Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
“Our” pair of Cooper’s Hawks (or their descendants, of which there have been many) hatched a pair of chicks that recently fledged and have been exploring their world:
New Hawks – standing tall
Sometimes they perch together:
New Hawks – companions
Their world contains many interesting things, not all of which are visible to the human eye:
New Hawks – curiosity
I’ve spotted a parent hawk circling high overhead while the youngsters practice their flight skills near the treetops. If you listen carefully, you can hear a hawk calling from far above:
New Hawks – parent overhead
We’ve seen them hopping from branch to branch, testing their wings, and by now they can launch from a standing start:
New Hawks – liftoff
New squirrels emerge at about the same time, with equivalent levels of experience:
New Hawks – curious squirrel
Right out of the nest, new hawks know what to do, if not quite how to accomplish it:
New Hawks – vs New Squirrel
That little squirrel instantly pasted itself to the bottom of the branch and escaped. This time, anyway.
Mary watched one hawk practicing its pouncing skills by attacking a pine cone. A talon wedged under a tight pine cone scale, to the extent that the hawk spent the next half hour flopping around the yard trying to part company with its personal Pine Tar Baby.
Perhaps the piles of Chipmunk Gibbage came from a new hawk practicing its regurgitation skills …
Go, new hawks, go!
Taken with the Sony DSC-H5, sometimes with the 1.7x teleadapter, under ambient light, hand-held, sometimes braced against the frame of a partially open door.
The Little Machine Shop 5200 lathe package includes DROs on the cross slide and compound cranks. The readouts report the position of the crank, not the slide position, which isn’t a major problem on a lathe.
Unfortunately, the compound collides with the DRO on the cross slide:
LMS Mini-lathe – compound vs DRO
That is a major problem on a lathe.
When you can’t turn the cross slide more than 45° from parallel with the bed, you cannot set the compound to the (typical) 29° degrees required for (traditional) thread cutting. That’s measured perpendicular to the bed, so it would be 61° on the compound rest scale, if the scale went that high:
LMS Mini-lathe – compound way
This mess doesn’t have a trivial fix, because the DRO body under the (non-removable) display doesn’t quite clear the compound screw:
LMS Mini-lathe – compound vs DRO – bottom
As nearly as I can tell, removing the entire DRO is the only way to slew the compound beyond 45°, but the DRO replaced the usual manual scale around the cross slide knob, so there’s no analog backup.
The DRO mounts to the cross slide with three screws, so you can’t rotate it 90° to the side to get better clearance:
LMS mini-lathe – DRO mounting screws
The other four screws presumably mount the DRO encoder housing to the outer shell.
The setscrew sticking up from the sleeve anchors it to the cross slide shaft. The slit milled into the shaft captures the end of the setscrew:
LMS mini-lathe – cross slide leadscrew shaft
The knob slides over the shaft, with a screw in the end holding it in place by friction against a split lockwasher; you can apply enough torque to turn the knob under the lockwasher in either direction.
Removing the DRO doesn’t produce more cross slide travel, because the DRO body sits flush with the back side of that large disk.
I think the cross slide knob collides with the compound DRO, but I put it all back together without any further exploration.
Actual 6 inch DROs based on linear encoders seem to run $40-ish and other folks have fitted them to their mini-lathes. Verily, I don’t do much threadcutting, so I’ll just put this mess on the far back burner.
That DRO ticks me off every time I look at it, though…
I reported this short cycle time to the Dutchess County DPW, back in September 2015, and got this response:
Thank you for contacting Dutchess County DPW about this matter. I will ask our Traffic Engineer to review the signal timing at CR 104/CR110 (New Hackensak/Jackson) to see if adjustments can be made. The primary factors used to set the current signal timing are operational efficiency, safety and Level of Service for motor vehicles. If there are signal timing adjustments which can achieve these goals and provide for safer passage of bicycles, we will explore those options. I will also ask our Traffic Engineer to investigate altering the sensitivity of the detector system to detect bicycles. I will share our findings with you. Thank you again for your comments.
Robert H. Balkind, P.E.
Deputy Commissioner
Dutchess County Department of Public Works
Emphasis mine, of course. Translation: “It’s not going to happen.”
I pinged him a few weeks later:
That review has not been done yet. I will advise you when our investigation is complete.
So, here’s what the signal timing looks like these days…
T = 0.000 s – Green
Jackson at New Hackensack – Signal Timing 2016-06-27 – 0218
On the positive side, a pair of big long wheelbase recumbents stopped in the middle of the lane seem sufficient to trigger the traffic detector!
T = 3.150 s – Yellow
Jackson at New Hackensack – Signal Timing 2016-06-27 – 0407
We’re definitely not fast enough off the block; Mary thought she had time for a sip of water. We started rolling less than two seconds after the green appeared, which is as fast as one should enter an intersection around here.
T = 8.000 s – Red
Jackson at New Hackensack – Signal Timing 2016-06-27 – 0698
T = 13.000 s – Opposing Green
Jackson at New Hackensack – Signal Timing 2016-06-27 – 13 s – rear
In round numbers, it takes us about 15 seconds from a cold start to reach the far side of that intersection. I can do it in a bit less, but Mary can’t, even though we’re in pretty good shape for the shape we’re in.
Our standard dishwasher loadout changed a while back, so I ran off more protectors to fill the bottom rack. The crystalline look of natural PETG is probably wasted in there, even though it puts the old, rather yellowed, PLA protectors to shame:
Dishwasher Rack Protectors – old PLA new PETG
Dollops of silicone sealant hold them in place: the bigger the blob, the better the job.
We don’t activate the drying heater, so the plastic doesn’t get exposed to absurdly high temperatures. As nearly as I can tell, those PLA protectors remain in fine physical condition, even though they’re turning an odd color.
The support structures peeled out easily with a fingernail pull:
Dishwasher Rack Protectors – 0.20 mm PETG bridging – detail
PETG doesn’t bridge well, as shown by the gaps between the support ridges. Those 0.20 mm layers seemed skimpy for lightly supported PETG, so I ran another set at 0.25 mm:
Dishwasher Rack Protectors – 0.25 mm PETG bridging – detail
Not quite enough improvement for a Happy Dance, although fine for the application.
We look forward to seeing what grows in those little crevices…
It turns out that the dual-core Intel Atom Inside an old Dell Mini 10 isn’t up to the demands of rendering modern web design; disk I/O speed has nothing to do with the CPU’s (lack of) ability to chew through multiple layers of cruft adorning what used to be straightforward static HTML.
So, equipped with Linux Mint / XFCE, it’s now found a new purpose in life:
SnowWhite back in action
In truth, an Atom isn’t quite up to the demands of modern 3D printing, either, at least in terms of processing a huge G-Code file into a layer-by-layer path preview. Fortunately, Pronterface doesn’t generate the preview until you ask for it: arranging the UI to put the preview on a separate tab eliminates that problem.
The Mini 10 can dribble G-Code into the printer just fine and looks much cuter than the hulking laptop in the background.
The support holding the two big drawers below the bottom shelf of our long-suffering Whirlpool refrigerator broke off. Having previously repaired and then replaced the tab holding the strut in place, then added metalskid plates to the bearing surfaces, I’m getting pretty good at fighting this particular bit of entropy to a standstill:
Refrigerator strut – clamped glue joint
Adding a few more clamps always make me feel good:
Refrigerator strut – many clamps
Although a good solvent-bond joint should be as strong as the original plastic, that’s not saying much: I expect the end of that strut will break off again. Perhaps the central web is wide enough for a few small screws?
The Smell of Molten Projects in the Morning 