The laser only melts the metal, it doesn't move it out of the cut, and some metals can react in plain air to sputter back at the emitter optics. Gas is forced under pressure at the cut to clear it. With some metals and/or thinner (speaking from a commercial perspective) stock, you can get away with plain air at high volumes of normal air compressor pressures with good results if it's very clean and dry. Others require specifically reactive or nonreactive gases or blends of gases. That's true for most hot cutting processes though. Plasma usually consumes plain air to blow out the cut, oxy-fuel uses excess oxygen to reactively blow out the cut. Lasers with their extremely narrow kerf are more finicky, which can mean a surprisingly high consumables cost.

More powerful lasers also require the optical hardware to operate in a light vacuum. Plasma generated by dust and air itself will ablate expensive parts. Home shops shouldn't need that much power though. If you're cutting thick or difficult metals regularly, that's not a hobby any more.

Then there's the byproducts. Organics get blasted into all kinds of random organic-ish things that aren't great to ingest or emit at ground level near neighbors (PSA, this is true for current hobby lasers). Cutting metals can get really exciting if you don't clean often and thoroughly enough to prevent a critical mass of finely powdered byproducts. An iron or aluminum fire will wreck your laser. An iron and aluminum fire will wreck your laser, and whatever it's sitting on, and the concrete below that.

So, please do not come around these lasers with remaining eye unprotected.

I think one should work with these things only using some cameras and never directly.

I've experimented with using a 60 W (I think*) laser on a small steel bracket, and even with the beam holding on a single point for a minute, it made a barely visible dot that you couldn't feel by running your finger over.

* It was nearly a decade ago, but I looked up the relevant hacker space and unless they changed the model, it was 60 watts.

The bracket was around 1cm by 5cm, and around 1mm thick.

The fiber marking LASERs at work are 1064nm, and at a mere 20w output, will absolutely eat away at steel with no problems.

Edit: I should note there are CO2 metal cutting LASERs, but they are at very, very high output powers to overcome that reflectivity barrier. You need 500w 10600nm to cut through what a 30w 1064nm could cut. My 80w CO2 barely cuts through heavy-duty aluminum foil, and in many spots it isn't a full cut. A 20w marking LASER at 1064 would obliterate the foil.

Everlast now has a laser welder, too.

I think it’s only a matter of time before we get cheaper cutter in the 1kw range.

At the moment, there are expensive-but-affordable home CNC laser cutters, typically for a small number of thousand EUR/USD. The more powerful ones can do a very neat job cutting (up to a few mm of) plywood. There are also CNC plasma cutters, which do a good but slightly rough job of cutting sheet metal, and are relatuvely large and complex beasts. I guess a highly-powerful laser, of the type envisioned, would offer the best of all worlds: relatively neat and quick cutting of all materials on the same compact machine.

They might also replace handheld plasma cutters (and welders?) too.

Edit: It easily cut 1/4 ply (used 2 passes, but judging by the burn marks behind it I only need one). And that's without air assist to clear to the smoke which would help it cut deeper.

The world of home laser apparati is kinda wild.

On the other hand, wood cutting is done with CO2 lasers which have a significantly longer wavelength of around 10 micrometers. This wavelength is absorbed very well by wood and most plastics, but is mostly reflected by metals. Additionally, the longer wavelength means that it cannot be focused to as small of a point, which reduces the maximum power.

Chip lasers would still be bound to the same wavelength limitations, so you couldn't cut both materials with the same laser. What you could do, though, is have a machine with two lasers - especially if they are very compact - and select which one to use depending on what material needs to be cut.

I don't see how a laser would be an improvement on any of those, unless the goal is to CNC with extremely tight tolerances, but even then... it seems unlikely that you'll be CNCing blocks of metal, more likely just 2D cuts from sheets of metal, which is pretty limiting.

It would certainly be fun for some stuff, but I think the danger level of a super high power laser detracts from the fun.

More powerful cutting lasers would definitely be great for industrial use cases.

I love the idea of using a mill/router instead, but IMO the more complicated process is fundamentally more dangerous. If a reflected beam can conceivably pierce the enclosure though, hmmm...

> We even used it to cut through steel. As the bright, beautiful beam carved a disc out of a metal plate 100 μm thick, our entire lab huddled around, watching in amazement.

So... it can cut steel! As long as the steel is 100 μm thick. I think that is literally about the thickness of a piece of paper.

This newfound information does not weaken my skepticism about the hobbyist laser mill that can cut through (useful) pieces of steel coming to a garage near you any time soon, unfortunately.

I'm just extremely skeptical of a steel-cutting laser mill being cheaper than a $10k CNC any time soon. Hardware is hard. Semiconductors are even harder.

As far as I can see, the article provides no indication of when they even hope to bring this product to market, which is never a good sign.

https://xkcd.com/678/

I would love to be wrong.

And really, that video is unsatisfying. They show a mark on the steel, but -- as far as I saw -- they never even show any steel piece that was cut out by the laser. Maybe the steel melts at the very surface... but it may not be possible to cut through a piece of steel unless it is thinner than a piece of paper? We have so little real world information on this laser.

* for wood, faster and/or thicker cutting, vs. existing CNC laser cutters

* for sheet metal, neater/cleaner cutting (i.e. cleaner cuts, higher tolerances, less subsequent prep-work needed) perhaps in a smaller neater machine, vs. existing CNC plasma cutters.

Especially within the context of a CNC machine, I wouldn't be overly concerned about safety - all of the more powerful CNC laser cutters I'm aware of already come with an exclosure - both for laser safety, and to constrain smoke (before it's vented safely).

If it can cut metal, it'd probably not suffer much in the way of limitations on other materials... .

LLMs and self-driving cars mean that KITT from Knight Rider will finally become a reality.

All I need is for Webb to discover a planet a long time ago, in a galaxy far, far, away, and my childhood fictions will have been realised.

The article says they are planning 10kW or 10^4W and a petawatt is 10^15W so 11 zeros by my calculation. Could be a while. I think Helion, who are now talking about turning on their fusion to electricity gizmo this summer, may get there first.

Fans of The Peripheral, I see.

All kidding aside, the sticking point for this, as it already is for high-power LED lighting, is cooling.

It still takes a fairly significant bit of hardware to cool stuff.

I'm curious how hard it is to model light interactions on this scale. What does it take for the authors to come up with modeling solutions for their scaling problems? Is this something one PhD in light physics could do, or do companies and people develop expertise in teams over decades?

It seems like ASML only requires the solution to one wave diffraction interaction (with the mask), but modeling the standing wave and higher-order modes would require much more mathematics. Is it even possible if the solutions for each interaction are probabilistic?

Also, are solutions in practice mostly bounded by the kinds of semiconductor features (holes) they can build?

One application for high efficiency large aperture lasers is powering long distance aircraft, either by beaming from ground stations along the way or from laser stations in space. The ability to forego fuel entirely would be quite attractive, allowing potentially unlimited range.

Another use for that is holography. If you can make a wafer-scale array of phase locked emitters overlayed with the ability to either attenuate or phase-shift each one, it becomes a nice (one color) holographic display. I'm assuming high divergence from each one.

Obviously impossible to defend against weapons like Russia's tsunami-triggering underwater nukes. But seems to rule out anything air-based?

it mentions the following list of priorities for the plan, and with a bit of creativity it is possible to imagine ways how cheap, powerful, compact lasers could benefit each of them in turn. Probably, and more mundanely, the researchers are doing their best to justify their research grants by connecting the outcomes to the political priorities of the day.

"Hence, Japan will consolidate the following fundamental technologies in particular, which function as core technologies in the real world, for new value creation in individual systems. - Robotics: technology expected to be used in various fields such as communication, social service/work assistance, and manufacturing - Sensor technology: technology that collects information from humans and all kinds of “things” - Actuator technology: technology related to activating mechanism, drive, and control devices in the real world, as well as the results of information processing and analysis obtained in cyberspace - Biotechnology: technology transforming sensor and actuator technologies - Human interface technology: technology using augmented reality, affective engineering, neuroscience, etc. - Material/nanotechnology: technology that leads to differentiated systems through enhanced functionality of various components, such as innovative structural materials and new functional materials - Light/quantum technology: technology that leads to differentiated systems through enhanced functionality of various components, such as innovative measuring techniques, information/energy transfer technology, and processing technology."

The laser in the article took 30 seconds to cut a a tiny disc out of a sheet of steel 100 micrometers thick. That isn’t going to change things on the theft front. There are much more effective methods available today.

If you live at a lower end apartment complex, you might be shocked at how many doors your key will unlock, if you apply a light turning pressure and thrust the key in and out a few times.

I know I was. But it came in handy when one of my xmas gifts had been delivered to the front office, and I needed to pick it up after hours in order to make a redeye flight the next morning. I just used my own room key to open the office (setting off the alarm!), grabbed the package with my name on it, and closed and relocked the door and was gone before anyone investigated.

I've tried this on higher-end tumblers and I could really feel that it wasn't going to work. It's just about the el-cheapo builder-grade locks (which comprise a shocking percentage of home and apartment locks).

P.S.: Trying to think of the rare examples that don't qualify, and so far:

1. Direct violent response, e.g. booby-traps.

2. Denial by destroying whatever the attacker wanted, or delaying them enough that time renders it valueless.

3. Denial by making the effort unprofitable, even with no hard time limits.

You get the bike for < 20 Euro and then, presumably, at some point in the future it gets stolen again, but you already got 20 Euros worth of use out of it.

Even in 2011, when I made the mistake of spending £90 for a new bike… well, the pedals came off while riding it due to metal fatigue.

Adjusting for inflation*, I'd expect similar build quality from a bike that "only" cost €200 today.

* hard to do when there is also a currency switch, especially when the exchange rate has changed so much

It would take work to remove it cleanly and/or cost money/time to paint over it.

So it's less attractive.

Ps. You'll still need to take care of your 10 k. Bike ofc.

Ps. 2 : engraving isn't done anymore since bike frames got smaller.

People in countries where bicycling is not seen as a personality statement generally ride very cheap city bikes around town. Last time I bought one of those in college it cost about 30eur. Is not like in SF where people ride 3000 dollar bikes to work because omg you’re a cyclist now and this is your whole identity.

No, it definitely does not!

750k bike thefts for 17.7m people. So roughly one per 25 people...

It famously has more bicycles than people!

There is also a growing segment of people who use $5,000–$15,000 cargo/kid carrier bikes as replacements for their cars, which is cool. But I do worry about theft of those increasing. Right now I rarely hear of it, but it seems likely to increase. Even the motors and batteries in them cost thousands alone.

1. Can be opened without tools

2. Can be opened with generic tools (e.g. hammer, hacksaw)

3. Can be opened with specialized tools (e.g. lock picks, liquid nitrogen)

Yeah, I'm currently waiting on parts to replace my door handle. As far as I can tell, the would-be thieves snapped off the handle to make an opening, and used that to cut/snap the rear part of the lock (which normally can't turn because of the lock cylinder) and then substituted in a screwdriver to convince the car that the lock had turned.

Thankfully--and puzzlingly--they were unprepared for the steering-wheel lock and only managed to mangle and bend it (probably with a small bolt-cutter) before they gave up or were spooked off.

'Course, after that they would have still been unable to start the car for other reasons, but it still saved me the cost of fixing a torn-up ignition.

Anything more powerful than a low end projector pointer is now classified as a weapon and regulated as such.

This at least keeps these units out of the hands of the general public and those that can't use them responsibly.

It sucks for those of us who used them responsibly just like we did with RC aircraft before drones became an item on a discount shelve. However at least I know that if I go to a large event or even a demo I won't be permanently blinded by some idiot.

I mean it seems like a fine law, but it's a law so now I never have to think about it again seems strange to me. Running stop signs is outlawed but I still look before I cross the street.

Same goes for guns. Someone can aquire one illegally but fines are high and there is no impulsive purchase at the next Kiosk so circulation is low and you aren't going to get "lasered" by 20+ green lasers at some event.

https://www.harborfreight.com/power-tools/grinders/angle-gri...

(Cost plus a $30-$80 battery, of course, but after you pick your color, you've already got batteries). Your locks are far more protected by the social contract that would compel a stranger to stop someone who was cutting a bike lock with an angle grinder, and would cause a thief to fear arrest, prosecution, and jail time, than by actual physical security.

If you want a laser, you can already get a 100W CO2 laser tube for $500:

https://www.amazon.com/Cloudray-W2-Dia-80mm-Engraver/dp/B08G...

It won't do a great job of cutting steel - it will engrave it, but to cut steel you'd need O2 assist gas and a lot more power. It will trivially cut ABS, foam, fabric, wood veneer/thin plywood, cardstock, etc. if attached to a pair of mirrors on an XY stage such as might be borrowed/scaled up from a 3D printer. And, of course, it will trivially scar corneas with the invisible danger of its 10,600 nm laser light. Good news, though - ordinary polycarbonate safety glasses are opaque to the extreme IR light.

Laser safety is serious business, but highly-collimated, tightly-focused laser light is not likely to be produced by a cheap, portable laser. The above tube produces a 100W beam with a diameter of about 8mm. I wouldn't put my eyeball in the path, but a safety shutter can sit in the beam indefinitely and merely get warm. You melt steel by focusing a this 8mm beam down to a infinitesimally tiny spot, and beyond the focal length of your optics it diffuses to something no more dangerous than the source 8mm beam, and beyond that it's no more dangerous than an average lightbulb - albeit one that you have no aversion response/blink reflex to. I always wear my goggles near our cutting lasers, but I know lots of guys who have worked around 20W-20,000W CO2 lasers their entire lives and they're pretty cavalier with regards to laser safety.

Finally, a welding mask is a neutral-density filter, suitable for the (very approximately) black-body radiation produced by a welder, and is much less effective than laser safety goggles at blocking high-intensity light of a very specific color. Depending on the sensor, an auto-darkening mask might not even trigger when exposed to a laser! You'd want a set of 940nm laser safety goggles for this, which is far more exotic and dangerous than a 10um CO2 laser.

That being said, it's all a function of how much light enters your eye. You can look up the exposure limits of the human retina for different frequencies of light to get an idea how little is actually required to damage your eye. The target or sensor should be sensitive enough to react to that much incident light. Just putting something behind the glasses and have it not get cut is definitely not enough to indicate no eye damage.

Laser light is different: a laser is effectively a point light source, and it will focus inside the eye to a single point with enough brightness to cause a burn. It can cause a permanent blind spot, and damage may happen faster and be less noticeable at first.

but laser light, like staring into the sun, will destroy the delicate structure of your retina, and it doesn't grow back.

VR-headset will be better, because it can attune brightness and contrast so you can see even at dusk.

In this example, crossfire would require someone to walk into an airsoft combat area without any eye protection and for someone in that game to not realise and shoot them anyway. In the face.

Stray BB bullets don't go very far. They're not rifled, and they have fairly low muzzle energy compared to real weapons — I've shot myself with one at zero range, and even on bare skin it stings rather than penetrates.

1. Spending your life in a nuclear bunker and only interacting via the internet — including pure VR and/or remote controlled robot bodies.

2. Let your government poke around everywhere to make sure nobody is making (or importing) their own nukes, regardless of what anyone says about the 2nd amendment.

But I guess this is inevitable, so we'll just have to devise better solutions than the simple locks which in their working principles have not changed in the past 100+ years.

Great market opportunity around the corner I suppose.

Cutting through a chain is going to take much much longer. There are already quicker methods to cut or break bike chains and disable locks.

Also cutting with a laser requires very precise alignment with respect to distance. You won’t be free handing this. You’d need to mount it on a device capable of precisely adjusting the distance as you cut deeper into the metal.

Plus it’s relatively easy to engineer laser resistant materials.

For plasma and oxy-fuel cutting, the hot gases coming out the nozzle acts as the mass that blows it off. Air carbon arc gouging uses regular compressed air to push the molten material away. I imagine industrial laser cuttings dealing with anything larger then a few millimetres thick would use compressed air as well.

I doubt a smaller more efficient laser is going to change that reality.

There are some interesting niche uses for laser sniper carbines[1], IMO, like punching holes in a gas tank and igniting the vapour, burning out security camera sensors, cutting power/communications lines, starting fires inside distant buildings by firing through windows, etc., but I think most useful laser weapons will be mounted and computer-controlled so that they can track moving targets long enough to do meaningful damage.

[1] It's not a rifle if the bore is smooth.

Could this architecture be faster then current 3d printers?

The article explains that it has potential for welding and other commercial applications but it's not quite there yet. It did cut through 100um steel. It's a good read.

There is focus on the industry, but surely this should be regulated as it seems to be of dual use.