I’m surprised that nobody has mentioned the robot fire at Michigan State Champs yet.
Yeah. (https://www.chiefdelphi.com/forums/showpost.php?p=1581366&postcount=11) 
I call this match “New Drivers Vs. Intake Rivets”
This is a good lesson about planned failure points though. If we had used steel rivets instead of aluminum rivets to hold the tubes making up our intake together, the intake would have broken further along, and in a far harder to repair place.
As you can see in the picture, 2781 flipped over, we (234) lost a bumper, and 461 dislodged a defense. I believe this all was in auto. Thank goodness, we were able to replay this match because of the field fault.
Edit: Also, all 3 of us are on the same alliance. This was at the off season event Boiler Bot Battle so correct colored bumpers were not required.
Unless those wires shorted to each other, that’s probably not what was happening. Assuming everything was taped over as it should be per the rules and the inspection checklist, the loosening of your connections was increasing the resistance at those connections. Higher resistance with the same current (P=IR) means more power dissipated. More power dissipated means the main breaker heats up, and since it’s a thermally tripped device, it opens faster shutting down your robot.
Either that or it’s really loose and wiggly and you intermittently lose power. Even a fraction of a second will reset the RoboRIO and Radio.
It’s a problem that pops up occasionally that often gets the attention of FTAs and Robot Inspectors due to the random shutdown. A bit of diligence when tightening and some lock washer (split, toothed, or other) will typically prevent such a problem. Although the extra pounding induced by the game this year may have challenged even those preventative measures.
About a week into October, the pancake on our gearbox decides to come unattached somehow. And second match at WMRI our robot’s autonomous decided to go the wrong way and she ended up mounting one of the other teams’ robots XD
Impressive, I don’t think 125 broke any of the 3d printed parts on the bot last year. This included the entire intake roller being 3d printed.
Guess it just goes to show you that 3d printed parts aren’t inherently weaker it’s all in how you design and use them.
We were relatively lucky. We only blew a few pneumatic tires and stripped out the plastic VEX versahubs.
I mean, they are inherently weaker than injection molded parts - it’s just a matter of if they are designed and built strong enough for the application. But your general point is right, I’m just being pedantic.
In general, this thread has many examples of design practices that are potentially fine being written off as universally bad ideas that can’t work, or design concepts being blamed as the root cause of problems that they may not have been. 14 tooth pinions do not universally fail in 3 CIM gearboxes, for example, nor do plastic hubs.
If making a general rule and building very conservatively works for your team, that’s great, just be sure not to give things out as general advice without a solid grasp of the “why”.
You’re right, the intent of “inherently weaker” was “inherently too weak”. They are weaker.
Gotta have a solid frame for a game like this and for antics like these:
Yes, the other team was slightly terrified we were stuck in high gear at the wrong moment without bumpers and nearly landed on them.
They also learned to keep the battery strapped in a bit better since it disconnected and flew from the robot.
We did drop to the floor once from about 4 ft from the ground during testing but I can’t remember any other serious drops on the field. Most of the climbing failures just resulted in a slow and uneventful descent. I was more worried about drive components cracking than frame this year.
I’m slightly surprised BillFred has’t posted his sign from Palmetto in this thread, for a while the CDF was not an option, they were ALL broken! Granted this was before HQ had the metal reinforcements put in them, but still, 1/2 inch polycarb is pretty hard to break. Other stories of motors releasing their magic smoke, I’m sure there are plenty, as well as connection losses because of wires jarring loose, I know we had that problem in the finals in Palmetto. We also had a minor issue with our 3d printed drive pullys, we used an elongated version of the rhino track this year with some Breco-flex belts and originally we 3d printed pulleys for them to run on, needless to say, those didn’t make it through one competition, we replaced the 4 corners with metal pulleys for our second regional.
The best part about this one was that the breaker was hit when the bot wasn’t on the batter. It was pushed on just before the end of the match.
As an aside, earlier in the competition we also got to see first-hand 330’s main bot do a recreation of the Einstein gymnastics after flopping on their back. That was cool to see in person
1058’s tougher times this year:
Reading Match 1
Our climber is a double linkage designed so the gas shocks keep it down before a piston pushes them past the center point so they take over and extend up. We some drop tests and drive tests over defenses and it worked fine. First match in autonomous the climber extends upwards after driving over the rockwall and we all had this moment of, “This is very bad”. The root cause was a knot of surgical tubing was preventing the assembly from collapsing all the way. We adjusted the knot, added a small strip of painters tape as a temporary fix, and added a piston lock as a permament solution for the rest of the season. We also destroyed about 8 halves of the AM 6" Pnuematic Tire Hubs over the course of the event. We had plenty of spares but we knew we had to start making our own hubs asap. This meant our addition of a shooter was delayed by an event so we were testing and tuning it during the New England Championship.
Any match on Caver
Our robot had about 80 official matches on it by the time St. Louis rolled around. We added our working shooter at the previous event and made a better intake for Champs hoping to see our cycling improve. Little did we know most of our matches there had some major failure or breakdown leaving us disabled or limping on the field. The plates holding our gearboxes had bent into the robot so we threw the gearbox chains the first match. Our climber had be used so much that after we scaled the tower the hook snapped and sent the robot the ground. We sheared the bolts on our metal hubs. Our pnuematic system acted funny so one or two shots barely lifted the ball out of the robot. We had some miscommunication on changing some autonomous modes (completely my fault) so we didn’t fully cross some defenses. We had some odd electrical problems leading to brownouts. It was a rough event knowing we had potential to do so much more but it was the point where Stronghold was getting the better of our robot.
Mainely Spirit
We had some odd pnuematics issues later in the day where near the end of each match we had no air left in our system but once we enabled in the pit the compressor turned on immediately. When we got home we diagnosed the compressor was shot. After the fourth or fifth cycle of needing air it refused to turn on even with a signal from the PCM.
Stronghold is the hardest game I’ve played and I’m amazed how well our machine has held up over 136 matches plus practice and demos.
Of course, how could I forget about our wonderful match in the Iowa regional?
First, the buildup, so we had a camera on our robot, but it was A.) not implemented, and B.) Worthless even if it was because our flywheel didn’t have a working encoder until last month (When I eventually became electrical lead).
So it was removed middle of the regional.
Turned out that caused something to go wrong in the teleop programming, so after a successful breach of some static defense, teleop starts and the robot turns, and keeps turning, in a circle, because it was going in donuts and left a mark on the field that was not removed by closing ceremonies.
Fun fact: There is no full video of that match, anywhere. There is video of the robot on it’s own going in donuts, but there is not any of the full match.
Where was that video from this season of the robot at a scrimmage blasting down the field wall and going into a crowd? That looked pretty intense too. Nobody got hurt thanks to some nice dodging skills but that could have gone badly.
During a practice run, our autonomous decided to drop out intake, (Which also functioned as our obstacle manipulator.), and run over it. The 1/8" aluminum was bent and we lost a mini cim as the wires were pulled out of it.
And how could we forget at Mainely Spirit when our practice robot, 8501/1056 sheared one of our gearbox output shafts in our second to last qualification match. Because of this we epoxied the shaft back together and had to sit out our last qualification match while we waited for the shaft to dry. We ended up still being the #7 alliance captain with 5687 and 910 (319’s practice robot’s drivebase) and making it to semifinals, beating the #2 seed with 319 and 133, two crazy powerful robots despite numerous electrical faults. We tacked those up to a loose PCM power cable that shorted out every match, eventually burning up the 20 amp PDP fuse that powered the VRM, and subsequently, the radio.
Our team ran into the same problem and we didn’t even think to check the wires until the end of the regional.
We don’t have any pictures, but during our first district event, the casing for one of our drive CIMs decided to disconnect from the shaft face and internals. There was about an inch of exposed CIM innards and the motor had a hard time functioning. We were unable to remove the CIM during competition and ended up using the largest C clamp we could possibly find to keep the casing attached to the face and gearbox.
Lol Our Team (5842 Royal Robotics) Used 2 100 lb pressurized gas shocks like those found in cars to hold a hood or tailgate up, for our scaling mechanism. Unfortunately however we never got to scale in a match with these.
At the South Florida Regional last year we were using our new amazing “safe” way to torque out the shock till they locked in the match starting position when something happened… ill put it short. "Do not over extend gas shocks!"
The end of the 100lb shock shot out right in the direction of where one of our students had just been standing where it ricocheted of a steel knight helmet ( leaving in a sense a bullet ricochet mark) and shot through another pit putting a hole in their team toolbox lid and the disappearing forever.
We now know if FRC does not allow gas shocks in the 2017 season we were the reason why…