Motorama 2015 Part II: The Pitfalls of Used Überclocker Ownership and Unmodeled Dynamics of Candy Paint & Gold Teeth

Monday, March 30th, 2015 @ 20:17 | Candy Paint & Gold Teeth, Events, Überclocker ADVANCE

Let the masses rejoice, for I’ve decided to go ahead and post my Motorama 2015 trip summary, having been satisfied by the newbies, finally *ahem*. I mean, that and the fact that I have backed up way too many other things to talk about waiting on you guys to finish. One of the consequences of writing up an event report a month and a half post facto is that everything I remembered from the event and wanted to talk about really really badly is now… wait, what was I talking about again? See what I’m talking about?!

Additionally, for the past month or three, I’ve been working some black projects which will be excellent once revealed, but I feel like need a little more time to develop before I drop the bass. Rest assured it will not be disappointing.

Anyways, onto the event!

thursday

In the last day or two of the run-up to Motorama, I focused on putting Überclocker Advance back together. I elected to keep it mostly apart until the end to encourage me to keep finding bugs on RageBridge 2. The intention was to assemble 2 or 3 motor controllers and run them at Motorama, but by Thursday I’d mostly declared this to be infeasible given the time constraints and me helping with a certain other build that was going on at the time.

At the least, I wanted to supplement Clocker’s top armor, since for the two weeks prior, I had been watching the development of a new rather formidable looking 30SC class hammerbot, Iron Golem. This, coupled with Mike Jeffries’ Nyx 2 which sports a hammer module, and the prevalence of other 30SC class hammerbots last year, made me think that the 1/8″ Garolite (FR-4 circuit board material with a different name, basically) was getting uncomfortably light.

Clocker was already very close to the weight limit, so the most armor I could afford weightwise was only around 6 ounces.

The top armor was made of some 1/16″ Grade 5 titanium using the outline of the existing top armor, but with the “clock” logo removed. I think with the more flexible, layered Garolite on top and the springy titanium on the bottom, the bot will be fairly impregnable from the top. The bottom, of course, is still made of the juicy 1/8″ Garolite.

On Thursday, here I am continuing to develop Ragebridge 2 firmware. As mentioned previously, I was working on getting the “Combine” mode working – where both channels are synchronized in hardware to drive one motor. This was pretty easily. All it entailed was detecting the presence of the “Combine” mode jumper, and it if was present, just writing over the final output PWM duty cycle of channel B with that of channel A. The measured current of channel A is assumed to be half of the total current.

This isn’t the most robust implementation – I probably still want this routine to sample both currents but still adjust both channels identically in the future. But this was tested to be working great for the time being.

That was important because…

…I was helping get this wildcard moonshot design out the door. This is the Adam & Dane combo-baby robot, Fission Product.

Originally conceived as a shufflebot ring spinner, it evolved into a shuffling hydraulic crusher – really, the first ever hydraulic 30lb class bot that I’m aware of, even if it actually weighs 45 pounds. What’s a “shufflebot”, you ask? In short, the rules of some robot combat leagues in the U.S. and abroad allow a 50% weight bonus for designs which do not rely on wheels, but also aren’t “walking” designs in the sense that actuators/motors have to stop and reverse to take a step. They sort of “shuffle” around the arena, usually by legs that are driven off continuously-rotating cams. Topologically speaking, many “walking” kit robots and hobby robots would be categorized as shufflers under this rule.

Probably the most legendary “shufflebot” was Son of Whyachi, which was originally designed to take advantage of a loophole that existed in the Battlebots rules regarding this type of drivetrain. This is actually the cleanest video I can find of Son of Whyachi moving around, at an event that occurred after Battlebots ended its TV run, but its most well known match is probably the Battlebots Season 3 final, even though it spent most of the time twirling.

In various manifestations, the “shuffle” cam-driven leg sort of became a category of drivetrain in its own right in the few years after. Today, it’s more done for its own sake – not because it confers innate advantages, but usually because of the weight bonus or aesthetic appeal. Interestingly enough, Fission Product isn’t even the only 30lb shufflebot. Clocker has actually faced one before at Motorama 2013, and would do so again this time.

So what does this have to do with Ragebridge? Well, the combine mode was needed to run the “Short Mag” motor that used a closed-loop hydraulic circuit as a gear reduction. This hydraulic system had no reservoir or valves – just a single hydraulic motor backdriven as a pump, to run both ways. For reference, the same motor is usually used on drivetrains of 120lb-220lb combat bots.

friday

I spent most of Friday just tidying up loose ends and packing. Here’s the final “fleet shot”:

I also spent some time assembling DeWuts. 8 of them total: 2 for prizes, 3 to fulfill a sale (“Want to just bring it to Motorama?”), and the rest for us as spares.

It’s time to hop on eBay yet again and clean everyone out of DeWalt transmissions! I’m also reaching the last 15 or so DeWut part kits in existence. It’s almost time for another production run… can you believe it? I’ve sold tens! Literally tens! That’s more than one.

 

With Friday night settling, here we have the van tetris arrangement of this year. Compare with Dragon Con… I think I’m going to start an album made just of van tetris photos.

By this point, I’ve well-figured out that Clocker fits best on top of a pile of standard 18″ aluminum suitcases secured to the rear left passenger headrest using a spare accessory V-belt. Everything else sort of grows off of that. Shown also is my new rolling toolbox which consolidates the van service tools and robot service tools, and adds some other goodies as well like my acquired-in-Shenzhen DSO Quad.

The MIT group was so large this year with several new builders it actually warranted two cars leaving on Friday night (and another group of new builders actually left the day before to attend the antweight tournament). I led the Top Heavy Express with Dane driving the rest of the group.

saturday

Snow was briefly in the forecast for the overnight trip during the middle of the week, which caused me concern, but as the weekend drew close, the forecast was pushing the snow later and later until it was starting (in Harrisburg) late in the day Saturday. So the trip was made in the clear to very little fanfare. I’ve gotten to the point where New York City (4 hours) is mentally no longer a road trip, so Harrisburg warrants a Mountain Dew or something before starting.

Despite my repeated attempts to violate the speed limit with 6 people and 8 +/- 1 robots onboard (hah), Dane somehow was less than half an hour behind me by the end, despite leaving a full hour and a half later. Luckily, the entire event was running 2 hours late, so it all worked out! I think…

 

The pre-event driver’s meeting to brief everyone on what to do if your robot lights the event organizer on fire, etc.

Before we start, a clean shot of the finished Fission Product. The racket it makes on the arena floor when moving around always attracted attention.

And now for the new builders, just as promised now that you guys finished your websites.

This rotund beetleweight named Midnight Samosas for reasons I can no longer recall is the work of Ciaran, and features a rather unique chopped up Colson wheel for armor which apparently worked extremely well. I was confused when he asked me for a Colson wheel due to my association of them with colsonbots. The wedge is servo-actuated and most of the frame is hand-formed aluminum.

It’s refreshing to see new builds like this shaped like a thing, because inevitably the rest of us (myself included) settle into what’s easy and what works, which makes everything huge and square.

Next is LUC-I built by Rebecca, which is another beetleweight pusher-wedge design with a servo arm. The formed spring steel front loop is actuated, and the frame is a one-piece 3D print (See? I’m passing my bad habits to all the new builders!). The surround armor is UHMW polyethylene, made with a remnant spare frame rail from the retired Null Hypothesis.

Both of these new builds used my “Uncle Chuck’s Somewhat Free Robot Motors” that I picked up in Shenzhen! Somewhat-free because in the end I had to ship the whole lot back home using express airmail since Seg-Thing took up a majority of my luggage weight, and the price of the shipping made them no longer free.

 

Clocker and Candy Paint situated on the pit table before clearing safety inspections.

Before leaving, I whipped up this combination weapon cage & mounting fixture. It clips to the front of the bot and over the blade, preventing it from rotating. And if I turn it upside down, it’s a work stand, since Candy Paint is built “upside down” and has a bottom lid instead of top.

Walking around a bit, here’s the new version of my classic nemesis, Nyx, now revamped into a version with attachable modular weapons. This one is the hammer module.

Seen here are the lifter and crusher modules. As usual, Mike has documented this build in detail in an Instructable. Gee, maybe I should start doing that…

Another bot I was looking forward to facing – Magneato, the successor to Jack Reacher, which Clocker battled in 2013 also. The flywheel mechanism is more powerful and reliable , and the two towers in the back helped it self-right.

Here’s Iron Golem in person. The speed with which it was built was quite impressive, but Jim might have overlooked a slight design flaw… sadly, the hammer bar broke off on the very first hit in his very first match. It lasted through all the testing too…

Oops.

Next up, we have our star attraction…

harborfreightbot

Your eyes aren’t deceiving you.

First blurrily hinted at in the background of this picture from the 5 inch hub motor dissection, HarborFreightBot is exactly what it sounds like: an attempt to build a robot only using the highest grade Chinesium oozing from the aisles of Harbor Freight. This thing deserves a build post all on its own… kinda…

The ingredients list ended up being:

  • 69651 18v cordless drills x2
  • 90991 4 inch rubber caster
  • 67255 14″ electric chainsaw – yes, it was only running on 18v, so it didn’t really do much besides sound somewhat menacing.
  • 69459 2-piece cutting board set
  • 67068 steel dustpan
  • 61451 ammo box, plus some miscellaneous zip ties and tape from the store.

Non-Harbor Freight parts include a RageBridge for the drive (along with a single Botbitz 85A for the saw) and a borrowed Hobbyking radio set, since Harbor Freight does not yet sell any wireless device I would trust attaching to something with a fucking chainsaw on it.

The construction of HarborFreightBot was mostly performed by our Chilean operative Joaquin, who once again came to MIT for his summer vacation in January and like last year, came to Motorama to help out.  It was finished for the new assbotsRatbots” class at Motorama, the existence of which is the pinnacle of inside jokes gone horribly, hilariously wrong. Yes, we’ve managed to formally codify the building of trashy troll entries. I’ve been planning HarborFreightBot for a while, and have even sort of gotten the local HF branch “in on it”, so I might attempt a more serious entry in the future. This one was just to get some of the newbies’ feet wet in building stuff.

So that’s all for introductions and prep. Now, how did I do?

Candy Paint & Gold Teeth

Candy Paint officially had just one match this time, against megaRon even. Disappointing? Not at all.

With no place to test it to full power safely, especially not hitting something, I just sort of entrusted all the parts and materials to hold up. In fact, I now think everything worked a little too well – so well that I pushed the spin motor beyond the sane limits I had designed it for and cooked it. What did this match look like? Have a look…

Basically, I found out that Candy Paint can in fact self-right after being flipped. Dramatically.

Hey, that was the design intent! I think the switch to the higher-torque motor winding helped in making it dance, but ultimately, me going full throttle to bounce around so many times overheated the motor and fused the windings into a molten cake. With the weapon motor completely toast, there was no point in continuing, so I pulled Candy Paint out of the tournament.

As much as I love the custom motor, I think I’m going to try and design in a stock (or very lightly modified) pancake motor like a giant quadrotor motor or something, such that spares can be easily had. This would be in the queue for the as of yet unplanned future version of the bot alongside other design changes like not using welded aluminum (interesting adventure, but no more please…), and instead using a thin skin pan made of welded 4130 chromoly steel or similar.

Überclocker

Clocker took away 3rd place this time, losing once during the winner’s bracket and getting all the way back to the finals in the loser’s bracket. I’m amazed it even lasted that long, due to the following issues:

  • Despite taking things apart to look at them, I didn’t go as far as to replace some highly stressed parts. For instance, I know the front leg anchoring screws take a lot of shit, and I should have replaced them after every event. But I did not…

 

…and in Clocker’s first match, a tap into the wall that shouldn’t have done anything instead broke the 1/4″ socket cap screw clean off under the head. Oops.

Fortunately, this matchup was against the guys, it IS actually an assbot non-assbot VLC. I came so close to losing against a discarded traffic cone…

I started the match basically running a figure 8 track around the cones, then decided to go on the offensive and see if I could beat one with the other by picking it up. Bad mistake with that initial wall rush…

While this repair was easy, more serious issues started cropping up, namely the wheel hubs began stripping out and disintegrating with each match. Basically…

  • The nylon hex rod material I substituted in place of Delrin, because it was more easily obtained, wasn’t strong enough to withstand the torque of the robot drivetrain in a walled arena.

At Dragon*Con 2014, I at least had to drive carefully to avoid running off the stage. Well, with the arena walls, I could freely reflect off them, some times even carrying an opponent straight into them. The torques on the wheel hubs are way, way higher as a result. I first noticed one of the wheels freely spinning after my 3rd match:

Oh dear, that isn’t good at all. The hex is shaved completely off.

The temporary solution? Take some #4-40 long screws and drill them as deep into the wheel assembly as possible, taking advantage of the fact that the E-clips don’t completely encircle the hex bore area.

I did this to not only the rear right wheel which had already failed, but to a lesser degree on another wheel which was showing signs of the hex material deforming.

It only started go downhill from there. After each match, another wheel disintegrated, and the E-clip flanges even started cracking apart. I started doing extremely hackish things, like drilling this screw down at a 45 degree angle…

After an intense match with Nyx, the “shock absorber” of the front left leg finally gave out. In this match, Nyx, using its hammer module, also managed to sever the left side drive chain.

This chance alignment of the bots was probably even better than the one we were hyped about, which we nicknamed the “money shot”:

By alleged (ahem) design coincidence, Nyx’s hammer reaches exactly halfway across Clocker’s frame on a broadside, hitting the DeWut gearbox running the lifter almost square in the middle!

I lost the match to Nyx after the chain and leg broke loose, but before that, it was one of the most intense blows-trading matches we’ve been involved in to-date… and Nyx & Clocker always make for entertaining matches.

The wheel cracking was getting kind of desperate – I had to drive even longer screws, spaced by washers, into this wheel to keep it together. Next, I had to cobble the front leg back together somehow, else Clocker would not be able to lift successfully. Dane put together a “peg leg” using, of all materials, the leftover HDPE cutting board from HarborFreightBot which we brought “just in case”.

Clocker’s last match in the losers’ bracket finals (basically, 3rd place fight) was using the peg leg, which worked – kind of. We should have put another HDPE layer and bound them together, because the thin 3/8″ material just bent out of the way when faced with lifting another 30lb bot. Halfway through, one of the hacked wheel hubs tore apart and shifted sideways, derailing the chain and basically destroying the hex hub completely.

With the 3rd place, the last hurrah of this tournament for Clocker was in the 30SC rumble. For the rumble, I just removed the rear right wheel and cut the chain….

Three-wheeled and peg-legged! The 30SC rumble was a bit sparse since it was so late in the day on Sunday, but nevertheless, I teamed up with Pitter Patter to pull off basically the first robotic execution…

Clocker is gonna need a lot of love to get running again for Dragon Con. I’m probably going to rethink the decision to move to single-supported legs. They’ve held up okay considering how much wall-bouncing happens, but I don’t like them as much because of the lesser support width. I think even adding a ‘finger’ from the frame to double-support them locally could help. Else, when it comes to wheel hubs, I should probably make them from aluminum with a plastic center bushing for weight purposes.

Dragon*Con 2015 might be the last event for Überclocker Advance, at least this version of it. Why? I’m now hankering for a shuffler too. An extra 15 pounds would be wonderful for Clocker to pick up any opponent with. Based on all the chatter between builders, 2016 might be the year of the shuffler. One problem to overcome with shufflers is that they tend to move slowly, even if the legs are actuating quickly due to the “bounce” between adjacent leg hits. I have a few ways to get around that which I want to explore.

Here’s all of Clocker’s matches again, in order.

Nyx & Clocker could probably be a robot martial arts movie:

And now, some other photos from the event!

HarborFreightBot, post-match after facing an assbot made of an angle grinder with a lawn mower blade on it. Actually quite formidable, but we were all expecting the cast aluminum grinder gearbox to detonate on a hit and send the lawnmower blade flying – which I am sure is the opposite of what someone with a sense of survival would want. The leftover chain oil in the saw also greased the floor up.  Before this, HFB had a much more hilarious match against the very manifestation of assbots – Spring Cleaning, cobbled together from garage trash.

Fission Product being readied for another match – top and front armor was being added here to prevent wandering robot fingers. Watching it move around, when it could, was entertaining for all. However, because the cam-drive legs were relying on the Delrin leg material alone – without bearings, the legs sometimes jammed, leaving the bot one-sided. It was a good concept which will be making a return later. The hydraulically driven claw proved a bit too slow to catch people, one of the historic downsides of crusher type designs. If you can’t trap someone, you can’t really get at them.

Jamison preparing Attrition for its finals match. Attrition cleaned house, going 4 wins and no losses.

LUC-I takes some damage after a match. It ran as a wedge for the most part due to the “one plastic gear” metal gear servo stripping out. In the rumble, it finally blew up. Hopefully Rebecca will bring it back better than ever!

Clocker doing a Clocker thing against Power of Metal in their first match.

…and I’m not sure exactly what’s going on here, but it probably has to do with robots fighting.

I was content with Motorama 2015, despite not “doing well” – though most would probably argue 3rd place is well, which I would agree with too – basically, I’m not too interested in winning so much as being entertained and being a resource to the newbies especially. Clocker is showing its age a little (this is its fifth event), so it’s time to think about that refresh!

 

Recently

  • Überclocker 5.0: In Which I Actually Have to Build the Bot, Not Just Talk About It
  • Überclocker 5.0: The Big Post of Designy-Stuff
  • The Overhaul of the Future Begins Now: Überclocker 5.0 (Also, Welcome Back to Robots)
  • Operation RESTORING BROWN Part 7: The Epilogue; or, Dragon Con 2019
  • Operation RESTORING BROWN Episode VI: Return of the Van Lights; the Conclusion
  • Late Stage #PostmodernRobotics: Welcome to Your Waifu is Trash, the Robot Dumpster Fire
  • Operation RESTORING BROWN 5: The Road to Reassembly
  • Operation RESTORING BROWN: The Paintening
  • Operation RESTORING BROWN Part III: An Electric Interlude
  • Operation RESTORING BROWN: Romance of the Rusty Roof Rail Repair
  •  

    5 Responses to “Motorama 2015 Part II: The Pitfalls of Used Überclocker Ownership and Unmodeled Dynamics of Candy Paint & Gold Teeth”

    1. SOI Sentinel Says:

      What I think their Fission Products bot needs is a hydraulic intensifier, such as a Minibooster brand one. They’ll bypass at fairly low pressure (quick grab) then switch to high pressure once there’s some back pressure. 10000PSI of hydraulic power is well within reach.

    2. the chuxxor Says:

      Even the smallest Minibooster looks really really heavy…

      They plan to bring it back with an accumulator to take advantage of the full cylinder travel.

    3. Tamber Says:

      I’m sure you could bash one together designed for lightness; I mean, surely life gets a little easier when it only has to survive a few events, rather than operate flawlessly for a few megacycles, right? ;)

    4. SOI Sentinel Says:

      I’m not sure I’d want to bash together something that operates at 10kpsi unless you try to waterjet your opponent. Smallest one is about a pound and a half.

    5. senso Says:

      Waterjet as a weapon would be sweeeeeet.