Battlebots 2020 Episode 2 review Ghost Raptor vs Shatter

It’s great to see Chuck Pitzer back with Ghost Raptor. The unusual duel weapon design of Ghost Raptor combined with Chuck’s highly aggressive driving always made for a great fight in the first couple of seasons of Battlebots. However this year’s Ghost Raptor does look like a slightly rushed design with not a whole lot of armour. The drive pods way out on each side of the robot look particularly exposed and the first pictures I saw of Ghost Raptor had me worried. But Chuck is an experienced builder so maybe there will be more to Ghost Raptor than I first thought.

In contrast to the lack of armour on Ghost Raptor, Shatter is absolutely covered in soft, shock absorbing UHMW polyethylene armour. I think Adam Wrigley has made a mistake in describing the polyethylene as “ablative armour” implying that the armour is meant to be damaged and break off the robot to absorb energy from attacks. I am sure Adam was using this term to try and persuade the judges that any cosmetic damage to the armour is intentional and should not be counted against Shatter because the polyethylene will get torn up by anything sharper than a spoon. However I think that by repeatedly using the term ablative to describe the armour caused the judges to be hyper aware that Shatters armour will be damaged and the judges attention will be drawn to any marks made to the armour. In fact, before the fight even started, during the in-arena function test Shatter lost several reflective stickers from the polyethylene due to the stickers not being able to bond to the polyethylene. Chris Rose then stated that Shatters ablative armour was coming off as intended. I am sure if Adam didn’t use the term ablative to describe the armour then Chris would have just said that the stickers were falling off, which sounds a lot less serious. But Shatter has both an axe weapon and mecanum wheel based omni-directional drive so it will always be one of my favorites.

The fight starts with Ghost Raptor spinning up the bar spinner and Shatter taking an axe swing right into Ghost Raptors weapon after waiting for the weapon to get up to speed. This seems to be the end of the bar spinner as Ghost Raptor doesn’t spin the weapon again but instead concentrates on using its lifter to try and turn Shatter over. Shatter on the other hand continues to loose all of its stickers plus the 3D printed top covers. Again this is only cosmetic parts falling off Shatter but again Chris announces that Shatters ablative armour is being damaged.

After Ghost Raptor pins Shatter against a wall but is unable to flip Shatter over Shatter is able to turn and land a couple of hits on Ghost Raptor. At this point it seems that Shatter has lost one of its wheels as it seems to only be able to move diagonally. But it doesn’t matter as Ghost Raptor overheats and begins to burn it’s drive motors. After both robots are stuck against the screws Shatter wriggles free off-camera (hmmmm. And where did all the smoke go? I think the fight was paused here and Shatter dragged off the screws) Ghost Raptor is counted out. Winner: Shatter.

So, despite Shatter not being hit by Ghost Raptors bar spinner it was stated that Shatter took damage due to the cosmetics falling off. Maybe Adam should stop using the term ablative armour to describe the polyethylene armour. Ghost Raptor on the other hand looked to be right on the edge design wise with all the lightening holes drilled all over the chassis and lack of armour showing that Chuck was battling with the weight limit when building Ghost Raptor. Also Ghost Raptor loosing both drive motors spectacularly during a pretty standard shoving match make me think that the electrical design is pushing the limits of what the components are capable of a little too hard. So in conclusion I think Shatter has some minor teething issues while Ghost Raptor may need a major rebuild.

Battlebots 2020 Episode 1 review Fusion vs Mad Catter

Fusion is the anticipated dual weapon robot built by Reese Ewert of team Whyachi. Fusion has both a horizontal spinner on one end of the robot and a vertical spinner on the other making it look like a robot designed by a 5 year old. It is well built and very compact with the Whyachi robots being known for being quite pricey with no expense spared in either materials, components or manufacturing techniques.

Mad Catter is more of a classic 4wd vertical spinner design but does have a self-righter that doubles as a lifter. This shows a lot of thought has gone into the design as weight has not been wasted on a single purpose self-righter. While Mad Catters creator Martin Mason brings a lot of energy to his on-screen antics Mad Catters driving is more subdued. This might be more to difficulty Mad Catter seems to have turning.

The fight starts with a classic box rush from Mad Catter. However a box rush on Fusion is only going to hinder the front facing weapon unless it can be pushed back into the wall. In this case the vertical weapon was hindered but the horizontal spinner was allowed to get up to speed. Despite the vertical weapon on Fusion being stopped Reese still insisted on driving Fusion forward into Mad Catter rather than turning around. Since Mad Catter was sluggishly turning to face Fusion it would have been the perfect time to use the horizontal weapon. The seconds after the box rush shows the lack of awareness from both drivers. Fusion should have turned to use the faster spinning weapon and Mad Catter should have driven forward to put more distance between itself and Fusion before attempting the sluggish turn to face it’s opponent.

It doesn’t take Reese long to realise he is using the wrong weapon and turns to use the horizontal weapon. This is when we get to learn why having both a vertical and a horizontal spinner on the one robot is a bad idea. Normally when a robot hits with a horizontal spinner it is jerked sideways due to the hit. With the vertical spinner creating its own gyroscopic forces the sideways motion the robot twists around an axis perpendicular to the axis of both weapons and Fusion flips over. Fusion can’t have both weapons spinning at full speed at the same time and hit something with either weapon without the other weapon flipping Fusion over. The large shocks and twists Fusion takes seems to cause the electronics to short circuit. Seeing how little smoke was produced for the amount of fire that came out of Fusion my guess is that it wasn’t a battery fire.

Battlebots 2020 Episode 1 review Skorpios vs Bloodsport

Skorpios looks very similar to last year and is the classic overhead saw robot with a very solid wedge. Bloodsport on the other hand has been completely redesigned from last year. It is still an overheard horizontal bar spinner but no longer has the circular chassis but now looks like a smaller Son of Whyachi. I really liked Bloodsport last year but it had some reliability issues.

The fight starts with the usual box rush from Skorpios but Bloodsport was able to avoid it. Bloodsport is obviously afraid of Skorpios front armour and instead attempts to drive around the side to attack Skorpios’ wheels. Zachary Lytle was able to keep Skorpios pointed at Bloodsport so Justin Marple had to drive Bloodsport into Skorpios wedge. The hit didn’t damage the front armor but instead spun Skorpios around and Bloodsport was now able to attack the right wheel of Skorpios. The only damage made during this exchange was that Bloodsport lost 2 of its wedglets. This repeated a couple of times until Bloodsports weapon slowed and Skorpios was able to shove Bloodsport into a wall. However the overhead saw was damaged and Skorpios was not able to use it.

Zachary looks very confidant in the front armor of Skorpios and it looks well deservered as Bloodsports hits many times but does no damage. That is until Bloodsport caught the edge of the wedge and starts to peel it off. At this point Justin Marple is heard saying that Bloodsports weapon was overheating and needed to cool. This is not something I have heard very often with most robot drivers just repeatedly yelling “Spin Up!!” and damn the smoke, especially if the other robot is showing signs of damage. But Skorpios is not looking good with the right hand wheel clearly not working and smoke coming from the hole that had opened up on that side of the robot. Bloodsport starts the weapon up but does not use it and waits for Skorpios to be counted out.

This was a great fight between two solid robots. It really looked like it could have gone either way for most of the match. It also really shows how aggressive a driver can be when they are confident that their armor can hold up.

Battlebots 2020 Episode 1 Sporkinok vs Rusty

This is the opposite of the last match with two first time robots. Sporkinok is an axe or lifter robot. In this fight Lilith Specht decided to remove the axe and just run as a lifter. Rusty is a pure junk robot that lives up to its name. Rusty is also an axe robot built by David Eaton out of junk he found on his parents farm.

Sporkinok started the fight well despite having some control issues with the robot. This was probably due to the wheels being so far behind the centre of gravity of the robot. Sporkinok was able to flip Rusty onto its side and it took David a while to get Rusty upright. It seems that there is a several second delay from when David fires Rusty’s axe to when Rusty actually responds which is seen later in the march as well. After the first flip Sporkinok is not able to lower it’s lifter completely and also seems to have even more trouble driving. This allows Rusty to haphazardly drive around to attack the rear of Sporkinok and bring it’s axe down on the drive Sporkinok. I am not sure if Rusty’s axe actually did any damage but Sporkinok was unable to turn or reverse after this attack and become stuck against the side of the arena. K.O. to Rusty.

Battlebots 2020 episode 1 review End Game vs Tombstone

The first main event of 2020 is between two classic robots. End Game is a basic vertical spinner but with a beefed up front wedge for this fight and Tombstone is a basic horizontal bar spinner. Both of these robots are the simplest form of their respective class of spinner and robots like these two are in every competition. It is very strange then that any horizontal bar spinner with exposed wheels is called a Tombstone clone but vertical spinners are not called a clone of any other vertical spinner. This is probably because Ray Billings has been competing with Last Rites/Tombstone for such a long time and has always made an impression. In fact I still have clear memories of Last Rites from Robogames 2007 and the noise it made. I just don’t like calling robots Tombstone clones and feel that some new builders feel restricted from building a horizontal bar spinner in case they might be accused of making a clone. I think if you see another robot you like and want to copy the design then go ahead. But that doesn’t have anything to do with this fight.

With a classic match like this the tactics used by the drivers is well known. End Game wants to attack Tombstones wheels or rear while Tombstone wants to get to the side of End Games wedge. Both Ray and Jack are known for being good drivers who are confident in their robots so this was a fight I was looking forward to like the Saw Blaze vs Whiplash fight. But unlike Saw Blaze vs Whiplash both End Game and Tombstone are extremely hard hitting so I was predicting this fight to come down to the first hit and was expecting a few seconds of strategic manoeuvring.

This fight was quick, lasting just 10 seconds. It turns out that the first hit was a front on hit by both robots in the centre of the box. No careful manoeuvring or sizing up. Just spin up the weapon and drive straight at the other guy. Tombstone hit the wedge on End Game and was turned around to face away from End Game. Ray tried to spin Tombstone around to hit End Games side (Ray always spins right to do this?) but End Game continues to drive forward. End Game hits Tombstones right wheel and both robots are thrown into different walls of the arena. While End Game immediately recovers and drives straight at Tombstone again Tombstone shudders forwards and backwards but doesn’t get away from the wall which also prevents the weapon from spinning. Perhaps the hit on the wheel damaged something or Tombstone is caught on the wall. We don’t get a chance to find out as End Game hits Tombstone in the side again and flips Tombstone over the wall and out of the arena.

Before and after the match the End Game guys like to explain how much they studied Rey’s driving to develop the best tactics to beat Tombstone. After watching this fight a few times I guess the strategy they developed was to drive straight at Tombstone and hit him fast and often.

Battlebots 2020 Episode 1 review Smeeeeeeeeeeeeeeee vs Sharkoprion

Smeeeeeeeeeeeeeeee, like Huge, is another example of a successful hobby weight class robot being scaled up to a heavyweight. I think for featherweights and lower as being a hobby robot while heavier robots as being being more professional due to the cost and labour involved in building them. But due to several density differences in the componants designs that work well in the hobby classes may not work so well in the prefensional classes. By density I mean things like the weight of a motor suitable for an antweight as a percentage of the robots total weight might be smaller percentage of the weight in a heavyweight. Likewise a battery that can handle the current draw of the antweight motor is easier and cheaper to buy than a battery for a heavyweight as the heavyweight motor can draw an exponentially larger amount of current. And not only do the electronics need to be larger and heavier for the current draw but also the wire and connectors. 0.5g of wire in an antweight getting jostled by an impact is different than 5kg of wire in a heavyweight. Fasteners like screws and bolts are proportionally larger in an antweight where a large 3mm bolt holding in a motor in an antweight would be at least 30mm in a heavyweight if scaled up directly and be deemed too heavy.

Basically parts of the robot that most people don’t consider behave completely differently when the size of the robot is scaled up to a larger weight class. The forces involved in an antweight are completely different relative to a heavyweight. A ridiculously wide antweight does not require much torque to turn on the spot but a 3m wide 100kg robot takes a lot of effort to turn. This isn’t even taking into account to speed that the wheels will need to turn to get the robot to turn at a reasonable rate. If Smeeeeeeeeeeeeeeee is 3m access then the wheels would need to run at 17km/h for Smeeeeeeeeeeeeeeee to make a complete turn in one second, and 1 second to turn around is not great.

So Smeeeeeeeeeeeeeeee is a comically wide wedge and it works well as a beetleweight. It is basically two smaller robots attached to each end of a long piece of steel that acts as a wedge. While this works well for a 1.3kg robot I am not sure that a 3m wide 100kg robot will be able to turn. The massive increase in the moment of inertia to turn the robot may be too much and Smeeeeeeeeeeeeeeee will just not be able to turn. Sharkoprion on the other hand is well known to be a quick and manoeuvrable robot. I know the idea of Smeeeeeeeeeeeeeeee is that it is so wide that other robots can’t get around it but we’ll see if it can turn fast enough to make that idea worthwhile.

When the fight starts Sharkoprion does not try to out manoeuvre Smeeeeeeeeeeeeeeee despite Smeeeeeeeeeeeeeeee demonstrating that it has trouble turning but instead attacks the centre of Smeeeeeeeeeeeeeeees wedge. Edward Robinson backs Sharkoprion off and then attacks the wedge of Smeeeeeeeeeeeeeeee a couple more times before trying to get around to the rear of Smeeeeeeeeeeeeeeee. However Sharkoprion seems to have its right drive fail after driving over one of Smeeeeeeeeeeeeeeees weapons. The wheel doesn’t look damaged and the drive chain is still in place so I am not sure what happened.

Smeeeeeeeeeeeeeeee continues to try an push Sharkoprion around but does not seem to have any power. Smeeeeeeeeeeeeeeee either struggles to turn or when it does get in a position to push the spring loaded wedge prevents it from doing so. The fight continues for 3 the full 3 minutes with Sharkoprion spining to hit Smeeeeeeeeeeeeeeee with its tail or tries to bite the wedge and Smeeeeeeeeeeeeeeee trying unsuccessfully to wrap around Sharkoprion. The fight ends up going to Smeeeeeeeeeeeeeeee by a judges decision due to the lose of Sharkoprions drive being counted as damaged by Smeeeeeeeeeeeeeeee. Interesting to note how much the damage counted in this match over control.

Battlebots 2020 Episode 1 review Lok Jaw vs Captain Shrederator

This is a battle of the old guys. Donald Hutson and Brian Nave have been around for many years and have really refined their robots. Lock Jaw is a great modular robot that Donald can modify quickly to match any opponent. Donald is also a thoughtful but aggressive driver who can also addapt quickly to a changing situation during a fight. Captain Shrederator on the other hand has been having reliability issues the last few years. The reporting pre-fight explains that Brian is using an older version of Captain Shrederator but the shell design doesn’t look so old to me. Perhaps they are referring only to the mechanical/electrical design of the internals.

Donald has chosen to use the angled anti horizontal spinner bash plate for the rear of Lock Jaw. The design of the plate is the same as one we tried a long time ago on Scarifier and is made to deflect a horizontal spinner upwards. It is a proven design but Donald will need to be careful to not let Captain Shrederator around the sides of the robot as the bask plate does not extend far beyond the wheels of Lock Jaw. This shouldn’t be a problem as I said earlier Donald is a careful driver while Captain Shrederator is hard to orientate and has trouble with trying to manoeuvre around other robots.

Donald doesn’t start the fight with a box rush but instead carefully approaches Captain Shrederator to make sure that Lock Jaw connects with the bash plate. Donald was probably not worried about Captain Shrederator getting its weapon up to speed as the idea of the bash plate was to cause Captain Shrederator to damage itself. The shape of the plate worked as Captain Shrederator has thrown into the air. However the plate of Lock Jaw broke and exposed one of the rear wheels and the second hit damaged a tyre.

Donald must have seen that the damaged plate was not going to work and instead spins up to go for a weapon on weapon hit. The hit launched Captain Shrederator across the arena and into a corner of the box. The corner of an arena is the worst place a horizontal spinner can be when trying to start a weapon. Captain Shrederator as walls on 2 sides and Lock Jaw on another preventing preventing the weapon from starting. One of the changes Brian made to Captain Shrederator was to swap the weapon speed controller for a contactor. This means the weapon is either on or off and the current draw for the weapon motor trying to start must have been huge. This is obvious by all the smoke.

Brian manages to escape the corner and move to the centre of the arena but Captain Shrederators weapon is cooked. Lock Jaw is also having weapon issues but this is now a pushing match which Captain Shrederator can’t win. After a short time Captain Shrederator is caught on the screws and Donald holds back waiting for the ref to begin his count.

Battlebots 2020 Episode 1 review Malice vs Axe Backwards

This was a fairly straight forward and predictable match. Malice is a a basic horizontal spinner that is known for having what is described on the Battlebots website as a horizontal drum spinner. It is just a medium diameter thick disc but whatever. Axe Backwards is a lot more interesting as it is one of the only “full body” drum robots. Full body refers to the design of drum surrounding the whole robot with the motors, electronics and batteries being inside the drum with the wheels running on the drums axle out of the ends of the drum. The name of Axe Backwards refers to 2 tails shaped like axes sticking out the back of the robot that allow the drive motors to have something to push against but give the robot the appearance of a overhead thwack bot.

A great part of the full body drum is the same as a shell spinner like Megabyte or Spinderella. Since the weapon surrounds the robot all the weight of the armor can be put into the weapon. The problem with a full body drum like Axe Backwards is the wheels are so exposed unlike a shell spinner where the wheels are under the shell. Since the wheels are on the same axis as the drum the wheels diameter has to be larger than the drum + teeth to maintain ground clearance. This also means that the wheels also stick out in front of the drum and when fighting a horizontal spinner the wheels will be hit first. In fact Kurt Durjan needs to be very precise with his driving for Axe backwards to hit with its weapon before a wheel contacts the opponent. Unfortunately the worst opponent for Axe Backwards is a horizontal spinner like Malice.

The fight went exactly as I thought it would with Malice taking out one of Axe Backwards wheels in the first hit. Malice also damaged a fuel supply for a flame thrower that was mounted inside one of Axe Backwards axes but this didn’t seem to be critical. What seemed to be more critical is that the power switches for Axe Backwards, which must be mounted where they can be quickly accessed are in the axes where it was hit. But every time that Malice attacked Axe Backwards a wheel was the first thing to be hit. After 1 minute Axe Backwards was counted out.

Battlebots 2020 Episode 1 review Sawblaze vs Whiplash

I great fight to start the season. Both Sawblaze and Whiplash are very well driven and very reliable. Jamison Go has a very aggressive driving style with Sawblaze constantly ramming opponents to push them into walls and pin them. Matthew Vasquez of Whiplash is not as aggressive but has excellent situational awareness and is able to quickly take advantage of any weakness that his opponent shows while accounting for any problems with his own robot. I was looking forward to this match with my prediction being Whiplash to win.

When the robots are revealed a point is made of Whiplashes tall vertical front armour plates. The idea is that these will counter the overhead saw of Sawblaze. I was a little doubtfull of these as the tall armour would only protect from a head-on attack. After both robots attempted to manoeuvre around each other and then made a couple of head-on attacks it became clear that the tall armour also allowed Sawblaze to win the wedge game. Jamison repeatedly drove Whiplash into the walls but seemed to wait for a pin before using the saw. However Matthew realised that while being pushed from behind that Whiplash would be able to use its weapon as an overhead saw the same as Sawblaze and managed 3 or so hits on the top of Sawblaze. It was great to see the rapid adaptation to the driving and attacking style of Whiplash when it became clear that Sawblazes wedges where always going to get under Whiplash.

But Jamison Go was always on the attack and did not give Whiplash more than a couple of seconds to recover and it was always going to be Sawblazes fight. Once Sawblaze was able to get to the sides of Whiplash then is was very quick to cut into the top armour and take out the drive of Whiplash. Aggressive and powerful driving with effective wedging really won this match.

New Z probe

The last thing I wanted to do for my UP Plus 2 rebuild was to change the Z probe. The original Z probe is a micro-switch that is attached to the hot-end before calibrating. The UP Studio software has a calibrate function to allow bed leveling before printing. The probe would be manually attached, the calibration routine run via software and then the probe removed before starting the print. The same procedure can be done through the Marlin firmware I am running and G code commands via Octoprint but it is a pain. I want to be able to just press print and forget it.

My old Prusa Mendel printer was capable of auto bed leveling via an RC servo retracted Z probe. This worked well with the only problems being the twisted frame of the machine causing issues towards the end of its life. I considered mounting the servo Z probe to the Up but wanted to try something new. I had seen non contact sensors before and wanted to give them a try.

After looking around the internet I settled on an LJ18A3-8-Z/BX 8mm Inductive Approach Proximity Sensor. This sensor will trigger when a piece of metal is brought within 8mm of its end. Since the bed of the Up is made of aluminium this is perfect. The sensors 8mm activation is repeatable to within ± 25um which is suitable for printing at 0.1mm layer height when printing on a raft like I do. The sensor is bolted onto the side of the hot-end adapter and can be manually adjusted without much difficulty.

Inductive probe mounted to the side of the hot-end adapter.

The wiring was a little interesting as the sensor runs off 12V but the Arduino take 5V inputs. There is a lot of discussions around the internet about how to interface these sensors with the Arduino with much arguing about voltage dividers vs opto-couplers vs transistor logic. I had a look at the sensors data sheet and saw that the sensor has a NPN Current-Sink output which means that the sensors output is grounded when active and floating otherwise. I checked that the Arduino input is set with a pull-up resistor and the firmware input is active low then connected the sensor output via a diode with the anode to the Arduino to prevent any 12V from getting through and that is it. Enable auto bed leveling in firmware and add the G code to the slicer software and now the Up is print and forget.

The last thing was to add a print job cooling fan and duct to the hot-end. This fan and the hot-end fan are both controlled via the RAMPS. The hot-end fan is on the third power output of the RAMPS and is set to turn on when the hot-end temp is above 50 degrees. The second fan is switched on via a relay from a spare output pin on the Arduino and is controlled via G code. All the wiring is brought up to the extruder via the existing Up wiring harness.