we need a steeper approach angle for the front wheel and remove the LR extension, your information and examples will help.

1 Like

I was referring to what looks like pocketed 1x2 that holds up the red angle in the middle. You removed a lot of material and I’m worried they might just crumple. I’ve heard that the slide latches can break when loaded wrong (lifting up on the bumpers)
For reference this is the CiT WCD that I cadded for the offseason, I’m also concerned about my part designed for that purpose tbh.


here’s some very compact gearboxes from 33 for example

I have read that the pins on 25H are slightly longer and rub against each other on double sprockets.

1 Like

@Van_DanderWide you and your team have put a lot of thought into this design. You will learn a lot from this exercise.

Access to an electronics panel installed on your belly pan will depend greatly on the mechanisms you do or do not install on top and their design details. For some mechanisms, you may be better to install the electronics panel on a vertical surface on the side of the mechanism. Regardless of location, I would recommend mounting the electronic components on a light, nonconductive subpanel. This allows the control system to be assembled in parallel with the chassis. It can be pre-tested before installation. This way of working reduces the time spent waiting for some other group to finish their work.

This post was flagged by the community and is temporarily hidden.

7 Likes