Author: Ed

The canonical beam alignment target seems to involve tape stuck on the mirror bracket:

With a full-power beam burned through it:

The roll of “white masking tape” supplied by OMTech turned out to be knockoff tapeless sticky adhesive film. After sticking a length to the mirror bracket, the white backing tape peels right off, leaving the adhesive film stuck to the bracket. Well, my tapeless sticky roll was running low, so this roll won’t go to waste.

A laser cutter can make intricate paper doodads, so I conjured better targets from the Vasty Digital Deep:

They’re burned into an ordinary manila file folder in “dot mode”: 2 ms pulses at 30% power separated by 0.25 mm. The 1 mm graticule locates the beam relative to the center, which is pretty close to the actual center of the opening, because the outer 17 mm cut fits neatly into the 17.5 mm hole. The label tells you where it goes and which line should point up.

Your mileage will vary, but the general idea is to have a disk held in place by actual masking tape:

Admittedly, orienting the graticule requires a bit of dexterity, but getting it pretty close is pretty easy.

Set the laser to fire a single 10 ms pulse when you press the front-panel button, thereby toasting a spot at the most intense part of the beam:

Repeat to record the beam position at all three mirrors:

The focal point target serves to verify the focused beam size and its alignment with respect to the aiming laser spot:

That target came from a scrap of cardboard while I was figuring out how to make the things.

All in all, OMTech did a pretty good job of aligning the beam, although the red laser dot needed a nudge. Now I have a record of where the beam was before I mess with clean the mirrors and lenses.

The SVG image as a GitHub Gist:

The OMTech laser cutter has six access hatches, each with one or two latches. These are not locks, although you do need a triangular “key” to turn the latch plug:

Being that type of guy, I want all the latches to have the same plug orientation when they’re closed, so that I can hold the key one way, poke it into any latch without thinking too hard, and have it fit onto the plug:

A quarter-turn clockwise (remember clocks with hands?) then releases the latch:

Inside the hatch, the closed position corresponds to a tongue capturing a flange around the cabinet opening (not shown):

After the quarter-turn, the tongue releases the flange:

So, we’re not talking high security here.

As delivered, the plugs had more-or-less random orientations when they were closed and some required a counterclockwise quarter-turn to release.

It turns out the latches aren’t a complete unit that simply drops into a hole in the hatch:

I sympathize with whoever must assemble ten handfuls of parts into ten latches on a production line and I also understand why orienting the plug wasn’t on that person’s to-do / QC checklist. I further understand why two cylinders lacked the big toothed washer under the nut; it’s not essential to the function and nobody will ever miss it.

The plug has a triangle on one end (for the key) and a square on the other (for the tongue), with one triangle point aligned to a side of the square:

To my way of thinking, that point must be upward, as shown in the first picture, when the latch is secured.

The cylinder can fit into the square(-ish) hatch hole in four possible ways, but its symmetry allows only two unique orientations. It must look like this in order to put that point upward when the plug is maximally counterclockwise (my finger is pointing upward):

So I devoted a pleasant half-hour to reducing the latch entropy.

The screw attaching the tongue to the plug also controls the friction of that spring against the plug as you (well, I) turn it. All the screws now sport a dab of Loctite to ensure the tension remains mostly constant (at least for a while), as do the two large nuts lacking corresponding toothed washers.

The “key” has no marking to indicate its “point-up” orientation, so I stuck a snippet of label on one side, with a jaunty red highlight marking the point. Something better will surely occur to me, but it’s no longer in the critical path.

The best place for the OMTech laser cutter seems to be snuggled at base of the chimney, venting into the long-disused fireplace through the steel plate adapting a long-gone wood stove to the opening:

The short run of flexible tubing allows some give-and-take at the cutter’s vent outlet. The elbow on the duct fan’s output terminates in a blast gate to cut off the draft blowing up (or down!) the flue with the fan off.

The cutter arrived with a huge high-speed axial blower screwed to its output baffle:

The noise from that fan had to be heard to be believed.

The cylindrical exhaust duct attached directly to the motor with four screws, only two of which matched holes in the baffle plate:

A trial fit revealed the assembly rattled something awful: those two screws let the duct vibrate against the baffle. Match-drilling two more holes into the baffle let me mount the duct with three screws and, in combination with the foam gasket, it is now solid and quiet.

A quick check shows the duct fan draws 10 to 11 m/s through the baffle at full throttle, roughly 400 CFM. That’s pretty close to the flow measured through a long pipe and, with only 6 ft³ of stink inside the laser’s cabinet, ought to exhaust the fumes just fine.