Engineering contraptions
I finished applying all the layers of paint, sparkles, and clearcoat to the Zipp 2001 frame a couple months ago, and finally worked up the motivation to sand and polish the clearcoat. I did 2 coats of dark grey base coat, 2 coats of the Keirin flake topcoat, and probably 5+ coats of the clear topcoat so I’d have enough meat to sand down the high spots and polish without hitting the sparkles…
I got a Wacom MovinkPad 11 drawing tablet a few months ago as a portable replacement for my ancient Wacom Bamboo USB drawing tablet. Overall it’s been great. I’ve been using it for illustrating blog posts and making engineering sketches for work and personal projects. I’ve enjoyed Clip Studio for actual “art”, but haven’t found a program I like as an engineering…
I replaced the little 5-axis mill with a router recently, and needed a low-profile vise system for holding large workpieces. I’ve heard good things about the SMW Mod Vise, but they are designed for fixture plates, not 100mm pitch t-slots like the router has. And 3 sets to cover the hole table with serrated inserts would be pretty pricey. I came up with a simple design that uses these…
Adhesive bonds. Don’t worry, this isn’t becoming a finance blog. I do a lot of bonding parts together between my work and personal projects. At this point I have some (I think) reasonable default choices - Loctite 401 or 380 for plastics and elastomers, Loctite E-20HP or 3M DP420 for rigid metal & composites, Loctite 648 for metal slip/press fits, etc. But for important or unusual…
This is the beginning of a series of posts about robot actuation. The intent here is not to prescribe some specific architecture or solution, but to talk about the fundamentals, with as little bias as I can manage. Hopefully you learn something. Here’s an ostensibly straightforward question for anyone who’s spent time thinking about robot actuators. Take a look at the three cartoon…
I needed to convert a DT Swiss hub from using a 15mm thru-axle down to a 12mm thru axle to fit on my commuter bike. You can buy the parts to do this, but instead of spending $50 on the parts, I spent some quality time with the lathe and a log of 7075 to make a new set.
The original saddle clamps on the Zipp 2001’s were not very well designed, and the clamp that came with my frame (like many) was broken. The originals have 2 major problems: 1) The screws that clamp the rails on the saddle are tightened counterclockwise, not clockwise - I’ve seen stripped out screws because of this on a couple clamps. You think you’re loosening the saddle clamps,…
I think the Zipp 2001 is one of the coolest bicycles ever made. These were only produced from 1992-1997, until the UCI banned bikes that weren’t the traditional “double-diamond” shape from professional racing. Probably because they were too fast. Image taken from this 1993 Zipp Catalog The early 90’s was when frame builders started seriously using carbon fiber for its…
3D printed expanding fixture for op 2
Split rear axle means 2 rear brakes Laser cut stainless steel bracket from Send Cut Send Using the lathe as a drill press to countersink the holesRe-using the old brake rotor Time for some plumbing
It has been an eventful year. I'm finally back to working on projects. After some firmware fiddling, the new drive units are finally spinning, driven by the kart's original gen 2 Prius-based inverter:
This isn't one of my usual project blog posts, but I owe my dear readers some context for the lack of project content. The last year did not go like I expected. Part 1: I ride my bike a lot In mid May last year my friend Sam and I decided to ride the Great Divide Mountain Bike Route, a mostly off-road route from Banff, AB to the Mexico border in Antelope Wells, NM. We'd been talking about it…
All bonded together A torsion-less bar, if you will
A fun afternoon design project: Grip geometry still needs some fine-tuning
As I've mentioned in previous posts, robot land has an annual car-and-other-vehicle show. Last year I brought the trike, but the year before that I brought the electric kart I worked on with Bayley. Picture from Bayley Before the event, the kart hadn't been touched since the pre-Covid days. Bringing it back to life and ripping around the parking lot was a great reminder of two…
Finally retrieved it from MITERS, where it's been hanging from the ceiling for the last six years or so Prius inverter based motor drive will be the next thing to be redone after it drives again:
Why haven't I been building anything? I was off on an Adventure.
Slapped a mini cheetah motor drive on there to make it go
Missing a rotorMachining a new shaft for the rotor: Improvised chip deflector for roughing nice Other side done bearing and lock ring test fit Next to the stock shaft. Going for a normal press vs the straight knurl on the stock shaft. I don't own a 2d printer and didn't plan far enough ahead to print the drawing at work, so this happened
Single-point keyway broaching
Dane tricked me into building something dumb ice skate in a garbage pile at mitersbridgeporting out a skate holder
Stators shrink-fit into the housings: Machining fixture turned into a guide fixture for the shrink-fit operation Band heater wrapped around the housing to keep it hot while installing the stator: Borrowed the pneumatic powered hydraulic press at work to make sure the stators didn't get stuck half way down:
Motor bracket laser cut and bent from .188" steel by SendCutSend: Crufted NEMA-56-size Clearpath servo motor - integrated motor drive, plugs straight into single or 3-phase mains. 1:1 belt to the spindle: Pile of junk being replaced:
Printed plastic expanding mandrel to hold the part from the inside: Around .005" TIR on the as-printed surface after clamping from the ID. Not too bad. This all gest machined away anyways. Checking the lock ring fit after machining the threads:
After brining it to the BD car show in 2022 and remembering what a blast it was to drive, I got interested in working on the Big Kart again. It's getting new everything, starting with drive units based on the newer Ioniq hybrid HSG. Real blog post when I've made more hardware progress, but for now enjoy this picture of some 3d-printed aluminum motor housings with internal water cooling…
Picked up from a technician at work who'd pressed many a bearing: Put a magnet in the back side of the press tool so it sticks to the bottom of the arbor press ram Tool sticks to the ram and automatically squares itself! No need for 3 hands to hold the part, the press tool, and the arbor press lever:
3d printed dovetails - the AXA sized holders are just big enough to fit two screws into a 20x40mm aluminm t-slot extrusion:
More little drone motors, flywheels, and motor drive mounting brackets: New MJF printed structure to hold everything: This is almost everything, shooting for 1kg total when it's done, not much left for the protective shells, foot, and control PCB.
Robot land has an annual car-and-other-vehicles show. Last year I brought the hybrid car powertrain go kart, but this year the kart's geting some upgrades (stay tuned), so I brought the electric tricycle instead. The trike has been hanging from a miters wall unused since 2019. The last time it was ridden, someone crashed it into a curb and bent the head tube…
There will be a proper blog post
It balances! The inertia changes significantly as the "body" moves up and down the leg - for the controller, I got feedback gains with LQR for the maximum and minimum extension cases, and linearly interpolate between those gains as the leg extends.
Running off a bench supply and offboard micro, controlling with a joystick. There's an IMU attached too, so hopefully soon it'll be doing some 1-DOF balancing:
Another CPW, another round of tesla coiling. Last time the primary capacitor caught on fire, so I helped Michael and Austin cobble together a new one out of the Eurofarad caps the coil originally used. I designed some bus bars and MITERS undergrads got them waterjet cut for us. 216 M6 socket head cap screws and 36 crimps later: Successfully ran 2 evenings with 0 capacitor…
Still needs a motor, of course, but it's taking shape Someone asked what the step on the inside of the reaction wheel was for:
I installed a DRO on the Clausing. I got Aikron scales and console - the LCD console, a 1 micron slim magnetic scale for the cross slide, and 5 micron standard width scale for the carriage. So far I'm very happy with it - the user interface is better than some US-made, much more expensive DRO's I've used, and being able to store a library of tool offsets is a real game changer -…
Not long after my masters thesis went online, clones of the mini cheetah actuators started showing up on AliExpress/Alibaba. Within another six months, there were clones of the whole robot. A dream outcome for the project - early on in the project I joked that it would be great if I could buy the stuff I designed from China cheaper than I could make it for (hence the original name,…
One of the Chinese mini cheetah clone manufacturers was nice enough to send me one. Stay tuned for a detailed (non-destructive) teardown.
Once again Mike's tesla coil worked flawlessly - at least until someone tripped over the extension cord powering a fan cooling the primary capacitor bank. We didn't notice until the capacitor bank burst into flames.
Here's the detailed post to go with the pictures and videos I've been dribbling out over the last few months. Last episode the mechanism was mostly finished, and was just missing the guide rollers that constrain the screw to linear motion and react the motor torque on the screw. Looking at the CAD cross section, there are 3 sets of 3 rollers ride in shallow axial grooves on the screw. The…
Gremlins eradicated - USB hubs are bad, don't use them: "Only" ~3.3 meters in this test - not maxing out the motor/mechanism yet, there's still some motor control tuning to do. And probably another round of transmission ratio optimization once the actual robot mass and inertia get finalized.