Big Data EVT

“8 oz” you said. This dang bar, man.


Battlebots in the Cloud

Big Data EVT drive

To cele­brate moving to the Bay Area, and by “cele­brate” I mean “forgot that I regis­tered for a robot combat event in a farm show complex in Harris­burg back when I was still living on the East Coast and now have to beast one so I can fly it into afore­men­tioned Penn­syl­van­ian snow pile,” I started living in the Google Work­shops around Janu­ary with a 12.4 oz chunk of hard­ened steel, a fever­ishly drafted new design, and dreams of brush­less synergy.

Big Data bar

The concept for my “new1” beetleweight (3 lb) Big Data is simple: a “high” volt­age2 sensored brush­less drive­train coupled to the beater bar from Cake ReMix. With a verti­cally-hitting weapon like the bar, the robot has to hit the wall in order to right itself to the correct orien­ta­tion.

This means that the robot needs be able to drive upside-down. The drive­train was designed around ridicu­lous sensored inrun­ners.

Maxon 200118

These Maxon EC 22 200118 motors are tuned for mo’ torque. They’re filled inside with globs of hundreds of turns of magnet wire thin­ner than hair to create a 400 RPM/V motor that’s only 22 mm in diam­e­ter. To give you an idea of how “cool” these are wound, they have 11.7 Ohms of phase-to-phase resis­tance.

The first order of busi­ness is to take the gear­head off, since it was so well jammed on the motor that I couldn’t get it off even with a heat gun and much vice grip­ping.

Maxon meets hacksaw

The connec­tors are huge!

Bare motors

These were well used bargains from eBay. On one motor, a hall sensor which detects the rotor posi­tion was stuck at zero volts all the time. The motor had some visi­ble rust on it, so I pulled it apart to have a look at the damage.

Oh shi-


Pulling off the ass of the motor had the exact same tugging and twang­ing as pluck­ing three pairs of hairs out of your head. The hall sensor board also termi­nates the six leads out of the wind­ings3, and I had pulled it clean off.

Sensor water damage

Also that round magnet, shaft, and a sensor all defi­nitely have rust on them. Time for some shady repairs.

Maxon winding "repair"

With tweez­ers I tied more magnet wire onto the remain­ing stubs and soldered them in place. Pencil eraser for scale.

As for the dead sensor… well, let’s see if you can figure out what I did.

Sensor repair 2

Sensor repair 1

I may or may not have tried to sand down the new sensor to make it thin­ner and had to repeat this whole repair because elec­tron­ics don’t work like that. >.>




Also I took a trip to Squaw Valley woooo North­ern Cali­for­nia.


Wire it up and it runs. Ship it!

Oh and what are they connected to, you ask?


These are Corn­flakes, version 3.0 of the beloved Corn­troller series. Same good ol’ ingre­di­ents, new pack­ag­ing.

The motor bell is a new design. The face of the bell has magnets pressed in that grip the steel weapon bar and trans­mit torque through fric­tion. This allows slip when the bar is hit, reduc­ing damage to the motor in the worst case, or prevent­ing sensor­less driver desync in the best case.

Green sauce

The motor can is liter­ally glued in with “green sauce,” or Loctite press fit retain­ing compound. It’s anaer­o­bic methyl acry­late adhe­sive that acti­vates in the absence of air in tight spaces. There’s also a less viscous version of green sauce that can wick into slip fits through capil­lary action.

Big Data EVT

The whole thing was designed in a week and built in a week. It was seri­ously tremen­dously rushed job, with most of the design work put towards reduc­ing manual tasks (more water­jet!) and cutting many corners (inac­cu­rate mass model­ing, miss­ing holes and fasten­ers).

I usually spend months design­ing and commit­ting thoughts to paper notes, before trans­fer­ring to Solid­Works features over a week or two. But in this case, the whole process was compressed down to a week. I had doubts about it before even send­ing off the water­jet files.

Anyways, ship it (to Harris­burg)!

Dan, Aaron, and Jim setting up

Everything about this robot sucks

Seri­ously. Even when not damaged, the weight distri­b­u­tion is so far forward that Big Data could use hardly any of its beefy drive, the HobbyK­ing 6S controller could sync only to the weapon motor once every ten tries, and even when it did the magnetic clutch didn’t grip the beater bar tightly enough to spin it well.

The real clincher was how frag­ile every­thing was. The drive motors were only face mounted with three M2 screws each to a frame rail, so after the first hit the mount points stripped out and both motors were hang­ing out in free space.

Using a 5-face box made of 1/4 in 7075 was also a bad idea, as was putting the weapon between two unsup­ported rails that easily snapped off. Vari­ous screws were loaded in shear. Heck, even the drive gears didn’t mesh that well. Every­thing that could be crappy about an honest attempt at a non-ass bot, I made crappy.

But the Corn­flakes held up fine, so I taped the drive motors back in, fixed the weight distri­b­u­tion by remov­ing the beater bar, and tossed it once more unto the breach.

This is one cloud tech­nol­ogy that failed to disrupt.


On the other hand, Sheboy­gan.


  1. Remem­ber I’m back­track­ing to early 2014. []
  2. At 6S: six lithium poly­mer cells in series, for a maxi­mum of 6 × 4.2 V = 25.2 V. []
  3. Config­ured in a “wye” termi­na­tion, if you cared. []