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: Repairs

If it used to work, it can work again

Bafang BBS02: Speed Sensor Nut Reshaping
A Bafang BBS02 (for a friend’s upright bike) arrived with a deformed speed sensor nut:

Bafang BBS02 – Deformed speed sensor nut – end view

It traveled halfway around the planet while trapped underneath the motor and, if it rode in the top layer or two of containers, the combination of pressure and heat would be irresistible.

The plastic was stiff and I couldn’t force the nut over the connector using as much force as seemed reasonable:

Bafang BBS02 – Deformed speed sensor nut – test assembly

On the upside, the nut just compresses the silicone washer between the connector and the sensor to make a waterproof joint, so it need not have perfect threads or a uniform shape. Once the nut is in place, it will likely never be removed and should never bother anyone else.

Being unwilling to apply a hot-air gun near the cable, I decided to try slowly cold-forming the nut inside a mold:

Sensor Nut mold – solid model

The gap isn’t a kerf: the two halves meet to form a cylindrical pocket. The smaller holes fit a pair of brass tubes keeping the halves lined up while I arrange things:

Bafang BBS02 – Deformed speed sensor nut – clamp detail

A pair of swivel-pad clamps apply the pressure:

Bafang BBS02 – Deformed speed sensor nut – compression clamp

A few days of squashing made it round-er, whereupon I applied the clamp directly against the remaining high point with the other side cradled in the mold. It still doesn’t slide over the connector body, but I’m not in a rush.

Bafang tech support generously sent a speed sensor extension cable from which I can extract a good nut, which will require cutting and splicing the cable from the motor.

I’m still hoping gentle suasion will prevail.

The OpenSCAD source code tucks into the overall file producing various useful bits:
```
// Mold to reshape speed sensor nut

SensorNut = [0,14.4,13.0];
SensorMold = [SensorNut[OD] + 2*WallThick,SensorNut[OD] + 2*WallThick,SensorNut[LENGTH] + WallThick];
MoldSides = 20;
RodOD = 1.6;

module NutMoldBlock() {

    difference() {

        translate([0,0,SensorMold.z/2])
            cube(SensorMold,center=true);

        translate([0,0,WallThick])
            rotate(180/MoldSides)
                PolyCyl(SensorNut[OD],2*SensorNut[LENGTH],MoldSides);
        translate([0,0,-Protrusion])
            rotate(180/8)
                PolyCyl(SpeedOD,2*SensorMold.z,8);

        for (i=[-1,1])
            translate([i*(SensorMold.x/2 - WallThick/2),SensorMold.y,SensorMold.z/2])
                rotate([90,0,0])
                    PolyCyl(RodOD,2*SensorMold.y,6);
    }
}

module NutMold() {
    gap = 1.0;

    for (j=[-1,1])
        translate([0,j*gap,0])
            intersection() {
                translate([0,j*SensorMold.y,0])
                    cube(2*SensorMold,center=true);
                NutMoldBlock();
            }
}

… snippage …

if (Layout == "NutMold")
    NutMold();
```
I haven’t worked on a safet y bike in years!
2021-06-10

Dripworks Mainline Pipe Clamp

This is laid in against a need I hope never occurs:

It’s intended to clamp around one of the Dripworks mainline pipes carrying water from the pressure regulator to the driplines in the raised beds, should an errant shovel or fork find the pipe.

It descends from a long line of soaker hose clamps, with a 25 mm ID allowing for a silicone tape wrap as a water barrier.

The solid model has no surprises:

Dripworks Mainline Clamp - build view — Dripworks Mainline Clamp – build view

The OpenSCAD source code as a GitHub Gist:

	// Dripworks 3/4 inch mainline clamp
	// Ed Nisley KE4ZNU 2021-06

	Layout = "Build"; // [Hose,Block,Show,Build]

	HoseOD = 25.0;

	TestFit = false; // true to build test fit slice from center

	//- Extrusion parameters must match reality!

	/* [Hidden] */

	ThreadThick = 0.25;
	ThreadWidth = 0.40;

	HoleWindage = 0.2;

	Protrusion = 0.1; // make holes end cleanly

	inch = 25.4;

	function IntegerMultiple(Size,Unit) = Unit * ceil(Size / Unit);

	ID = 0;
	OD = 1;
	LENGTH = 2;


	//———-
	// Dimensions
	// Hose lies along X axis

	Hose = [200,HoseOD,HoseOD]; // X = longer than anything else

	NumScrews = 2; // screws along each side of cable

	WallThick = 3.0; // Thinnest printed wall

	PlateThick = 1.5; // Stiffening plate thickness

	// 8-32 stainless screws
	Screw = [4.1,8.0,50.0]; // OD = head LENGTH = thread length
	Washer = [4.4,9.5,1.0];
	Nut = [4.1,9.7,3.3];

	Block = [30.0,Hose.y + 2Washer[OD],HoseOD + 2WallThick]; // overall splice block size
	echo(str("Block: ",Block));

	ScrewMinLength = Block.z + 2PlateThick + 2Washer.z + Nut.z; // minimum screw length
	echo(str("Screw min length: ",ScrewMinLength));

	Kerf = 1.0; // cut through middle to apply compression

	CornerRadius = Washer[OD]/2;

	ScrewOC = [(Block.x – 2*CornerRadius) / (NumScrews – 1),
	Block.y – 2*CornerRadius,
	2*Block.z // ensure complete holes
	];

	echo(str("Screw OC: x=",ScrewOC.x," y=",ScrewOC.y));

	//———————-
	// Useful routines

	module PolyCyl(Dia,Height,ForceSides=0) { // based on nophead's polyholes
	Sides = (ForceSides != 0) ? ForceSides : (ceil(Dia) + 2);
	FixDia = Dia / cos(180/Sides);
	cylinder(d=(FixDia + HoleWindage),h=Height,$fn=Sides);
	}

	// Hose shape
	// This includes magic numbers measured from reality

	module HoseProfile() {

	NumSides = 12*4;

	rotate([0,-90,0])
	translate([0,0,-Hose.x/2])
	resize([Hose.z,Hose.y,0])
	cylinder(d=Hose.z,h=Hose.x,$fn=NumSides);
	}

	// Outside shape of splice Block
	// Z centered on hose rim circles, not overall thickness through center ridge

	module SpliceBlock() {
	difference() {
	hull()
	for (i=[-1,1], j=[-1,1]) // rounded block
	translate([i(Block.x/2 – CornerRadius),j(Block.y/2 – CornerRadius),-Block.z/2])
	cylinder(r=CornerRadius,h=Block.z,$fn=4*8);
	for (i = [0:NumScrews – 1], j=[-1,1]) // screw holes
	translate([-(Block.x/2 – CornerRadius) + i*ScrewOC.x,
	j*ScrewOC.y/2,
	-(Block.z/2 + Protrusion)])
	PolyCyl(Screw[ID],Block.z + 2*Protrusion,6);
	cube([2Block.x,2Block.y,Kerf],center=true); // slice through center
	}
	}

	// Splice block less hose

	module ShapedBlock() {
	difference() {
	SpliceBlock();
	HoseProfile();
	}
	}




	//———-
	// Build them

	if (Layout == "Hose")
	HoseProfile();

	if (Layout == "Block")
	SpliceBlock();

	if (Layout == "Show") {
	difference() {
	SpliceBlock();
	HoseProfile();
	}
	color("Green",0.25)
	HoseProfile();
	}

	if (Layout == "Build") {
	SliceOffset = TestFit && !(NumScrews % 2) ? ScrewOC.x/2 : 0;
	intersection() {
	translate([SliceOffset,0,Block.z/4])
	if (TestFit)
	cube([ScrewOC.x/2,4*Block.y,Block.z/2],center=true);
	else
	cube([4Block.x,4Block.y,Block.z/2],center=true);
	union() {
	translate([0,0.6*Block.y,Block.z/2])
	ShapedBlock();
	translate([0,-0.6*Block.y,Block.z/2])
	rotate([0,180,0])
	ShapedBlock();
	}
	}
	}

view raw Dripworks Mainline Pipe Clamp.scad hosted with ❤ by GitHub

2021-06-09

Microscope Stage Positioner: MakerBeam Rebuild MVP

Over the course of half a decade (!), the 3D printed arm on the XYZ positioner I use with the stereo zoom microscope sagged:

Microscope Stage Positioner – PETG creep angle

It’s about what you’d expect from a plastic beam carrying a big lump of brass and steel:

Microscope Stage Positioner

The near side of that arm (the -Y end) drooped about 5 mm below than the side nearest the Z axis slide, so it was time for an update.

Having some MakerBeam ready to hand, this didn’t take long:

Microscope Stage Positioner – Makerbeam overview

Protip: before dismantling a fitted slide, mark one end so you know how to put it back together. Bonus points for taking a picture:

Microscope Stage Positioner – slide marking

Double bonus points for writing a blog post.

Rather than fight with the existing fine-pitch USA-ian screws, I drilled out their threaded holes:

Microscope Stage Positioner – Y slide drilling

And epoxied 3 mm brass inserts in their place:

Microscope Stage Positioner – Y slide M3 inserts

Those holes match up with a pair of corner cubes normally appearing on the end of the beams:

Microscope Stage Positioner – BHCS mods for Makerbeam

It turns out M3 button head cap screws will slide into the beams if you file the slightest angle on opposite sides of the button, although a small bag of tiny tee nuts should arrive in a while.

Then a variety of brackets spliced everything together:

Microscope Stage Positioner – Makerbeam detail

Although it looks strictly from industrial, it actually wasn’t much better than the plastic edition and, in fact, the beam supporting the XY slides sagged about the same 5 mm. The plastic upright post also contributed a bit of wobble.

It turns out that the extruded aluminum beams have plenty of longitudinal and torsional stiffness, but all those flat steel fittings don’t.

There’s a way to work with the beam strengths, rather than against them, but that’s a story for another day …

2021-06-07
Bicycle Helmet Mirror: Stalk Repair

The mirror on my bike helmet snagged on a mesh fence while walking the bike to Mary’s garden:

Helmet Mirror – bent stalk

One of the good things about building your own stuff is that you have all the parts when something breaks:

Helmet Mirror – damaged parts

The decorative M2 screw and insert pulled out of the ball. The rim of the nail set punch (intruding from the top) just barely caught the edge of the stub inside the ball, so a few taps could extract it. A Dremel cutoff wheel peeled the crumpled end off the stalk.

Reassembly proceeded without incident:

Helmet Mirror – installed

The bizarrely blurred mirror over on the left comes from the Pixel phone camera app deciding this was a Portrait, applying a background blur, and running into trouble with those hard edges in the foreground. The camera app has a distinct Portrait mode that, perhaps, I inadvertently engaged while fumbling around.

2021-06-04
Aceco FC1002 Battery Re-replacement

My old Aceco FC1002 frequency meter stopped working without being plugged into the charger. It runs from a quartet of NiMH cells taped into a tray I made seven years ago:

Aceco FC1002 – hacked battery

One of the cells was completely dead and the other three will blink LEDs for the rest of their lives.

The Eneloops have trickled down from the DSC-H5 and still seem perfectly fine for ordinary use.

The faceplate bears the scars of its cracked acrylic (?) coating, so I pushed it out, traced the outline on a flat piece of polypropylene clamshell packaging, cut it out, and stuck it in place with tapeless sticky:

Aceco FC1002 – polypropylene faceplate

That removes the branding, but IMO improves the appearance.

It should continue working for another half decade or so!

2021-06-03
Dripworks Micro-Flow Valves: QC FAIL
We recently installed a Dripworks drip irrigation system for Mary’s garden and, of course, pre-assembled the emitter / dripline tubing, fittings, and supply / filter / plumbing for each of the beds in the Basement Shop. A few days after burying the main lines, plumbing the filter + pressure regulator, and plugging in half a dozen bed assemblies, Mary noticed some emitter tubes weren’t delivering any water and other beds seemed too dry.

N.B.: We bought everything directly from Dripworks. This is not counterfeit crap from a sketchy Amazon seller.

I cut the dripline just downstream of the Micro-Flow valve on a completely dry bed, whereupon no water emerged. Cutting the supply tube just upstream of the valve produced a jet squirting halfway along the bed. I tried and failed to blow air through the valve: it was completely blocked despite being in the “open” position. I installed another valve and the emitter tube started working properly.

I sat down at the kitchen table with a bag of unused valves and peered through them (the pix are through the microscope):

Dripworks valve – mostly open lumen

That’s one of the better-looking valves, with only a little mold flash in the lumen.

Partially occluded lumens were more typical:

Dripworks valve – partially occluded lumen

Quite a few were almost completely obstructed:

Dripworks valve – mostly occluded lumen

For lack of better instrumentation, I blew through the valves and sorted them by effort:

Dripworks valve – sorted by blockage

Two of the valves in the group on the left are completely blocked, with the others mostly blocked.

The middle group has enough mold flash to produce noticeable resistance to the air flow. I think water would have more trouble getting through, but the emitters would at least look like they’re delivering water.

The group on the right has mostly unblocked valves, with visible mold flash but little restriction.

I have no way to measure the actual water flow, so it’s entirely possible the QC spec allows considerable blockage while still delivering enough water to the emitters. More likely, the spec assumes a clear lumen and the mold flash is a total QC faceplant; it’s obviously not a controlled quantity.

Well, I can fix that:

Dripworks valve – drilling

That’s a 2.3 mm drill going straight through the valve body. I drilled the valves from both ends and blew out the swarf:

Dripworks valve – drill swarf

That produced twenty valves with clear lumens. Of course, the drill leaves a slightly rough interior surface, but it’s now much easier to blow air through them.

We hadn’t installed the driplines in two beds with three emitter tubes per bed. I cut out those six unused valves and sorted them by resistance:

Dripworks valve – six samples

Both of the valves on the left are blocked, the three on the right are mostly OK, and the one in the middle is partially blocked.

With two dozen repaired valves in hand, we returned to the garden, I cut 22 valves out of the installed driplines and replaced them under field conditions. Returning to the Basement Laboratory, I blew the water out (*), sorted them by resistance, and produced a similar distribution, albeit with no pictorial evidence. Although we have no immediate need for the used valves, they’re drilled out and ready for use.

In very round numbers, you should expect:
- A third of Dripworks valves will pass (close to) the expected flow
- A third will have a minor flow restriction
- A quarter will have a severe flow restriction
- One valve in ten will be completely blocked
Plan to drill out all the Micro-Flow valves before you assemble your driplines.

AFAICT, none of the other ¼ inch fittings we used have any interior flash, so it’s only a problem with the valves.

We are, as the saying goes, not amused.

(*) If you will eat a peck of dirt before you die, I’m well on my way.
2021-06-01
Felco C7 Cable Cutter: Spring Repair

The back of the Pliers & Cutters drawer produced an ancient Felco C7 Cable Cutter minus its spring:

Felco C7 cutter – missing spring pin

That’s an M4 screw serving as a size test for the hole where the other pin used to be.

Surprisingly, Felco still exists, still makes the C7 Cable Cutter, and actually sells a replacement spring as part number C7/10. Unfortunately, their online sales apparatus and cart seem broken: I put the spring in the cart, but found no way to pay for it. Worryingly, the usual Terms & Conditions link produced a modal dialog with one word: TEST.

So I got a spring (part number 5/11, available only as a pair in kit 5/91) for a Felco C3 cutter (no, the numbers do not match) from Amazon. Later I found a sketchy seller offering a sketchy C7/10 spring that might fit correctly or could be total trash.

Felco swaged the original spring pins into the handle, a manufacturing technique I certainly cannot duplicate, but an M3 screw will just barely fit inside a 4 mm stud, so I made some measurements:

Felco replacement spring pin – dimension doodle

Fitting action to words:

Felco C7 cutter – replacement spring pin

That started as a 1/4 inch rod of no particular provenance and is reasonably close to the actual dimensions.

The spigot on the screw end is threaded M3 and is just barely shorter than the thickness of the handle, so the button-head screw can pull it snug:

Felco C7 cutter – button screw

And then the spring just snapped into place:

Felco C7 cutter – spring installed

It it obviously grossly excessively too long, but that really doesn’t matter for the number of power-on hours it’s likely to see during my administration. In truth, it feels pretty good in the hand after releasing the latch and having it expand smoothly.

If I ever run across a C7/10 spring, it’ll be an easy swap.

2021-05-25