Sunday, January 20, 2019

Big Disk Energy Post-event Report: Norwalk Havoc January 2019

Norwalk Havoc January 2019 is BDE's best showing so far. Officially, BDE went 4-2.

BDE was completed a full day in advance to make sure that I would get enough sleep before the event. I wanted to make sure that I would be able to stay calm and make good driving decisions in matches. I made sure that BDE's batteries were charged when I got to the event then waited for my first match.

BDE's first match was against Smee, a lifter robot with a large wedge in the front. I was not sure how BDE would fare against wedges, the whole idea behind BDE was to get a big enough weapon moment to do damage at lower weapon speeds to allow for better bite. I was hoping that I could get bite on the wedge but if my blade isn't sharp enough to dig into the wedge then it would deflect and just send the robot flying. BDE only has one blade and it is getting is dull from all the previous hits dealt to opponents and the arena so I wasn't sure how it would play out.


As anticipated, BDE did not get bite directly off of the wedge so there were just a few glancing blows until I got a hit in on the very back of the wedge, grabbing onto the back corner and breaking the wedge out of Smee. I was pretty excited after this match as it was BDE's first ever victory.

The next match was against a robot made by the same builder as Smee named Faster Owen. This was a unique robot that used a single weapon motor for translational movement by accelerating and decelerating the weapon. After talking with the builder, we realized that we both just wanted to see some high energy weapon on weapon hits. To make sure the weapons would collide, I inverted BDE into an overcutter by putting the tongue on the top of the robot and ran upside-down.


Sadly, we didn't get any direct hits because I couldn't find an angle to get to the blade without hitting a wheel on Faster Owen. I hit a wheel and the bar on Faster Owen was on the ground so it could no longer spin up. He asked for a second hit but that didn't help so Faster Owen was counted out.

BDE's next match was against Squishy. BDE has gone against Squishy at some point in every event so far. This time, Squishy's vertical module was functional and had a 1lb minibot called hockey puck. This match was going to be tough for BDE. My plan for this match was to ignore hockey puck and get as many hits in as possible on Squishy's vertical supports in an attempt to stop the weapon.


This match took a toll on BDE and was the first time the weapon has ever mechanically been stopped. There were a lot of intense hits from both robots here. It looks like I won a good amount of hits at first because the robots shot apart horizontally, rather than BDE being launched vertically if Squishy won the hits. The first hit flipped squishy upside down, which meant that he hit my blade downwards. These downward hits were hard enough to strip the 4 10-32 screws that hold the blade onto the hub off and disconnected the blade from the hub. The hubmotor still spun but the blade was dangling on my robot, held on by the tongue. After my blade was pulled off, the match became very one-sided as hockey puck could tee BDE up for Squishy to get better bite and launch me. The remainder of the match was me trying to keep the blade pointed towards squishy and avoiding electronics damage. I only lost one wheel in the remainder of the match.

After this fight I went over to see the damage I did to squishy. It turns out that I gave the vertical supports enough damage to cause some severe rubbing inside the motor. If Squishy's weapon had stopped in that match, it is quite likely that it would not have started again.

I got high centered on the tongue during this match because some of the hits rotated it back to the least optimal position. The next version of BDE will need a keyway or spline to prevent the tongue from rotating.

BDE was not out of the competition yet though, just pushed into the loser's bracket.

I was able to cross-thread some M5 screws into the holes holding the blade on to keep the weapon together for the remainder of the competition. I made a spare hub because this was a likely failure mode but I forgot it at WPI.

The next match was against Killer Taco, a beater bar robot made out of 3d printed rubber. I was relatively confident in this match but the weapon assembly did take some serious damage and I was worried that another weaker vertical hit could break the blade off again because I only had 4 cross-threaded M5s holding the blade on.


BDE's blade was no longer balanced after the Squishy match so getting up to speed was a little bit difficult from the vibrations but after getting to speed I was able to deal enough damage to stop Killer Taco's weapon and remove a side plate. The tongue was a little bit of a hindrance in this match because it was loosely spinning (I must have forgotten to tighten it or not tightened it enough). I nearly got high centered on a piece of debris from the arena but was saved by Brenda the brick (Brett was destroyed by So Ratchet). Later in the match, the weapon was stopped because the loctite holding my stator in place broke free and let the stator fall and rub on the weapon hub. I bought some green loctite that will work much better than the red loctite that was holding the stator on here.

I pulled the weapon assembly apart after this match and re-glued the stator. After a quick test and tightening the center bolt better so the tongue won't rotate, BDE was ready for the next fight.

The next fight for BDE was against So Ratchet, a horizontal midcutter that uses a box wrench as its blade. This robot has more reach than BDE and a higher blade so to get close I'll either have to tank a hit or get around the side of So Ratchet.


I was relatively confident in BDE's front armor so I went head first into So Ratchet. This approach didn't work too well as BDE just got hit to the side and chunks were removed from the frame. I got lucky with a side hit on one of the wheels then got a hit in hard enough to stop So Ratchet's weapon. With no weapon and one wheel, So Ratchet was unable to move. I gave So Ratchet a few more hits to see if the weapon would spin up then called it and So Ratchet got counted out.

BDE didn't take any damage that required repairs, but the frame did get a little torn up from that wrench. I just had to charge the batteries for the next match.

BDE's final match was against Wasp, another robot from WPI. Wasp is simple yet reliable as the weapon and drive rarely stop on this bot. My plan for this match was to bite into the titanium weapon supports and try to bend them inwards to stop Wasp's weapon. To do this, I needed to keep the weapon speed low and get solid bite.


I got three hits in on Wasp before my weapon was hit vertically, stripping the screws out of the hub again and letting the weapon fall off. I was kind of expecting this to happen because the weapon screws were still M5s cross-threaded into the stripped out 10-32 holes. At this point, BDE's weapon was retired for the event because I did not have the spare hub and there was no space on the hub to drill new holes. It looks like Wasp lost connection somehow (There was 2.4ghz interference at the event and Wasp was using a Hobbyking radio) and since Wasp was a robot from WPI, I tapped out strategically because BDE could not be repaired but Wasp could, giving our team a better chance of winning the event.

I wanted to do some drivetrain testing so I also put BDE, without a weapon, into the rumble at the end of the event. I just charged the lipos again then threw it in (on top of Roll Cake, to see it get launched)


Sawsloe came at BDE at the beginning of the match, sawing on the frame and doing some damage to one of the wheels. I sat in one spot going in circles until another WPI bot, Meili, came by and removed both wheels with one tap each. After that BDE just sat for the rest of the match while Meili and Roll Cake duked it out.

After the rumble I began to consider wheel guards on BDE because of how fast Meili took the wheels out.

Here is BDE post-event



I decided that the frame is going to be retired, as well as all of the components in the weapon module. The inner hub's tapped holes are stripped out and the outer housing was bent downwards after the Squishy fight because Squishy was upside down, giving downward hits.

BDE V4 is planned to compete at Motorama.


Big Disk Energy V3

After thinking on the way back after Norwalk Havoc December 2018, I came to the conclusion that to become more stable, BDE's 'triangle of stability' is going to have to grow I didn't have enough time or want to make a new frame to change the rear wheel contact points, so I opted to make a titanium piece that bolts on under the blade to move the front skid further forward.

I didn't CAD this addition out, I just did some weight calculations to see how much weight needed to be shaved off of BDE for this titanium tongue. This piece weighs 12g which isn't much but BDE is only 2g under weight, with a top screw missing. I had to shave weight somewhere on the robot.

I decided that the piece to take weight out of was the outer weapon housing. There was space for 6 holes and I decided to face a little bit of material off the top to get a little more weight out.


These changes freed up 14g, just enough to add this tongue and put the last top plate bolt back on.



I also changed from 2 3s 450mah tattau lipos to 2 3s 500mah bolt lipos so I can optionally charge the pack an extra volt higher and have a higher current capacity (while spinning BDE up you can see the blade RPM dip then come back up from the voltage drop across the pack) This had no impact on the weight since these packs weigh the same.

Here is BDE ready to fight again


Big Disk Energy Post-event Report: Norwalk Havoc December 2018

Big Disk energy went 0-2 for a second event, but this time it was due to me making poor decisions while driving.

I only got 3 hours of sleep before this event because I was working on a 1lb scaled copy of BDE called LDE (little disk energy). With the Norwalk Havoc rule set, multibots are allowed a 1lb weight bonus so with a 3lb robot you can add a 1lb minibot with no weight penalty. LDE was very last-minute and I just used a 28-26 stator I had on hand for its weapon, which was far too fast at 1000kv. I called it at 5AM when it detonated a second weapon ESC. Here is an in-progress picture of LDE


RIP LDE

In BDE's first match, I had to go against Nitro Hornet, a wide drum spinner robot. I didn't come into this match with much of a plan other than spin up and hit the side plates of the other bot. About a minute into the fight, my ESC browned out and wanted a low throttle to restart. My trims were too high so low stick on my controller wasn't enough but since I was panicking I didn't realize this. Shortly after Nitro Hornet's weapon also stopped and it became a pushing match. BDE is not made to push robots and I was constantly over-steering so Nitro Hornet won this match with control and aggression.


Immediately after this match I put BDE into the test box to test the weapon. It came back to life as soon as I adjusted the trims.

In BDE's second match, I had to face Squishy. I was worried about this match because I was not sure how the weapon would fare against his vertical module with a wedge. His vertical module was having problems, however and he ended up running his undercutter module. BDE's blade is at the prefect height to go over Squishy's blade and hit his weapon belt so I was confident I could stop his undercutter module.

In this match I had a weapon ESC failure because I pushed it too hard. I think one of the phases either disconnected or completely died on the controller which caused the weapon to jitter. Instead of cutting my weapon throttle and playing it safe after stopping Squishy's weapon, I gave it full throttle and tried to get my weapon to spin up. This caused the ESC to detonate in my robot.


This was just a bad call, I could have turned my weapon off and likely won the match on damage. The only thing that was damaged in this match was my weapon ESC, which is now dust. I had another spare ZTW 40A ESC that I soldered in and BDE was once again ready to fight, but I was out of the bracket.

Between the semifinal and final matches, grudge matches were being allowed to pass the time as the finalists worked on their robots. I brought BDE up but was the only robot ready to fight. After waiting a few minutes to see if any other robots are ready, I asked to fight Brett the Brick, the house robot. Brett's builder accepted my challenge so I fought the house robot.


When BDE's left drive motor stopped, I was worried that I had killed the Botkits motor but actually one of the wires just broke off of my drive ESC. I stopped BDE when the lipos popped out of my top plate. Since BDE was so close on weight, I had to remove one of the top plate screws to make weight. This made it possible for the battery to fall out if BDE accelerated fast enough.

After watching this match I realized that BDE needs to become more stable. Because the weapon isn't perfectly level, quickly turning is enough to gyro the blade into the floor and launch the robot. I also need to drive at a lower weapon speed to minimize gyro-ing and stay calm during matches.

Big Disk Energy V2

After having BDE completely together and fighting for 2 matches and a rumble, I was able to conclude that the poor CG placement of the robot made it very hard to maneuver in combat and that my wheels had to change because the lite flites and fingertech hubs fell off every match. BDE V2 is going to address these issues with a new frame and a change from lite flite/fingertech wheels to the botkits D2 foam wheels.

I knew that I wanted the next iteration of BDE to be short but wide so I could push the electronics behind the wheels to move my CG back while also being able to fit blade's full diameter under the robot. Since most of my CAD was based off of a single sketch, changing the frame shape was relatively simple. 

Through a little bit of trial and error changing different parameters, I was able to get the CG of the robot about halfway between the front skid and the rear wheels. This means that about half of the robot's weight is on the wheels. I checked the V1 CAD and in that the CG was far closer to the front skid, leaving only 1/4 of the robot's weight on the rear wheels.


The second part of this iteration is switching to the improved botkits D2 wheels. I ordered them a week after Franklin and they arrived shortly after. These wheels use a clamping hub to attach to the motor shaft and hold the wheels on with two bolts, so they should be much more reliable. The reason I did not use these at first is because this wheel system is much more expensive than lite flites and only come in a 2.75" diameter, while BDE V1 used 2" wheels. 

Because of this, I needed a longer front skid bolt head to keep the blade from hitting the ground. I decided to machine a bolt myself this time so I can polish it to get a lower coefficient of friction on the front skid, further improving the robot's drive-ability.


After re-machining the frame and switching wheels, BDE was once again ready to fight.


Thursday, January 3, 2019

Big Disk Energy Post-event report: Franklin Institute 2018

Big Disk Energy went 0-2 at Franklin Institute, mostly due to CG and electronics issues. Although I intended for this robot to deal massive hits, I did not fully anticipate the forces that the electronics would see because I thought the softer UHMW frame would dampen any hits the robot gives or takes. That was not the case, the frame was far too rigid to preform this function. I did not put any other mechanism in the robot to dampen the shock the electronics see so in the first fight, after a minute of bouncing around the arena, the pads on the output terminals of the weapon ESC tore off from the inertia of the ESC.

In BDE's inaugural match, it had to go against Silent Spring, one of the most successful robots in the northeast. After a few blade to blade hits, Silent Spring was able to remove both of the snap wheels from BDE after I made a bad move and exposed the wheels. The snap hubs immediately fell off, leaving me to question their reliability. I kept the blade spinning and due to the gyroscopic forces from the weapon and the torque of the hub motor, the blade kept spinning and hitting the arena floor and walls, shooting BDE around the arena. After about a minute of flying around, the pads tore off of my weapon ESC, disabling it. BDE was then counted out.


In BDE's second match, it had to fight Squishy, a robot built by another member of the WPI Robotics Club. After breaking the weapon ESC in the first match, I realized that I forgot to bring the spare motor controller and frantically looked for a replacement. This proved to be difficult due to the size constraints from the frame and because it had to be able to work on 6s. I could not find a replacement ESC but was then shown that the top layer of fiberglass could be scratched away on the broken ESC, revealing an internal copper trace that could be soldered to. I successfully repaired the weapon ESC just before the match started and quickly plugged everything back together. I, however, accidentally plugged the weapon PWM connector in the receiver backwards. Squishy also had some electronics issues with the weapon but was able to get them fixed before the match.

Later in this match the snap hubs failed a second time, and a wheel fell off without being hit, then Squishy tore the other one off. BDE was counted out and Squishy won the match.


BDE was able to be fully repaired from the second match by flipping the weapon PWM connector and re-tightening the snap hub setscrews. At the end of the event, there was a 3lb rumble that BDE participated in. At the beginning of the match, I spun the weapon up but was quickly forced into the wall by a wedge. BDE bounced off of two walls, put a gouge in the arena wall, then on the final hit both motors shot out of the robot. That mesh is made out of 1/8" thick steel! 


Overall, I was satisfied with BDE's initial performance. The weapon module and blade design worked exactly as I wanted. I added holes in the side of the robot in the initial design for a plate to prevent the motors from coming out but after testing with the motors glued in, I thought that the glue would hold up. the next version will have the plates to prevent the motors from coming out. The snap hubs failed in both matches so the wheels will be changed to the Botkits D2 wheels. The Botkits wheels are about 3x heavier but are far more reliable because the wheel is bolted on to the hub and the hub uses a clamp mount to the motor shaft, grabbing along a wider area with more force. The ESC will stay the same but will have to be mounted better and the leads will have to be battle hardened for shock resistance.

The main problem with BDE at this event, however, was the poor CG placement. Due to the heavy weapon module and blade, the CG of this robot is very close to the front, meaning that there is more weight on the bolt dragging on the ground than the wheels. This makes the robot turn slowly and get very little traction in reverse, causing a slow reverse acceleration. The focus of the design for the next version of BDE is going to be making it drive better by creating a new frame with better motor and electronics placement for a more favorable CG, putting more weight on the wheels.