In our last episode, Pop Quiz moved. I finally confirmed that the drive gearboxes are at least adequate to move the bot around as a weapon platform, possibly even doing some light pushing. Today, however, was a day of epic progress on Ãœberclocker, since I was able to get on the waterjet cutter. In the same session, I also cut Pop Quiz’s top and bottom armor plates out of a raw carbon fiber panel. Drivetrain parts for LOLriokart are currently in limbo due to a bank fart, but some additional components are on the way, and said fart should be resolved during Monday’s regular financial industry business hours.
So now, the projects in order.
Pop Quiz 2
Semi-finished CF panels for Pop Quiz. These are made from .041″ thick carbon fiber, and still require countersinking of the screw holes before I can mount them. Normally, delamination is a concern with these composite materials (like CF, fiberglass, and Garolite). I dug up one of the waterjet bricks and taped the carbon fiber to it. The bricks add many more points of support over the thin steel slats in the tank, and so I was able to run these parts with only the slightest delamination on one hole. Pretty good if you ask me, and SIGNIFICANTLY better than TB4′s garolite plates (which were run directly on the slats and suffered from huge delaminated splotches).
Activating the “low piercing pressure” option also helped.
Here’s a quick video of the jetting process. 6 megs, Quicktime .MOV, straight off my not-camcorder.
The 1/4″ 2024 plate I ordered originally for TB4.5MCESP1 is on its last legs of usefulness, and I’m going to milk as many parts out of it as I can possibly manage. The first step is to properly lay out your parts to conserve material.
I spent the better half of the afternoon carefully arranging all the aluminum parts as closely together as they could be put without interfering. The jet nozzle is 0.03″ across, so anything closer together and the parts will be damaged by the cutting action. My custom now is to space things about 1/16″ apart.
Also, taking advantage of part mirroring or rotational symmetry helps. As the picture shows, the front “legs” of the bot nested nicely into eachother.
To make the process go faster, I also merged every part into every other one. That is, the whole thing up there is one contiguous mass of metal. The parts themselves are joined by tiny tabs about 1/32″ wide. This prevents the parts from jumping around after they are cut out – a mutual support network. I’ll just brak the tabs off when I want to separate the parts.
The added side effect? After the jet is done piercing the holes, it only has to pierce the part in one place, then just make a round trip, following all the outlines one by one. There is no longer a need to pierce and cut each part. It speeds up the process by however many pierces you are able to reduce (each one is a few seconds long, plus traversing between parts)
Next up was the half inch aluminum slab. I originally wanted to cut SP1′s arm links out of this stuff, but decided to save it for another day. This particular 2024 slab is actually finely polished on one side. That means it’s very very shiny. I actually did mirror the left and right parts this time, which is why they don’t nest as well as the quarter inch parts, because I wanted the shiny side out.
I should have cut with the cardboard-adhered side down, because waterjet droppings toasted the finish on some of the holes. Oh well – those are INSIDE the bot!
The “round trip” around all these parts took an hour.
At the end of the day, all the finished parts. I tried to finish the top and bottom plates for Ãœberclocker also, but the waterjet got cancer.
On the first cut with Pop Quiz’s carbon fiber, the jet nozzle spat out its abrasive feed tube. Uh oh – bad, because some water would inevitably be squirted up the tube. The entire abrasive feed system must be absolutely dry in order to not clog during operation.
After consulting with one of the shop managers and totally blasting out everything with compressed air, the jet then proceeded to cut and pierce all the aluminum parts without incident. I was actually up and ready to leap upon the Pause button as soon as I saw sand begin to fly out of the tube, but it wasn’t needed.
However, about two holes into the 1/16″ garolite for Uberclocker… which should have been trivial by comparison and taken about five minutes, the jet ditched the abrasive tube again. I cleaned it up as best as I could, but machine repair is not under my jurisdiction, and so I called it quits for the day, and left a note for the shop managers. I hope they don’t bite too hard.
If it really is waterjet cancer, then the parts will have to wait until it is uncancerfied.
By the way, the shine on the edge of the 1/4″ aluminum parts is from the reflection off the extra extra shiny half inch aluminum.
So concludes bot work for now. I have plenty of parts to keep me busy, and if worse comes to worst, Ãœberclocker’s cover plates are simple and rectangular – I’ll just kick it old skool with calipers, a drill press, and selective sawing.
Because much of the work on LOLriokart is contingent on my drivetrain parts arriving, the progress has been a bit sparse and mostly related to the front end with steering and braking.
I ran into a slight quandary regarding the front brakes on the kart. Researching go-kart designs on the Intertubes, most brake designs seem to use the rear drive sprocket as a single large discÂ brake.
Simple and effective, but I wanted something different. The Etek controller already has regenerative braking built-in (as well as a non-regen dynamic braking). Adding another rear axle brake would just be redundant, though a mechanical brake is still necessary to fully stop the vehicle.
I wanted front brakes for several reasons. One, to separate the mechanical and electrical brakes. Two, to have the option of slamming both pedals at once to perform epic burnouts (braking the rear axle would just… well, burn out the motor).Â And three, to easily cause weight transfer during slowing and turning to initiate drift. You know, all the important stuff.
I had several options in front of me that had varying degrees of engineering involved.
- Simple tire brake, kind of like the fender brakes on scooters. However, the thought of running a plate of metal against a pneumatic tire sort of put me off to the idea. I didn’t want to mess with trying to curve metal that was stiff enough to keep its shape under braking pressure either. These could be cable actuated.
- A “squish brake” idea that revolved around the brake pad of a bike, mounted on a sliding joint. Spring and cable magic would allow me to slam this brake shoe against the interior of the wheel rim. Sort of like a sideways drum brake. This idea needed the most custom engineering and probably careful tuning.
- A bicycle coaster brake, modded to become the actual hub of the wheel. A coaster brake is an interesting gadget – it has three modes, “Drive”, “Coast”, and “Brake”, all actuated by different angular positions or velocities of the drive sprocket.
They are based on a short screw actuator and interior cone clutches and brakes. The wheel hub spins independently of the sprocket, with no freewheel mechanism in between, or any other mechanical connection except when the sprocket moves. Essentially, when the sprocket turns one way (let’s say “forward”), the screw actuator forces the clutch cone one way, squeezing it into its mating cone inside the hub. This transmits power to the hub.
When the screw turns the other way for a far enough distance, e.g. backpedaling, the clutch cone is pushed in the other direction past the “no engagement” point, and spreads out a set of fixed brake pads, usually attached to the dead axle. This, of course, produces intense internal friction and the brake is able to function.
This actually seems very applicable for a flywheel powered weapon in a ‘bot.
Some cable trickery could imitate the motion of the sprocket. I really only need the “Brake” part, and so could set up a simple short-throw lever. However, a 3/8″ threaded axle is standard on pretty much all bike rear ends. There was NO way it was going to hold up under the weight of the kart, since they would be single-supported. I would have had to severely mod the brake mechanism to fit on a larger axle.
- A custom derivative design of the coaster brake, essentially making my own cone brake using some big brass rounds.Â Conveniently enough, a few large tapered roller bearing shells would have made great brake surfaces, and I would even have integrated said tapered roller bearings into the wheel hub (massive overkill!), but it was the actuation and associated leadscrew trickery that stopped this idea dead. There was no way I was going to buy a foot of 1″ leadscrew ($$$$) to use 3 inches. (Or maybe I could just cut my own threads on the lathe)
- SoÂ I finally broke down and ordered some band brakes with included rotors designed for use on electric scooters, by real engineers no less. They would just be tucked next to the wheel, actuated by cable.Â Who knows, maybe I can have some fun with the mechanism anyway.
I never knew I would put so much effort into making something stop for once.
Anyway, a few minutes with an angle grinder and I have this large hole in the basket. I’m having a bit of fun deciding between sitting fully in-basket or using the back half as a seat with pedals mounted on the front crossbar where the casters used to be. The former is more associated with shopping cart riding and adds to that surrealism, but it would suffer from a very high center of gravity. The latter allows more natural, car-like control, but just turns this into another overpowered go-kart.
Regardless, I need a hole in the basket to route important mechanical linkages and cables.
Oh, and the jagged metal edges will of course be ground off and the edges themselves covered by something plushy, so it doesn’t saw my legs off in operation.
A collection of coaster brakes that I dug up & purchased. Maybe these will work for a bike some day, but I need a more severe-duty setup for now.
Let’s start on the other parts in the mean time. I decided to go the overkill route and make the steering forks from aluminum stock bolted to the front crossbar. All these parts are made from 1″ x 2″ aluminum barstock (I didn’t stay at a Holiday Inn Express last night, nor did I save 15% on my car insurance by switching to Geico, but I did bust $40 on a whim for a giant stick of metal).
Totally overkill, a waste of stock, and not designed according to stresses, but… do I really care?
Each steering fork will be retained by a giant 1/2″ bolt (through the original caster mounting hole) and four little 1/4″ bolts.
Upper and lower parts of the fork complete. I don’t have bolts long enough to sandwich them together, so this is just a gimmick picture. I really should have threaded the bottom part of the fork, in retrospect, but a locknut will do just as well.
Bushings will be installed in the axle holders so they can pivot on the cap screw kingpin. Then I’ll somehow link everything together to a common steering arm!
Nothing to see here, just shoving a wheel next to the thing to see how it fits!
It’ll be easy to create an axle mount which sets the wheel higher than the steering fork – for epicly low riding awesomeness. Decision to come.
In other news…
Today (last night?!) was epic MITERS cleanup night. The short way of describing it is – four guys, five hours, and a little more than six trips to the dumpster.
About 100% more floor space exists in the room now – with actual space to work on the tables. All the tables had some degree of crap piled on them before. It was absolutely unreal.
Here’s my personal handiwork – materials processing. The tubing sorter we got from the Solar Car/Formula SAE cleanout was put to use in sorting tubes. Alot of redundant or useless material, like galvanized conduit tubing and the body sections of hockey sticks, were thrown out completely. Useful wood was saved in the wood pile for further processing, and alot of scraps which contaminated the stock bins were thrown out also.
All in all, I think we tossed a few dozen dollars’ worth of metal. D’oh.
No pics of the “Before” setup, but take my word for it – the literal pile of random materials extended past the line between the shelf corners.
Stuff too short for the sorter was packed into bins. These bins were pre-existing, but I just went through and piled them more accurately.
This weekend should be filled with plenty of bot work. More updates to come!