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.

Tag: Improvements

Making the world a better place, one piece at a time

Kenmore 158: Frame Pivot Pins
The entire Kenmore Model 158 sewing machine tilts on a pair of pivots extending from the rear of the base, just below the top surface. Mary’s slightly more recent machine has all-steel pivots:

Kenmore 158 – steel pivot pin

The older crash test dummy machine has two-part pivots, with a plastic housing molded around a steel pin:

Kenmore 158 – plastic pivot pins

Obviously, plastic was the wrong material for the cross pins that rest in the base, leading to the all-steel redesign. Sears no longer stocks replacement parts for those pins, sooo …

Both machines have a large plastic base that’s gradually disintegrating. The plan is to embed the machine frames in countertops, with those cross pins resting on plastic plugs set flush with the surface.

The frame sockets aren’t quite 1/4 inch in diameter; the rest of the hardware uses hard metric sizes, so they’re most likely 6 mm. A 15/64 inch (5.95 mm) drill bit fits snugly and a length of 0.228 inch (5.79 mm) drill rod fits loosely. The round pins are 18 mm long from the shoulder.

The square section is 8.5 mm wide, 9.5 mm tall, and 16 mm long. I have no idea what that mysterious tab on the end is supposed to do.

The cross pins are 5 mm diameter, a scant 15 mm end-to-end, stand 3 mm proud of the central block, and are centered 11 mm out from the edge of the block. I’d make them longer, to distribute the machine’s weight over more of the plugs in the countertop when it’s tilted back.

I can’t duplicate the newer forged steel pins and, for sure, they’re not good candidates for 3D printing. Perhaps:
- Saw off 16 mm of 3/8 inch (9.5 mm) square stock
- Blind drill 16/64 inch for the 0.228 main pin
- Cross drill #12 for a 3/16 inch pin
- Epoxy everything together
- File off the sharp edges
For the moment, the crash test dummy sits happily on the three legs that the designers thoughtfully cast into its frame.
2014-07-04
Silicone Caulk + Desiccant = Win!

After doing the second batch of quilting pin caps, I dropped the newly opened silicone caulk tube into a jar with some desiccant, which worked wonderfully well. Unlike the usual situation where the caulk under the cap hardens into a plug after a few weeks, the tube emerged in perfect condition. In fact, even the caulk in the middle of the conical nozzle was in good shape, with just a small cured plug on either end; it had been sitting inside a cloth wrap with no sealing at all.

Here’s what it looked like after finishing the last of the most recent caps:

Silicone caulk tube with silica gel

The indicator card says the humidity remains under 10%, low enough to keep the caulk happy and uncured. Well worth the nuisance of having a big jar on the top shelf instead of a little tube next to the epoxy.

Although I thought the desiccant was silica gel, it’s most likely one of the clay or calcium desiccants.

2014-06-28
More Quilting Pin Caps

Pinning the top of Mary’s latest quilt used more than 1600 pins: three boxes of specialized quilting safety pins, plus straight quilting pins tucked into all the 3D printed / silicone filled caps. Less than a quarter of the quilt top fits on the table:

Quilt top with pins

Although Mary doesn’t need them right now, I made another batch of 100 caps for her next project:

Quilting pin caps – 4 x 25 – on platform

I tweaked the OpenSCAD source to build a 10×10 array:

Quilting Pin Cap – 10×10 array

But it turns out that a 5×5 array of caps, duplicated four times, works out better:

Quilting Pin Cap – 5×5 array

Slic3r takes far longer to process the larger array than to make four copies of the smaller array.

Half an hour later, they’re ready for silicone fill. In retrospect, natural PLA wasn’t a good choice for this job: there’s no way (for me) to take a picture of translucent silicone in crystalline PLA atop waxed paper on a white cutting board under fluorescent light…

On the upside, however, you can see exactly how far the pin goes into the cap:

Quilting pin in translucent cap

2014-06-18
Subaru Forester: Human Factors

Our Forester has three knobs that control air direction / speed / temperature. Knobs are much better than buttons, because you can adjust them without looking. At least, that’s the ideal situation.

Here’s the setting for airflow to the footwell:

Subaru Forester – Airflow knob – feet – daylight

Here’s what it looks like with airflow to the cabin:

Subaru Forester – Airflow knob – face – daylight

The knob has no tactile position indicator. That greenish rectangle, located in one of seven symmetric dimples that camouflage its position, is barely visible in normal light, invisible with sunglasses, and not apparent to the touch.

Well, if conspicuous is what you want, I can fix that:

Subaru Forester – knobs – highlighted

Fluorescent tape will fade quickly, but it’ll last until something better comes along. Perhaps a small pointer epoxied onto the knurled surface, extending around to the indicator?

2014-06-12
Pink Panther Woman: Extruder Contamination

The Pink Panther Woman is my reference standard (*) for smooth perimeters and zitless filament retraction:

Pink Panther Woman – left

That’s vastly improved since the Thing-O-Matic’s last attempt:

PPW – outie zits

Done in natural PLA, as it seems the previous version also walked off:

Pink Panther Woman – natural PLA

The attentive reader will note an odd red stripe on the left leg of the black PLA version. Here’s a closer look:

Pink Panther Woman – black with red contamination – detailPink Panther Woman – black with red contamination – detail

I had recently changed from red to black PLA and, as usual, purged the extruder with a few hundred millimeters of black filament, until it emerged pure black. Alas, I forgot to wipe the outside of the nozzle:

Pink Panther Woman – black – contaminated nozzle

That red blob produced the red tab on the neck, as you can see if you look carefully at the first picture.

There are very few visible imperfections in either object: the state of DIY 3D printing is pretty good.

(*) Does anyone know of similar male figures suitable for this purpose? That torso seems to be about the extent of Thingiverse’s offerings.

2014-06-06
Makergear M2 Build Platform: Moah Powah!

A surplus Mean Well PSP-600-48 48 V 12.5 A power supply just arrived, I dialed it back to 40 V, and swapped it with the 36 V brick I’d been using to drive the M2’s improved heated build platform.

The improved platform was designed for a 30 V supply that would run it at about 150 W, which took slightly less than forever to reach operating temperature.

With the 36 V supply set to 38.6 V, the platform drew 6.2 A at room temperature, which worked out to 6.2 Ω and 240 W. It was a tad pokey getting up to temperature

At 40 V, the platform starts at 6.3 A / 6.3 Ω / 250 W from a bit over room temperature and drops to 5.8 A / 6.9 Ω / 232 W at 70 °C.

At about 250 W, the platform takes about three times longer to reach operating temperature than the extruder, but it doesn’t require calling down to the engine room for more coal before maneuvering. I must run some numbers on it, now that I have a power supply with a useful range.

There’s obviously an upper limit to the peak power the PCB traces under the glass can handle, but it runs at the same average power (to produce the same average temperature) and, at least so far, hasn’t shown any signs of distress. The few additional watts at 40 V won’t make any difference.

Note that you must use an external DC-to-DC solid state relay, because the Rambo controller board can’t handle anything over 24 VDC and high current loads tend to melt its Phoenix-style connectors. When you add the SSR, replace the HBP connectors with Anderson Powerpoles, use fat wires, and be done with it.

M2 HBP SSR Wiring

The M2’s Marlin firmware uses bang-bang control and tends to overshoot the setpoint; I’m not sure a few degrees makes all that much difference, particularly because it’s not measuring the temperature at the top of the glass plate.

2014-06-05
Makergear M2: Platform Leveling with Cart Coins

It turns out that an array of Cart Coins and Cart Releasers make a fine thickness test pattern and become useful tchotchkes when you’re done:

Cart Coins – printing

They’re a bit easier to see in the digital realm:

Cart Coins – platform layout – layer 1

The trick is that they’re both eight layers thick at 0.20 mm/layer. With the platform aligned exactly right, all the objects should measure exactly 1.60 mm thick.

The blue numbers give the thickness measured across the stem, just above the hole, on each object:

Platform Leveling – Initial

The green numbers are the skirt thickness: 22 = 0.22 mm.

The platform has a tilt of 0.20 mm from +Y to -Y and is just about perfect from -X to +X.

The M3x0.5 adjusting screws under the (improved) platform, seen from the front (-Y) end of the platform:

M2 – Improved HBP – bottom view

The silicone plugs inside the springs are slightly compressed, so the springs are only decorative. The platform is rigidly mounted on the plugs, with only very slight compliance, and I haven’t leveled the platform in a few months.

Tightening the “north” adjusting screw by 1/6 turn lowered the +Y end of the plate by about 0.05 mm and tilted the +X side slightly higher:

Platform Leveling – Adjustment 1

The skirt thicknesses are now in blue, too.

Tightening the “north” screw an additional 1/6 turn and tightening the “east” screw 1/6 turn produced an almost perfect result:

Platform Leveling – Adjustment 2

The thicknesses don’t vary quite randomly, but I think further adjustments won’t make much difference: the total range is only 0.12 mm = 1.53 to 1.65 mm. That’s pretty close to the limit of my measurement ability on the plastic pieces.

Notice that the skirt thread, which should be exactly 0.2 mm thick all around, really isn’t. I’m going to see whether a two-layer-thick skirt measures a more consistent 0.40 mm.

2014-05-30