937X Final Robot Explanation | VEX Push Back

Added: 2026-05-12

[00:00:00]Hi, I'm Brandon.

[00:00:01]>> And I'm Lucas.

[00:00:02]>> And this is our world's robot explanation. So, we took this bot to two comps being states and worlds. Um, we made quite a bit of changes after states and some people wanted us to make an explanation video, so here it is. Um, we did pretty good with this bot. We triple crowned our states. um world. Our skills unfortunately didn't work even though we had 119 max with Prague, but we still managed to reach semis in our division even though we unfortunately lost to to a battery disconnect and we won our design or our division's design award.

[00:00:33]So yeah, I'll start off with the drive.

[00:00:37]Oh shoot, we ran a pretty conventional drive being um 450 RPM on 3.25 in wheels. Um, something unique we did is we ran Lathe 48s, which were just from previous robots, but this allowed us to fit the two barrier cross wheels, which are wrapped tractions pretty snugly because of the zip tie heads. And this just dropped the weight of the drive in general. Um, all the wheels we ran structural screw joints, which are just standoffs and then two screws in opposite directions. And then this binds the drivetrain together. For the rest of the bracing, we just did one high high strength axle in the front and then one full length C channel in the back which house like our liner and the main mounts for like the entire intake. And we found that like this was enough. So yeah.

[00:01:20][snorts] Um something unique we did as well is with the sleds. You can see here we have this like 4x8 piece for the middle section, but the front and the back half of the drivetrain are with drilled standoffs instead. This is just because we wanted to save plastic. Um, especially with the main design consideration with this bot, which is be able to fit a nineball capacity while having full um centered odon pods with retract as well. So, this is just necessary because the pods need made out of poly to like actually retract properly. [snorts] And we found no issues with this because they were just bent tangent to the wheel and they're pretty smooth. Um, something else we also did unique with the drivetrain is all of the motor axles didn't have shaft collars and instead we just had um, at least for this motor sack here, we had a press fit uh, 1/8 inch nylon standoff and then for all of the uh, axles inside the drivetrain, they were just capped off by the slid piece. This just dropped some weight because we didn't have to use shaft collars or have to maintain the shaft collars in case they came loose or anything. and the press fit nuts like uh kept the axle in pretty well as well. So, a major change we made for Worlds as well with the end of the drive was by bending this in on both sides. And then this just created an antifunnel if you've seen on many other teams before and this just makes winging like have a little bit more tolerance because the goal can slide on the outside. So, we found that helped and yeah, overall it was a pretty good drivetrain. We did have to do a motor stack in the front which is a little bit unconventional but this is just to fit the odome retract in the back as well.

[00:02:58]And now moving on to the intake which is probably the most unique part of this bot. And a lot of people asked us how we got our intake so fast. And the way we did this was by having 22 watts connected on the entire S. So the intake motors which are mirrored on either side were relatively low compared to the drivetrain just to drop COG. But this also allowed us to gear the entire intake. So from both of the motors, they powered this central larger 3-in roller which ran at 900 RPM and then all the other consecutive stages ran at 1350 RPM. And the way we were able to achieve um storage with nine blocks on S was with like these special rollers that we made in early season. You can see it in our early season explanation as well.

[00:03:40]They're basically just flex wheels gutted um with zip ties through all the spokes and latex tubing is press fit onto those zip ties. And what this allows is the latex tubing has relatively low friction on all the balls. So, um they could store easier compared to like normal bands, but the RPM difference made it enough so like the intake was relatively fast with the scoring as well.

[00:04:04]And then for the rest of the intake, it was relatively basic. It was just like a tilter design. Um, and then we had this piston hood at the end to store all the blocks. The piston hood had this like heat bent section down just to reach deeper into the blocks. And then the flat edge prevent us from being like winged back from on the goal as well. So if you want to demo the storage as well.

[00:04:28][snorts] Yeah. So you can see here even though it did slow down a bit at the end, our storage with nine was like um really good and especially after states we made a change where we ran zip tie compression instead of mesh and this just helped like loosen everything up for scoring. It is our long goal scoring.

[00:04:56]So yeah, that was how we were able to like tune it really really fast since we only built this bot like I think like 3 weeks before our states and had it like tuned about the same speed. So yeah, and then um going on to the rest of the bot, we had this like mid goal aligner.

[00:05:15]It was pretty basic. Um it was a little bit worse than our old Rean robot as well since it was shorter, but this is just to make space for the mid goal score as well. And um wings we ran pretty traditional for states, which is just one wing in the back. But for world, we decided to switch over to dual wing since we thought like its capability in match play was a lot better for being able to catch balls during like ping pong like interactions and stuff. And we use this like pretty well at worlds I would say. Um long go aligner was pretty basic. We just ran long longer um since this helped us keep straight on the goal. But something different we did for worlds was we used these gears. And what this allowed us to do was to not tip up on the goal because the gears would basically hit like the underside of the goal because this is an issue we had at states where in one match we got like shoved up onto the top of the goal and this like we didn't want this to happen basically at worlds as well. Um other than that the bot is just a pretty basic Sbot. Do you want to explain like the distance sensors and odon pods?

[00:06:20]>> Yeah sure. So, um you can reference back to our other explanation video from the start of the year for most of the same information, but um we have like the same four distance sensor setup that we had on our first bot at the beginning of the year in November. It's just one every 90°, one on each of the axes. And so, we positioned them a bit higher this time just so that they didn't interact with the balls on eye level. And so, this allowed us to get like less noise in our readings and less incorrect readings. And um yeah, one of the we just do distance sensor resets. And so we make sure that our distance sensor resets are always at like really known locations on the field and are always like we attempt to get the distance sensor as perpendicular to the wall as possible just that we get the most amount of refraction back from the sensor into the or back from the light source back into the sensor itself. And so this is something that we've used throughout the season and we've been able to get like pretty consistent autos and skills runs because of it. Um [snorts] that comes in addition to the the really nice ODOM pods that we have. And so um Brandon already mentioned it before, but since they are zero offset, we don't tend to get as much drift from the ODOM when we're doing point turns. And so um this just makes it easier because every time we point turn, we don't necessarily get as much error build up. And so having them be um centered tends to help with that. Um not entirely sure why, but that's what we've figured out through our testing. And same 30° offset as before for the wheels using the inserts.

[00:07:56]And so um one thing that we would have hoped to have done better this year is to try and like get off of ODOM. Um, and we've known teams that have like successfully done this like just using raw IME val encoder values from the motors and um, distance resets. They were able to get like consistent enough autons and skills performance. And so I think that was one of the big limiting factors of this robot. just having to build around like this huge floor space on the bottom of the bot in order to fit the ODOM was just very like incon like was not very conducive to like um say our performance during skills with like the park clear and other sorts of obstacles.

[00:08:38]>> Yeah. Um uh also going on to like more of the unique aspects on this bot despite it just being like a regular S.

[00:08:46]Um, you may have seen already as well when we like started up the code that the mid goal dcore like fired up. This is just because our mid goal dcore is linked to the odometract. This just helped us save a little bit of weight.

[00:08:59]Um, and something unique we did which was our innovate submission actually is we had this um latch on the side which is just this like plate metal bar with like a cutout on one of the holes. And the idea for this was that um it would be like a one-time deploy. So the uh the zip tie, there's like a zip tie on both of the strings that the Odon pods are connected to and then the latch would just basically like go under the zip tie and hold it there. And then this would mean that like at the start of the match, the Odon pods would stay retracted, but our driver could still use the middle goalie score however he wanted. And then for skills, we just um we just disabled the latch by like zip tying it back to the piston. So that was pretty unique. Um as well, a lot of the design structure on this bot was um like designed in mind to like have things not really bend as easily because on our old bot especially as well there we had a lot of issues with consistency especially if we started running it like a lot through skills testing especially autons and taking it to multiple comps. So, like our ideology with like the aligner, which is why we this is why we switched to a high strength axle aligner. It saved both plastic and it was just stronger overall. And then to make sure this didn't bend, we boxed it through the C channel with um a tapped collar.

[00:10:25]You may not be able to see it from inside, but this basically meant that like one side of the collar, which is the normal uh screw for the set screw, we used for this screw, and then we drilled another hole on the other side.

[00:10:35]And this just allowed us to box it through the cross brace C channel while not like having to add like a bunch of additional bracing and stuff. And then the main um intake uprights as well was this diagonal C channel, but it had a bunch of like flange cuts as well, as you can see, for this shorter vertical C channel to go across. And then we drilled a hole for like a drilled standoff to basically um center these two vertically. And we just did this through CAD just to make sure um this was 90° relative to the bot. And we mainly did this because um we found that like with a snacky style Espot doing good wings on it was really hard with like the diagonal C channel. So we wanted vertical mounts for the top which is what you can see here. [snorts] Other than that um this bot did weigh a lot at around almost 17 lbs I think. But this was just intentional because we really didn't want anything to break on this bot if we had as we had with like prior robots in the past. And we're pretty successful with this. We didn't really change anything throughout worlds too much and everything was pretty consistent. So yeah, overall we're pretty happy with this robot. Um I mean there's not really much we could have done for skills sadly because we just didn't have a new field to test on and all of our goals at home were worn down.

[00:11:54]So the friction was definitely less.

[00:11:55]>> You want to talk about the eyelash?

[00:11:57]>> Oh yeah. Um, one mech we had to do for skills especially was this like um this eyelash mac mech um on the bottom with this like block of foam. And what this basically did was with our tilter geo if you want to drop it. Um there wasn't actually enough compression between like the floor of the tilter and the the last rollers. And this was good for match because we could just automatically drop two balls into the goal without like pressing our intake. But it was bad for skills because there just wasn't enough compression. So, uh, we had this mech with the block of foam which basically just like added more compression into the the tilter floor. And this like kind of worked, but we just added more zip ties as like a flooring as well for the ball to drop onto, which we tried to make consistent for skills, but it just really wasn't tunable.

[00:12:49]And yeah, I think >> yeah, we had um before we had like a color sensor here like mounted somewhere. I think it was here. And so um we used the color sensor like in an attempt in skills to grab the entire stack of four right after the first um park clear. And so that would allow like we thought that would increase the consistency since um you might have seen on some of the China routes that like if they grab all 10 blocks they will like put four or seven fully into the mid goal and then once it sees the other color the the color sensor will stop it.

[00:13:23]But one of [snorts] the issues we had specifically with our robot is just since there is so much of a gap here and there's not too much contact by the time it saw the red, it was very like inconsistent in where the placement of the final blue block would be. And so um what we found was that sometimes the red would score or sometimes the last blue wouldn't score just as a result of our mid goal design. And so we ended up getting rid of the the the color sensor and just opting to grab just the one blue ball from the stack and then grab the other three later on. Um some other issues with the skills route. Um like the the intake on the park clear was also like not ideal for um grabbing the middle at first. I don't know like what the entire issue was but >> so our at least from our testing I don't know if a lot of other teams like XP like managed to like bypass this issue was when we tried front clear especially on our old robot with our recon robot we found that like with the three wide aspects of the ball set up against the wall it was a lot better for the middle ball to be um pulled in first because it just made the funneling on the side too like a lot easier but What it this issue created was that like our match loading became much worse because even with like the match loader funnels. If the flex wheels like in this middle were too centered then the flex the floating stage started like bounce and then they would start like mixing down in this bottom. And this really like messed up our skills at the match loader because we needed our control bonus to be all like six of the same color from both of the match loaders. So, we like kind of combed this issue by doing um half cut flex in the middle where does this have enough grip like normally, but um when match loading on like the higher lip, they would like split outwards and like center the ball. And this like kind of worked, but we found that like our tuning for this just didn't really work at Worlds for a reason. Maybe it's like the new blocks just being like a slip, like having um a slipperier surface, but it just really wouldn't like funnel in the sides anymore after grabbing the middle, and they just wouldn't um grip at all. So, that was one of the issues on this bot that we just never really got to tune or understand why it wasn't it was an issue. Um, as well, I think overheating was kind of a bad issue for this bot just because of the weight. our drivetrain like overheated a bit and then the intake obviously as well due to the 22 watt. But I mean that wasn't like too big of an issue considering the trade-off for like the speed and consistency was like a lot better. So [snorts] yeah. Did >> we ever talk about the logo?

[00:16:08]>> Oh yeah, logo. Um logo we just had a piston on the side here just to raise it up. And then um another thing we did for this bot was have the ramp be a lot higher than our old bots. And this didn't really change the pickup, but it meant that the balls when they were exiting the bot basically just exited at the level of the logo plastic. So we found out like that was a little bit better for the consistent seven as well.

[00:16:36]And yeah.

[00:16:38]Do you have anything else or >> um Yeah, there's not too much. Yeah, it's a pretty Yeah, it's a [clears throat] pretty simple bot, but a lot of people just asked us to make an explanation because it was a pretty clean bot and it worked well. So, yeah, thank you to everyone that we met throughout the season, especially HCU since we had around like four teams also build our robot after we gave them the CAD after states. But even though like they it didn't really help our design too much, it was it was nice talking to them and helping them out with their robots as well. And yeah, we're all seniors, so we've all graduated. But good luck to everyone competing in Override. [snorts]