Unfortunately, US Rt 44 is the only way to get from Adams Fairacre to the Stop & Shop, so we ride this section once a week. I’m towing the trailer with two bags of groceries:
A palimpsest of patches shows that they really don’t have staying power:
Gravel and a manhole cover keep cyclists off the shoulder:
We take the lane well before we encounter this section, because veering into traffic doesn’t work well:
The longitudinal crack marks the edge of the original paving. This is a common hazard on Dutchess County roads, as many were widened by simply paving the original shoulders into the travel lane, without actually rebuilding the substructure.
Our Larval Engineer stopped by, on her way to a half-year co-op job out around Route 128, and devoted a few days to merge-sorting / triaging her possessions. Having shown her the HP 74754A plotter project, she later dropped a bag o’ stuff on my desk without comment:
The perforated pen holder stuck to the plotter case (hey, it would still fit!) in front of the carousel with a bit of foam tape on an angled bracket you can’t quite see. It held 15 pens at the ready: I really used that plotter.
The doodle on the yellow sheet sketches a bulky adapter between the spindle nose thread on the Sherline CNC mill and a plotter cartridge. The flange-less pen body might just fit into the spindle bore, but I remember concluding that machining pen bodies or adapters wasn’t worth the effort. Now it’s a simple matter of some OpenSCAD source code and a few hours of hands-off production, so perhaps I should re-think that.
No dates on anything, but I got the Sherline in 2004. The pen holder probably dates back to the late 80s, shortly after I got the plotter. Most likely, I gave her the bag o’ stuff and told her to make something interesting; it could still happen…
The central contact seems to be double-insulated from the chamber with glass (?) seals in a soldered-in-place assembly:
That might be rosin left over from soldering, but you’d think they would have rinsed it off to reduce the leakage. Some cleaning will be in order.
A picture in The Fine Manual for the CD-V-710 Model 5 Radiation Survey Meter showed that the circuit board used point-to-point wiring, with the range switch soldered directly to that bent metal contact:
Another page gave some useful values and a simplified schematic:
Never fear, the manual also has the full schematic; they don’t write manuals like that any more.
The chamber bias voltage was +22.5, from one carbon-zinc battery available back in the 1950s. You can still get 22.5 V batteries at about ten bucks a pop, but 24 V from a pair of cheap & readily available 12 V A23 alkaline batteries should be close enough. There’s no current drain, so the batteries should last their entire shelf life.
The “HI-MEG” resistor represents a trio of glass-body resistors selected by the range switch:
R5 = 100 GΩ → 0.5 R/h
R6 = 10 GΩ→ 5 R/h
R7 = 1 GΩ→ 50 R/h
As the saying goes, if you must select R7 in an actual emergency, you should sit down, put your head between your legs, and kiss your ass goodbye.
The steel-wall chamber responds only to gamma radiation, with a nominal current of 5 pA at 0.5 R/h. However, given an op amp like the LMC6081 with 10 fA bias current, maybe building an electrometer-style amplifier that can respond to background gamma radiation or maybe secondary gamma rays from cosmic ray air showers would be feasible; I haven’t done anything like that in a while and even a faceplant would be interesting.
Alas, radium-226 and its progeny, including radon-222 decay through alpha and beta emission that’s specifically excluded by the can.
The thing has absolutely no affordance for hand-holding, so perching it on the Walkway handrail 200 feet over the Hudson required tamping down my usual risk aversion.
Both images have been slightly contrast-tweaked and lightly compressed from the original data, but not by enough to matter here. Generally, I apply ruthless compression to keep the image size under control, so these look a lot better than the usual pix around here.
I picked up a pair of 125 mm OD white LED rings for the hand magnifier project from the usual eBay source, which arrived with the expected level of build quality:
But, hey, all the LEDs lit up more-or-less uniformly.
With 20 mA in each of 13 parallel strings of 3 white LEDs, the ring should draw 260 mA. It’s nominally a 12 V device sorta-kinda intended for automotive “angel eye” use, where the actual battery charging voltage runs around 14 V. The 180 Ω ballast resistors seem to be sized with that in mind:
The reciprocal of that 45.5 mA/V slope is 220 V/mA = 220 Ω, which is close enough to the actual (we presume from their marking) 180 Ω resistors for comfort.
Driving it at 14 V to get 250 mA dissipates 3.5 W and makes it pleasantly warm.
For use with a magnifying lens, I think it deserves a brightness control. Perhaps hacking a bigger trimpot with a knob onto a cheap & tiny boost converter will suffice.
Given a hint that the Sienna’s left rear ABS / speed sensor had failed, we took a look:
She removed the wheel under field conditions using only in-the-car tools for practice, with the jack stand and wheel chock because we weren’t really beside the road. It turned out that breaking The Last Lug free required bouncing her full weight on the wrench handle, which is what we expected based on previous experience.
Yes, I pointed out the inadequacy of that footwear. Yes, she loosened the lugs before jacking the van.
With the van up, the first look showed the ABS diagnostic blink code was dead on:
That bit of tubing in her fingers should contain a pair of wires, which was a bit of a puzzle.
The connector remained snapped onto the sensor head, but the whole affair came out easily enough:
We thought those wires seemed very tightly twisted, too. I guessed that a clip holding the sensor head in place had gone missing, allowing it to rotate in place.
Which was partially true, as the “missing” wires were very very very tightly twisted inside that flexible tubing and, thus, much shorter than they should be:
Lining up the removable parts:
The sensor head should be firmly glued onto the back of the wheel hub, with no clips or screws holding it in place, as we found by comparing it with the right rear wheel. That slightly rough gray ring just outside of the central cylinder was the adhesive…
She soldered longer wires to the pigtails on the connector and applied heatshrink. The hyper-twisted wires under the car got un-twisted a bit, straightened, cleaned up, then rejoined to the connector with pair of gel-filled beanie compression splices and more tubing to ease the strain.
We buttered up the sensor head flange with JB Kwik epoxy, squished it back in place for a good seal, spun the hub to make sure the sensor fingers weren’t hitting anything, then she practiced ten minutes of meditation while holding it in place and awaiting a firm set.
It turns out that the sensor head is not a replaceable part: it’s factory-bonded to the back of the hub and should never, ever come loose. Given that this one had made maybe a dozen orbits and was finger-loose in the back of the hub, with some dust & crud visible inside the hub where it shouldn’t be, replacing the wheel hub is in the plan.
Also, we still don’t know why different versions of “the same cable” have such a huge price difference; despite their sensor attribute, they definitely don’t include the sensor head.
After repairing the cable, she put the wheel back in place, reset the ABS codes, drove the van around the block, found a patch of sand to check out the ABS braking, and reported normal operation.
We’ll replace both the cable and hub, then declare victory.