Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
Our Larval Engineer acquired a free bicycle to get around at school: a Rollfast “girl’s bike” dating back to 1972 with 105 miles on the odometer. She completely dismantled it, cleaned everything, reassembled it in reverse order, and added a rear rack & panniers. Having touched every part of the bike, she’s now in a much better position to fix whatever may go wrong in the future.
It was an inexpensive bike to start with and we left everything as-is, with the exception of the brake pads. You’re supposed to bend the brake arms to align the pads with the rims, a technique which I didn’t like even back in the day. So we swapped the OEM pads with worn-but-serviceable Aztek pads sporting spherical washers:
They’re way grippier than the old pads, even on those chromed-steel rims. I had a bike with steel rims and old pads; given the slightest hint of water, it didn’t stop for squat. With any luck, the Azteks will at least slow this one down.
Although she wanted to take the Tour Easy, the bike must live outside under the apartment stairs all year and, frankly, that’d kill the recumbent in short order. Forgive me for being a domineering parent; when she has a good place to store a spendy bike, it’s hers for the taking.
We haven’t figured out how to mount the GPS/APRS tracker + radio and antenna. The evidence suggests she prefers to travel incognito from now on…
So I bought a lurid green $8 Tomodachi Santoku knife at K-Mart, which was the first non-stick-coated Santoku-shaped knife I’d seen since that comment. It’s made by Hamilton Forge Ltd, one of those generic names that doesn’t produce any search results worth mentioning and so probably isn’t a real company:
Tomodachi Santoku knife
The knife has a huge steel blade with a solid plastic handle injection-molded around a short tang, which put the balance point maybe 50 mm out into the blade. I didn’t like the feel when I waved it around in the store and really didn’t like how it behaved on the cutting board.
The way I see it, I can fix a too-light handle…
Pursuant to that post, I have a bag of tungsten electrodes, some complete with a glass seal:
Tungsten electrode with glass seal
Wrapping some masking tape around the glass, tapping it with a hammer, then sliding the tape-with-fragments into the trash got rid of the glass. The bulbous tip seems to be a stainless steel tube welded around a thin tungsten shaft, so I clamped it in the vise and whacked it with a chisel; tungsten is strong-but-brittle and cracks easily:
Fracturing tungsten electrode
Of course, whacking a tungsten rod didn’t do the chisel the least bit of good, but it was about time to sharpen that thing anyway.
Why use tungsten electrodes instead of, say, ordinary drill rod? Tungsten has about the highest density you can get without going broke, getting poisoned, or dying of radiation exposure. That useful table gives elemental density in g/cm3:
aluminum = 2.7
iron = 7.9
lead = 11.4
gold = 19.32
tungsten = 19.35
osmium = 22.6
Can’t afford gold, not even I would put a lead slug in a kitchen knife, and I had the electrodes, so why not?
Waving a neodymium magnet over the handle convinced me that I could drill a hole slightly more than two inches deep without hitting the tang. I briefly considered drilling half a dozen smaller holes, but that started to look like a lot of work and I don’t have any suitable gun drills.
The business end of the electrode measures 1 inch long and 0.1375 inch in diameter. A hexagonal cluster of seven rods fits neatly into a round hole about 3×0.137 = 0.413 inch in diameter: quite conveniently a nice, long Z drill. So I clamped the knife between two strips in the drill press vise and had my way with it:
Drilling knife handle
Actually, I spot-drilled with a center drill, then used a long step drill, stopping with the 3/8 inch step just kissing the low side of the handle, to get the hole mostly on center, before running the Z drill down about 2-1/8 inch. The handle walls became so thin that they flexed around the drill to produce an undersized hole, so I reamed it with a hand-turned 7/16 inch drill and the electrodes fit with no room to spare:
Tungsten electrodes in knife handle
Yeah, that’s a crack in the top electrode: tungsten is brittle.
A dollop of epoxy atop the electrodes should seal them in place forever. I clamped the knife (in its color-matched scabbard) with the angled end of the handle water-level, so the epoxy settled in a neat, symmetric blob that looks better in person than it does here:
Epoxy seal over tungsten weights
The epoxy forms a plug over the ends of the electrodes and (probably) doesn’t extend very far down between them, but they’re firmly jammed in a snug hole and (probably) won’t ever rattle around.
Seven electrodes weighed 32 g and, figuring the missing plastic rounds off to slightly over nothing, the handle now has 60 g of additional weight out toward the end, producing a knife weighing 185 g that balances near the narrowest part of the handle. It’s somewhat heavier than I’d like, but I can cope.
The edge came from the factory reasonably sharp; a few passes over the sharpening steel touched it up nicely.
Early results: it cuts cheese perfectly, drifts to the right in melons, cuts wafer-thin slices from a loaf of my High-Traction Bread, and dismantles fruit with some clumsiness. Overall, I like it, although I could do without the bright green color in a big way.
After our Larval Engineer tweaked the code to track the maximum speed for the current run, so that the color always hits pure blue at top speed and red near standstill, we can prove it happened: we have a video! It’s much less awful than the First Light video, but with plenty of cinéma-vérité camera shake, lousy focus, and bloopers:
Longboard In Action
That’s a frame extracted from one of the raw videos files using ffmpegthumbnailer:
for t in `seq 0 10 100` ; do ffmpegthumbnailer -i mov07117.mpg -o Longboard-$t.jpg -t $t% -q 10 -s 640 ; done
This view of the 3D printed case shows the power switch and the Hall effect sensor cable snaking out of the truck just below the near axle:
Longboard RGB LED Electronics – right front view
She filled the case corners that pulled up from the build platform with a thin layer of epoxy, getting a plane surface by curing it atop waxed paper on the shop’s surface plate, to keep the polycarbonate sheet flat. I didn’t have any acorn nuts to top those nylon lock nuts, alas.
The 4-cell Li-ion battery lives in the slice between the white aluminum plates, where it takes about four hours to charge from 3.0 V/cell. The Arduino Pro Mini lives behind the smoked polycarb sheet, where its red LED adds a mysterious touch. Maybe, some day, she’ll show the 1/rev pulse on the standard Arduino LED for debugging.
A view from the other side shows the hole for the charger above the circuit board, with the Hall sensor out of sight below the far axle:
Longboard RGB LED Electronics – left front view
Yes, the cable to the LEDs deserves better care. She learned that you must provide strain relief at cable-to-component junctions, which we achieved by pasting the wires to the board beside the LED strip with double-stick tape. The rest of the LED strip interconnections live atop similar tape strips. There’s nothing much protecting the LEDs or their delicate SMD resistors, butit works!
Actually, one red LED in an RGB package went toes-up and wasn’t revived by resoldering its leads. So we jumpered around the package, subjecting the remaining two red LEDs in that string to a bit more current than they’d prefer, and that’s that.
There’s a whole bunch not to like one could improve in both the mechanics and electronics, butit works! If you’ll grant it alpha prototype status, then I’d say it’s Good Enough; this is her project and she’ll learn a lot from how it works and how it fails, just like we all do.
Having gone to great pains to put the center of the contact studs on the GPS+voice case exactly at the center of the screws on the back of the radio:
HT-GPS Case – Wouxun KG-UV3D rear view
I now discover why Wouxun used 7 mm square pads on the batteries: the springy contacts hit the pack so far off-center from the studs that they very nearly miss the heads on the 4-40 brass screws I’m using as contacts. This family portrait shows the radio, the battery pack, and the GPS+voice case:
Wouxun KG-UV3D – battery contact locations – GPS case
The lines on the masking tape highlight where the spring contacts touch the case and barely kiss the screw heads:
KG-UV3D contact marks on GPS case
Squinting at the marks on the battery case contacts (you can’t see it in the pictures), the contact line is maybe 2.5 mm beyond the centerline of the square pads. How this worked on the first case I built, I have no clue. For this version, I deliberately filed the heads a bit less and recessed them into the case a bit more; obviously, that was the wrong thing to do, as the connection was intermittent at best.
For the purposes of getting things working, I wrapped snippets of copper mesh tape (from NASA, according to the surplus blurb, with conductive adhesive) around thin chunks of conductive foam, then put them over the studs. The scars in the plastic came from an abortive attempt to get the springs far enough into the case surface to kiss the very edge of the studs:
Copper mesh on GPS case contacts
There’s no point in having a contact patch on the near side of the radio springs, because nothing ever touches there. So the right thing to do is simply move the contact studs to the far side by 3 mm, centering them around the actual contact point. That means changing the PCB layout by the same amount. That’s easy enough to do, but … drat!
When I took the case apart to boost the mic gain, I replaced those neatly filed studs with unfiled pan head 4-40 brass screws from the same parts stock. The heads were tall enough to touch the radio spring contacts closer to their centers and make perfect contact. Not elegant, but better than that copper braid tape.
The one thing I do not like about the Wouxun battery packs: the radio contact pads are flush with the pack surface, so there’s absolutely no protection against casual shorts when the pack isn’t on the radio. The packs also sport four bare round contacts on their outer surface that mate with the charger, two of which make direct contact with the battery; those sit inside a shallow molded recess that helps prevent inadvertent shorts.
I assume there’s a protective circuit inside the pack that turns off the current on a dead short, but I am most assuredly not going to test that assumption. When the packs aren’t on the radio (which they never will be, effective immediately), they sport a strip of tape across those radio contact pads.
After our Larval Engineer allowed as how OpenSCAD’s learning curve was rather too steep, I punched a few holes in the solid model of the case for the Longboard Ground Effect Lighting controller:
Longboard Case Solid Model – with holes
Those rounded corners sucked the Kapton tape right off the build platform as the massive shape shrank. The top layer was the worst offender, with 1.4 mm of clearance (shown with that tapered scale) under one corner:
Longboard case – warped corner
The warping doesn’t matter much, because the case will be compression-loaded by screws and wave washers in the corners. We may need to fill or level the warp to keep the polycarbonate cover flat, though.
I thought about putting a support structure in the rectangular power switch opening, then decided to just try it and see what happens. It turned out fine; this view looks up toward the as-printed top of the opening (the camera’s barrel distortion makes the curve on the bottom surface look worse than it is):
Longboard case – switch hole overhang
Four stacked lithium cells produce upwards of 14.8 V, considerably more than those poor 12 V LED strips prefer to see, so I had her take some current vs. voltage data. She figured out how to convert 10-bit ADC values into battery voltage, after which she could, if she wanted to, beat her Arduino sketch into limiting the maximum PWM duty cycle to hold the LED power dissipation down to a reasonable number. Right now, it’s set to a fixed 25% and is way bright.
Longboard with variable RGB LED Ground Effect lighting
A truly crappy First Light video taken in the driveway is there. She’s been doing the Happy Dance all day… and promises to document the whole project in gruesome detail.
I designed the GPS+Audio case around the TinyTrak3+ board in my radio, which has two square, blue-plastic trimpots. The case worked fine for that board. Then I printed the case for the next bike and that TT3+ didn’t slide neatly into place:
TinyTrak3+ trimpot overhang
Turns out that one of the three TT3+ boards uses plastic trimpots and the other two have metal trimpots bent to fit the existing holes (so they’re not a drop-in replacement), with a very slight overhang beyond the edge of the PCB.
So I attacked the case with some riffler files and carved a notch above the PCB slot. No pictures of that, lest you think I’m a butcher of lovely 3D printed objects. Next time: build the notch into the case’s solid model.
Most likely, this is the only instance of those pots causing anyone a problem…
Our Larval Engineer has been diligently procrastinating on her summer project to add ground effect lighting to her longboard. I’m hereby depriving her of the opportunity to learn enough OpenSCAD to build a case from scratch:
Longboard Ground Effect Lighting Case – exploded view
This is upside-down from its in-use position, but she’ll have it in this orientation on the bench. Four 10-32 screws clamp the whole affair together and hold it to a bottom aluminum plate with threads to suit; that plate also gets bolted between the longboard and the rear truck.
The general idea is that four 2 A·h lithium prismatic cells live in the bottom slice with their protection circuit, sandwiched between two aluminum plates that should protect them from all but catastrophic impact. The circuit board (which ought to be a PCB, but we’ll go with hand wiring for the first iteration) gets clamped in the recess between the two upper slices, above the upper aluminum plate. A polycarbonate sheet on top provides visibility for the Arduino blinky LED inside and shows off the circuitry to one and all.
I think a ridge on each long wall should suffice to hold the cells against the end wall; we don’t have the cells in hand to figure that out yet. She gets to add internal partitions, cable cutouts, and suchlike.
Oh. “Ground effect lighting” means ten RGB LED strips glued under the longboard deck. Her innovation is to make the LED color depend on the speed, which can range upward to scary-fast. It’s a simple matter of software, using a Hall effect sensor for input. This will look much better after dark, but she’s pretty much nocturnal anyway.
The Smell of Molten Projects in the Morning 