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
Before doing another spring constant test with the old Harbor Freight scale, I found deployed my cheap calibration weight sets to verify it displayed the right numbers:
It’s spot on for all weights above 1 g, although I must tap the pan to settle on the reading from above for it get the last 0.1 g right.
Below 1 g, it’s the wrong hammer for the job; I expected no better from it.
One of my very first projects, after setting up my very first home shop in our very first home, was building an overly elaborate prototype board with five (!) linear power supplies:
The components come from the mid-70s and the shop happened around 1980, so it’s been ticking along for nigh onto four decades. Of late, the supply voltages became erratic and I eventually popped the top:
Yeah, linear pass transistor regulators driven from bulk cap storage, hand-hewn bridge rectifiers, and multi-tap transformers. Everything mounts on screws tapped into the 1/8 inch aluminum chassis, with power transistors on a huge finned heatsink attached to the rear panel. The thing weighs 11.6 pounds = 5.3 kg.
Not a trace of firmware to be found. Heck, surface-mount components hadn’t yet come into common use.
The circuitry lives on a crudely etched phenolic board:
There may be a schematic somewhere in my collection, but it hasn’t surfaced in a long time. I’m mildly surprised I didn’t tuck it inside the case, which may have been a life lesson yet to be learned.
Based on my recent experience with the Tek AM503, I wiggled the two metal-can regulators and the ceramic (!) regulator, gingerly plugged in the line cord, flipped the switch, and all the supply voltages once again work perfectly.
Whew!
The outer doorknob on the kitchen pantry became very loose and sloppy, with the screw holding the inner knob on the shaft remaining snug. Obviously, something else was wrong inside the door.
A spring clip should retain the outer knob in the escutcheon:
The flange holding the clip has worn away, letting the clip fall loose. A side view shows the problem:
Yes, the knob’s chrome plating is in sorry shape after six decades of wear. I’d rather keep using a solid knob, instead of force-fitting some contemporary half-assed / cost-reduced junk into the door.
Reference: beausage. I say it “beau-sage”, the beauty that comes from usage.
The shaft consists of three triangular rods, with the setscrew on the inner knob pressing against the smaller rod to lock all three of them in place and eliminate all rattle & play:
A tapered pin (!) locks the three shaft rods into the outer knob:
Some doodling, most of which turned out to be irrelevant, captured the essential dimensions and suggested how to replace the flange:
The stock is 11/16 inch O-1 oil-hardening rod, forever to remain unhardened:
I drilled a few holes to get up to 1/2 inch, the largest drill bit I have and just barely clearing the the boring bar.
With the hole bored out to fit the end of the knob, cut it off:
Trial-fit the ring on the knob with the spring clip:
Reinstall the shaft, tap in the retaining pin, then epoxy the ring in place with the knob supported from below to eliminate having to fiddle with the spring clip:
Add a few dots of oil here & there, reinstall the parts in reverse order, and the knob works perfectly again. Still looks heavily used, of course, but that’s OK.
They definitely don’t make ’em like that any more …
The MPCNC has bCNC running on a Raspberry Pi, with a Samsung EVO MicroSD card serving as the “hard drive”:
The picture also shows a defunct Sandisk Extreme Plus killed by continuous video recording in my Fly6 bike camera. I later replaced the EVO with a video-rated Samsung card which has been running fine ever since, albeit with the occasional crash-and-reformat expected with “action” cameras.
With that as background, a different Samsung EVO card from the same batch has been running the MPCNC’s Raspberry Pi for about a year. Over the course of a few days last week, the RPi went from an occasional stall to a complete lockup, although waiting for minutes to hours would sometimes resolve the problem. As I’ve learned by now, it’s not a software crash, it’s the controller inside the card suffering from write amplification while trying to move data from failing sectors.
Applying f3write to the card shows the problem:
The write speed started out absurdly high as the card’s write cache fills, then slowed to to the flash memory’s ability to absorb data, and eventually ran out of steam during the last few files.
But, as you might not expect, f3read reported the data was fine:
sudo f3read /mnt/part
F3 read 7.0
Copyright (C) 2010 Digirati Internet LTDA.
This is free software; see the source for copying conditions.
SECTORS ok/corrupted/changed/overwritten
Validating file 1.h2w ... 2097152/ 0/ 0/ 0
Validating file 2.h2w ... 2097152/ 0/ 0/ 0
Validating file 3.h2w ... 2097152/ 0/ 0/ 0
Validating file 4.h2w ... 2097152/ 0/ 0/ 0
Validating file 5.h2w ... 2097152/ 0/ 0/ 0
Validating file 6.h2w ... 2097152/ 0/ 0/ 0
Validating file 7.h2w ... 2097152/ 0/ 0/ 0
Validating file 8.h2w ... 2097152/ 0/ 0/ 0
Validating file 9.h2w ... 2097152/ 0/ 0/ 0
Validating file 10.h2w ... 2097152/ 0/ 0/ 0
Validating file 11.h2w ... 2097152/ 0/ 0/ 0
Validating file 12.h2w ... 2097152/ 0/ 0/ 0
Validating file 13.h2w ... 2097152/ 0/ 0/ 0
Validating file 14.h2w ... 2097152/ 0/ 0/ 0
Validating file 15.h2w ... 2097152/ 0/ 0/ 0
Validating file 16.h2w ... 2097152/ 0/ 0/ 0
Validating file 17.h2w ... 2097152/ 0/ 0/ 0
Validating file 18.h2w ... 2097152/ 0/ 0/ 0
Validating file 19.h2w ... 2097152/ 0/ 0/ 0
Validating file 20.h2w ... 2097152/ 0/ 0/ 0
Validating file 21.h2w ... 1322894/ 0/ 0/ 0
Data OK: 20.63 GB (43265934 sectors)
Data LOST: 0.00 Byte (0 sectors)
Corrupted: 0.00 Byte (0 sectors)
Slightly changed: 0.00 Byte (0 sectors)
Overwritten: 0.00 Byte (0 sectors)
Average reading speed: 43.04 MB/s
Obviously, the card’s read speed isn’t affected by the write problems.
Assuming the actual data & programs on the card were still good, I slurped the partitions:
sudo partimage save /dev/sdf1 mpcnc_boot.gz sudo partimage save /dev/sdf2 mpcnc_partition.gz
And wrote them back:
sudo partimage restmbr mpcnc_boot.gz.000 sudo partimage restore /dev/sdf1 mpcnc_boot.gz.000 sudo partimage restore /dev/sdf2 mpcnc_partition.gz.000
Unshown: a finger fumble requiring MBR restoration.
Having forced the card controller to reallocate all the failed sectors, the card works now fine and runs at full speed again. This won’t last long, but it’ll be interesting to see how it plays out.
While I was at it, I wrote the partitions to a new-ish / unused Samsung EVO Plus card, now tucked under the MPCNC’s monitor in case of emergency.
An old SFF Optiplex with an SSD may be a better fallback.
The ceramic valve core from our kitchen faucet certainly qualifies for a spot on the bottom flange of the I-beam across our basement serving as a display case / collection area for shop curiosities, mementos, and the like. I am, if nothing else, a creature of fixed habits, because the spot where the core belonged already had one:
The core on the left dates back to the 2016 replacement, so they’ve apparently decided plastic will work fine for the handle socket.
Having the ceramic core fail after two years suggests the manufacturing process needs attention, though. I can still wring the slabs together, though, and they’d need a drop of oil to serve as bearing surfaces.
This compact fluorescent lamp seems to have survived nearly two decades of use in a desk lamp:
It had plenty of starts, although maybe not so many total hours, as the other CFLs you’ll find mentioned around here.
I swapped in a similar CFL and we’ll see what happens.
The clamp holding the magnifying lamp (with a fluorescent ring light!) over the Basement Laboratory Desk finally fractured:
Gorilla Tape held the broken parts together well enough to determine how it used to work:
The two parts used to be 11.2 mm thick, but it fit on a random chunk of half-inch aluminium plate so perfectly as to constitute a Good Omen:
I decided the saw kerf would make up the difference, because, sheesh, we’re talking pot metal here.
Lay out the center, use a transfer punch the same diameter as the lamp pivot to get the proper spacing, give it a whack:
The alert reader will note I came that close to drilling the hole through the wrong side of the angle.
And, yes, extrapolating the vertical edge downward suggests the large hole-to-be will intersect the small hole-in-being. This is deliberate: the clamp screw through the smaller hole fits into a recess around the lamp pivot shaft to keep it from sliding to-and-fro. I had to convince myself, but it really did work out OK.
Pay some attention to clamping it at right angles to the spindle so the big hole goes through more-or-less in the right direction:
The masking tape serves as a depth reminder:
Set it up in a machinist’s clamp, bandsaw in twain, file the kerf reasonably flat, clamp the halves together, then bandsaw the clearance slot:
The clearance kerf wasn’t nearly as on-center as I wanted, which doesn’t really matter, but I filed a bit more diligently on the shallow side while clearing up the slot:
Introducing the new parts to Mr Disk Sander roundified them enough to pass inspection. These angular bits obviously require a bit more attention to detail:
The lamp originally had a fancy knob on the screw which never worked particularly well, so I replaced it with a nylon locking nut to maintain a reasonable amount of pressure:
The far end of the screw has a square shaft fitting into a square hole in the lamp arm, making it easy to torque the nut enough to make the pivot grip the shaft properly; if I ever find my Belleville washer stash again, I’ll add one. I should cut the screw off, too, but that’s definitely in the nature of fine tuning.
A pleasant morning of Quality Shop Time!
The obligatory doodle with dimensions, some of which turned out to be completely incorrect:
The Smell of Molten Projects in the Morning 