Category: Machine Shop

Mechanical widgetry

Vacuum Tube LEDs: Ersatz Tube Sockets

Even vacuum tubes destined to be decorations need sockets:

Vacuum Tube Bases - solid models — Vacuum Tube Bases – solid models

They’re entirely plastic, of course, but they match the dimensions of “real” tube sockets pretty closely. The bosses around the pins have hard-inch dimensions, so you (well, I) can unleash Genuine Greenlee Radio Chassis Punches on sheet metal.

All the key dimensions come from a table, so you can build whatever sockets you need. These four seem to cover the most common relics of the Hollow State Empire:

T_NAME = 0;                                             // common name
T_NUMPINS = 1;                                          // total, with no allowance for keying
T_PINBCD = 2;                                           // tube pin circle diameter
T_PINOD = 3;                                            //  ... diameter
T_PINLEN = 4;                                           //  ... length (overestimate)
T_HOLEOD = 5;                                           // nominal panel hole from various sources
T_PUNCHOD = 6;                                          // panel hole optimized for inch-size Greenlee punches
T_TUBEOD = 7;                                           // envelope or base diameter
T_PIPEOD = 8;                                           // light pipe from LED to tube base
T_SCREWOC = 9;                                          // mounting screw holes

//    Name      pins     BCD   dia  length   hole            punch       env  pipe screw
TubeData = [
    ["Mini7",     8,    9.53, 1.016,   7.0,  16.0,        11/16 * inch,  18.0,  5.0, 22.5],
    ["Octal",     8,   17.45, 2.36,   10.0,  36.2,    (8 + 1)/8 * inch,  32.0, 11.5, 39.0],
    ["Noval",    10,   11.89, 1.1016,  7.0,  22.0,          7/8 * inch,  21.0,  5.0, 28.0],
    ["Duodecar", 13,   19.10, 1.05,    9.0,  32.0,         1.25 * inch,  38.0, 12.5, 39.0],
];

Given that the tubes lack electrical connections, I omitted the base keying: plug them in for best visual effect.

The hole through the middle passes light from a knockoff Neopixel on a 10 mm OD PCB:

Vacuum Tube LEDs - Octal base - top — Vacuum Tube LEDs – Octal base – top

Seen from the bottom, each base traps a pair of 6-32 nuts for chassis mounting and has a Neopixel press-fit in the middle:

Vacuum Tube LEDs - Duodecar base - bottom — Vacuum Tube LEDs – Duodecar base – bottom

Those recesses require support structures:

Vacuum Tube Bases - solid models - support — Vacuum Tube Bases – solid models – support

The Miniature 7-pin socket has the least space for the 10 mm OD Neopixel PCB and shows the thin layer between the bottom of the pin holes and the top of the openings.

Vacuum Tube Base - Mini7 - solid model section — Vacuum Tube Base – Mini7 – solid model section

You see half of the eight holes in the “7 pin” socket, because it has the eighth hole where a standard socket has a gap between pins 1 and 7.

Somewhat to my surprise, punching the support spiders out with a 6-32 stud (grabbed in the drill press) worked perfectly:

Vacuum Tube Base - nut trap overhang - detail — Vacuum Tube Base – nut trap overhang – detail

They look like I intended to build tiny decorations:

Vacuum Tube Base - support structure - detail — Vacuum Tube Base – support structure – detail

The cookies held on tenuously, then released with a loud bang! as I gradually increased the pressure. A PETG support structure in a blind recess wouldn’t pop out nearly so well.

The OpenSCAD source code as a GitHub gist:

	// Vacuum Tube LED Lights
	// Ed Nisley KE4ZNU January 2016

	Layout = "Sockets"; // Cap LampBase USBPort Socket(s)

	Section = true; // cross-section the object

	Support = true;

	//- Extrusion parameters must match reality!

	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);

	//———————-
	// Dimensions
	// https://en.wikipedia.org/wiki/Tube_socket#Summary_of_Base_Details

	T_NAME = 0; // common name
	T_NUMPINS = 1; // total, with no allowance for keying
	T_PINBCD = 2; // tube pin circle diameter
	T_PINOD = 3; // … diameter
	T_PINLEN = 4; // … length (overestimate)
	T_HOLEOD = 5; // nominal panel hole from various sources
	T_PUNCHOD = 6; // panel hole optimized for inch-size Greenlee punches
	T_TUBEOD = 7; // envelope or base diameter
	T_PIPEOD = 8; // light pipe from LED to tube base
	T_SCREWOC = 9; // mounting screw holes

	// Name pins BCD dia length hole punch env pipe screw
	TubeData = [
	["Mini7", 8, 9.53, 1.016, 7.0, 16.0, 11/16 * inch, 18.0, 5.0, 22.5],
	["Octal", 8, 17.45, 2.36, 10.0, 36.2, (8 + 1)/8 * inch, 32.0, 11.5, 39.0],
	["Noval", 10, 11.89, 1.1016, 7.0, 22.0, 7/8 * inch, 21.0, 5.0, 28.0],
	["Duodecar", 13, 19.10, 1.05, 9.0, 32.0, 1.25 * inch, 38.0, 12.5, 39.0],
	];

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

	Pixel = [7.0,10.0,3.0]; // ID = contact patch, OD = PCB dia, LENGTH = overall thickness

	Nut = [3.5,8.0,3.0]; // socket mounting nut recess

	BaseShim = 2*ThreadThick; // between pin holes and pixel top
	SocketFlange = 2.0; // rim around socket below punchout
	PanelThick = 2.0; // socket extension through punchout

	//———————-
	// 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);
	}

	//———————-
	// Tube cap

	CapTube = [4.0,3/16 * inch,10.0]; // brass tube for flying lead to cap LED

	CapSize = [Pixel[ID],(Pixel[OD] + 3.0),(CapTube[OD] + 2*Pixel[LENGTH])];

	CapSides = 6*4;

	module Cap() {

	difference() {
	union() {
	cylinder(d=CapSize[OD],h=(CapSize[LENGTH]),$fn=CapSides); // main cap body
	translate([0,0,CapSize[LENGTH]]) // rounded top
	scale([1.0,1.0,0.65])
	sphere(d=CapSize[OD]/cos(180/CapSides),$fn=CapSides); // cos() fixes slight undersize vs cylinder
	cylinder(d1=(CapSize[OD] + 23ThreadWidth),d2=CapSize[OD],h=1.5*Pixel[LENGTH],$fn=CapSides); // skirt
	}

	translate([0,0,-Protrusion]) // bore for wiring to LED
	PolyCyl(CapSize[ID],(CapSize[LENGTH] + 3*ThreadThick + Protrusion),CapSides);

	translate([0,0,-Protrusion]) // PCB recess with clearance for tube dome
	PolyCyl(Pixel[OD],(1.5*Pixel[LENGTH] + Protrusion),CapSides);

	translate([0,0,(1.5*Pixel[LENGTH] – Protrusion)]) // small step + cone to retain PCB
	cylinder(d1=(Pixel[OD]/cos(180/CapSides)),d2=Pixel[ID],h=(Pixel[LENGTH] + Protrusion),$fn=CapSides);

	translate([0,0,(CapSize[LENGTH] – CapTube[OD]/(2*cos(180/8)))]) // hole for brass tube holding wire loom
	rotate([90,0,0]) rotate(180/8)
	PolyCyl(CapTube[OD],CapSize[OD],8);
	}

	}

	//———————-
	// Aperture for USB-to-serial adapter snout
	// These are all magic numbers, of course

	module USBPort() {

	translate([0,28.0])
	rotate([90,0,0])
	linear_extrude(height=28.0)
	polygon(points=[
	[0,0],
	[8.0,0],
	[8.0,4.0],
	// [4.0,4.0],
	[4.0,6.5],
	[-4.0,6.5],
	// [-4.0,4.0],
	[-8.0,4.0],
	[-8.0,0],
	]);
	}


	//———————-
	// Box for Leviton ceramic lamp base

	module LampBase() {

	Bottom = 3.0;
	Base = [4.0inch,4.5inch,20.0 + Bottom];
	Sides = 12*4;

	Retainer = [3.5,11.0,1.0]; // flat fiber washer holding lamp base screws in place

	StudSides = 8;
	StudOC = 3.5 * inch;
	Stud = [0.107 * inch, // 6-32 mounting screws
	min(15.0,1.5*(Base[ID] – StudOC)/cos(180/StudSides)), // OD = big enough to merge with walls
	(Base[LENGTH] – Retainer[LENGTH])]; // leave room for retainer


	union() {
	difference() {
	rotate(180/Sides)
	cylinder(d=Base[OD],h=Base[LENGTH],$fn=Sides);
	rotate(180/Sides)
	translate([0,0,Bottom])
	cylinder(d=Base[ID],h=Base[LENGTH],$fn=Sides);
	translate([0,-Base[OD]/2,Bottom + 1.2]) // mount on double-sided foam tape
	rotate(0)
	USBPort();
	}
	for (i = [-1,1])
	translate([i*StudOC/2,0,0])
	rotate(180/StudSides)
	difference() {
	# cylinder(d=Stud[OD],h=Stud[LENGTH],$fn=StudSides);
	translate([0,0,Bottom])
	PolyCyl(Stud[ID],(Stud[LENGTH] – (Bottom – Protrusion)),6);
	}
	}
	}

	//———————-
	// Tube Socket

	module Socket(Name = "Mini7") {

	NumSides = 6*4;

	Tube = search([Name],TubeData,1,0)[0];
	echo(str("Building ",TubeData[Tube][0]," socket"));
	echo(str(" Punch: ",TubeData[ID][T_PUNCHOD]," mm = ",TubeData[ID][T_PUNCHOD]/inch," inch"));
	echo(str(" Screws: ",TubeData[ID][T_SCREWOC]," mm =",TubeData[ID][T_SCREWOC]/inch," inch OC"));

	OAH = Pixel[LENGTH] + BaseShim + TubeData[Tube][T_PINLEN];
	BaseHeight = OAH – PanelThick;

	difference() {
	union() {
	linear_extrude(height=BaseHeight)
	hull() {
	circle(d=(TubeData[Tube][T_PUNCHOD] + 2*SocketFlange),$fn=NumSides);
	for (i=[-1,1])
	translate([i*TubeData[Tube][T_SCREWOC]/2,0])
	circle(d=2*Nut[OD],$fn=NumSides);
	}
	cylinder(d=TubeData[Tube][T_PUNCHOD],h=OAH,$fn=NumSides);
	}

	for (i=[0:(TubeData[Tube][T_NUMPINS] – 1)]) // tube pins
	rotate(i*360/TubeData[Tube][T_NUMPINS])
	translate([TubeData[Tube][T_PINBCD]/2,0,(OAH – TubeData[Tube][T_PINLEN])])
	rotate(180/4)
	PolyCyl(TubeData[Tube][T_PINOD],(TubeData[Tube][T_PINLEN] + Protrusion),4);

	for (i=[-1,1]) // mounting screw holes & nut traps
	translate([i*TubeData[Tube][T_SCREWOC]/2,0,-Protrusion]) {
	PolyCyl(Nut[OD],(Nut[LENGTH] + Protrusion),6);
	PolyCyl(Nut[ID],(OAH + 2*Protrusion),6);
	}

	translate([0,0,-Protrusion]) { // LED recess
	PolyCyl(Pixel[OD],(Pixel[LENGTH] + Protrusion),8);
	}

	translate([0,0,(Pixel[LENGTH] – Protrusion)]) { // light pipe
	rotate(180/TubeData[Tube][T_NUMPINS])
	PolyCyl(TubeData[Tube][T_PIPEOD],(OAH + 2*Protrusion),TubeData[Tube][T_NUMPINS]);
	}
	}

	// Totally ad-hoc support structures …

	if (Support) {
	color("Yellow") {
	for (i=[-1,1]) // nut traps
	translate([i*TubeData[Tube][T_SCREWOC]/2,0,(Nut[LENGTH] – ThreadThick)/2])
	for (a=[0:5])
	rotate(a*30 + 15)
	cube([2ThreadWidth,0.9Nut[OD],(Nut[LENGTH] – ThreadThick)],center=true);

	if (Pixel[OD] > TubeData[Tube][T_PIPEOD]) // support pipe only if needed
	translate([0,0,(Pixel[LENGTH] – ThreadThick)/2])
	for (a=[0:7])
	rotate(a*22.5)
	cube([2ThreadWidth,0.9Pixel[OD],(Pixel[LENGTH] – ThreadThick)],center=true);

	}
	}

	}

	//———————-
	// Build it

	if (Layout == "Cap") {
	if (Section)
	difference() {
	Cap();
	translate([-CapSize[OD],0,CapSize[LENGTH]])
	cube([2CapSize[OD],2CapSize[OD],3*CapSize[LENGTH]],center=true);
	}
	else
	Cap();
	}

	if (Layout == "LampBase")
	LampBase();

	if (Layout == "USBPort")
	USBPort();

	if (Layout == "Socket")
	if (Section) {
	difference() {
	Socket();
	translate([-100/2,0,-Protrusion])
	cube([100,50,50],center=false);
	}
	}
	else
	Socket();

	if (Layout == "Sockets") {
	translate([0,50,0])
	Socket("Mini7");
	translate([0,20,0])
	Socket("Octal");
	translate([0,-15,0])
	Socket("Duodecar");
	translate([0,-50,0])
	Socket("Noval");
	}

view raw Vacuum Tube Lights.scad hosted with ❤ by GitHub

2016-02-10

Sears Sewing Table Hinge Covers

The extension surfaces on the Sears sewing table in the Basement Sewing Room unfold from the top, leaving the hinges exposed:

Alas, quilts snag on the squared-off ends of the hinges, a situation that is not to be tolerated…

This protective cap isn’t as small as we’d like, but it must be that thick to cover the hinge, that long to cover the squared-off ends, and that wide for symmetry:

Sears Sewing Table Hinge Cover - solid model — Sears Sewing Table Hinge Cover – solid model

Two neodymium magnets fit in the holes and secure the cover to the all-steel “bronzed” hinges:

Sears Sewing Table - hinge covers — Sears Sewing Table – hinge covers

We’re not sure how well that will work in the long term, but early returns seem promising.

It could be slightly narrower left-to-right and maybe fewer vertices should be oriented differently.

The OpenSCAD source code as a GitHub gist:

	// Vacuum Tube LED Lights
	// Ed Nisley KE4ZNU January 2016

	//- Extrusion parameters must match reality!

	ThreadThick = 0.20;
	ThreadWidth = 0.40;

	HoleWindage = 0.2;

	Protrusion = 0.1; // make holes end cleanly

	inch = 25.4;

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

	//———————-
	// Dimensions

	Hinge = [7.0,52.0,6.0];

	TopThick = 3*ThreadThick;
	PlateThick = Hinge[2] + TopThick;

	NumSides = 8*4;

	//———————-
	// 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);
	}

	//———————-
	// Build it

	difference() {
	hull()
	for (a=[0:7])
	rotate(a*360/8)
	translate([Hinge[1]/2,0,0])
	scale([1.5,1.5,1])
	sphere(r=PlateThick,$fn=NumSides);
	hull()
	for (k=[-1,1])
	translate([0,Hinge[1]/2,k*(Hinge[2] – Hinge[0]/2)])
	rotate([90,0,0]) rotate(180/8)
	PolyCyl(Hinge[0],Hinge[1],8);

	for (i=[-1,1])
	translate([i*Hinge[1]/2,0,-Protrusion])
	PolyCyl(4.8,2.5 + Protrusion,8);

	translate([0,0,-PlateThick])
	cube(2*[Hinge[1],Hinge[1],PlateThick],center=true);
	}

view raw Sewing Table Hinge Covers.scad hosted with ❤ by GitHub

2016-02-09

Hollow State Electronics: Desk Decorations

I did a lightning talk / show-n-tell last Tuesday at the MHV LUG meeting and covered one end of a table with the Neopixel-lit bulbs & vacuum tubes & hard drive platters I’ve been playing with:

MHVLUG – Hollow State Decorations – Lightning Talk

Some of the posts won’t go live for a week, but here’s a peek into the future:

Vacuum Tube LEDs – IBM 21HB5A Beam Power Tube – violet amber phase

Dang, that came out well…

2016-02-06
Vacuum Tube LEDs: Halogen Lamp Success

The Neopixel-illuminated halogen lamp looks much, much better than I expected:

Vacuum Tube LEDs – halogen lamp – red phase

The ceramic ring screws down around the socket shell and pulls it up against the base; the threads have only as much precision as required to keep it from falling off. I may need to add a leveling shim just so I don’t have to explain why it’s always crooked.

2016-02-05
Knurled Metric Inserts
These seem like they ought to come in handy for fastening things to 3D printed objects:

Kurled Inserts – M2 M3 M5

The assorted screws come from the Small Can o’ Small Screwlike Things, all harvested from various dead bits of consumer electronics:

Kurled M3 Inserts

These would benefit from a heated staking tool that slides them into the hole parallel to the axis and flush with the surface. Such things are commercially available, of course, but for my simple needs something involving a cartridge heater, a wall wart, and a drill press may suffice.

It would be better if the inserts had actual knurls, rather than splines. So it goes.

For the record (thread x length x Knurl OD x Body OD):
- M2 x 4 x 3.5 x 2.8
- M2 x 6 x 3.5 x 2.7
- M3 x 4 x 4.5 x 3.8
- M3 x 8 x 5.0 x 3.9
- M5 x 10 x 7.5 x 6.9
The actual measurements seem to vary within ±0.02 of nominal and I doubt the manufacturing consistency justifies any assumption tighter than ±0.1 mm.

The M3 inserts really do have two different ODs.

The M5 insert was listed as “7 mm OD” and measures 7.5 mm, which suggests a typo in the description.

The polygonal hole adjustment I use produces dead-on diameters for small vertical holes:
```
HoleWindage = 0.2;

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);
}
```
So an ordinary cylinder() with the nominal knurl OD or a PolyCyl() with the nominal body OD should suffice. Horizontal holes can probably use a plain old cylinder() with the nominal body OD, because they need reaming anyway.

Perhaps a dab of epoxy would bond better with the plastic around a nominal-size hole than forcing the insert into an undersized hole or heat-bonding the insert. Some experimentation is in order.

Ten bucks for the entire collection (five bags of 50 inserts each = 250 little brass doodads = 4¢ each), shipped free halfway around the planet, seemed reasonable, given that inch size knurled brass inserts run anywhere from 50¢ to upwards of $2 a pop and a Genuine Helicoil 4-40 insert sets you back just shy of a buck.

An Amazon vendor offers 4-40 inserts for $0.24 each in single quantities, but with $9.25 shipping. [le sigh]

Inch-size inserts with knurled rings intended for ultrasonic bonding seem to be 5¢ to 15¢ on eBay. I think the straight-side versions will work better than the tapered ones for heat or epoxy bonding.

It knurls my knuckles that we here in the US haven’t gone solidly metric. Yes, I have a goodly assortment of metric hardware in addition to the harvested fasteners shown above, but it definitely wasn’t cheap & readily available.
2016-02-03
Filament Drive Jam

So, while printing the first pass of the halogen lamp base, this happened:

Lamp Base – wrecked print

The first layer went down fine, but the filament stopped feeding after laying down the small linear patch along the right side. The wrinkles come from me peeling it off the platform while it was still hot and flexy.

Although feeding PETG at 75 mm/s for infill worked so far (I mean, sheesh, look at all the stuff I’ve made in the last year), this involved a fairly large expanse of filament and maybe, just maybe, the high flow rate cooled the nozzle enough to increase the extrusion pressure and eventually strip the filament.

I shoved the filament hard enough to get it feeding again, bumped the extrusion temperature to 260 °C and started another print, whereupon things went swimmingly for the first 12.2 mm. Alas, the filament jammed again, just below the top of the hole for the USB adapter, where you see the odd line in the middle of the finished base:

Lamp Base – USB port

Because it’s now printing a relatively thin cylinder at relatively slow speeds (less infill per perimeter), the “feeding too fast” argument falls flat on its face: obviously, something else is wrong.

Removing the fans showed a bit of plastic on the drive gear teeth, but nothing too terrible:

M2 Filament Drive – jam front view

The witness mark on the planetary gearbox output shaft still lines up with the mark on the gear, so the tiny grub screw hasn’t come loose. Note the slight misalignment between the bottom of the filament drive and the hot end inlet; I’ve already snipped the filament and done some retraction.

A small struggle involving needle nose pliers dragged this classic gouged filament from the drive:

Stripped PETG Filament

This spool of PETG filament started out at 1.70 mm, but this section measures 1.80 mm. That’s at the high end of the ±0.05 mm tolerance around the nominal 1.75 mm, but, frankly, I don’t take the tolerance too seriously.

Undamaged filament from the spool didn’t push smoothly through the drive, so I reamed out the entire path with a 2 mm drill (actually, a #46 drill = 2.05 mm). I don’t recall if I did that before mounting the drive, but even if I did, I’d expect some crud and distortion to accumulate after a while; it’s been running without much attention since last March.

Reassembling the drive and feeding the filament to just above the hot end showed a slight misalignment:

M2 Filament Dive – misaligned front view

I cured that by loosening the screws and rotating the whole drive slightly clockwise:

M2 Filament Dive – realigned front view

Viewed from the side, the drive positions the filament slightly too far to the rear:

M2 Filament Dive – alignment side view

I didn’t (think to) check if the hole in the snout has become bellmouthed, but it wouldn’t take much. In any event, the filament fed into the hot end without incident, so maybe there’s enough slop to cover that misalignment. Maybe I should add a small shim behind the drive?

With the filament drive working again, I had Slic3r chop the bottom off the solid model of the lamp base and create the G-Code for just the top section, which printed without any problem at all.

I drilled eight holes in the bottom surface of the new ring, slobbered epoxy around the ring and tucked it into the holes, used a pair of brass rods to align the two parts, and clamped them together while the epoxy cured:

Lamp Base – clamping

I should be using black PETG anyway, so we’ll call this one a prototype and move on.

So that’s where the line came from…

2016-02-02

Monthly Science: Hard Drive Mood Light Thermal Coefficient

Having that knockoff Neopixel fail from overheating prompted me to measure what was going on. Because the LEDs sink most of their heat into the package leads, the back of the LED strip should be the hottest part of the package and the Mood Light’s central pillar should be pretty nearly isothermal. Despite that, I figured I should measure the temperature closer to the back of the strip, sooo I drilled a hole for the thermocouple…

Clamp the whole Mood Light to the Sherline’s tooling plate with the pillar sides mostly square to the axes and line up the spindle 2 mm behind the LED strip:

Mood Light - aligning thermocouple hole — Mood Light – aligning thermocouple hole

The two clamp pads are CD chunks, under just enough pressure to anchor the Mood Light.

Screw the cap in place (to match-drill both holes at once) and drill a 2 mm (#46, close enough) hole down past the top LED:

Mood Light - drilling thermocouple hole — Mood Light – drilling thermocouple hole

I tucked the Mood Light into a box to ward off breezes, jammed one thermocouple into the new hole, let another float over the top platter, then forced the Neopixels to display constant grayscale PWM values (R=G=B) while recording the LED and air temperatures every five minutes:

Hard Drive Mood Light - temp vs power data — Hard Drive Mood Light – temp vs power data

That was easier and faster than screwing around with automated data collection. The data has some glaring gaps where I went off to do other things during the day.

I turned those numbers into a graph, printed it out, puzzled over it for a bit, then annotated it with useful numbers:

Hard Drive Mood Light - temp vs power data - graph — Hard Drive Mood Light – temp vs power data – graph

That first little blip over on the left comes from a minute or two at PWM 32; the cooling time constant works out to be a bit under 10 minutes. The warming time constant looks to be somewhat longer, but not by much.

Eyeballing the endpoint temperatures for each PWM value, feeding in the current measurements, and creating a small table:

VCC	5	V
Current	0.057	A
Package	0.285	W
Total	3.42	W
PWM	Duty	Nom Power	Failed LEDs	Net Power	°C Rise
0	0.00	0.00	0	0.00	0
32	0.13	0.43	0	0.43	6
64	0.25	0.86	0	0.86	12
85	0.33	1.14	1	1.04	16
128	0.50	1.71	1	1.62	24
192	0.75	2.57	1	2.47	35
255	1.00	3.41	4	3.03	42

The same blue LED that failed earlier dropped out again, plus another package (on a different strip) went completely dark shortly after I clobbered the LEDs with full power at PWM 255. The Net Power column deducts the power not used by the failed LEDs, under the reasonable assumption that the total heating depends on the number of active LEDs.

All the failed LEDs worked fine when they cooled to room temperature, so, whatever the failure mode might be, it’s not permanent. The skimpy WS2812B datasheet says bupkis about a protective thermal shutdown circuit, although it specs an 80 °C maximum operating junction temperature. I’ll stipulate a 20 °C temperature difference from junction to thermocouple at PWM 255, but that doesn’t explain the first blue LED failure at PWM 85.

Methinks these knockoffs will be much happier operating in the mid-30s.

Turning the last two columns of that table into a graph (minus the PWM 0 line to let the intercept float around) looks like I’m faking it:

Hard Drive Mood Light - Temperature vs Power — Hard Drive Mood Light – Temperature vs Power

The Y intercept is off by less than 1 °C, which seems pretty good under the circumstances. The kink at PWM 85 shows that I probably didn’t allow enough time for the temperature to stabilize after the blue LED failed.

So, in round numbers, the thermal coefficient for a dozen knockoff Neopixels on a plastic pillar inside a stack of hard drive platters works out to 14 °C/W.

The raised sine waves in the Mood Light produce a long-term average PWM half of their maximum PWM. They’ve been perfectly happy with MaxPWM = 64 pushing them barely 6 °C over ambient, so they should continue to work fine at PWM 128 for a 12 °C rise… except, perhaps, during the hottest of mid-summer days.

Obviously, I should jam a thermistor inside the column and have the Arduino wrap a feedback loop around the column temperature…

2016-02-01

Category: Machine Shop

Vacuum Tube LEDs: Ersatz Tube Sockets

Sears Sewing Table Hinge Covers

Hollow State Electronics: Desk Decorations

Vacuum Tube LEDs: Halogen Lamp Success

Knurled Metric Inserts

Filament Drive Jam

Monthly Science: Hard Drive Mood Light Thermal Coefficient