Off-Season WCD

Instead of changing over to 1/16" tubing, if you’re running 1/8" wall tubing and can mill a high quality 1.125" hole in the tube, these two parts might both be un-necessary
image

Bearing blocks are only needed on 1/16" wall or if you can only machine a 1.13X" hole.

If I were to manufacture and test this design, I’d invest in being able to machine a nice bearing hole, remove those two parts, lengthen the standoff bolts so they go through both plies of the tube, and increase the diameter of the standoffs until they were 0.030" from brushing the chain instead of trying to package 4 standoffs into the design.

It looks like you’re planning to tap holes in the exterior plate and use loctite; make sure you have at least 4 threads engaged, not counting the first two threads on the bolt (bolts are rolled, rule of thumb don’t trust the last two threads). If I were doing the design, I’d shift the standoffs out until I could use nylock nuts, and upsize all the way to my team standard 1/4" bolts.

I’ve noticed on most west coast drives teams use bearing blocks even with 1/8” wall tubing. The times I don’t see bearing blocks is when teams choose to do chain in tube which we don’t want to do since there’s a risk factor. The reason I put in bearing blocks was I noticed when looking at teams CAD or technical binders, I’d notice bearing blocks even when they would run 1/8” tube and had the capability to machine a precise hole. Our team does however just drill holes that are precise and they end up being snug but never on a drivetrain since our team has only used the kit bot.

Also, wouldn’t lengthening the standoffs to go through both sides cause clamp crush to the tube weakening it? I’d need a Spacer and it would be hard getting a spacer that far in. If I were to move the standoffs further apart so I could use nylocks, then all the clamping force would be on the edges and I feel like the gearbox would flex more towards the top.

Wouldn’t removing the bearing block require a bearing on both sides of the tube which would cause the shaft to be over constrained?

In general: keep in mind the risk profiles. What’s the goal for this iteration? What is the cost of mechanical failure? What is the goal of the next iteration? What is it’s cost of mechanical failure?
I’d rather break something in the offseason and add weight inseason than build conservative in offseason and continue cutting weight inseason.

We use them at the gearbox because the WCP SS kit gearbox mounts off them, rather than because they’re strictly necessary. Then we can re-use the practice bot kit the following year.

You’d want a spacer for 1/16" tube, should be okay on 1/8" tube :slight_smile:
I’ve run the “pair of 1/4-20 through bolts” mounting system on 1/8" 6061 tubing on three teams over ten years with no tube crushing or failures traceable to that choice.
If you try it with 8, you might have a small enough stress area under the bolthead to be able to damage the tube.

You’re not wrong on principle, but you are at a different order of magnitude. You’re already clamping the edges of the plate. If that’s 1/4" AL plate, a half inch difference in location won’t matter, especially if the standoff can get wider. I’d say the same for 3/16" Al, maybe start paying more attention at 1/8" Al or with a plastic motor plate. Remember, the motor output forces themselves aren’t actually very large.

If you want to go to a conventional 4-standoff design and keep dual plates, you could make both gearbox plates plastic (PC, nylon, delrin).

If you’re worried about strength under shock forces as the robot drops over a ledge, maybe go to 2x NEOs to reduce the weight/lever arm or add a support strut across the space between the two gearboxes?

Nope, just keep one in the motor plate and one on the wheel side of the tube.

Even if you keep all three, if you are using flat plates, standoffs, and through holes there will be some assembly slop that can allow all three bearings to get aligned during assembly, despite being overconstrained in the classical design sense. YMMV, drive fatigue failures are generally reported at the ~4th event of the year; my teams have generally been 2-event, maybe 3. I’ve never had a fatigue failure in the drive despite accidentally overconstraining that shaft several times (with flat plate assembly stacks).

Thank you so much. I don’t want to cause the drivetrain to have a failure since then the team would likely not approve of it when the issue was a small issue that could’ve been a easily fixed. But what I am thinking of doing is don’t use the bearing blocks but order them. So if using just screws doesn’t work, I can always install the bearing blocks.

1 Like

Chain in tube now a days really doesn’t have any risk factor and is super easy to put together. We’ve done it for the past 3 years with 4" wheels + 16t sprocket, 5" wheels + 16t sprockets, and 6" wheels + 17t sprockets(dropped off lvl 2 every match for 8 events)—all with regular 25 chain—and have never once broken a chain on our practice or comp bot.

4 Likes

Wow, that’s impressive. The thing that concerns me is attaching stuff to the frame since the chain is right below the drive rails. If we use bolts or rivets there’s a chance of hitting the chain

We put 2 rows of holes to avoid rivets and bolts hitting the chain:

For the top we put 2 rows of tapped holes:

4 Likes

Here is a video series that was put together describing our CIT WCD chassis.

http://team2363.org/2018/05/2363-builds-a-drivetrain/

Is that 1/8 tube? Looks amazing and the two rows of tapped holes is a great solution I love it!

Wow, that’s beautiful. Sadly, we don’t have a (working) cnc router that could do that for us. If we can manage to get it to work that would be cool but drilling and tapping all those by hand would be a pain the rear. Our cnc router isn’t exactly a cnc router but is/was a cnc plasma cutter. Our lead mentor modified it so it can be a cnc plasma cutter, cnc router, and a cnc engraver but we’ve never tested it. The last time we tested the cnc plasma part of it, it worked but it also set the fire alarm off so there’s that. Also, what are the holes in the side of the tube for?

Yes, 1/8" wall tube with half depth pockets.

Ah that is unfortunate. We cut our tubes on our CNC router and then power tap the holes with a hand drill.

Which holes on the side are you referring to?

the bigger side. The one where the bearings are seated. Also, why is one of the gearboxes on a tube mounted to the drive rail while the other is directly attached to the drive rail?

It’s for elevator clearance. This DT makes a bit more sense when you’re looking at the whole robot.

2 Likes

Support the gearbox around the motor mount holes, it’ll give you enough support. I’d use the top and bottom motor mount holes, still using the side holes like you would attaching the motor to a plate.

The ones in white were spare DT encoder mounting holes, the ones in yellow were for our bumper frame, and the ones in blue were for mounting our elevator.

Like Troy said, we had a sideways elevator and in order to accommodate it we had to offset one of the drive gearboxes.

Yeah, I watched a video of your bot and I understand what you mean. So the stage wouldn’t come crashing down on the gearbox.

@juju_beans are you guys using a fixed c-c on the drive? I think your tapped holes in the top of the tube is a really creative approach to still be able to attach to the tube.

Yes we do fixed C-C with a ~18thou adder.

Only L-gussets for now :slight_smile: But we’ll have T’s in before build season as well.

1 Like

Hello,

This is cool overall, but there are a few things suggestions I’d like to make.
First: the gussets are pretty overbuilt, as are the 2x3 tubes you have between the wheels (I think you could easily get away with using 2x1x1/16).

Second: the 2x1 tubes are a good bit thicker than they need to be, plenty of teams use .1" thick tubing to great effect. I think it would be really interesting if you were to run structural analyses on the use of 2x1x.1 and 2x1x.125 “C” channel on the front and/or rear. In this application, tube may not be all that useful since the front and rear tubes aren’t having to deal with torsional forces. I saw someone in the comments suggest using 1/16" rectangular tubing, I really don’t think that’s a good idea unless you used 1/16" steel (which could be an interesting idea to test).

Third: The gearboxes are a cool design, but you do need more standoffs. You mentioned the reason for there being two was that you didn’t know where you could put additional standoffs. I think this is one of those areas where you either add standoffs or the gearbox just doesn’t work. A pair of .25" bolts just isn’t enough to hold two 2.8lbs motors. Plus, in a gearbox, things have to line up perfectly so this is an area where overbuilding is really good (to a certain extent). I might also suggest looking at ways to integrate your gearbox into the frame a bit more so you can get rid of that outer gearbox plate. Lastly, from personal experience, making a 180 degree gearbox and getting the motors out of the centre of the robot thus freeing up a good 6" is really nice.

I made some quick suggestion edits in CAD, so here that is:
https://a360.co/2WjyWOV

I have been working on a few of robot concepts over the summer partly to develop WCD chassis ideas for my team, if you have suggestions, I would love to hear them.

Artemis is a high pivot arm. Chassis and gearbox will be redesigned when I have the time:
https://a360.co/2Wx34sW

Athena is a 3-stage elevator lift concept using channel rather than tubes:
https://a360.co/30AXzYP

Hyperion is a scissor lift concept (work in progress) my team had a scissor lift back in 2018 (Spartronics 4915 2018 CAD Release - THEMIS (scissor lift)) that worked really well. Scissor lifts are great since they don’t have too many complex parts, are easy to assemble, and if designed properly, can be incredibly light:
https://a360.co/2p9k7Tc

1 Like