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.
Cruisin’ the streets
A snapping turtle headed toward the beaver pond on the Dutchess County Rail Trail:
At this time of year and phase of the moon, she is most likely in search of a good spot for a nest and her clutch of eggs. Being an aquatic creature, she and her progeny surely benefit from Team Beaver’s engineering.
Today I Learned: snappers are the New York State Official Reptile.
In the process of installing a Bafang BBS02 mid-drive motor on a friend’s diamond-frame bike, I discovered, once again, how little anybody cares about the colors inside cables:
The cheerful rainbow on the right is the stub end of the Bafang display extension cable I built into the previous adapter.
The new cable on the left seemed like it might match the canonical colors:
It comes heartbreakingly close:
Brown and Orange connect as the naive user might expect, which does reduce the likelihood of incinerating the motor controller / USB adapter / laptop by connecting the 48 V battery directly to the logic-level electronics.
However, White wasn’t on the original menu, Green is now TXD, and Black has become, comfortingly, GND.
Verily, it is written: Hell hath no fury like that of an unjustified assumption.
This socket connector has a watertight shell making it extremely difficult to mate and unmate with the pin connector on the bike. Watertightness being unnecessary, a little razor-knife action seems in order:
Visually, they’re both green-ish, but sometimes the Pixel camera accentuates any differences.
Last June, NYS DOT surprised us by clearing the Japanese Knotweed along Rt 376 just north of Maloney, then applying enough defoliant to keep it knocked back this Spring:
A year earlier, they clearcut the overgrowth beyond the guide rail from Red Oaks Mill southward, but without defoliant, and the Knotweed is off to a good start:
I’ll do my part, inadequate though my pruners may be:
I got most of the growth in front of the guide rail extending across the shoulder, but must wait for another weekend morning to hack back the main stems.
Unfortunately, Knotweed control requires nearly continuous clearcutting and defoliation to prevent new growth.
We have just started rolling from Overocker Road and the traffic signal on Burnett Blvd at Rt 55 (on the far left) has just turned green for the single car on the sensor loop:
Much to our surprise, 17 s later the signal is still green:
As usual, the unmarked sensor loop doesn’t detect bicycles and the control doesn’t take our clearing time into account, so the signal turns yellow 5 s later (after 22 s from turning green) while we’re still in the intersection:
After another 6 s, though, we’re through the intersection and lined up on the right side of Rt 55, just as the Rt 55 signal turns green:
Note that the Burnett Blvd signal remained green for 22 s, much longer than in bygone years, and the green-to-green time is now 28 s. We got through the intersection without any difficulty, although the green-to-red clearance time remains scanty.
Those of long memory may recall my writeup of the timing in early November last year. That was with many cars triggering the sensor loops, so the timings from a trip last July with a single truck-and-trailer tripping the sensor may be more relevant. Or take your pick from other timings done during the last six years; there’s plenty of data to show something’s new and different.
Mary recently discovered a reason why NYS DOT may have suddenly changed the signal timing at the Burnett intersection after all those years:
During the incident, a black Nissan Titan, driven by a 51-year-old male resident of Lagrangeville, collided with a bicycle, ridden by a 58-year-old male resident of Poughkeepsie, in the area of the crosswalk on the southeast portion of the intersection, said the Town of Poughkeepsie Police.
The bicyclist sustained serious injuries and was transported to MidHudson Regional Hospital.
https://dailyvoice.com/new-york/putnam/police-fire/bicyclist-seriously-injured-after-crash-with-vehicle-in-area/798453/
The crosswalk mentioned in the article appears in the last picture.
The cyclist died of his injuries shortly after that article went live.
Mary knew him. He was one of the gardeners near her plot in the Vassar Community Garden who lived in the apartments a few hundred yards from that intersection, didn’t own a car, and, for years, rode through that intersection to the grocery store at the far end of Burnett Blvd (across another of DOT’s intersections). Everyone knew him as a nice, considerate guy.
When DOT tells you “Clearance times are determined based on speed, intersection dimensions, grade, and reaction time and cannot be adjusted” they don’t add “Because not enough people have died to get our attention.”
Death is the only thing that will convince NYS DOT’s engineers to change the signal timing at an intersection.
As far as I can tell, all of the other intersections along our usual routes still have the same inadequate clearance times. Evidently, the bicyclist death toll isn’t high enough to get their attention and evidence here doesn’t matter there, because motor vehicle traffic cannot be delayed, even for a few seconds, merely to protect the most vulnerable “users” of their facilities.
We’ve been bicycling all our adult lives and haven’t been killed yet, despite NYS DOT’s complete lack of attention. Our experiences justify my cynicism and bitterness.
I eventually figured out why no NYS DOT staffer will accompany me on bike trips along their “safe for all users” roads. If they did, they’d be unable to deny knowing how hazardous their engineering designs & maintenance practices are in real life, should the question come up in a court of law.
If you think that’s not the case, then let’s go riding together …
Road design, build quality, and attention to details matter, even though drivers and, yes, cyclists share some of the blame.
We’re riding southbound on the recently opened section of the Empire State Trail, just south of Hopewell Junction, and are approaching a dog walker totally face-sucked by her phone in the middle of the path:
Mary has been dinging her bike bell for the last few seconds and finally manages to break through:
The dog walker leans against the fence while pulling on the leash as hard as she can, as if she knows the dog poses a threat:
Which it does:
The leash is too long for close-quarters work:
Nice teeth, doggie:
Surely, the dog just wants to lick me to death:
Tell me again how well-trained this dog is:
Seven seconds after the first picture:
The dog also lunged at the pair of bicyclists following us, so perhaps this is how she makes sure it get its exercise during the walk.
As mentioned earlier, the Bafang brake sensors on Mary’s Tour Easy require a magnet on the brake levers to activate the switches. They arrived with disk magnets that did not suit the levers, so I used neodymium “bar magnets”:
That worked for a few rides, but the alignment turned out to be entirely too critical, because the magnetization is through the bar’s thin dimension, rather than along its length, making the field weakest in the direction of the switch.
Magnetic field visualization film shows the field null along the thin edge of the bar:
That’s a slightly shorter magnet from a different toothbrush head, cemented edgewise into a holder conjured from the vasty digital deep:
The field is much more uniform on the flat side of the bar:
Some double-sided foam tape snuggles the sensor and the magnet together on the brake lever:
I coated the magnet with JB Plastic Bonder urethane adhesive in the hope of filling any gaps in its nickel coating caused while extricating it from the toothbrush head.
The rusty screw head in the upper right positions the lever at the proper distance from the grip to suit Mary’s hand. An earlier version of the holder shows the alignment:
The switch trips (opens) with the lever roughly parallel to the grip, again with the earlier holder:
A detailed view of the gap with the lever at the tripped position:
The levers have enough travel to prevent accidental trips due to light finger pressure, which turned out to be a problem with the original end-on alignment.
The brake pads don’t quite touch the rim when the switch trips, so the motor has plenty of time to shut off before the brakes take effect. It also stops when the pedals stop turning, so we should not see any disagreement between motor and brakes as to the bike’s momentum.
The wider base on the new mounts makes them much more stable on the levers, although I don’t like having them stick up so far. Mounting everything underneath the levers would look better, but any problems will be more obvious with everything in plain sight.
I may affix the magnets directly to the levers with Plastic Bonder if the foam tape doesn’t live up to its reputation. Removing them would be more challenging; a shot with a small chisel should suffice.
After nearly two and a half years, this happened:
Yup, it broke just where I expected!
The front fender on Mary’s bike suffers a bit more stress than you might expect, as she must wheel it through high grass to her Vassar Farms garden plot and the low-hanging spray flap can snag on the taller greenery.
Re-slicing the original model, printing the result, and installing it took about an hour:
Affixing the strut with duct tape and a cable tie looks déclassé, but continues to work better than anything else I’ve tried: simple, flexible, easily readjusted, totally nonfussy.
At least I now use black outdoor-rated double-stick foam tape, so life is increasingly good …
The Smell of Molten Projects in the Morning 