pew pew lasors!
This is all very well, but when do we get the sharks with frickin' laser beams on their heads?
The British Army has successfully destroyed flying drones for the first time using a high-energy laser mounted on an armored vehicle. If perfected, the technology could form an effective counter-measure against drone attacks. The UK Ministry of Defence (MoD) announced that soldiers from 16 Royal Artillery had conducted tests …
Energy weapons such as lasers are regarded as a reasonable solution as these can continue to fire for as long as there is sufficient power, whereas guns run out of ammunition and missiles are soon used up.
In a portable deployment on a truck in the field, there isn't too much difference between running out of supply of ammunition or running out of a supply of power.
A gun can continue to fire as long as there is sufficient ammunition, whereas lasers can run out out of electricity supply whether its stored in cells or generated from fuel.
In what world do you live where you think that an Abrams is going to have any battlefield mobility whatsoever with batteries ?
As far as planes are concerned, yes, we have experimental, one-person pseudo-gliders with long wings covered in solar cells, but if you think that will in any way be efficient on the battlefield, you have read too much science-fiction.
I'll believe your words the day I see a Caterpillar 930G survive a day's work on batteries.
Hey, don't get me wrong. I'd love to see that. I'd love to know that we have fusion that works.
But I'm aware that we're not there yet.
Interesting link. I learned something, there.
I still don't see how that can apply to tanks, which work best on level terrain.
There will be no regenerative braking to help much, and they're not supposed to take on 65 tons of ore at any point.
But it's nice to know somebody got this working.
I wonder what it will cost to replace the batteries when the time comes . . .
The Challenger, I recall being told, has 30 seconds braking. I inferred that if it then let the brakes cool it would have 30s again, until it needed it's brake pads replaced.
Why do you think a tank wouldn't have electrical braking?
Plausibly worth having just for going downhill, but the bonuses of providing power for the boiling vessel and being alternately a motor along the lines of the Formula 1 car might be appreciated.
I don't know what the clutch life on tanks is, I suppose there's a torque converter, but you might even find a fully electric transmission is useful, as on trains, heavier and faster vehicles running on tracks that they are.
Get clever enough, and you run a combustion unit at its most efficient speed to generate power, making it smaller and quieter.
That was included in a book of science puzzles for children (published by Knight, I think) fifty years ago:
"A milk float leaves a dairy on top of a hill every morning and charges its batteries as it goes down hill. After dropping off the milk and collecting the empties, it has just enough charge to get up the hill and back to the dairy at the end of its rounds. Where does the energy to power it come from?"
Though of course in that case the answer was "milk tankers", not "geological processes".
The answer is always "gravity"
The sun has a lot of gravity. That pulls material together with enough force to create nuclear reactions. Those reactions produce emissions across the EM spectrum. Those emissions, upon hitting the earth, make grass grow. Cows eat grass. Cows produce milk. Cows get milked at top of hill. Gravity makes the milk flow downhill (on its own or pushing the vehicle)
Probably more accurate to state that it is Potential Energy. Gravity is the mechanism that facilitates the return of that Potential Energy. All the mass had to get to the top of the hill at some point in time and that process required energy.
There are a few processes that appear to go counter to this principle, for example a pipe with a gradient and a coating of steadily decreasing strength hydrophobic coating from bottom to top will transport water up the sloped pipe with no obvious energy input.
Actually, the answer is "sunlight".
The dairy might be at the top of the hill, but the cows are probably not. So they walked up to the dairy to be milked, fuelled by the grass that has fixed the Sun's energy as glucose, cellulose and starch.
Even if the farm is on a plateau, the water in the milk fell as rain onto the plateau, fuelled by the Sun again!
Batteries, no.
Electric propulsion has a lot of advantages, that's why ships use it. For example:
You can run your engine(s) at their most efficient RPM for the power demand, increasing your range for a given fuel load.
Redundant engines are easier to arrange, as you don't need complex driveshafts - just run fat cables and chonky (solid state) relays.
Auxiliary stuff like turrets, autoloaders etc are electric anyway.
Isn't the next generation M1 Abrahms going to be hybrid ?
I can't remember if it's only for short range silent mode (except for big clanking metal tracks presumably) or the whole drive train is diesel-electric (because insanely complex multi-mode drive trains are really cool - look at the Tiger tank)
The Abrams was given an APU to handle idle ekectrical requirements post the 1991 first Gulf War (or really the second Gulf war because the Iran/Iraq one deserves to be called the first). The APU means that you don't have to run the turbine engines to generate power for boiling a cup of coffee or running the thermal sights while holed up. As a result the fuel consumption of the Abrams in the field is now pretty much on a parr with that of contemporaries like the Challenger 2 and the Leo 2.
Larger tanks often have an auxiliary engine to generate electricity so they don't have to run the main engine to do so. IIRC, the Centurion used a 1,000 cc Austin or Morris engine to do the job, though their crews often complained that it was the most unreliable part of the entire vehicle so they'd end up having to use the 27 litre main engine to generate electricity. Smaller tanks like the Scorpion didn't, which is why their nice E-type engines were often shagged out in fairly short order due to running them at full speed all the time to keep the turret's batteries charged up.
I know that Ferdinand Porsche was obsessed with electrical drive systems, typically involving air-cooled petrol engines, but they seldom worked that well. Whereas Merritt's regenerative steering did: its only fault was the fancier controls he tried to implement when it made its debut on the Churchill which just added more things that could break (and did), which is why British tanks reverted to brake levers thereafter, but still connected to the new steering unit that could do fancy stuff like turning on the spot; shown to good effect with Comets managing to navigate the twisty little roads in German villages where the Shermans got stuck. He wryly observed that the Germans copied the system from a captured Churchill and their implementation on the Tiger et al made it more complicated. With the anticipated results.
> Which has for a long time been MY suggestion: turbine-electric powertrain for "hybrid" motor vehicles.
Sorry but you didn't think of it first.
Google "series hybrid" - there are multiple examples of such motor vehicles in production.
I think the issue is the total amount of energy required. Sensors don't need much energy, whereas a high-energy laser system needs a lot of electrical energy.
Whether it's easier to pipe that electrical energy down wires, or generate it on-site using fossil-fuel-powered generators, I don't know.
It looks like it uses the vehicle's onboard power systems, presumably diesel. https://www.linkedin.com/posts/serkan-k-b4b841a8_laser-air-defense-activity-7221442716847509504-t7JM/
"Power and Thermal Management:
- Integrated Power Supply: The system is designed to draw power from the vehicle’s onboard power supply, with additional provisions for dedicated power sources to ensure sustained operation."
Quote "... with additional provisions for dedicated power sources to ensure sustained operation."
So that to me looks like they can only ensure sustained operation when using additional power sources.
Which in turn likely means using the vehicles power system only allows for non-sustained, operations. Likely charging something up, then fire, then wait to charge again.
But if you want sustained fire, time to get the extension cord out!
Not really an issue, as this is only a demonstrator anyway.
Let's add that it only needs to fire for mere seconds at a time, and can operate with a much lower power energy source if it has chunky capacitors, although that will impact fire rate. DragonFire (50KW output) is said to cost less than £10 per shot. If you estimate the energy use that's either very expensive electricity - perhaps consistent with the cost of supplying and operating diesel generators in the middle of an ocean - or it includes other maintenance costs.
Definitely waaay easier to re-supply than a gun requiring some specific calibre ammunition with a long lead time and limited global manufacturing capacity.
And much, much cheaper than missile defence systems where each missile easily costs 3 orders of magnitude more than it's target drone.
This post has been deleted by its author
The principle difference is that ammunition is "application specific" (as it were), while fuel isn't. So the logistics of delivering the material needed to destroy the UAS are very different: the truck delivering the munitions _also_ needs fuel, while the truck delivering fuel just needs the fuel it's carrying (i.e. it can consume its own cargo to move).
In the field there would be a layered defense to protect vital assets, the more important, the greater the level of defense. As a first line of defense, expensive missiles are being used to take out inexpensive drones. Ammunition implies small arms or guns (like the Phalanx used on ships), that is typically a last resort since it requires a close range engagement.
My understanding is that a field deployed laser would be taking drones out before they are in range of kinetic weapons. The real comparison is Lasers vs missiles. Right now we are spending 10X the cost of drone to take each one out, and the drones can be manufactured in much higher volumes.
It's easier to generate more electricity than is it to replenish conventional weapons. The math is spend more up front for a Laser weapon that reduces the need to use more expensive defenses. If assets are attacked by a drone swarm (think thousands of drones) there won't be time to wait until they are in range of conventional weapons.
Duty cycles are typically at least 10% and shot times in the 10 to 15 second range for these smaller systems. ALso figure the E to O efficiency of a typical laser is in the 50% range, so its 2X input power for shot power. The reality is generating the power needed to run the system is trivial compared to the development work that goes into the rest of the system.
It sounds like effective laser range is actually pretty similar to auto-cannon/Phalanx range, I guess better accuracy per shot but then an auto-cannon can unleash a lot of shots quickly if needed. They both suffer from the need to have an accurate trajectory at 'launch', whereas a missile can alter course in flight giving it much better effective range. Lately we also have these precision-guided shells that are sort of half-way between missiles and dumb kinetic shells so you can perhaps imagine a future auto-cannon achieving better range with those, but there's obviously no hope of equipping pure photons with a guidance system :)
The positive is that since laser ammunition is just electricity the sky (and general safety/collateral issues) is really the limit to their ability to engage either very large numbers of targets at close range or to devastate relatively static targets at very long range (weather conditions quickly becoming the limiting factor). It's a really interesting technology overall and arguably comes at a very convenient time when drone technology has up-ended established defensive doctrines.
and the best efforts will be made to 'neutralise' those vital assets. Somehow, I can see this as a laser battle station meant to protect the whole unit, being attacked by 50 - 100 drones, and trying to protect itself until the lights go out.
Stupid question (genuine one too) here but what happens when they miss? I assume with a conventional weapon we have an ultimate loss of propulsion and gravitational forces bring the projectile back to earth - hopefully without collateral damage.
A laser suffers some bending due to Earth's gravitational pull on a big scale but nowhere near enough to "bring it down". Thus it heads out through the atmosphere getting weaker and dispersing as it goes. Does it have enough force to damage a satellite if it comes in to contact as it goes? I'm assuming the chances of doing so are slim, but how slim?
A ground-based laser intended for short-range battlefield use isn't going to do any damage in space. (Potential collateral damage to people being blinded or burned, but the straight trajectory probably makes that easier to avoid than with ballistic ammunition).
- For a shot straight upwards, assuming a beam divergence angle of 0.1 degrees the energy will be spread over a diameter of hundreds of metres by the time it reaches low-earth-orbit. Most shots are not going to be straight upwards so typically that area would much greater still.
- Satellite that low in orbit have an insanely fast ground-speed (in km/s) so are not going to be in the beam for any meaningful time unless intentionally tracked.
- Without careful selection of wavelength atmospheric gases are likely to absorb a lot of energy, and even with careful selection atmospheric dust/vapour will disperse a lot.
Interestingly the first two points mean that although satellites are very spread out and space is huge, it's not unlikely that some satellites would travel through some beams during an intense battle. Maybe that has implications for intelligence collection/3rd parties monitoring ground battles/etc if the lasers have a better time punching through clouds than flashes from explosions for example, I don't know.
In principle a powerful enough ground laser could damage a satellite of course. Obviously, when that powerful enough laser is invented some nut will put it up in space for the reverse effect.
I seem to recall the situation where at one point in time the MoD's latest naval helicopter had one set of tyres rated for wheeling it in and out of the hanger, and a different set of tyres rated for take off and landing!
I have seen it written (so it must be true!) that with a laser, if you can see the target, you can kill it.
But I suppose it depends on how well you can track it, and if it's near enough to avoid too much plasma defocussing on the way... but of course, the nearer it is, the faster the radial tracking needs to be.
Now protect your other eye! ---->
"seems to be "correct" English"
apparently, if a mistake or error is widely enough perpetrated it becomes valid[1]. Of course the Earth must at some point have been flat because almost everyone thought so. Must have been a huge upheaval when it became spherical.
[1] Similarly to 'learnings' in its complete neglect of grammar is the almost ubiquitous and utterly ungrammatical "attendee" and 'escapee". The 'ee' ending grammatically indicates the object of a sentence (the thing or person to which something is done -- as in 'employee') The correct grammatical ending for the subject (the thing or person that performs the action, as in 'employer') is 'er' or 'or', so the correct spelling would be ;'escaper' and 'attender', regardless of whether a majority of the uninformed say otherwise. And yes, grammar does matter. Without it, language becomes ambiguous and incapble of expressing nuances so the clarity of communication suffers.
Likewise, Columbus wasn't a visionary by thinking the Earth was round when everyone else thought it was flat; he was an idiot for believing it was far smaller than everyone else and refusing to listen to them. Had there not been a previously-unknown continent in the way, he and his crew would have died.
Well in the search for an easy route to India (specifically not via land routes over nations that weren't always friendly or around the entire of Africa), all "forrnners" were obviously pretty much indistinguishable from each other and all of them were definitely savages because they didn't speak civilised languages... i.e. they didn't speak Latin, Old Spanish/French/English/Italian/German or, possibly, some other European language at the time which speaking might have grudgingly elevated the speaker from "savage" to something marginally more respectable than a stray dog.
Pretty much since the invention of a sea faring boat people have known the earth is a globe. The size of the globe was calculated by smart people quite recently in comparison (i.e. ancient greeks).
Where much of the lagging of the solar system was the understanding that the geocentric "model" had so many flaws that it was near impossible to predict anything and beyond the relationship between the Earth and Moon and this is where the heliocentric model came into play. This could accurately predict the relative movements of the celestial bodies and also deeply upset apologists/theists who were stuck unable to comprehend that the Earth is not the centre of the universe. It's the standard exceptionalism mentality where these kinds of people have a need to be special, and in this case a need for Earth to be special. Earth is special of course... because it's our home and where we evolved to live and survive, which is very handy because 99.999999999999999999999999999999999999999999% (whatever) of just our galaxy is incredibly inhospitable to humans.
Mike 137,
Obviously you have a point about the horrible sounding "learnings".
You're mostly wrong about "attendee" though. To quote from your post:
The 'ee' ending grammatically indicates the object of a sentence (the thing or person to which something is done
So I well might be an attender at my Mother's birthday party. But at most work-related meetings and seminars I feel that "attendee" is totally correct. As the object of the sentence - I objected to being sentenced to attend said snoozefest, but was overruled. I did not volunteer - I was voluntold.
Language is for communication. It literally does mean what people think it means, because that's how communication works. If you don't like it, go learn Esperanto or something.
And "learning", meaning "knowledge or information gained from study", has been around for centuries.
They've had it since the 1960's (when my father worked on one of the projects in the California desert) the problem at that time was that you needed one 48ft railcar for the laser, a second 48ft railcar for the generators and capacitor banks, and a tank car for the diesel fuel supply. While you could only fire one shot, it would vapourise a one inch hole through a three inch plate on the other side of the valley (yes, they were using a mountain range as a backstop.)
It's already deployed in the US: https://en.wikipedia.org/wiki/High_Energy_Laser_with_Integrated_Optical-dazzler_and_Surveillance.
The 2030s is quoted for the naval laser - which is incorrect. Dragonfire has already been deployed on a warship for trials and purchased - It's supposed to be first deployed in 2027. It should be quite an easy retrofit as well. They're supposed to go on all the escorts - I don't know if they'll also be fitted to the carriers and / or auxilliaries (like Phalanx already is).
The US have started to deploy microwave anti-drone weapons on their Arleigh Burkes, rather than lasers.
Those drones are in some cases (shown on videos in the news) flying directly over people's houses in suburbia. If they really are the size of a car, what do you think happens if you "neutralize" a drone that big over someone's house? This isn't Star Trek, they don't simply vaporize when hit. They fall, and from the height they are apparently flying fall with a LOT of momentum!
And if there happens to be an airplane flying 10 miles away in line of sight behind that drone and even 1% of the beam misses the drone you'll have 1% of a 15 KILOWATT laser which is far more than needed to permanently blind both pilots in a commercial jet.
"Shoot them down" is the easy answer, but it isn't so easy in the real world and that will be doubly true if you do so with a laser. This would be a great tool in a warzone where you either don't care what happens when a drone crashes or figure it will cause less damage falling than if it gets to where it wants to go, and you have situational awareness of all of your manned flights in the area. It isn't practical to use around civilians.
Apparently you did miss something.
The guy I was replying to literally said "send them to New Jersey", which is main locale where the "SUV sized drones" have been reported flying around the US.
So yes there absolutely was a suggestion (though perhaps somewhat tongue in cheek) that they could be used to shoot down those drones of unknown provenance flitting about the skies of the US.
Remains the case that there is nothing in the article though, and whilst you may have been replying to the earlier comment, the distance between that comment and your post meant that this was not immediately (or easily) obvious, and it would appear that I wasn't the only one to miss that connection;)
I am pretty sure the comment you thought you were replying to was made tonge-in-cheek.
Citation needed on properly operated and maintained nuclear reactors blowing up. Hint: military reactors, mostly in submarines, have never blown up, despite assorted Bad Things happening. See further USS Thresher, USS Scorpion, assorted Soviet/Russian subs. Civilian reactors have had problems, but tend to not explode; even at Chernobyl, and despite the very best efforts of the improperly trained crew on site, the best they got was a chemical explosion and fire. Windscale, Three Mile Island, and Fukushima didn't even manage that, though Windscale came close.The guys in charge at Windscale weren't quite as incompetent as the guys in charge at Chernobyl and the resulting fire was... quite spectacular. Not an explosion, though. The small reactors fitted to objects such as NASA's Mars rovers and some space probes are extremely unlikely to explode; I would expect that an AFV would get something along those lines, just bigger, as it wouldn't have to fly to Mars. If they can be made to work reliably on Mars, even the average squaddie will have to work at it to break them. I wouldn't say that the squaddie couldn't break them if they tried hard enough, but they'd have to put some effort into it. Famously Arnie Schwarzenegger was, for reasons of incredible idiocy, assigned to tanks for his time in the Austrian Army and managed to break two before being assigned elsewhere. (Arnie has since bought an old M-47 just like the ones he broke when he was a squaddie; his personal tank is still operational, so either he's learned something or he really did break things on purpose.) (Exactly how someone Arnie's size managed to fit inside the driver's compartment of an M-47 is best left to the imagination.)
Please note that there is one, and only one, way to have a nuclear reactor, any nuclear reactor, including carbon-moderated death traps like at Windscale and Chernobyl, to explode like a nuclear bomb: put a nuclear bomb in it and light the bomb off. Nuclear reactors _cannot_ explode like nuclear bombs, physics says so.
The 'nuke reactors can explode' meme is a particular pet peeve of mine.
Fun fact, it's actually really hard to make a pile of uranium blow up. That's why it took the massive and very expensive Manhattan Project several years to figure it out.
Second fun fact, after the first two bombs they couldn't do it again! (They had literally run out of weapons-grade material. For the next couple of months, the US wasn't actually a nuclear power. Obviously they kept that bit quiet.)
Sadly <a href="https://en.m.wikipedia.org/wiki/Third_Shot”>not true</a>. The only argument at the time was whether to use them as production allowed, or to save them up for use in Operation Downfall.
There were yield problems at the Hanford site due to the Wigner effect and excess <sup>240</sup>Pu, but they were mostly in 1946 and eventually solved or worked around.
A nuclear explosion isn't the only concern. An explosion sufficient to break containment can spread some really nasty highly radioactive stuff all over the place, creating a far worse situation than whatever a drone was intending to do.
Chernobyl managed to be pretty bad without a nuclear explosion. Worse in some ways than the aftermath of the Trinity site.
This post has been deleted by its author
The USN is working on infrared and ultraviolet. The way to spot them would be the pretty chain of cyan bubbles where air or water is being ionized by the beam. That, and stuff blowing up. Despite Star Trek/Wars etc coherent light weapons and directed energy weapons in general are invisible in a vacuum and damn hard to see directly in air or water. What you see is side-scatter, and indicates that the weapon isn't very efficient. Coherent light weapons, and directed energy weapons in general, aren't cutting weapons, they're methods for delivering a lot of energy to a small point on the target, heating that point up to where stuff vaporizes rapidly. This is commonly referred to as 'causing stuff to explode'. The side-scatter wastes power en route to the target so that stuff doesn't explode as much as it could if the weapon was more efficient. Research on making lasers more efficient continues.
Note that you want to have a combination of a fairly large main mirror, a lot of power, and as low a frequency as possible (red light is better than infrared, green light is better than red light, ultraviolet is better than green light, x-rays or gammas would be best if only there was a reliable method of producing them that didn't involve nuclear explosions) plus a clear line of sight. Note also that visible light works nicely in air, less nicely under water, while infrared and ultraviolet and even x-rays and gammas, depending on frequency, might waste power hammering through air/water but would work Really Well in vacuum.
Realistically you don't want laser weapons on AFVs, except as last-ditch point-defense systems. You want laser weapons in large aircraft well away from the battle area so that they have long lines of sight, or mounted high up on ships to give long sight lines. Put the AFV on top of a hill and well away from trees and the laser will work well... until every weapon the opposition has targets that hill and possible targets start making smoke. No, mirrors will not work. And spinning the projectile won't either.
Ack. that should be as _high_ a frequency as possible. High freq= low wavelength. You want as short a wavelength as you can get. In times gone by I did some work on a space war game which never got published, in part because those behind it wanted realism, not pew-pew-pew. The space battleships had Really Big coherent ultraviolet lasers (1 metre main mirrors, 100 Exawatt power level at the main mirror, liquid helium cooling) which would make things uncomfortable for anything inside a light-second or so; at two light seconds aiming the things got interesting. If you got a hit the target would notice, but first you gotta get the hit. And you had to use short pulses, and not many of them per minute, or you'd overheat the weapon. And you'd have to replace the main mirror every ever so often anyway. You know, little things never mentioned in Star Trek/Wars. (No, x-ray lasers need not apply. The Bad Guys had x-ray lasers; they a distinct chance of having a slight accident with the fusion bombs which pumped the lasers, typically resulting in No More Bad Guys.)
> directed energy weapons in general are invisible in a vacuum
As pointed by, IIRC, Arthur C. Clarke: in one story, our protagonist is in a suit on the Lunar surface and sees a massive glowing beam coming from an installation in the distance. At first he thinks this is a laser, then realises it can't be, no air to scatter off or to ionise. Turns out to be a stream of molten regolith, moving at a fair old lick. Still not something you want to be hit with.
Hmm, according to what Vader's grandson was able to do to a blaster bolt, maybe they are not any form of laser (no character ever says they are) but are firing gobs of red hot liquid stuff that roils and fizzes when stopped in mid-air? Hand-held railguns whose bolts are entirely molten at the muzzle?
Anyway, back to giant lasers...
Backscatter is the enemy of that, I suspect. While a laser will - to a degree - also 'bounce' off a target, you'd have to be mightily unlucky to get enough energy reflected onto one spot to do harm (human eyeballs not withstanding I guess) but the scatter from a knock-out microwave beam bouncing back from the target would almost certainly affect a ton of 'friendly' RF sensors over a fairly unpredictable area simply by overloading them. Might not break them, per se, but I doubt anyone really wants their kit to suffer temporary 'blindness', especially when there's enemy hardware incoming.
Microwave weapons are in use already. The US Navy are fitting them to their Arleigh Burke destroyers - instead of the lasers the RN are using. I'm sure I saw an announcement about the US army deploying a few early production truck based Microwave anti-drone weapons as well. But perhaps they were test units?
Most armies also have GPS and radio jamming capability. Doesn't knock the drone out, but confuses it about where it is and stops the operator directing it.
There are also various rifle and machinegun sights that will work out the lead distance for you to shoot down drones. They give you a point to aim at in front of the drone, to make it easier to shoot them down.
That won’t work, part 1: reflective polished foil would be highly visible, both to normal sight and to radar. The drones would be seen coming from a long distance.
That won’t work, part 2: as noted above, military laser weapons are not cutting weapons, they are methods of delivering lots of energy to a small spot very quickly. Mirrors, including reflective polished foil, will not work. The surface of the foil would become rapidly expanding plasma, ruining the reflective capability. Worse, that’s rapidly expanding, very hot, plasma in air. You now have a metal fire. Metal fires are unhealthy. For an example of what a metal fire can do, look up USS Belknap/USS John F. Kennedy collision. Also, the air surrounding the reflective foil would tend to become a plasma; the laser just burned a hole in your foil without having to touch it. A thicker coating would be a lot more resistent to this effect. A thicker coating would be a lot harder for the drone to carry, and every gram of coating (recall that the foil has to cover the entire drone) is a gram not available as airframe, weapons, fuel, etc.
That won’t work, part 3: Telsa Cybertrucks are nice shiny stainless steel; they aren’t as reflective as highly polished foil, but they are sure shiny… until they encounter a few raindrops. You think that maybe it’s possible that there might be a bit of weather? Or perhaps some dust, some leaves, other items, messing up that nice shiny finish? Perhaps the drone had best avoid clouds.
That won’t work, part 4: items which reflect visible light do not necessarily reflect infrared and vice versa. Visible light is a quite small segment of the electromagnetic spectrum. There are lots of ways to generate coherent infrared, microwaves, and even ultraviotlet. There are tunable lasers which can use different frequencies.
That won't work, part 5: high energy laser weapons, especially those using ultraviolet frequencies, ionize the air as they go. A way to detect a high-energy ultraviolet laser is the stream of electric blue-white spheres in the air along its track, and the aroma of ozone. The laser just delivered a massive electric charge to the drone. I hope that your drone is hardened against EMP. Hmm. That takes up space and weight which could be used for more useful things.
There's a good discussion on this on Reddit.
Seriously, do you really think that such a simple solution hasn’t been thought of before?
1. Visibility? Who cares? A swarm of fast moving drones doesn't depend on stealth.
2. Metal fire? Okay, but even if that's true, it's a metal fire heading at 100m/s towards enemy troop positions. Sounds like not an entirely wasted endeavour.
3. Any weather that will mess up your drone will also, by the same token, attenuate the laser beam quite rapidly. I bet these tests were made on a clear day.
4. Sure, but you know what does reflect infrared - really well, in fact? Aluminium foil, that's what.
5. If you're relying on EMP effect, why are we even talking about the laser at all? Also, you know what's a simple and low cost way to protect electronics from EMP? Metal shielding, such as foil.
Do I think no one has thought of it? Actually no, I think probably everyone concerned has thought of it. But they're not saying anything, because they're having too much fun playing with their toys and no one wants to spoil it. It's tomorrow's problem.
Lasers ain’t the only things which will be shooting at the highly visible targets. A laser vehicle woud be well to the rear of the Forward Edge of the Battlefield; everything forward of it would see the drones and engage as they go by. They don’t have to kill the drones, just put holes in the foil.
The burning drones would fall out of the sky when the fire reaches the control surfaces and/or engine. Aluminum burns at 1200 Centigrade. It won’t take more than seconds. As the highly visible drones would be detected tens of kilometres away, the burning wreckage would be falling on top of the opposition’s troops. The drone opperators ain’t going to be making many friends amongst their own infantry.
The laser vehicle ain’t a drone, and the weather problem is what happens when water droplets (from rain, or clouds, or icing) are on the surface of the foil, both making the foil considerably less shiny and providing easily heated fluids to add to the plasma sheath.
Aluminium foil burns easily, and doesn’t reflect ultraviolet that well. There’s a reason why research into ultraviolet lasers, despite the problems with them, continues.
The EMP is just the froasting on the cake.
You haven’t thought this through. Foil won’[t work. Mirrors have actually been tested as laser defenses, the results were not pretty. Foil is a nice, simple, cheap solution which will not work.
If you had all this electricity a decent linear motor hefting a half brick at a drone might be pretty effective or even using the power to instaneously convert an enclosed volume of air into plasma could put a decent bit of welly into any old bit junk lying around.
I wonder whether a Falconry unit might be cheaper and just as effective.
You'd be surprised how little energy a laser requires to be effective for this use. Certainly less than a motor. Weighs a lot less as too.
You don't need a continuous beam, you need a very hot, very intense beam for a fraction of a second.
In terms of cool down time and recharging, a rotating phalanx of capacitors would solve this problem. You have one filling up while another is in use. Dissipating heat can be done with water or just loads of thermal mass and a fan.
A falconry unit would be shit compared to a laser turret and nobody wants to bunk in a barracks full of stinky dead mice and sweaty anoraks.
A thick film shield of Aluminum Oxide Ceramic and Tungsten Nitride along with an outer layer organic polymer that I happen to have the formula of will KILL the effectiveness of the laser as that substance combination will have 4000C for the Aluminum Oxide Ceramic layer and 4200C for the Tungsten Nitride Ceramic and the organic intumescent polymer layer which interacts with the lower layers will combine to create a heat shield that not even a 50,000 Celcius plasma arc system can penetrate! A 15 KW laser will be able to create about 2000C and my suggestion needs a 50,000 C plasma heat source that is hotter than the surface of the Sun to penetrate!
I think this weapon is DEAD in the water already!
V