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.
Mechanical widgetry
The Bafang battery charger uses an AC line cord “binocular” connector with what must be the weakest spring contacts ever made, which finally annoyed me enough to fix:
Also, the case now sports four thick fuzzy felt feet to keep it from sliding around quite so easily.
Another customer-does-the-last-ten-percent product …
With the Bafang BBS02 and all its gimcrackery on the Terry Symmetry buttoned up and ready to go, I took a few closeout pictures for future reference.
The motor has a sheaf of wires sticking out of the bottom crying out for a protective covering:
Although cameras tell only the truth they’re allowed to see and can be made to lie by omission, sometimes their latent truth was completely invisible to eyewitnesses in real time.
I only noticed the mis-routed shift cable when I looked through the last set of pictures.
It should pass through the plastic channel under the metal tab holding the cable guide to the bottom bracket shell:
Normally, aiming the cable into the channel is no big deal. In this case, I had to undo the shift cable, remove the left crank, loosen the motor and rotate it out of the way, nudge the cable with a small screwdriver, then reinstall in reverse order.
Dang, that was close …
The original BBS02 reaction spacer for Gee’s Terry Symmetry didn’t work quite the way I expected:
The motor evidently vibrates enough to propel the block forward, shearing the double-sticky foam tape which was never intended to resist force in that plane. I thought the block was located at the point where the motor casing was tangent to the frame tube, so as to equalize the forces in both directions, but … nope.
A revised design based on measurements informed by new knowledge:
The upper curve is now symmetric and the whole block mounts more rearward under the bottom bracket lug, where some tedious work with a machinists square located the real tangent point:
The motor sure doesn’t look like it’s tangent, but a dry fit showed all the curves laid against the case and tubes.
The brazing fillet means the step fitting the downtube can’t sit snug against the edge of the lug, but most of the reaction force should go through the section into the lug, near the center of the block.
A crude marker will keep track of any motion:
I think the symmetric curve against the motor has enough projection to keep the block from wandering off, even if I haven’t gotten the location exactly right.
Stipulated: Hope is not a strategy.
The OpenSCAD source code:
MotorOD = 111; // motor frame dia
MotorOffset = 10.0; // motor OD tangent wrt lug edge
ShiftSpace = 6.0; // motor to frame space
LugLength = 25.0; // length of section over BB lug
Spacer = [5.0 + LugLength,DownTube[ID]/2,4*ShiftSpace];
SpaceAngle = 0*atan(1.8/Spacer.x); // tilt due to non-right-angle meeting
echo(str("Spacer angle: ",SpaceAngle));
module MotorSpacer() {
difference() {
translate([LugLength - Spacer.x/2,0,0])
cube(Spacer,center=true);
translate([0,0,DownTube[ID]/2])
rotate([0,90 + SpaceAngle,0]) rotate(180/FrameSides)
cylinder(d=DownTube[ID],h=DownTube[LENGTH],$fn=FrameSides,center=true);
translate([DownTube[LENGTH]/2,0,DownTube[ID]/2 - DownTube[LENGTH]*sin(SpaceAngle)/2]) // concentric with ID
rotate([0,90 + SpaceAngle,0]) rotate(180/FrameSides)
cylinder(d=DownTube[OD],h=DownTube[LENGTH],$fn=FrameSides,center=true);
translate([MotorOffset,0,-(MotorOD/2 + ShiftSpace)])
rotate([90,0,0]) rotate(180/48)
cylinder(d=MotorOD,h=2*Spacer.y,$fn=48,center=true);
}
}
Nothing like actual riding to reveal what needs more thought!
With a generous dollop of JB Plastic Bonder left over from a set of Bafang brake sensor magnets, I tried coating the ersatz plate cap of a triode tube:
That’s the result after leaving it hanging upside-down while it cured to push all the drips to the top.
For comparison, the uncoated cap back in the day:
Seeing as how the urethane is an adhesive, not a coating, I’d say it looks about as bad as expected.
As with all 3D printed things, one must embrace imperfections and striations, rather than endlessly strive for perfection.
Now, if I had a resin printer …
The truck side marker lights I’m thinking of using as daytime running lights have a pentagonal lens, so they should have a pattern with a bright central beam surrounded by five lobes. The one on Mary’s Tour Easy produced an oddly shaped blotch on the garage wall, so I ran the others though a simple test setup:
The lights sit horizontally in a small vise to keep them level and in the same position, although in no particular rotational orientation, and 100 mm from the graph paper. It’s running at 6 v to keep the brightness down enough to avoid blowing out the image. All of the images were exposed based on the central spot, so the surrounding paper gives some idea of the relative brightness: darker paper = brighter LED spot.
The front view of the lights comes from the stereo zoom microscope, with the wires gripped in a Third Hand and rotated to put the (inverted) TOP label where you’d expect it. They’re all roughly at the same position and pretty nearly lined up with the lens axis. The bubble-looking thing behind the central pentagon is the lens on the Piranha LED package, which should be centered but rarely is. You can see the dark orange square of the amber LED chip in some of the pictures.
Without further ado, the nine truck side marker lights that aren’t on her bike:
Side Marker E has a blob that looks like a cataract atop the LED lens, but it might be a mold imperfection.
Obviously, paying a buck a light doesn’t get you much in the way of build quality these days.
Having seen a few bikes with amber “headlights” and being desirous of reducing the number of batteries on Mary’s bike, this seems like an obvious first step:
It descends from the fairing flashlight mount with an entry to suit a 15 mm truck side marker body:
LightBodies = [
["AnkerLC90",26.6,48.0],
["AnkerLC40",26.6,55.0],
["J5TactV2",25.0,30.0],
["InnovaX5",22.0,55.0],
["Sidemarker",15.0,20.0],
["Laser",10.0,30.0],
];
The rest of the code gets a few cleanups you’d expect when you compile code untouched for a few years using the latest OpenSCAD.
The markers are allegedly DOT rated, which matters not for my use case: SAEP2PCDOT.
The mount is grossly overqualified for a wide-beam light with little need for aiming:
Eventually, the marker should slip into a prealigned cylindrical holder, with a dab of epoxy to keep it there.
The lights are a buck apiece, so there’s no reason to form a deep emotional attachment. They are the usual poorly molded and badly assembled crap, although the next step up from a nominally reputable supplier is a factor of five more expensive.
It’s generated for the left side of the fairing, although I think having a pair of them would improve conspicuity:
Being automotive, it runs from a 12 V supply, which comes from a boost converter driven by the Bafang 6 V headlight output. The absurdity of bucking a 48 V lithium battery to a 6V switched headlight output, then boosting it to 12 V to drive a single amber LED with a 1.5 V forward drop does not escape me.
It’s possible to slice the lens off (using a lathe), remove / replace the resistor, then glue it back together, which would be worthwhile if you were intending to drive it from, say, an Arduino-ish microcontroller to get a unique blink pattern.
Given the overall lack of build quality, it might make more sense to slap a condenser lens in front of a Piranha LED.
Bonus: contrary to what you (well, I) might expect, the black lead is positive and the white lead is negative.
Although Gee’s Terry Symmetry is sized for female bodies, I managed to ride it up and down the driveway while watching the power display:
| Voltage | 52.5 | ||||
| Rated Current | 24 | ||||
| Max current | 18 | ||||
| Power | Power | ||||
| PAS | Assist | Amp | Calc | Observed | Ratio |
| 0 | 0% | 0.0 | 0 | 0 | ~ |
| 1 | 4% | 0.7 | 38 | 26 | 69% |
| 2 | 6% | 1.1 | 57 | 52 | 92% |
| 3 | 9% | 1.6 | 85 | 78 | 92% |
| 4 | 13% | 2.3 | 123 | 104 | 85% |
| 5 | 20% | 3.6 | 189 | 182 | 96% |
| 6 | 30% | 5.4 | 284 | 258 | 91% |
| 7 | 50% | 9.0 | 473 | 453 | 96% |
| 8 | 85% | 15.3 | 803 | 675 | 84% |
| 9 | 100% | 18.0 | 945 | 900 | 95% |
The variations in the last column suggest my data-taking is … wobbly, at best.
I think the displayed power does not come from actual current and voltage measurements, because recalculating the power using the nominal 48 V battery value produces an unnatural agreement:
| Voltage | 48 | ||||
| Rated Current | 24 | ||||
| Max current | 18 | ||||
| Power | Power | ||||
| PAS | Assist | Amp | Calc | Observed | Ratio |
| 0 | 0% | 0.0 | 0 | 0 | ~ |
| 1 | 4% | 0.7 | 35 | 26 | 75% |
| 2 | 6% | 1.1 | 52 | 52 | 100% |
| 3 | 9% | 1.6 | 78 | 78 | 100% |
| 4 | 13% | 2.3 | 112 | 104 | 93% |
| 5 | 20% | 3.6 | 173 | 182 | 105% |
| 6 | 30% | 5.4 | 259 | 258 | 100% |
| 7 | 50% | 9.0 | 432 | 453 | 105% |
| 8 | 85% | 15.3 | 734 | 675 | 92% |
| 9 | 100% | 18.0 | 864 | 900 | 104% |
The motor controller may measure the actual winding currents while generating the BLDC waveforms, but the values may not be available to the display at the end of the cable. If Bafang documented the commands & responses, we’d know for sure, but they don’t.
Those assist values come from Mary’s Tour Easy, a much heavier bike than the Symmetry, but the first few levels work well in my limited tests. The highest levels may be too peppy for Gee’s normal routes, but having some serious boost in reserve can defang (hah) the worst hills.
IMO, the bike would burn rubber at the motor’s full 24 A current …
The Smell of Molten Projects in the Morning 