Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
A week or so ago, the scroll ring on the Kensington Expert Mouse trackball at my left hand failed completely. Unlike the previousrepair attempts, tweaking the IR emitter-detector pair positions did nothing. Tried it on three different PCs and five different operating systems with the same result: the ring stayed dead.
Fortunately, this one was a warranty replacement for the dead Unit 1 I bought some years back and was still within its 5 year warranty, so when I contacted Kensington tech support with the story they immediately shipped a replacement. It just arrived and works fine.
The scroll ring detents seem much smoother on this one, so I haven’t taken it apart to remove the magnetic latch and don’t know if they’re using a different quadrature sensor. One can but hope.
Kensington Expert Mouse – ball bearing
For what it’s worth, an absolutely brand new ball barely moves on those three jeweled bearings (one marked with the yellow oval in the picture). Just rub the ball on one side of your nose to add some skin oil: shazam it spins like glass on ice.
They don’t mention that trick anywhere in the meager instructions…
Update: Eight years in the future, a real fix appears!
The 120 m 50 V BUZ71A that served as the crash test dummy while I got the thing working:
BUZ71A-overview
A detail of the interesting area near the origin:
BUZ71A-detail
The datasheet drain resistance values are the maximum values, so they’ll generally be higher than what I measure.
A plastic-encapsulated W7NB80 with a 1.9 (!) drain resistance, due to its 800 V (!) rating:
W7NB80-overview
Hold the gate voltage constant at 10.0 V and step the temperature from 0 °C to 50 °C:
W7NB80-Temp
I haven’t figured out how to get the actual temperatures from the Gnuplot input dataset to the graph without knowing them in advance. The “index” is simply the 0-origin block number, which conveniently (and coincidentally) lines up with the 0 °C to 50 °C temperature range.
An overview of a 400 m 200 V IRF630:
IRF630-overview
The juicy part:
IRF630-detail
And the variations with temperature:
IRF630-Temp
A 1.5 200 V IRF610, another high-resistance transistor:
IRF610-overview
The temperature variations:
IRF610-Temp
The winning entry for high resistance, though, is the 500 Ω (!!!) BSS127 that emerged from a paper on current sensing using mirror FETs for temperature compensation. It has a 600 V rating, but I have no idea why such a high drain resistance makes any sense in a SOT-23 package. They’re obsolescent and I won’t buy any just to have ’em around.
Just for completeness, a 1 1% resistor:
Resistor – 1.0 ohm
And a 100 m 1% resistor:
Resistor – 0.1 ohm
It turns out that the wire leads I soldered on contributed 6 m to the total, so the tester actually reports the truth! I checked that by passing 1.000 A through the resistor, which put 100 mV at the base of the resistor pins, then measuring 106 mV at the end of the wire leads. One can quibble about voltmeter accuracy, but it’s pretty close and much better than the ohmmeter accuracy at that resistance.
The firmware forces 0.0 for drain current identically equal to 0.0 (it’s a floating point number cast from a 10-bit unsigned integer) to avoid numeric explosions. The next few points away from the origin show the effect of small errors on small measurements; the voltage resolution is 15 mV and the current resolution is 2.5 mA; you can actually see the steps near the origin.
All in all, a fun project…
Need the datasheets? Ask your favorite search engine for, say, IRF610 datasheet. That should do the trick.
New this year at the Trinity College Firefighting Home Robot Contest will be a Checkout Table, where teams can verify that their robot meets some initial specifications (Section 2.5 of The Rules). The overall size should be the easiest spec to check; I just glued up a pair of suitable Bounding Boxes:
Trinity Robot Contest – bounding boxes
Robowaiter robots must fit in the smaller cube, which is 30 cm on a side. Firefighting robots must fit in the larger box, with wheeled / treaded robots inside the 31 x 31 x 27 cm outline and walkers within the larger 46 x 31 x 27 volume.
Next step: fluorescent orange paint over a white shot coat to kill the lettering.
… look like they ought to fit some sort of tubing and, indeed, a bit of rummaging produced a hank of suitable thick-walled plastic stuff. Heating one end until it went clear and floppy, then jamming it over a syringe’s Luer fitting produced a workable flushing tool:
Syringe with tubing to fit R380
I folded a tissue, laid it over the sponges and wipers at the printhead park position, then pushed the ink tank carrier over the tissue to absorb the spray. Squirting three syringes full of 10% ethanol through the head cleaned out a few of the blocked jets, but didn’t produce a complete fix.
Next up: homebrew window cleaner, diluted about 1:3 to knock back the ammonia concentration.
I dropped that lens cap and the sheet-metal disk popped out; evidently the acrylic caulk doesn’t really count as an adhesive. Cleaned out the residue, ran a thin layer of urethane adhesive around the rim, and applied some clamps:
Re-clamping the cover
Cleaned out the inevitable urethane bubbles that emerge from even the most minute opening and it’s all good.
A few months after shaking off the previous fruit fly infestation, the worm compost bin has succumbed to another species of fruit fly that’s probably Drosophila melanogaster: much larger, breeds faster, and seems far more tenacious. Even though they’re completely innocuous, Something Must Be Done, but alas there are no insecticides suitable for a worm bin that produces vegetable garden compost. That reduces the situation to the Siege of Stalingrad: cut off their supplies and let them fight it out.
It seems that fruit flies and their progeny die slightly faster than worms; after three or six weeks without feeding, the flies will should be history and the worms will be eating the dead. Temperatures in the Basement Laboratory Vermiculture Wing will remain in the 60 °F range for the next month or two, so the fly egg-to-adult time will be longer than the usual eight days and this may not work as well as we’d like.
Assuming that succeeds, however, we’ll be freezing all the kitchen scraps that go into the bin to kill off the fruit fly eggs that arrive here from around the world. There seems no way to get fruits without fruit fly eggs, even with non-organic produce. Organic stuff, well, it’s worse than that.
I conjured up a Fruit Fly Escape trap that should I hope will lure flies out of the bin to their death, while keeping the worms inside.This won’t help much with the current extreme infestation, but may help dry the bin’s upper layer and, when we get the population knocked down, should exterminate the more adventurous survivors. Obviously, we’re breeding for stay-at-home fruit flies and, given their rapid-prototyping life cycle, they may evolve into tiny couch potatoes.
Anyhow.
Flies like heat and light, while worms vastly prefer cool and dark, so the general idea is to drill a hole in the bin lid, fit a long tube over it, put an LED ring light at the base, and run a flypaper spiral up the tube to a vent cap near the top. The first picture gives an overview, although it’s tough to see the vertical tube against the clutter: it’s clear with two red spirals, having started life as some weird-ass holiday decoration for the previous owners of our house.
Anyhow, the more interesting plastic bits look like this:
Fly Escape – solid model
The top ring is the vent cap, with a hole in the middle for a string supporting the sticky tape strip. The middle ring holds three sections of LED strip light that dissipate about 2 W from a 12 V wall wart; that’s enough heat around the tube to produce a slight upward draft. The riser tube at the bottom has an angled rim that compensates for the bin lid angle and holds the long tube vertical. The ring around the riser has a matching angle.
They fit into the lid thusly:
Fly Escape – Riser trial fit
Two beads of hot-melt glue, top and bottom, hold them in place and make an air- / worm- / fly-tight seal.
The inner tube holds the fly paper container and has a slight inward taper toward the top to wedge it in place:
Fly Escape – solid model – bottom
A similar view from inside the actual lid:
Fly Escape – Riser trial fit – bottom
That was the first pass at the dimensions; the tube walls didn’t quite join because I forgot to force the number of polygonal sides to be equal. It’s deliberately thin to make the walls springy, but everything must be Just Right to get both no fill and no space between the two perimeter threads.
The riser and LED ring, combined with festive spiral stripes along the tube and some silicone tape sealing the tubes together, produce a cheery nuclear glow that’s enhanced by the victims mired in the adjacent flypaper strips. A third strip runs up the middle of the tube:
Fly Escape in action
The vent cap on the top of the tube has a small hole in the middle to hold the string supporting the flypaper spiral exactly in the middle of the tube. This view is upside-down from the mounted orientation :
Fly Escape – Vent Cap
The alert reader will notice a red top plug in place of the vent cap in the first picture. This whole project happened over the course of a frantic afternoon, evening, and morning, with progressive product improvements along the way. For example, it turns out that some flies went pedestrian and walked up the inside of the tube, so there’s now a circle of screening inside that nice vented cap.
Having a 3D printer to hammer out custom plastic widgetry on a short schedule = win.
The OpenSCAD source code:
// Worm bin fly escape
// Ed Nisley KE4ZNU - March 2012
Layout = "Show"; // Build.. Show Riser Ring Cap
//- Extrusion parameters - must match reality!
ThreadThick = 0.25;
ThreadWidth = 2.0 * ThreadThick;
HoleFinagle = 0.3;
HoleFudge = 1.00;
function HoleAdjust(Diameter) = HoleFudge*Diameter + HoleFinagle;
Protrusion = 0.1; // make holes end cleanly
//-- Dimensions
RiserID = 47.0; // ID = transparent riser tube OD
RiserOD = 51.0; // OD = hole in lid (matches hole saw OD)
RiserHeight = 50.0; // wall height from lid
RiserSides = 4*8; // for consistency & symmetry
RiserBaseHeight = IntegerMultiple(5.0,ThreadThick); // stop ring height
RiserBaseID = RiserID - 2*1.0; // stop ring ID
LipOD = 59.0; // OD of lip mounted on lid around tube
LipAngle = 3.0; // angle for lip to make tube vertical
LipMinThick = IntegerMultiple(3.0,ThreadThick); // min lip thickness
LipAngleThick = LipOD*tan(LipAngle); // angled section thickness
LipThick = LipMinThick + LipAngleThick; // total lip thickness
RingClearance = 0.5; // space between ring and tube
TrapID = 23.0; // sticky tape container OD
TrapIDTaper = 2.0; // taper to hold container in place
TrapHeight = 45.0; // ... height
TrapWallThickness = 2*ThreadWidth;
TrapSides = 4*4;
TrapFlanges = 3; // number of support flanges
TrapFlangeThick = IntegerMultiple(3.5,ThreadWidth);
LEDThick = 2.5; // LED strip thickness
LEDWidth = 11.0; // ... width
LEDWireOD = 3.0; // power cable dia
LightID = RiserID + 2*LEDThick; // ID of LED collar
LightOD = LightID + 2*4*ThreadWidth; // ... OD
LightFlangeThick = IntegerMultiple(2.0,ThreadThick);
CapID = RiserID;
CapRingID = CapID - 2*1.5;
CapOD = CapID + 2*4*ThreadWidth;
CapBaseHeight = RiserBaseHeight;
CapHeight = 10.0 + CapBaseHeight;
CapSides = RiserSides;
CapFlanges = 3;
CapFlangeThick = TrapFlangeThick;
CapGuideID = 3.0;
CapGuideOD = CapGuideID + 6*ThreadWidth;
//-- Sticky tape container holder
module TrapMount() {
ODBot = TrapID + 2*TrapWallThickness;
ODTop = TrapID - TrapIDTaper + 2*TrapWallThickness;
difference() {
union() {
cylinder(r1=ODBot/2,r2=ODTop/2,h=TrapHeight,$fn=TrapSides);
for (i=[0:TrapFlanges-1])
rotate(i*(360/TrapFlanges) + 90) // align leg with thick side
translate([RiserOD/4,0,RiserBaseHeight/2])
cube([(RiserOD/2 - 4*Protrusion),TrapFlangeThick,RiserBaseHeight],center=true);
}
translate([0,0,-Protrusion])
cylinder(r1=HoleAdjust(TrapID)/2,
r2=HoleAdjust(TrapID - TrapIDTaper)/2,
h=(TrapHeight + 2*Protrusion),
$fn=TrapSides);
}
}
//-- Riser tube
module RiserTube() {
TotalHeight = RiserHeight + RiserBaseHeight;
difference() {
cylinder(r=RiserOD/2,h=TotalHeight,$fn=RiserSides);
translate([0,0,RiserBaseHeight])
PolyCyl(RiserID,TotalHeight,RiserSides);
translate([0,0,-Protrusion])
cylinder(r=RiserBaseID/2,h=TotalHeight,$fn=RiserSides);
}
}
//-- Angled lip around ring
// aligned with flat side downward at Z=0
module LipRing(Clearance = 0.0) {
difference() {
cylinder(r=LipOD/2,h=LipThick);
translate([0,0,-Protrusion])
cylinder(r=(RiserOD/2 + Clearance),
h=(LipThick + 2*Protrusion),
$fn=RiserSides);
rotate([LipAngle,0,0])
translate([-LipOD,-LipOD,(LipMinThick + LipOD/2*tan(LipAngle))])
cube([2*LipOD,2*LipOD,LipAngleThick],center=false);
}
}
//-- Collar to hold LED strip light
module LEDCollar() {
difference() {
PolyCyl(LightOD,(LEDWidth + LightFlangeThick));
translate([0,0,LightFlangeThick])
PolyCyl(LightID,(LEDWidth + Protrusion));
translate([0,0,-Protrusion])
PolyCyl(RiserID,(LightFlangeThick + 2*Protrusion));
translate([0,0,(LightFlangeThick + LEDWidth/2)])
rotate([0,90,90])
PolyCyl(LEDWireOD,LightOD);
}
}
//-- Cap to hold trap string and vent the tube
module VentCap() {
union() {
difference() {
cylinder(r=CapOD/2,h=CapHeight,$fn=CapSides);
translate([0,0,-Protrusion])
cylinder(r=CapRingID/2,h=(CapHeight +2*Protrusion),$fn=CapSides);
translate([0,0,CapBaseHeight])
cylinder(r=CapID/2,h=CapHeight,$fn=CapSides);
}
difference() {
union() {
for (i=[0:TrapFlanges-1])
rotate(i*(360/CapFlanges))
translate([CapOD/4,0,CapBaseHeight/2])
cube([(CapOD/2 - 4*Protrusion),CapFlangeThick,CapBaseHeight],center=true);
cylinder(r=CapGuideOD,h=CapBaseHeight);
}
translate([0,0,-Protrusion])
PolyCyl(CapGuideID,CapHeight);
}
}
}
//-- Handy routines
function IntegerMultiple(Size,Unit) = Unit * ceil(Size / Unit);
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 + HoleFinagle)/2,h=Height,$fn=Sides);
}
//-- Put peg grid on build surface
module ShowPegGrid(Space = 10.0,Size = 1.0) {
Range = floor(50 / Space);
for (x=[-Range:Range])
for (y=[-Range:Range])
translate([x*Space,y*Space,Size/2])
%cube(Size,center=true);
for (z=[1:10])
translate([0,0,z*Space])
%cube(Size,center=true);
}
//- Build it
ShowPegGrid();
if (Layout == "Ring")
LipRing();
if (Layout == "Riser")
RiserTube();
if (Layout == "Cap")
VentCap();
if (Layout == "Show") {
color("SkyBlue") {
TrapMount();
RiserTube();
LipRing();
}
color("Salmon")
translate([0,0,2*LipThick])
rotate([180,0,0])
LipRing(RingClearance);
color("Chocolate")
translate([0,0,(1.25*RiserHeight)])
LEDCollar();
color("Sienna")
translate([0,0,2*RiserHeight])
rotate([180,0,0])
VentCap();
}
if (Layout == "Build1") {
TrapMount();
RiserTube();
LipRing();
}
if (Layout == "Build2") {
LipRing(RingClearance);
}
if (Layout == "Build3") {
LEDCollar();
}
if (Layout == "Build4") {
VentCap();
}
Something weird is going on with the Northern Cardinals at our feeder. First a female missing a leg, now a male minus his head feathers:
Bald Cardinal – right side
A view from the other side:
Bald Cardinal – left side
A bit of searching with the obvious keywords produced that writeup, which suggests feather mites or other parasites. Given that this was in March, that cardinal is definitely not molting!
Those pictures are tight crops from a hand-held Canon SX230HS at dusk, through two layers of 1950-vintage glass. Sorry about that, but the bird spooks whenever I crack the door open for a better view.