Posts Tagged Improvements
Tek Circuit Computer: Cursor Fixture Adhesion
Posted by Ed in Machine Shop on 2020-04-03
After removing debris, flattening the top surface, and generally paying more attention to detail, the PETG sheet has much better adhesion to the fixture:
This time, I traced the inside of a drag-knife cut cursor to extract the blank from the stock and, yes, used new double-sided tape under the lower white protective film on the PETG.
Fewer air bubbles means better adhesion:
Spinning the 1/8 inch end mill at about 5000 RPM produced finer swarf at the Sherline’s maximum 609 mm/min = 24 inch/min pace, with less uplift. I suspect Moah RPMs! would be even better, constrained by melting the plastic into heartache & confusion.
Scribe the hairline with the diamond tool, ease the finished cursor off the fixture, scribble Sharpie into the scratch, and wipe
It’s Pretty Good™ when seen against an un-laminated bottom deck drawn with a Pilot V5RT pen:
The diamond point tears a slightly gritty path through the PETG, which then looks a bit more granular than a real hairline. I’ve been using four passes for emphasis; perhaps fewer would be better.
Clearing The Noto Font Clutter
Posted by Ed in PC Tweakage on 2020-04-02
The Noto (“No Tofu”) font family includes nearly All. The. Languages., which is certainly a noble goal, but I’m just not ever going to need fonts like these:
./NotoSerifTelugu-Regular.ttf
./NotoSansBengali-Bold.ttf
./NotoSansGurmukhiUI-Bold.ttf
./NotoSansGurmukhi-Bold.ttf
./NotoSerifTamil-Regular.ttf
./NotoSansOriyaUI-Bold.ttf
./NotoSerifSinhala-Regular.ttf
./NotoSerifSinhala-Bold.ttf
./NotoSerifMalayalam-Bold.ttf
./NotoSansTelugu-Bold.ttf
./NotoSansAvestan-Regular.ttf
… and so forth and so on …A bit of searching & listing identified the few I might ever use, so armor those against the coming catastrophe:
cd /usr/share/fonts/truetype/noto/
sudo chmod a-w NotoMono-Regular.ttf
sudo chmod a-w NotoSans-Bold*
sudo chmod a-w NotoSansDisplay-*
sudo chmod a-w NotoSans-Italic.ttf
sudo chmod a-w NotoSansGothic-Regular.ttf
sudo chmod a-w NotoSansMono-*
sudo chmod a-w NotoSans-Regular.ttf
sudo chmod a-w NotoSansSymbols-*
sudo chmod a-w NotoSerif-Bold*
sudo chmod a-w NotoSerifDisplay-*
sudo chmod a-w NotoSerif-Italic.ttf
sudo chmod a-w NotoSerif-Regular.ttfThere seems no regex-ish way of picking those out; next time, I’ll recycle the list as a script.
With armor in place, remove the rest:
find . -perm -u=w -type f -exec sudo rm '{}' \;Rebuild the font caches:
sudo fc-cache -v -fMaybe do such things near the end of the day, when you’re going to shut down anyway, because you’ll want to restart any programs using fonts in any nontrivial way.
Making the desired fonts read-only may confuse the next update involving the Noto fonts, but this setup (Xubuntu 18.04 LTS) is getting old and maybe something else will happen when I get around to installing a whole new release.
Upconverting Baby Wipes
As other folks have discovered, it’s straightforward to convert soft, soothing baby wipes into toxic sanitizing wipes by pouring harsh chemicals down the hatch:
Ending up with the proper dilution, though, requires knowing how much liquid the wipes already have, so you can account for it in whatever recipe you’re following.
Stand back, I’m going to use arithmetic!
Gut a new package of wipes: 552 g total weight, with 80 wet wipes weighing 536 g, so the packaging amounts to 15.5 g and each wet wipe weighs 6.7 g.
Hang five wipes in the breeze for a few hours to find they weigh 9.2 g. They’re still slippery, because of all the aloe & Vitamin E & whatever else Amazon specifies for the mix, but they’re dry. One dry wipe weighs 1.8 g, so all 80 weigh 150 g.
The block o’ wet wipes holds 536 – 150 = 390 g = 390 ml of water.
Should you want a 70% (by volume) isopropyl alcohol solution, pour 0.7/0.3 × 390 ml = 910 ml of 99% alcohol into the package and let it settle for a while. Each wipe will emerge dripping wet, but that’s not entirely a Bad Thing. Perhaps it’d be a good idea to start by letting the block dry out for a while, re-weigh, then calculate the alcohol dose from the reduced amount of water.
Bleach dilutions for sanitation seem wildly varied, but the jug of 8.25% sodium hypochlorite on the shelf says 1/2 cup to a gallon, a 1:32 volume ratio. Starting with 390 ml of water-like substance in the package, pour 12 ml of bleach into the hatch, let things settle, then squish it around for good measure.
None of the dosages seem particularly critical, given the slapdash way everybody applies wipes.
You should, of course, conspicuously mark the packages, so as not to apply toxic wipes to sensitive parts of you or your baby …
Sherline: Diamond Drag Engraving Tool Holder
Posted by Ed in Machine Shop on 2020-03-26
Although I shouldn’t have used a hardened shaft for the case, the rest of the diamond drag tool holder worked out well enough:
The dimension doodle shows what’s inside and gives some idea of the sizes:
From left to right, it’s an M6×1.0 setscrew to adjust the spring preload, a spring harvested from a cheap clicky ballpoint pen, a machined cap, a 3 mm rod (which should be a hardened & ground shaft, but isn’t) surrounded by a pair of LM3UU linear bearings, a machined coupler, and the stub of a diamond engraving tool’s shank.
Tapping 15 mm of M6×1.0 thread inside of the case took an unreasonable amount of grunt. Next time, brass.
The setscrew gets a little boss to hold the spring away from the adjacent threads in the case:
The little machined cap has a somewhat longer spring guide to prevent buckling:
The spring fits snugly on the slightly enlarged section inside the last few coils, with the rest being a loose fit around the guide. When the spring is fully compressed, it’s just slightly longer than the guide and can’t buckle to either side.
The cap gets epoxied onto the 3 mm rod with some attention to proper alignment:
The other end of the rod has a 3 mm thread, which would be a serious non-starter on a hardened rod.
The shortened diamond tool shank gets epoxied into the gizmo connecting it to the now-threaded rod, again with some attention paid to having it come out nicely coaxial:
The LM3UU bearings got epoxied into the case, because I don’t have a deep emotional attachment to them.
Unscrew diamond tool, push spring onto cap, drop rod through bearings, crank setscrew more-or-less flush with the end of the case, screw diamond in place with some weak threadlock, add oil to rod, work it a few times to settle the bearings, and it’s all good.
A quick spring rate measurement setup, with a brass tube holding the diamond point off the scale pan:
The spring rate works out to 230 g + 33 g/mm for deflections between 1.0 mm (263 g) and 3.5 mm (346 g), so it’s in the same ballpark as the diamond tools on the MPCNC and CNC 3018.
Note: WordPress just “improved” their post editor, which has totally wrecked the image alignment. They’re all set to “centered” and the editor says they are, but they’re not. It’s a free blog and I’m using one of their ancient / obsolete / unsupported themes, so I must update the theme. Bleh.
GRBL Error 33: Arc Coordinates vs. Decimal Places
Posted by Ed in Machine Shop, Software on 2020-03-18
The LinuxCNC G2/G3 arc command doc has this to say about numerical precision:
When programming arcs an error due to rounding can result from using a precision of less than 4 decimal places (0.0000) for inch and less than 3 decimal places (0.000) for millimeters.
So I normally set GCMC to produce three decimal digits, because its default of eight digits seems excessive for my usual millimeter measurements, and assume whatever G-Code sender I use won’t chop digits off the end in passing. Mistakenly setting bCNC to round at two decimal places showed what happens with fewer digits, as bCNC’s default is four decimal digits.
A closer look at the coordinates in the lower right part of the spreadsheets (from yesterday) shows the limited accuracy with two decimal digits:
The red blocks mark the first failing arc, where the relative error falls out of tolerance. If GRBL were less fussy (which it should not be), then the next arcs would proceed as shown.
Rounding to three decimal digits pushes the errors into to the third place, with the yellow highlight marking the worst errors:
As you should expect, the smallest arcs have the largest relative errors, although they’re now well within GRBL’s (and LinuxCNC’s, for that matter) limits.
Rounding to four decimal digits makes the errors vanishingly small:
So, by and large, don’t scrimp on the decimal digits … but we knew that already.
I’d been setting GRBL to produce three decimal places, but now I’m using four. Adding a few characters to each G-Code command reduces the number of commands fitting into GRBL’s buffer space, but bCNC normally keeps it around 90% full, so the path planner should remain perfectly happy.
Homage Tek Circuit Computer: Yellow Variation
Posted by Ed in Machine Shop on 2020-03-12
An on-sale pack of yellow Astrobrights card stock tempted me:
The somewhat wrecked cursor comes from my collection of discards, because I haven’t yet figured out how to mill the outline and engrave the hairline on raw stock.
The paper isn’t quite the same color as my Genuine Pickett Model 110-ES circular slide rule:
Nor, of course, are the ticks and legends nearly as fine as you get with real engraving, but it’s probably Close Enough™ for anybody other than a Real Collector™.
The Pilot V5RT ink bleeds less on Astrobrights card stock than on the previous, somewhat coarser, card stock:
An automagic color adjustment bleaches the yellow and makes the black ink much more visible.
Laminating the paper crisps the contrast a bit, although it’s more obvious in person:
You can see tiny air bubbles over the darkest part of the ticks and letters.
Batmax NP-BX1 Batteries
Posted by Ed in Electronics Workbench, Photography & Images, Recumbent Bicycling, Science on 2020-03-11
Having recently lost one of the year-old DOT-01 batteries, a quartet of Batmax NP-BX1 batteries for the Sony HDR-AS30V helmet camera just arrived:
The orange curve is the last surviving (“least dead”) Wasabi battery from the 2017-08 batch and the dark green curve just above it is another DOT-01 from 2019-02. The problem is not so much their reduced capacity, but their grossly reduced voltage-under-load that triggers a premature camera shutdown.
The Batmax batteries measure better than the craptastic Wasabi batteries, worse than the STK batteries, and should survive the next year of riding. As before, I have zero belief that Amazon would send me a “genuine” Sony NP-BX1 battery, even at six times the nominal price, nor that it would perform six times better.
Batmax is one of many randomly named Amazon Marketplace sellers offering seemingly identical NP-BX1 batteries: Newmowa, Miady, Powerextra, Pickle Power, LP, Enegon, and so forth. Mysteriously, it’s always cheaper to get a handful of batteries and a charger, rather than just the batteries, so I now have a two-socket USB charger:
Despite the “5 V 2 A – 10 W” and “4.2 V 0.6 A – 5 W” label on the back, charging a pair of batteries after a ride started at 700 mA from a USB 3.0 port. The charger makes no claims about USB 3 compliance, so I’d expect it to top out around 1 A from a generously specified port.
Blowback