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.

Category: Machine Shop

Mechanical widgetry

Dell 75UYF Battery Teardown: Part 1

One of the batteries on the ancient Dell Inspiron 8100 laptop died completely and our Larval Engineer reports the other battery isn’t far behind; it gets her from outlet to outlet and not much more. Pursuant to that comment about harvesting reasonably good cells from dead batteries to build an extended-life external battery for the Canon SX230HS camera, I made a preliminary pack probe.

The label says it’s a 14.8 V battery, so you’d expect four 3.7 V lithium cells in series. The 3.8 A·h capacity suggests parallel cells:

Dell 75YUF battery – label

Indeed, peeling off the label shows four cells pairs in series:

Dell 75YUF battery – under label

The case joint seems firmly welded together and resisted simple attempts to crack it open. I might run a slitting saw around the edge, although I’ll probably just crunch it in the vise because the patient need not survive the operation.

A single cell should have a 1.9 A·h capacity, although in an awkward cylindrical form factor. The 3.5 A charging current would drop to 1.7 A for a (string of) single cells.

The Canon SX230HS uses a single 3.7 V, 1.1 A·h prismatic “battery”, which means replacing that with a single external cell wouldn’t be a major win; the size difference shows how much lithium energy storage tech has advanced in the last decade or so. A pair of cells in parallel would quadruple the runtime, which might be enough. Three in parallel would be fine, although that would require attention to matching their capacity; the nominal 5.2 A charging current (1.5 × 3.5 A) seems aggressive.

Time to start mulling charging circuits…

2012-06-11
Evapo-Rust vs. Battery Contact Corrosion

A long-forgotten toy emerged from the heap bearing a trio of corroded NiMH cells between the usual plated-steel contacts:

Corroded contacts – original

The toy wasn’t worth salvaging, but I extracted the contacts and applied Evapo-Rust to see what happened. After an overnight soak, some corrosion remained:

Corroded contacts – after Evapo-Rust

Scrubbing with a stainless-steel detail brush removed the flakes and left reasonably clean metal behind:

Corroded contacts – after brushing

Although it’s not beautiful, I think the contacts came out as well as one could expect. The longer contact plate has holes, thinned sections, and some corrosion inside the spring; I’d be mildly tempted to rebuilt the whole thing with some nickel shim stock and a new spring.

If I were salvaging the toy, I’d dab vinegar on the wiring to neutralize the creeping potassium hydroxide, rinse the whole thing with water, and clean out the case. Instead, it joined the consumer electronics recycling box with a thud…

2012-06-10
Wouxun KG-UV3D GPS Interface: PCB Component Envelopes

This doodle just emerged from the heap:

TT3 and Interface PCB keepout dimensions

It gives the envelope dimensions for the two PCBs in the quasi-extruded GPS interface case:

HT-GPS Case – Trial fit – rear view

The overall idea was to subtract the envelopes from the interior of the solid case and print the shell:

HT-GPS Adapter Case – end view

Surprisingly, after minimal cleanout, the PCBs pretty much just slid into place and the setscrews locked them down.

Now I can throw out that scrap of paper…

The OpenSCAD source has the gory details; the actual dimensions don’t quite match the doodle, but that’s in the nature of fine tuning.

2012-06-08
New Aztek Brake Shoes

The front brake on my bike started sounding more gritty than usual on a recent ride, which led to pulling the pads off, which led to discovering that one pad had worn completely through:

Worn-through brake pad

The rim had a slight scuff where that aluminum tab stuck through, but nothing worth worrying about. The wear indicators aren’t reliable, because the pad curve matches 27-ish inch wheels and the Tour Easy has a 20 inch front wheel. If you align the pads to the outside of the rim, as I do, the inside edge gets light wear. So I let ’em wear, check them when the tire gets a flat, and this is the first time a pad has worn through. I think that means the front tire hasn’t had a flat in quite a while…

While I was at it, I replaced all the pads on both our bikes. The rear pads didn’t have nearly as much wear, which is about what you’d expect, although the wear indicator grooves have just about bottomed out:

Worn replaceable pads

Those are replaceable pads, which work quite well on the new brake arms. I suspect by the time I get around to needing new inserts (I bought a bunch, of course) they’ll be obsolete and unobtainable.

I file the pads flat to save a bit of time wearing them in:

Filed replaceable pads

I don’t hold with the notion of toeing in the pads to avoid squealing, vastly preferring crisp brakes with very little travel. Whatever the material is in Aztek pads, they don’t squeal after they’ve fitted themselves to the rim… but, of course, this new pair howled worse than the Freezer Dog when I got them out on the road.

Squealing brakes aren’t entirely a bad thing, as they scare the daylights out of oblivious pedestrians, but I’d rather use the bell. So I gripped a strip of fine sandpaper between pad and rim, gently squeezed the brake lever, and rolled the bike about two wheel revolutions. Repeat on the other side and the rim’s now nice and clean and grippy. Flip the sandpaper over, scrub the pad surface, and they don’t make a sound.

Ding!

2012-06-07
Belt Gluing

When I bought a new belt some months ago, I thought the lack of stitching meant it was made from a single strip of leather. Come to find out that it’s actually two strips glued together with something sticky that came un-done at the point where the belt passes through the buckle.

So I peeled a bit more apart, smoothed a thin layer of urethane glue (aka Gorilla Glue) inside, laid waxed paper on both side just in case the foam expanded beyond my wildest imagination, and clamped it together:

Belt clamping

The glue layer turned out just about perfect, with only a few blobs sticking out the sides:

Belt with urethane glue blobs

Those blobs snapped off easily enough and the belt works fine again. We’ll see how long this one lasts…

2012-06-05
NIC Backpanel Shortening

Actually, that NIC didn’t slip right into place, because its backpanel plate was sized for a full-height PC case. Unlike the cheap stamped steel you find these days, NetGear used much thicker metal that required an attack with the bandsaw, a hammer, and some files to clean up the raw edges.

But it fit pretty well after all that:

Shortened NIC backplate

You can just barely see the NetGear logo wrapped around the right-angle bend…

2012-06-03
Lenovo Headset Boom: Repair Faceplant

I picked up a Lenovo headset on sale and over the course of a few weeks the mic boom pivot worked itself loose, until I finally dismantled the left ear cup to see what was inside. Come to find out that the mic boom has a molded threaded section held into the cup with a simple nut and no locking mechanism at all:

Lenovo headset – OEM mic boom pivot nut

I think the metal washer was intended as a low-friction pivot atop the compliant silicone (?) washer underneath, but the net effect was that the nut unscrewed a little bit more every time the mic boom moved. By the time I got in there, the nut was completely off the threads.

The original nut left a thread or two showing, so I found a thicker replacement nut with a better grip. The obvious solution involves a dab of Loctite to jam the nut in position, but we all know that some plastics, most notably acrylic, react badly to threadlocker and tend to disintegrate. Although I considered just epoxying the nut in place, that seems so, well, permanent.

So I dutifully tested a dab of Loctite on an inconspicuous spot inside the ear cup, got no reaction at all, put a drop on the boom pivot threads, and reassembled everything:

Lenovo headset – replacement mic boom pivot nut

Alas, by the time I got back upstairs and hung the mic on the rack, the boom fell completely out of the earcup! Back in the Basement Laboratory, I dismantled the thing again and confronted this mess:

Lenovo headset- Acrylic plastic vs. threadlock

Huh. The ear cup isn’t made of the same plastic as the mic boom: one shrugs off threadlock, the other disintegrates.

That’s obvious in retrospect, eh?

The only threads that aren’t ruined lie completely within the ear cup frame, with just a stub sticking up around the wire. So I cleaned things up and did what I should have done originally: put a dab of epoxy inside the nut to bind the pivot firmly in place. A snippet of unshrunk heatshrink tubing around the wire provides a bit of strain relief:

Lenovo headset – boom pivot nut with epoxy

There’s no longer any space for the compliant washer in that stack, so we’ll see how long this lasts. The next repair will certainly venture far inside non-economical territory. I like the headphones, though.

Memo to Self: Check in an inconspicuous spot on the same material.

2012-05-30