The Smell of Molten Projects in the Morning
Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
If it used to work, it can work again
You can do it by hand, as I used to, or use recoverjpeg:
dmesg | tail
cd /tmp
sudo dcfldd if=/dev/sde1 of=pix.bin bs=1M count=100
recoverjpeg pix.bin
ristretto image00*
Nothing prizewinning, but better than no picture at all:
Note that you start by copying a reasonable chunk of the partition from the Memory Stick / (micro)SD Card first, to prevent a bad situation from getting worse.
Now I can remember the easy way the next time around this block …
A clicking sound from the rear of the bike suggested something was amiss as I rolled up the driveway after a recent ride. Spinning the rear tire produced this alarming sight:
Pulling it out of the gash shows it’s the base of a W5W (or something similar) automotive bulb:
Which seems perfectly designed to cripple a bike tire:
The gash cuts all the way across the tire tread:
The blue stuff is Schwalbe’s rubber / latex / plastic SmartGuard layer, all 5 mm of it hard at work separating the glass from the inner tube.
I cleaned the wound, filled it with silicone rubber, topped it with some duct tape, and it’s still holding air after a 13 mile ride. I think the gash cut through the rubber tread and SmartGuard layer, but didn’t affect the cords in the tire carcass, so keeping further road debris out of the gash should let the tire wear out more-or-less normally.
Putting duct tape on the tread will certainly help …
Five years later, the digits I painted with Rust-Oleum Rusty Metal Primer have weathered pretty well, while the original ink has fallen off the retroreflective sticker:
As before, I wiped off the crud with denatured alcohol and painted neatly inside the lines. The other digits on both sides still look as good as the day I painted them, with only a few bubbles and nicks.
Memo to self: Next time, buy a big sheet of 3M retroreflective film, make a stencil by vinyl cutting, paint the entire number in one shot, and be done with it.
I built a small plywood work table for the drill press:
Obviously, that was a long time ago. It’s a plywood scrap with a small cleat screwed to its bottom, upon which one can position / clamp / hold / finagle smallish workpieces without worrying about drilling into the surface.
The most recent batch of aluminum backing plates prompted me to finally replace that relic:
The mill vise under the plywood grips the cleat and the whole affair rides on a Sears “Drill Press Milling Attachment Stock No 27585” which is basically a simple XY table with hand dials. It’s not rigid enough for actual milling (which you should never do on a drill press, anyway, because the end mill will pull itself out of the Jacobs chuck), but it’s good for tweaking the position before you drill something.
One should never hand-hold workpieces while drilling.
Don’t do as I do, do as I say. OK?
Having figured out the geometry for two- and three-channel soaker hoses, I cranked out more clamps:
Actually, those are the remainder of two production runs devoted to reducing the amount of water sprinkling the garden paths. A 50 foot hose runs along both sides of one 14 foot bed, crosses the path, then continues along the adjacent bed. The hoses have (deliberate!) sprinkler holes along their porous rubber body and sometimes the layout puts a hole where it waters the path.
The blue silicone rubber strips provide a bit of sealing to prevent the absurdly high pressure water from streaming through the orange PETG clamps. It’s OK if the clamp leaks, but less flow is better!
I’m getting really good at making those aluminum backing plates and, in fact, I think it’s faster to run the blanks past the disk sander, then drill the holes, than to CNC-machine them. Could be wrong, but Quality Shop Time is not to be sniffed at.
Despite knowing the wire colors inside USB cables need not follow any particular convention, this still came as a surprise:
Yes, that’s a negative indicator on the meter: it reads -5.020 V.
No, I didn’t swap the test probe banana plugs on the other end.
A bit of continuity testing shows the green and white data wires are also reversed, so whoever assembled the cable simply soldered the proper wire color sequence backwards onto both connectors. As long as you don’t cut the cable to reuse the connectors, it’s all good.
Memo to Self: Stop trusting, always verify!
Even though I’m using what seem to be good-quality parts, one of the WS2812 RGB LEDs in a Glass Tile frame died:
It passed the Josh Sharpie Test:
After building the third Glass Tile unit, one of the LEDs didn’t light up due to an easily diagnosed problem:
A closer look:
Shortly thereafter, the Nissan Fog Lamp developed an obvious beam problem:
The WS2812 had the proper voltages / signals at all its pins and was still firmly stuck to the central “heatsink”:
It also passed the Josh Sharpie Test:
I’m particularly surprised by this one, because eleven of the twelve flex-PCB WS2812s in the Hard Drive Platter light have been running continuously for years with no additional failures.
The alert reader will note the common factor: no matter what substrate the LED is (supposed to be) soldered to, no matter when I bought it, no matter what it’s wired into, a WS2812 will fail.
They’re all back in operation:
Although nobody knows for how long …
Obviously, it’s time to refresh my programmable RGB LED stockpile!
The Smell of Molten Projects in the Morning 