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.

Category: Science

If you measure something often enough, it becomes science

Laser Test Paper Beam Alignment Targets

Having failed at making flexible plant tags, I figured using laser test paper to make laser test targets might work:

Test paper – target patterns – 2024-07-03

They descend from my original dot-mode laser beam targets:

OMTech 60W laser – beam alignment – 2022-03-22

The dots just barely punch through the back side (open in a new tab & zoom for more dots):

Test paper – target patterns back side- 2024-07-03

The plastic coating chars and buckles with each pulse, but remains in place:

Test paper – 2 shot – uncleaned – 2024-07-03

Wiping the surface removes the loose coating / ash / debris to expose the underlying charred paper core:

Test paper – 2 shot – wiped – 2024-07-03

Those are two pulses marking the ends of each axis, so the machine remains well aligned after the fourth-quarter tweak.

A single pulse shows the beam has a nice round shape with well-defined edges:

Test paper – 1 shot – wiped – 2024-07-03

In principle, the beam should be more intense toward the middle, but I suspect that’s beyond the paper’s ability to resolve the energy; the beam either burns through the coating or it doesn’t. In all those targets, the back surface of the paper remains undamaged.

Manila paper targets seem to have better energy resolution and take much less time to produce:

Beam Alignment – Mirror 2 detail – 2023-09-16

The black test paper will certainly come in handy for something, though.

2024-07-17
Tide HE Laundry Detergent: Shrinkflation

The most recent Tide HE Laundry Detergent bottles seemed smaller than the one we were about to empty and, indeed, they were:

Tide HE shrinkflation – bottle labels

Call it 9% smaller, based on the volume in liters. I suspect the price was also 9% higher, but that would require more digging in the file cabinet than seems justified.

Note that both bottles claim “64 loads”, each with an asterisk (well, a lozenge ◊ symbol) explained on the label:

Tide HE shrinkflation – new load bars

That’s the new chart. The old chart was more explanatory:

Tide HE shrinkflation – old load bars

Note the “just below Bar 1 on cap” weasel wording. The term “meniscus” enters the chat, although laundry detergent doesn’t have much in the way of surface tension.

One might reasonably assume the bars on the new cap have gotten shorter, so that the volume of detergent used for each load would be smaller.

One would be wrong:

Tide HE shrinkflation – cap capacity marks

The blue cap on the right is one we’ve been using for the last few years, because I put black tape at the level of the first bar to match our “Medium” loads. I cannot imagine how much dirt would require filling the cap to Bar 5.

The clear cap on the left is the new cap. I filled the blue tap to the top of Bar 5 with water and poured it into the clear cap, where it comes about 3/4 of the way to the top of the new Bar 5. Evidently, the amount of detergent required to get grubby clothes clean has increased by 33%.

The old cap holds just shy of 4 fluid ounces to the top of Bar 5:

Tide HE shrinkflation – old cap bar 5 capacity

The new cap holds 5.5 fluid ounces to the top of its Bar 5:

Tide HE shrinkflation – new cap bar 5 capacity

If you have really crusty clothing, you’re now using 36% more detergent per load.

The obvious arithmetic shows the old bottle holds 23 “Bar 5” loads and the new bottle holds 15.

To the limit of my measuring ability, both caps hold 1.3 fluid ounces to the top their respective Bar 1 levels. I cannot vouch for the “just below” level, but I suspect more accurate measurements would show the new caps have slightly lower volume at that level, juuust enough to make the “64 loads” weasel wording come out right.

As with all too many such claims, they lie.

2024-07-16
House Finch Nesting Attempts
Earlier this year, a pair of House Finches chose the a pine cone wreath hanging outside our front door for their nest.

One day a Starling attacked:
Starling Attack – IM_00052
Starling Attack – IM_00053
Starling Attack – IM_00054
There’s a Youtube video of the action following those pictures:

Ms. Finch suffered a peck to the head raising a few feathers into a small topknot, but seemed otherwise undamaged. The eggs survived unscathed and a month later they fledged a quartet of new finches:

House Finch chicks – pre-fledging – 2024-05-18

Yes, they’re surrounded by a ring of bird crap: finch chicks can aim and fire overboard, but they don’t have much range.

The same finch pair abandoned their second nest after a Brown-headed Cowbird added an egg and punctured both Finch eggs:

House Finch nest – Cowbird egg vs punctures

Their third attempt failed after four eggs when a Cowbird added a fifth:

House Finch nest – Cowbird egg with 4 finch eggs

A few days after that picture, something tore that nest apart and destroyed all the eggs:

House Finch nest – destruction with feathers

The scattered feathers suggest a major battle with severe injuries.

Three nesting attempts produced only four fledglings: a bad year for those two finches.
2024-07-15
Daisies Dancing
Morning KP provides considerable time to watch the goings-on in the back yard, including the wide variety of pollinators (formerly known as “bees”) in the stand of daisies just off the deck:

Daisy thumbnail – 348

I wondered if the flower heads tracked the sun or just sort of stood there, so I deployed the trail camera to take one picture every five minutes for a bit over 24 hours. Converting just under 500 images into a movie required this incantation:
```
ffmpeg -framerate 4 -start_number 75 -i IM_%05d.jpg -s 1920x1080 Daisies.mp4
```
The result of which now appears on Youtube:

Daisies dancing

The short answer: daisies don’t really track the sun, but they move more than I expected. The stalks carrying unopened flowers writhe all around, occasionally getting stuck on other stems and suddenly snapping free. I was particularly surprised at the number of bees going about their business just around midnight.

For whatever it’s worth, I had to put the open camera in full sunlight on a table out on the deck to dry out the water left from a recent rainstorm. Whether the water comes from diurnal pumping or a leak through the rim gasket, I cannot say, but it can’t possibly be doing the PCB any good.

I do not expect the video to go viral …
2024-07-12
Laser Test Paper: Outdoor Testing

“Laser test paper” is, of course, intended for testing lasers, but I thought it might make a outdoor plant tag. A while ago I tried some Trolase Thins acrylic for that job:

Plant tags – Trolase Thin – prototypes

Which turned out to be entirely too stiff, which wasn’t surprising given that Trolase Thin is intended for signage stuck on flat or slightly curved surfaces.

Despite being “paper”, laser testing paper is also too stiff:

Laser test paper – outdoor labels – 2024-06-22

The wrinkles and cracks on the left end of the tags shows the plastic coating makes it basically impossible to shape / bend the paper enough to wrap around a plant stem, then push it through the hole (offscreen to the left). I was not surprised too much by this discovery.

Those two strips now hang outside the kitchen window (left end upward), where they’ll get enough sun and rain to keep a plant happy, and I’ll see how well the engraved / damaged plastic coating stands up to that sort of abuse.

For Science!

2024-06-26
M5 Tee Nut: Test To Destruction

The mounting block under the electronics box for the new UPP battery has a recess for an M5 tee nut:

UPP Battery Mount – Block 5 Show View

As with the Terry frame mounts, I glued the modified tee nut in place with JB Plastic Bonder urethane adhesive, did a test fit on the bike, discovered the whole affair had to sit about 10 mm forward, put the new frame measurement into the OpenSCAD code, and ran off a new block.

Which gave me the opportunity to perch the old block atop the bench vise with the tee nut aimed downward between the open jaws, run an M5 bolt into the nut, and give it a good thwack with a hammer:

UPP Battery Mount – M5 insert adhesive test

Although the urethane adhesive didn’t bond uniformly across the tee nut, it had enough grip to tear the PETG layers apart and pull chunks out of the block.

As with the tee nuts on the Terry bike, this one will be loaded to pull into the block, so it will never endure any force tending to pull things apart, but it’s nice to know how well JB Plastic Bonder works.

I chiseled the PETG and adhesive debris off the tee nut, cleaned it up, slathered more Bonder on the new block, and squished the nut in place. After I get the electronics box sorted out, the whole affair will never come apart again!

2024-06-20
Laser Test Paper

A pack of Black Laser Engraving Testing Paper arrived and I put a few snippets to the test:

Laser test paper – miniature pattern overview

That’s the standard backlash test pattern shrunken down to a little over an inch wide, with the laser power reduced to the bare minimum. Despite that, the numerous holes show where the pattern concentrates enough energy to vaporize the paper.

The “paper” seems to be laminated between two black plastic sheets that smell terrible when engraved, so they’re probably some form of acrylic. The Amazon product description is, despite all the verbiage- remarkably uncommunicative of the actual materials involved.

The circular pattern is 10 mm diameter on the outside:

Laser test paper – miniature pattern detail

Those should be circles around the perimeter, but their distortion shows what happens when you try to move a hulking CO₂ laser head around a 1.5 mm diameter circle at 400 mm/s. Of course, the actual speed is nowhere near that fast along such tiny vectors.

The traces are about 0.2 mm wide, with obvious scorches where the beam starts and stops, which agrees reasonably well with previous measurements.

All in all, both the paper and the laser pattern look better than I expected, particularly as the results indicate the machine has no measurable backlash at all.

2024-06-11