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

Nike Cycling Shoe Latches: Resprung

The Nike cycling shoes I bought some years ago (at a steep discount when they got out of the cycling shoe biz) close with a ratcheting plastic strap rather than laces, so I bought a spare set of straps: the plastic part always breaks first. As it turned out, a coil spring inside each latch failed and the stub end (on the right side here) gradually worked its way between the latch tab and the frame:

Cycling shoe latch – broken spring

Eventually this got to the point where the latches jammed and I had to do something. The first step was to drill out the rivet holding the spring and tab in place:

Drilling latch rivet – magnetized bit

You’ll note the rich collection of swarf clinging to the drill bit, which indicates this one hasn’t been used since a lightning strike magnetized all the steel in the house. A pass through that demagnetizer shook off the swarf and prepared the bit for the next time.

Releasing all the parts shows the problem:

Nike cycling shoe latch – broken spring

The OEM springs used 24 mil spring wire that, surprisingly, matched a box of music wire in the Basement Laboratory Warehouse Wing. The spring coils have 5 turns that just clear the 3 mm rivet that I recycled as a mandrel; I think a 2.5 mm pin would produce a better fit. Not being a fan of rivets, I replaced them with 4-40 machine screws, even though the threads probably won’t do the aluminum frame any good at all.

A protracted bending and wrapping session produced a reasonable approximation of the OEM spring:

Latch spring – formed

It’s worth noting that each of those coils uses up about 55 mm of wire: 5 × 3.5 mm × π. Cut an excessively long piece from the music wire coil!

Trimming and shaping the ends to fit through the notches and around the outside of the frame shows that my wire-bending skills need considerably more practice. This spring (the second one I made) also shows that my beginner’s luck with the first coils wore off all too quickly:

OEM springs with homebrew replacement

But both springs fit and work fine, so I’ll call it done for now:

Repaired latch – nut side

Will a replacement spring break before the plastic strap?

Obviously, I need a CNC spring bender…

2012-09-24
Dead-on-arrival Lithium Cell

DOA Energizer CR2032 cell

The display on Mary’s Cateye Astrale cyclocomputer (remember cyclocomputers?) faded to gray, which meant a new CR2032 lithium cell was in order. I grabbed one from the heap, popped out the old cell, inserted the new cell, and … the display stayed blank.

Quick like a bunny, I reinserted the old cell to save the odometer (15524 miles) and wheel circumference (1475 mm) data; the display returned to dim gray.

The “new” cell, which came from an unopened pack, read 0.45 V with no load…

The cell didn’t have a date code, but the package sports a cryptic MU that might encode the date of manufacture or the date of packaging or the copyright date or something; the various search results aren’t forthcoming and the Energizer site gives no explanation.

I’m pretty sure I haven’t owned that package for more than a few years and it’s been in a shirtsleeve environment (plus the occasional hot van) ever since.

Another Energizer cell from a more recent lot, bearing CA on the package and YA on the cell, worked fine.

Being that sort of bear, I wrote the date and mileage on the previous cell (a Newsun, whoever they are, with a 3Y code), because the last time around the odometer value didn’t survive the cell change. The current total works out to 277 miles/month = 3300 miles/year, including winter downtime, which is fine with us; we mostly ride the bikes around town on errands and take the occasional tour.

2012-09-21
Multimeter Range Switch Contacts: Whoops!
One of my multimeters began reporting bogus values that improved by working the range switch back-and-forth, which suggested the switch contacts need cleaning. Taking the meter apart was easy, right up to the point where I removed the range switch from the PCB by compressing the four locking tabs on the central shaft:

Multimeter range selector switch

Just before taking that picture, the switch launched half a dozen spring contacts across the bench, my shirt, and the floor… I recovered four for the picture and later found a fifth smashed on the floor, but the last contact remains AWOL.

The contact in the middle, the oddly shaped one with small tabs on the ends, is a prototype replacement conjured from 6 mil phosphor bronze stock:

Multimeter range switch contacts

The little domes ensure a good sliding surface, but require two bends in the middle of the contact and some way to shape the metal into a dome. After a few experiments, I filed the end of a nail into a rounded chisel that worked pretty well:

DMM switch contact punch

The original contacts came from 3.5 mil stock and have considerably more flex; 6 mil stock is what I have.

I think I should make half a dozen contact springs to replace the entire set, a task requiring more time than I have right now. For the record, the overall process goes like this:
- lay out overall shape, slightly longer than needed
- cut center opening with abrasive wheel
- cut out contact
- punch contact domes (from back = dimples)
- bend to shape
- trim ends to length (not done in picture)
- dress raw edges (not done in picture)
Given the number of parts and the fiddly accuracy required to make the slot, this might be a good job for the Sherline, although clamping each little proto-spring down while getting the abrasive wheel in there seems daunting.

Perhaps cutting the slots and punching the dimples would work better before cutting out the contacts, with a sheet clamped on four sides? The center will be floppy, what with all the slots, but grinding slots on the middle contacts first might be helpful. Would adhesive under the sheet to hold down the middle gunk up the abrasive wheel?

So many projects …

Memo to self: Springs! Always expect springs!
2012-09-20
Black and Decker Pocket Power: Battery State

A Black & Decker Pocket Power emerged from the heap and refused to take a charge, which obviously calls for a teardown. The case has three screws, one lurking behind the label:

Pocket Power – case screws

The sticker over on the right says it’s five years old, which explains the whole problem right there; you can evidently buy new-old-stock units from the usual low-dollar sources that will arrive with a similarly dead battery.

Peeling off the rubber bumpers and prying the case open reveals the innards:

Pocket Power – internal layout

The battery pack looks to be an octet of ordinary NiMH cells; the label on the other side of the shrink wrap reports 9.6 V @ 1200 mA·h, which is about what you’d expect, with a date in mid-2007 that matches the sticker on the case.

The upper left corner of the main label has some interesting information:

Pocket Power – label specs

The tiny wall wart that came with the unit produces 12 VDC at 300 mA, which doesn’t match the INPUT spec at all. Perhaps the maximum current from the internal pack made its way to the label by mistake?

The label also shows the reason I got this thing: it can produce just enough 120 VAC power to run an arbitrary wall wart charger for a gadget that doesn’t charge from a 12 VDC source. Upconverting 9.6 VDC to 120 VAC, then downconverting it to, say, 14.4 VDC makes no sense whatsoever, unless that’s the only way to charge that particular gadget. Which has, I’m sorry to say, been the case every now and again.

I think the Model name has a typo: everything else suggests this is a CP120B. So it goes.

Unsoldering the leads and perpetrating the obvious tests produces these curves:

B&D Pocket Power

The black curve is the initial “won’t accept a charge” state with the wall wart and internal circuitry; the pack obviously has two weak cells. The curves in the lower left correspond to individual cells and series pairs that I discharged to 0.9 V/cell after the top curve ended.

The tiny stroke between the sets, way over on the Y axis, is cell pair BC (my arbitrary labels) that probably accounts for the sudden drop in the black curve. However, the orange curve also came from pair BC after charging for about 18 hours at 120 mA, so they’re not completely dead. Their capacity has dropped to about 700 mA·h, though, which isn’t good.

Soldering the pack back together and charging for another 18 hours at 120 mA produced the green curve at the top, which shows the same sudden dropoff at about 700 mA·h.

So I’ll put it back together again and let it charge for a while, but new cells will definitely brighten its disposition.

2012-09-19
Three-way Lamp Socket: Fuse Test
After un-bending the top of a pole lamp that suffered an untimely collision with the floor, I discovered that the entire stock of three-way bulbs in the heap had at least one burned-out filament each; I’d acquired them when Mom moved out of the Ancestral House, so they dated back a long time. So I figured I’d insert a decently sized single-filament bulb and be done with it.

Three-way lamp sockets have an additional tab contact between the usual central contact and the outer shell:

Interior of 3 way lamp socket

The shell forms the common contact for the filaments and the switch counts in binary: off / off, off / on, on / off, on / on. In principle, the tab sits low enough to not contact the shell of an ordinary bulb.

I was doing this in the Basement Laboratory Workshop Wing, with the lamp plugged into the outlet strip along the front edge of the bench; that way, I simply poked the power strip button to remove line voltage from the lamp while swapping bulbs. So I:
- turned the power strip off
- unscrewed the last dead three-way bulb
- threw it away
- screwed in an ordinary bulb
- turned the strip on
At which point all the fluorescent overhead lights in the Laboratory went dim, the shop resounded with a deep resonant groan, and the acrid smell of electrical death filled the air. Elapsed time less than a second, tops.

Come to find out that the socket’s contact tab stuck up a little bit further than it should, producing a dead short across the line:

Melted bulb base

Of interest: the branch circuit breaker didn’t trip, the GFI on the circuit didn’t trip, and the pop-out breaker in the power strip didn’t trip.

Huh.

I harvested the pole sections, the base counterweight, and the line cord. The rest of the corpse joined the bulbs in the trash…
2012-09-18
Current-Sense Resistors: Mind the Power

The bench supplies I use have current limiting, but the 10 mA meter resolution leaves a lot to be desired, so I conjured up a simpleminded 200 mA meter from a panel-mount meter and a 1 Ω sense resistor. That means it’s good for only 200 mA, so I insert it in series with the supply only when it’s needed. Lately it’s been reading more than a little bit high and I took it apart to find this obvious evidence of abuse:

Homebrew millammeter with burned sense resistor

The loose resistor sitting atop the chip shows what the burned resistor soldered in the circuit should look like.

The power supply has a 3 A current limit. No surprise: 9 W is more than the unfortunate 5 W resistor can handle.

It’s all better now …

2012-09-13
Compact Fluorescent Bulb Lifetime: Another Data Point

Each of the three chandeliers in the Poughkeepsie Train Station sports 36 bulbs in two rings. When the station opened in 1918 they installed those newfangled incandescent bulbs that were all the rage at the time. The color of the bulbs in this Wikipedia picture, dated October 2007, suggests that tungsten ruled for at least nine decades:

Poughkeepsie Train Station Interior

Since then, they installed chunky compact fluorescent bulbs that probably provide the same amount of light, minus the pinpoint highlights from tungsten filaments in clear bulbs. This view from below the central chandelier shows the layout and some detail of the carving & decorative sockets:

Pok RR Station Middle Chandelier – detail

In addition to being decorative, those chandeliers also give useful data on the reliability of compact fluorescent bulbs. With the contrast stretched the other way to make the bulbs easier on the eye, count the number of deaders in …

Chandelier 1:

Pok RR Station Chandelier 1

Chandelier 2:

Pok RR Station Chandelier 2

Chandelier 3:

Pok RR Station Chandelier 3

I took each picture from a vantage point showing all the deaders; the bulbs hidden behind the central dingus work.

Let us assume all 108 bulbs were installed at the same time and, given the number of deaders, haven’t been touched since then (although they’re not covered in fuzz, which suggests that they’ve been dusted within living memory). I was there in mid-afternoon, so the bulbs probably burn 24 hours/day and aren’t subject to early failure from frequent starts.

So, in no more than five years, 108 CFL bulbs have a 4.6% failure rate, which works out to 0.9%/year, more or less, ignoring any infant mortality. If they’ve been up there for the last 2.5 years, then it’s 1.8%/year. Replacing deaders since installation, of course, makes it worse than that.

Over the course of a decade, a compounded 0.9% failure rate will kill 9.4% of the bulbs. After 20 years, 20% will be dead. A 1.8% annual failure rate kills 20% and 43%, respectively.

Now, I’ll grant you that tungsten bulbs burn far more energy over that time, but replacing a percent or two of those complex and somewhat eco-hostile CFL bulbs every year cuts away a big chunk of the rainbows-and-pink-unicorns delight involved in Saving The Planet.

2012-09-06