Tag: Improvements

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

Alpha Geek Clock: Battery Refresh

A pair of D cells can power an obsolete / out of production C-Max CMMR-60 WWVB receiver for about five years and, having the plastic pieces for a blinkie at hand, junking the faded case in favor of a test lashup seemed appropriate:

C-Max CMMR-60 WWVB receiver – AA alkaline test setup

Given the fragility of that ferrite bar, I should conjure a wide D-cell base, a bar holder to cover the ends, and a PCB mount of some sort.

The receiver data pin drives the red LED of an RGB piranha through a 2.2 kΩ SMD resistor, so it’s visible in a dim room. Given that the thing flickers constantly during WWVB’s poor-reception daylight hours, reducing the LED current counts for almost everything.

The antenna has a cap under that heatshrink tubing, which called for a resonance check:

C-Max CMMR-60 WWVB receiver – antenna peaking – driver coil

The blue dingus is an RF sniffer driven three orders of magnitude below its frequency spec:

C-Max CMMR-60 WWVB receiver – antenna peaking – function generator

The antenna response peaks where you’d expect:

C-Max CMMR-60 WWVB receiver – antenna peaking – scope

Given the broad peak and typical tolerances, it’s spot on.

2021-11-15
CNC 3018 Tool Clamp Rehabilitation

The CNC 3018 Z-axis stage has a plastic clamp holding the spindle motor, so I just duplicated the motor diameter in the mounts for my diamond drag bit, cheap pen, and fancy pen holders. For obvious reasons, I tend to err on the small side for anything intended to fit into anything else, which led to each of the holders sporting a small strip of tape to soak up the difference.

While poking around the 3018, I once again noticed the clamp’s crappy fit around the holder:

CNC3018 tool clamp – top

The inside should be circular, but it’s definitely not:

CNC3018 tool clamp – top detail

The end of the 30 mm M3 SHCS bottoms out before the clamp closes, although I’ve managed to crank the screw tight enough to put enough of a dent in there to snug the clamp:

CNC3018 tool clamp – side

Some awkward scraping and filing eroded enough of the plastic to let a 25 mm SHCS close the clamp firmly around the holder:

CNC3018 tool clamp – revised

The tool holders now slide in easily with the screw released and fit firmly with the screw tightened a reasonable amount, minus the tape snippets shimming the difference.

If I had the courage of my convictions, I’d take it all apart, bore the clamp out to a circular profile, realign the clamp screw passage to suit, then rebuild all those tool holders for the new diameter; it now works well enough to tamp that project down.

2021-11-09

Clearing the Noto Font Clutter: Again

Installing Atkinson Hyperlegible reminded me to clear out the Noto font clutter in this (relatively nerecentw) Manjaro installation. Of course fonts now appear in slightly different locations with slightly different names, so this remains just a serving suggestion:

cd /usr/share/fonts/noto
sudo chmod a-w NotoSans-*
sudo chmod a-w NotoSansMono*
sudo chmod a-w NotoSansDisplay*
sudo chmod a-w NotoSansMath*
sudo chmod a-w NotoSansSymbol*
sudo chmod a-w NotoSerif-*
sudo chmod a-w NotoSerifDisplay*
sudo chmod a-w NotoMusic*
sudo chmod a-w NotoMath*
sudo find . -perm /u=w -name \*ttf -delete

Get rid of some other clutter:

cd ../TTF
sudo chmod a-w DejaVu*
sudo chmod a-w Inconsolata-*
sudo find . -perm /u=w -name \*ttf -delete
cd ../droid
sudo chmod a-w DroidSans-Bold.ttf 
sudo chmod a-w DroidSans.ttf 
sudo chmod a-w DroidSansFallback*
sudo chmod a-w DroidSansMono.ttf 
sudo chmod a-w DroidSerif-*
cd ../adobe-source-han-sans
sudo rm *otf

For unknown reasons, we now have two font cache updaters:

sudo fc-cache -v -f
sudo fc-cache-32 -v -f

Now font selection in, say, LibreOffice doesn’t involve paging through a myriad fonts in languages I cannot recognize, let alone read. Admittedly, Inconsolata does have more variations than I’ll ever use.

2021-10-21

Improved Mini-lathe Disk Turning Fixture

Unsurprisingly, the mini-lathe lacks enough stiffness to apply enough force to hold a disk in place while turning its rim:

Tour Easy Rear Running Light – end cap fixture – swirled adhesive

The old South Bend lathe had mojo, but those days are gone.

So drill and tap that fixture for an M3 screw, then stick some coarse sandpaper to it:

Improved disk turning tool

Snug the screw (a Torx T9 from the Small Drawer o’ Random M3 Screws) down on a rough-cut disk:

Improved disk turning tool – in use

Sissy cuts remain the order of the day, but the screw applies plenty of clamping force and doesn’t require the hulking live center.

2021-10-14

Dirt Devil Vacuum Tool Adapters

Being the domain expert for adapters between a new vacuum cleaner and old tools, this made sense (even though it’s not our vacuum):

Dirt Devil Nozzle Bushing - solid model — Dirt Devil Nozzle Bushing – solid model

The notch snaps into a Dirt Devil Power Stick vacuum cleaner and the tapered end fits a variety of old tools for other vacuum cleaners:

Dirt Devil Nozzle Bushing top view - solid model — Dirt Devil Nozzle Bushing top view – solid model

Having some experience breaking thin-walled adapters, these have reinforcement from a PVC tube:

Dirt Devil adapter - parts — Dirt Devil adapter – parts

A smear of epoxy around the interior holds the tube in place:

Dirt Devil adapters - assembled — Dirt Devil adapters – assembled

Building the critical dimensions with a 3D printed part simplified the project, because I could (and did!) tweak the OpenSCAD code to match the tapers to the tools. Turning four of those tubes from a chunk of PVC conduit, however, makes a story for another day.

The OpenSCAD source code as a GitHub Gist:

	// Dirt Devil nozzle adapter
	// Ed Nisley KE4ZNU 2021-10

	// Tool taper shift
	Finesse = -0.1; // [-0.5:0.1:0.5]

	// PVC pipe liner
	PipeOD = 28.5;

	/* [Hidden] */

	//- Extrusion parameters

	ThreadThick = 0.25;
	ThreadWidth = 0.40;

	HoleWindage = 0.2;

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

	Protrusion = 0.1; // make holes end cleanly

	//———————-
	// Dimensions

	TAPER_MIN = 0;
	TAPER_MAX = 1;
	TAPER_LENGTH = 2;

	Socket = [36.0,37.0,40.0];

	LockringDia = 33.5;
	LockringWidth = 4.5;
	LockringOffset = 2.5;

	Tool = [Finesse,Finesse,0] + [30.0,31.1,30.0];

	AdapterOAL = Socket[TAPER_LENGTH] + Tool[TAPER_LENGTH];

	NumSides = 36;
	$fn = NumSides;

	//———————-
	// 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(r=(FixDia + HoleWindage)/2,h=Height,$fn=Sides);

	}


	//——————-
	// Define it!

	module Adapter() {

	difference() {
	union() {
	difference() {
	cylinder(d1=Socket[TAPER_MIN],d2=Socket[TAPER_MAX],h=Socket[TAPER_LENGTH]);
	translate([0,0,LockringOffset])
	cylinder(d=2*Socket[TAPER_MAX],h=LockringWidth);
	}

	cylinder(d=LockringDia,h=Socket[TAPER_LENGTH]);
	translate([0,0,LockringOffset + 0.75*LockringWidth])
	cylinder(d1=LockringDia,d2=Socket[TAPER_MIN],h=0.25*LockringWidth);

	translate([0,0,Socket[TAPER_LENGTH]])
	cylinder(d1=Tool[TAPER_MAX],d2=Tool[TAPER_MIN],h=Tool[TAPER_LENGTH]);
	}
	translate([0,0,-Protrusion])
	PolyCyl(PipeOD,AdapterOAL + 2*Protrusion,NumSides);
	}

	}

	//———————-
	// Build it!

	Adapter();

view raw Dirt Devil Nozzle Bushing.scad hosted with ❤ by GitHub

The taper in the code almost certainly won’t fit whatever tool you have: measure thrice, print twice, and maybe fit once …

2021-10-06

Micro-Mark Bandsaw: Acetal Upper Blade Guide

There being nothing like a good new problem to take one’s mind off all one’s old problems:

Micro-Mark Bandsaw – acetal upper blade guide installed

It’s basically the same as the lower blade guide, except coming from a stick of 5/8 inch acetal. A scant 6 mm stem goes into the vertical square rod, with a flat matching the setscrew coming up from the bottom to hold it in proper alignment.

I came within a heartbeat of cutting the slot parallel to the flat.

It worked OK while cutting a chunk of stout aluminum tube: so far, so good!

The impressive chunk of hardware is the OEM blade guide, with the brass tube for coolant flow all over the bearings. It’s mostly intended for use with the diamond blade, so I’ll swap it back in when I finally get around to cutting some slate for base plates.

2021-09-29
Tour Easy Rear Running Light: First Light!
The rear running light definitely has an industrial look:

Tour Easy Rear Running Light – installed

The front of the light has plenty of clearance from the seat mesh:

Tour Easy Rear Running Light – installed side view

Out on the road, the 1 W LED appears about as bright as automotive running lights:

Tour Easy Rear Running Light – tunnel

The blink pattern makes it perfectly visible in sunlight, although I’d prefer somewhat larger optics:

Tour Easy Rear Running Light – sunlight

In shaded conditions, it’s downright conspicuous:

Tour Easy Rear Running Light – shade

At any reasonable distance, the 10° beam covers much of the road behind the bike:

Tour Easy Rear Running Light – distant

You may not know what the occulting red light represents, but something ahead is worthy of your attention.

The Arduino source code producing the two dits:
```
// Tour Easy Running Light
// Ed Nisley - KE4ZNU
// September 2021

#include <morse.h>

#define PIN_OUTPUT  13

// second param: true = active low output
LEDMorseSender Morser(PIN_OUTPUT,true,(float)10.0);

void setup()
{
    Morser.setup();

    Morser.setMessage(String("qst de ke4znu "));
    Morser.sendBlocking();

//    Morser.setWPM((float)3.0);
    Morser.setSpeed(75);
    Morser.setMessage(String("i   "));
}

void loop()
{
    if (!Morser.continueSending())
        Morser.startSending();

}
```
Looks good to me, anyhow.
2021-09-27