Archive for February, 2020
We bought the best-looking (pronounced “least bashed”) pair of hulking five-drawer industrial-strength HON Brigade Lateral File Cabinets from the local ReStore outlet’s assortment for Mary’s quilting fabric stash. They came with a steep discount, barely fit inside the Forester, caused minor interior trim damage, and should organize her entire stash.
One cabinet lost a foot nut at some point in its 16 year history:
The surviving foot nuts sported two weld nuggets apiece:
The hole had the remains of one nugget at the top left and looks like a manufacturing defect to me. Of course, we’re (at least) the second owners and the usual lifetime warranty no longer applies.
I can fix that.
Bandsaw a 1×¾ inch rectangle from 3/8 inch aluminum plate to match the surviving foot nut (which is steel, but aluminum will suffice for our needs). Break the edges, clamp in the Sherline, and mill a square protrusion to match the square-ish hole:
Drill a 17/64 inch hole (looser than the nominal F drill, because I’m a sissy) for a flat-head bolt from the Drawer o’ 3/8-16 Bolts, tap, and clean up.
A trial fit showed the nugget had to go before the nut would come even close to fitting flat into the hole:
The sheet metal around the hole had absorbed at least one mighty blow pushing the entire surface inward behind the front edge. To compensate, recess the nut’s front edge and slope the sides with a Dremel wheel to let the bottom face sit level:
Another trial fit showed the need for more recess:
Another spate of grinding made it sit mostly level on the decidedly non-level surface around the hole:
The beveled corners fit inside the swaged hole corners.
Grind paint / crud off the sheet metal and roughen the surface for good epoxy griptivity:
Stand the cabinet top-side-down to make the bottom level. I wish the basement had one more course of block, but it’s not to be.
Butter the nut with JB Weld epoxy, plunk it in place, apply excess epoxy to make a fillet around the edges, apply duct tape to guy the top of the bolt level-ish, and let it cure:
After the epoxy stiffened enough to hold its position, remove the bolt, file a crude ¼ inch hex, and saw a screwdriver slot to make it match the other feet:
Not the fanciest job I’ve ever done, but it now behaves just like the other ones and it’s all good. The HON Storage Files FAQ points to a Troubleshooting Guide showing how to level the thing with a hex socket from inside the bottom drawer.
The flat heads on those bolts are basically 25 mm OD steel plates calling for fuzzy felt bumpers on the Sewing Room’s wood floors. When properly leveled, the front will be ⅛ inch higher than the rear. Although they suggest a pencil should roll toward the back, the top sheet metal on this one may be sufficiently warped to confuse the issue; I have a long level well suited to the task.
The original dimension doodle includes metric offsets for cutting with a ¼ inch end mill:
All in all, a satisfying day in the Basement Shop …
Five years ago, robotic trash cans were a thing on Vassar’s campus, including this duo in front of Vassar’s library:
I infer the robotics did not work out as anticipated.
It seems reporting a frozen hydrant to the local fire company didn’t produce any meaningful action:
We didn’t have any fires in the neighborhood where it might have been a problem, but I’ll try the water department this year …
Oddly, the water department repainted most of the fire hydrants along most of the roads last year. This one apparently didn’t qualify, for whatever reason, despite being only slightly off Rt 376 on Sheldon:
When it’s not frozen, it’s not obvious …
For scale, the cylindrical part of the blade is 1.0 mm OD.
The blade with the longer face (left above and bottom below) has seen the most use and is definitely rounded at the tip:
Three unused blades have sharp tips:
From the top, the (nominal) blade angles are 60°, 45°, and 30°, generally indicated by yellow, red, and blue plastic caps. However, various eBay sellers disagree on how to measure the angle (up from surface / outward from axis) and which cap colors correspond to which angles.
The unused 45° blade bracketed by the two used blades:
The two lower blades have angles somewhere between 30° and 45°, suggesting slack grinder and QC tolerances. If the actual angle matters to you, buy an assortment (from one seller!), measure what you get, and don’t be surprised when the results aren’t anything in particular.
Perhaps, with careful attention to alignment in a non-pivoting / collet holder, one might scribe exceedingly narrow lines.
That’s the back of a sheet of carbon paper (remember carbon paper?), which is deep dark gray in normal light. It’s sitting on the sheet of 100 mil grid paper providing scale for small objects, atop the microscope stage positioner, with cold white illumination from an LED ring light.
Protip: even worn blades remain lethally sharp …
For the usual inscrutable reasons, updating bCNC killed the USB camera on the MPCNC, although it still worked fine with VLC. Rather than argue with it, I popped a more recent camera from the heap and stuck it onto the MPCNC central assembly:
This one has a nice rectangular case, although the surface might be horrible silicone that turns to snot after a few years. The fancy silver snout rotates to focus the lens from a few millimeters to infinity … and beyond!
If you think it looks a bit off-kilter, you’re absolutely right:
The lens image reflected in a mirror on the platform shows the optical axis has nothing whatsoever to do with the camera case or lens snout:
Remember, the mirror reflects the lens image back to itself only when the optical axis is perpendicular to the mirror. With the mirror flat on the platform, the lens must be directly above it.
Because the MPCNC camera rides at a constant height over the platform, the actual focus & scale depends on the material thickness, but this should be typical:
It set up a Tek Circuit Computer test deck within 0.2 mm and the other two within 0.1 mm, so it’s close enough.
The image looks a whole lot better: cheap USB cameras just keep improving …
One of the bird box entrance reducers I installed nigh onto a decade ago is still on duty, although downy woodpeckers definitely want a larger hole:
Another reducer had gone missing over the years, so I made one from a length of PVC pipe:
It started as 1-½ PVC pipe, 1-⅞ inch actual OD and should fit into a 1-½ hole, so I measured 1.5 × 3.15 around the circumference, bandsawed out the excess, draped it over a 1-½ Forstner bit, toasted it with a heat gun, and squashed it so it’s just a little bit bigger than the (enlarged!) hole in the box.
Now the entrance is 1-¼ (-ish), just like it should be:
The oblong hole required advanced manufacturing techniques:
The front face should be too slick for larger birds and the little ones will zip right into the hole:
The two starlings who’d been evaluating the box seem to have moved on; we doubt they’re now homeless.
The OpenSCAD source code as a GitHub Gist:
A 0.5 mm Pilot V5RT pen produces good-looking results on presentation-grade paper:
Peering through a measuring magnifier shows a bit more tremble in the traces, but they’re still OK:
The desk light off to the upper left casts shadows from the reticle on the three different sheets.
A closer view of the linear scales:
The pen lines seem to be 0.25 to 0.3 mm wide, with 0.4 mm dots at the end of each stroke.
For comparison, the engraved lines on my trusty K&E Deci-Lon slide rule are under 0.1 mm:
The digits look like they’re embossed into the surface with shaped punches, rather than engraved like the lines. Of course, I don’t know how K&E’s production machinery worked.
A closer view:
I think 0.1 mm is an aggressively narrow trace width, even for a laser engraver.