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.

Author: Ed

  • Mobile Amateur Radio Power: Check the Fuseholders

    The Yaesu FT-857 I have in the car has been not turning on lately, which I feared had something to do with being cooked inside a closed van for a week on the top level of a Camden parking garage during the hottest part of the summer.

    But, no, as it turned out, that had nothing to do with it: when I got the radio on the workbench, it powered up just fine. Back in the car, it’s dead.

    Which implies a power problem. The radio power comes from 10-AWG zip cord, through a pair of 40 A fuses, directly from the battery. The zip cord terminates in Anderson Powerpoles (of course) under the driver seat, mated to the end of the cable that came with the radio. That cable uses craptastic Molex connectors (equally of course) that are instantly suspect when any problems arise, plus a pair of smaller in-line 3AG glass fuses.

    Voltage at the Molex connectors: anything from 4.8 V to 11.9 V, depending on imponderable factors. Voltage at the Powerpoles: ditto. So maybe it’s not the Molex connectors, after all.

    The 40 A fuses are the kind the high-power automotive sound system folks use, complete with gratuitous goldish-plated everything. These I got surplus at a minute fraction of sticker price and mounted on the air filter housing, thusly:

    Engine compartment fuses for radio power
    Engine compartment fuses for radio power

    I plugged a 12 V bulb in place of the radio, then went a-measuring. Voltage downstream of the hot fuse: 0 V. Tah-dah, it’s a bad fuse!

    Nope, the fuse element is intact.

    The zip cord terminates in ferrules penetrated by 1/8-inch setscrews. Applying a wrench, I find that the setscrews are somewhat loose, although nothing catastrophic. Tighten all four screws and the radio turns on just fine.

    Case closed!

    Until the next day, when the radio doesn’t turn on. Reinstall the lamp, re-measure, once again find 0 V downstream of the hot fuse.

    Pull the fuse out again and it comes apart in my hand.

    Defective 40A fuse
    Defective 40A fuse

    Huh. That would explain everything.

    I suspect the fuse was marginally defective from the factory and finally failed after that prolonged heat wave. Living in the engine compartment isn’t easy under the best of circumstances, so I’ll give this one a pass.

    Being that sort of bear, I plucked a spare fuse from the ziplock baggie of fuses & bulbs that’s tucked into the van’s jack compartment, popped it in place, and the radio works fine again.

    Problem solved, for sure!

    Side note: those fuseholder screws go through the air filter housing, into nuts with Loctite, and I ruined the threads to absolutely prevent the nuts from coming off. You really don’t want a nut loose inside the engine air intake, downstream of the air filter and upstream of the throttle…

  • Silver-Soldering a Stainless Steel Measuring Cup

    Quite some years ago, the spot weld holding half of the handle to the side of my all-time-favorite 1/3-cup measuring cup broke loose. The minuscule weld nugget suggested that the spot welder got distracted; the weld on other side of the handle is perfectly bonded.

    I tried repairing it with silver solder and a torch, which simply proved that’s not within my skill set. I buffed off most of the residue and applied JB Weld epoxy, which lasted just fine until a few days ago. It’s a low-stress situation, indeed, but I’m not surprised that the epoxy didn’t really bond to a slightly scuffed stainless steel surface.

    So, this time around, I did it right: sandpapered off the epoxy, scuffed up the cup and handle by shoe-shining a sandpaper strip face up and face down in the gap, then silver-soldered the handle in place using my resistance soldering gadget (which I promise to describe some day).

    The setup was straightforward. Clamp the cup in the bench vise with soft copper jaws (hammered from ordinary pipe) that also grip one electrode from the soldering unit.

    Silver-soldered handle - left side
    Silver-soldered handle – left side
    Silver-soldered handle - right side
    Silver-soldered handle – right side

    I used a strip of fancy Brownell’s Silvalloy 355 silver solder ribbon (which is 56% silver instead of the chintzy 4% junk I normally use) with some truly toxic silver solder flux. About ten seconds of heat melted the solder and produced a pair of nice fillets along the sides of the handle.

    The flux washed off in hot water and a few licks with fine sandpaper cleaned things up just fine. The carbon electrode left a bit of schmutz on the handle, but nothing a Dremel brass brush wheel couldn’t solve.

    The inside has a bit of heat discoloration, but the sandpaper knocked that back reasonably well, too.

    Heat discoloration inside cup
    Heat discoloration inside cup

    The final product looked a lot better than these in-work pictures, but it’s tough to photograph subtle differences in a shiny silver object.

    Anyhow, as you might expect, we value function over form in this household.

  • Dry Ice Sublimation Rate

    For reasons I’m not at liberty to discuss, we had a cooler of dry ice pellets in the freezer for a few days. I used about a pound of it a time to mumble.

    I started with “10 pounds” of dry ice in a half-pound Styrofoam container with 1.5-inch thick walls; the total weights include the container. For what it’s worth, dry ice costs $3.50/pound under 10 pounds, then $2.75/pound over that. It used to be plenty cheaper in the old days, evidently, but everything else was, too.

    In between withdrawals, the cooler sat in the freezer and and the dry ice quietly sublimated; here’s how the weight varied between uses.

    Starting weight: 9.2 lb gross, so I lost quite a bit in transit. Which, as it happened, was about half an hour in a bike trailer during a rather hot afternoon.

    A) 7.2 to 6.7 lb -> 0.5 lb / 15 hr = 0.033 lb/hr

    B) 3.8 to 3.0 lb -> 0.8 lb / 11 hr = 0.072 lb/hr

    C) 2.7 lb to 2.0 lb -> 0.7 lb / 11 hr = 0.064 lb/hr

    I’m suspicious of that low number for the first stay, too. Maybe a side effect of having the cooler’s cavity nearly full of dry ice? Or the freezer ran defrost cycles for the other two?

    Anyhow, to a back-of-the-envelope resolution the cooler loses a bit over 0.05 lb/hour of dry ice. Call it 15 hr/lb.

    The temperature of sublimation is, according to Wikipedia, -109°F. The freezer is around 0°F: a differential of 109°F across 1.5 inches of Styrofoam. Assuming the cooler foam has R=4 with units of (ft^2·hr·°F) / (BTU·in ) and an internal surface area of 304 in^2, the cooler leaks heat at 38 BTU/hr. Call it 11 W.

    Cross check: Wikipedia says the enthalpy of sublimation at STP is 571 kJ/kg. Sublimating 0.07 lb = 0.031 kg requires 18 kJ (18 kW·s) and doing that over the course of an hour requires 5 W.

    Well, considering the rough-and-ready measurements and the fact that the freezer isn’t at STP and that I’m ignoring gas leakage and a bunch of other stuff, a factor of two error is spot on.

    If I were you, though, I’d double-check those calculations before leaping to any particular conclusions. Fair enough?

    When all was said and done, I found this thing in the bottom of the cooler. It wasn’t there when we started, soooo…

    Dry Ice Thing
    Dry Ice Thing

  • Stopping Bike Helmet Strap Creepage

    My bike helmet sports a mirror, microphone boom, and earbud, so I generally hang it from the top of the seat on my Tour Easy. There’s a convenient peg seemingly made for capturing the triangle of strap that normally goes over my ear and, up to the point where I set up this helmet, everything was good.

    Helmet hanging on Tour Easy seat
    Helmet hanging on Tour Easy seat

    After about a week, I noticed that the buckle was grossly off-center under my chin: the straps had shifted to one side.

    Come to find out that the front strap on this helmet passes through an opening across the central member, below the plastic covering. Judging from the teardown of an older helmet, Bell used double-stick tape to hold the strap in place. Applying a constant force in one direction (I’m a creature of habit, the helmet always hangs from its right-side triangle) gently pulls the strap through the passage.

    Front strap passing through helmet
    Front strap passing through helmet

    So I cut two slabs of closed-cell foam and jammed them into the opening atop the strap, one from each side, with a screwdriver. That forced the strap against the adhesive and mechanically wedged it in place.

    Problem solved!

  • Chipmunks Discover Agriculture

    Chipmunk atop sunflower
    Chipmunk atop sunflower

    One of the sunflowers in the garden started swaying wildly, despite having no breeze at all. I though it was a goldfinch plucking seeds, but a quick look through the binoculars showed a brown furry tail hanging below the topmost seed head, about five feet off the ground.

    Mary reports that this is one of five sunflower plants growing in a tight group near the garden fence; she thinks a chipmunk’s seed stash sprouted. This could be the start of something big: next thing you know, they’ll be planting seeds and harvesting crops!

    As we watched, the critter’s cheeks became more and more distended.

    Chipmunk stuffing cheek pouches
    Chipmunk stuffing cheek pouches

    Eventually, however, gluttony overcame common sense.

    Getting the last seed
    Getting the last seed

    Seconds after the shutter clicked, the sunflower head disintegrated, depositing the chipmunk on the ground with a rustle and a soft plop.

    No damage done, we’re sure, and that critter’s pantry should be stuffed full in short order. Next year Mary will probably do some extensive sunflower culling to get room for the rest of her crops!

    And, yes, the title is a riff on Bears Discover Fire…

  • APRS SmartBeaconing Parameters for Bicycling

    Setting relatively prime beacon times for the GPS-to-APRS trackers on our three bikes worked quite well, but I wondered how much better SmartBeaconing would be. The trick is getting the numbers right for typical bicycling speeds.

    Here’s some settings (from the TinyTrack3+ config program display) that seem to work reasonably well…

    SmartBeaconing Parameters
    SmartBeaconing Parameters

    The general idea is to beacon every 10 minutes at rest and about three / mile in motion.

    The only time I hit 3 MPH is up a really nasty hill, the likes of which I avoid with all due diligence. On the other end, 24 MPH is pretty much as fast as I can go for any length of time; faster, certainly, on downhills, but those are rare & precious commodities on most rides around here. The Slow and Fast parameters control both ends of that range. The beacon rate increases linearly below the Fast speed: 180 seconds at 12 MPH, which is roughly what I used for the constant-time setting.

    Note that the Rate parameters are actually periods. Rate is thing/time, period is time/thing. The period varies as 1/speed, while the rate varies directly with speed. See the SmartBeaconing writeup or the TinyTrak3+ doc for the algorithm.

    The Turn Slope parameter is the most confusing. It has units of degrees/MPH degree·MPH and serves to modify the Min Turn Angle so that you must turn more sharply at lower speeds to generate a beacon. This works better for vehicles with a wider dynamic range: our bikes tend to stay within 5-20 mph and a factor-of-four doesn’t affect the basic angle very much at all.

    My track through a residential area shows pretty good “Corner Pegging” for those settings and, in any event, it’s much better than the simple every-three-minutes beaconing I’d been using before. On the other hand, this is in a low-RF-traffic area with a digipeater about a mile away across the Northway, so very few packets get clobbered.

    APRS Track with SmartBeaconing
    APRS Track with SmartBeaconing

    Perhaps setting Turn Slope to 240 degrees/MPH degree·MPH with a Fast Speed of 24 MPH and a Turn Angle of 10 degrees would be slightly better. At top speed the minimum turn angle would be 10 + 240/24 = 20 degrees and nose-pickin’ speed relaxes the angle to 10 + 240/6 = 50 degrees. On the other hand, that track looks pretty good as-is!

    One problem with three bikes in close proximity (the track above is just me) is that we’ll all be turning at about the same time and, thus, sending beacons almost simultaneously. This will take a while to sort out, given that many beacons never make it to a receiver…

    [Update: A correction shows why the units aren’t what I expected.]

  • Getting More Clearance While Bicycling

    Some cyclists complain that motorists don’t give them enough room while passing. That’s less of a problem for recumbent bikes, but this gets me a lot more clearance:

    Bike trailer with propane tank
    Bike trailer with propane tank

    There’s one section of very nice and totally gratuitous 6-lane highway (NY Route 55 near the NYS DOT Region 8 HQ; I think they’re just showing off) where drivers normally edge over to the left side of the right-hand lane where I’m riding. With a 20-lb propane tank lashed to my bike trailer, most folks have no trouble whatsoever with a double lane change into the far left lane…