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.
Tweaking an LMS 5200 Mini-Lathe
A walk around Wappinger Lake brought me to the old penstock controls:
Still meshed after all these years:
The “new” penstock intake control, a pair of utterly practical and totally non-photogenic screw drives, sits just to the left of these relics.
A night view of the penstock from some years ago:
It still carries water to the recently refurbished power plant downstream on the right.
Those gears will remain meshed after everything else rots away …
Going from a 21 tooth gear to a 42 tooth gear means you must reduce the remaining train ratio by a factor of two for a given thread pitch. Here’s a 42-50-45-60 train, with the same -125 ppm error as the 21-50-60-40 train and no screw / washer clearance issues between the A screw and the C gear:
The original 60-40 CD pair has a 3:2 ratio, the 45-60 CD pair is 3:4, so that’s where the factor-of-two reduction happens.
The first pass at the solid model included a debossed legend:
With a printed gear in hand, I realized the legend must be embossed below the surface, so as not to rub against an adjacent gear; better modeling is in order.
The general idea is to set Inkscape’s (known-good) gear generator to the correct gear parameters (module 1 → 3.14 mm circular pitch, 20° pressure angle):
Save the outline as an SVG:
If you do like I did and neatly position the gear at the bottom-left origin, all SVG viewers will show only the Quadrant I arc, probably because Inkscape sets the SVG file to display it that way. I’ve made that mistake before and maybe, someday, I’ll remember.
Load the SVG into OpenSCAD, which will find the entire gear, no matter where it falls in the coordinate space, and spike it at the origin:
linear_extrude(8.0,center=false,convexity=5)
import(file="/the-source-directory/Mini-Lathe/Change Gear - 42 teeth.svg",center=true);
The linear_extrude( … center=false … ) keeps the bottom of the blank at Z=0. The import( … center=true … ) puts the 2D shape at the XY origin. Because OpenSCAD centers the bounding box, gears with an odd number of teeth will be ever so slightly off-center, which would matter a whole lot more in a fancier machine tool than a mini-lathe.
All of which produces a tidy 3D gear blank:
OpenSCAD ignores SVG holes, which isn’t a problem for me, because I’d rather punch the bore, keyway, and so forth programatically.
But that’s another story …
While not strictly necessary for metric threading on a USA-ian mini-lathe, a 32 tooth gear can produce reasonable approximations, so I printed a pair from a Thingiverse collection:
The model was designed for a slightly different mini-lathe, as it includes a short boss and thinner plate, but it did fit on the shaft:
The gear mesh seemed odd, though, and comparing it with a standard 30 tooth gear and a different printed 32 tooth gear (about which, more later) showed it was definitely not compatible:
Yes, the 32 tooth Thingiverse gear on the right is slightly smaller than the stock 30 tooth gear in the middle.
The larger 32 tooth gear (on the left, above) meshes better:
Both of those trains have a 600 ppm error, so they’re definitely suboptimal compared to the results with a 21 tooth gear in the train.
The real reason you need a 32 tooth gear is for exact 25, 50, and 100 TPI threads with a 1/16 inch leadscrew. I don’t foresee much need for those around here, but you can never have too many change gears …
Mini-lathes sold everywhere except the USA have hard-metric leadscrews with a 1.5 mm pitch, so they can cut metric threads without any trouble at all. USA-ian mini-lathes have hard inch leadscrews with a 1/16 inch pitch and require gymnastics to cut decent metric threads.
For inch threads:
For metric threads, it’s upside-down and converted:
Including a 21 tooth change gear at A or C in the train will get closer to metric threads, but the stud holding the B-C gears on my lathe imposes a minimum B gear size. Here’s a 1 mm thread with a 21-30-45-50 train:
Oops.
You can interchange the AB and CD gear pairs, but the CD pair also has a minimum spacing:
If it worked, the 1 mm thread would be off by -125 ppm, which is surely close enough.
Without the 21 tooth gear, the LittleMachineShop gear calculator produces a 45-55-50-65 train that fits much better, albeit with +875 ppm thread error:
Dropping the 21 tooth gear into the LMS calculator produces a (barely) workable 21-50-60-40 train with -125 ppm error:
The very very snug fit of the screw (omitted here) on the 21 tooth gear nearly hits the 60 tooth gear. Your mileage may vary, of course.
Various sources suggest the gears are module 1, 20° pressure angle, 12 mm bore, and 8 mm thick (mine measure a scant 7.75 mm), with a 3 × 1.4 mm keyway (the key itself is 3 mm square, but it’s half-buried in the shaft).
Update: Welcome Adafruit! The reshaped elbow shown here eventually got threaded adapters for the tubing and an awful paint job.
The floor lamp with the invisible / non-tactile controls moved to a different chair, where it didn’t have quite enough reach and too much height. Knowing what was about to happen, I spliced a JST-SM connector into the wire inside the tube:
After trimming off all the extraneous bits, the larger half of the connector (male pins) fits through the tubing and the smaller half (female sockets) barely fits through the bottom bushings.
It turns out half-inch copper pipe fittings (ID = 15.9 mm) almost exactly fit the tubing (OD = 15.7 mm):
A quick test showed the 45° (actually, it’s 135°, but we’re deep into plumbing nomenclature) positioned the lamp head too high and with too much reach:
So shorten the tube attached to the head and deburr the cut:
The 45° fitting is too high and a 90° fitting is obviously too low, so cut a 20° slice out of a 90° fitting:
Cut a snippet of brass tubing to fit, bash to fit, file to hide, buff everything to a high shine, silver-solder it in place, and buff everything again:
The 5/8 inch aluminum rods serve to stiffen the fitting, smooth out the torch heating, and generally keep things under control.
Wrap the obligatory Kapton tape around the butt ends of the tubes to fill the fitting’s oversize hole, put everything together, and it’s just about perfect:
I immobilized the fitting with black Gorilla tape, but it really needs something a bit more permanent. One of these days, maybe, a pair of setscrews will happen.
The additional reach required a little more counterweight on the far side for security, so I added the broken stub of a truck leaf spring. It should be secured firmly to the base plate, but no tool I own can put a dent in those three pounds of spring steel. Maybe it’ll merit a fancy enclosure wrapped around the base?
After cleaning the fog lamp lens enough to be encouraging, I made an LED “bulb” from four WS2812 RGB pixels:
The small threaded hole has an M3 setscrew to let the brass post slide up & down to adjust the LED position inside the fog lamp’s reflector.
Despite my poor experience with the PCB-based WS2812 LEDs, the strip-mounted ones have been ticking along in the hard drive platter lamp basically forever, at least after I tamped down the heat problem.
The brass hex rod has plenty of thermal conductivity, particularly clamped into an aluminum disk connected more-or-less well to the fog lamp’s base.
The two short wires linking the two LED strips (the purple wire is data into the first LED) hold them in place around the hex, despite their desire to straighten out, pull free of their adhesive, and fall off.
The general idea was to put the LEDs at about the same level as the halogen bulb filament, thereby spreading enough light to fill the reflector housing:
I drilled a hole through the hex as a cable “conduit”, turned the end into a nice rod, then machined a stub of aluminum to fit:
A pair of slots milled along the sides of the aluminum disk fit the housing’s locating features:
Nissan used an elaborate spring latch to clamp the halogen bulb’s sheet-metal base in place, but its 50 mil wire didn’t have nearly enough give for my chunky aluminum disk. My version of a spring latch came from a length of 24 mil music wire, which definitely beats the epoxy I was planning to use.
Heat transfer seems to be a non-issue, as the LEDs get barely warm to the touch. Until they drop dead, I’ll assume it’s all good in there.
Two screws hold the lens in place, but the collision seems to have stripped their grip on the plastic and they didn’t un-screw:
Jamming a utility knife blade under the screw head and prying upward while turning the screwdriver persuaded them out of their sockets, after which the lens popped out of its form-fitted silicone gasket with surprisingly little effort:
The lamp spent a week or so beside the road, out in the weather, and shipped a few drops of rainwater through the rectangular hole under the spring latch anchor. Some delicate cotton-swab action removed most of the grime without too much damage, but the reflective film on those corrugations won’t ever be the same again.
Now it’s just a simple matter of software …
While doodling a drag knife holder for the Sherline, I figured a 3/8 inch shaft would hold all the parts and fit neatly into a standard Sherline tool holder, which it did:
Having recently upcycled a 3/8 inch shaft from the Thing-O-Matic into a pen holder for the CNC 3018-XL, I cut off another section with an abrasive wheel, then tried to face it off:
Although the mini-lathe’s carbide insert gnawed at the shaft’s case-hardened shell, it obviously wasn’t making much progress against that step.
Back to the abrasive cutoff saw:
Which looked better, although it still wasn’t quite perpendicular to the shaft axis.
Back to the lathe:
Well, it’s better, but it sure ain’t pretty.
Put gently, the mini-lathe’s lack of rigidity doesn’t help in the least. The compound was a-reelin’ and a-rockin’ on every revolution and eventually turned a slight tilt into a distinct radial step.
Memo to Self: Dammit, use a brass rod!
