http://www.animalschool.net/images/Press/turkeys.jpg
Watch out!
1 Like
Yes, we got phoned up by the US Embassy when we just enquired of the manufacturer about some low sensitivity high speed thyristors. Fortunately I was able to direct them to our product pages in a well known electronics catalogue, and explained that we were replacing thyratrons but ordinary thyristors were too prone to gate damage. It seemed we would need to track them in and out. But about this time power MOSFETs became available that could handle the duty, and these were not subject to export controls.
I got the impression that the people who maintained the restricted export list werenāt very bright and did not realise that there were other ways to trigger a plutonium implosion bomb. Not that we were making themā¦just trying to propagate a sharp edged ultrasonic pulse through aluminum looking for slag inclusions.
2 Likes
This is exactly the kind of behavior that I donāt want to tolerate on my planet. These bureaucrats should die in a fire in a mass extinction of their ilk. If that option is not viable, they should be made powerless and aware of it.
They think they can control everything from switches to high-precision machining. They need to be beaten into submission from a down-to-up technology progress.
The technologies have to be liberated.
I donāt want to live in a world where a doctor has to go through paperwork and pay extra just to get a high-sensitivity thermal camera to see e.g. inflammation sites. A friendās doc has one.
When I realized I canāt buy fast switches for this reason, I came up, purely out of spite, with a design based on laser slapper detonators, all actuated via optical fibers from a single shared Q-switched laser. No syncing needed, and the differential propagation delay can be adjusted by the length of the fibers. Prevent this, Man-in-the-suit.
Shaped shock waves can have lots of uses in material engineering. Not just for the detection, potentially also for creation of metastable phases in just the right places within the bulk of the material. Or for measurements of high-pressure behaviors; the extreme pressure at the shockwave collision lasts for a very short time, but you can still get quite some data.
2 Likes
Too easy. Catapult?
1 Like
What about a catapult with guided payload? At the low speeds, even common microcontrollers could handle the realtime job.
1 Like
Yes!
Not quite the same, but I was just wondering about a Pi based system that would observe and compute āguessā predictions for the target paired with a second system that would do the same for recommended adjustments to the catapult mechanism. Result? Very satisfyingly inaccurate. For awhile ā¦ 
1 Like
Requires good shot-to-shot repeatability, a tight grouping of the trajectories of multiple identical shots. This will be a major problem, and the reason why Iād suggest to put the guidance package into the projectile itself.
Of course I can be wrong and the repeatability can be good enough.
Iād suggest an experimental setup with two or more cameras, if possible high-speed (the Casio Exilim ones should be enough), synchronized (can be as simple as a LED in the field of view, so the video streams can be aligned to its blinking). Then using OpenCV (or manual grind) to reconstruct the projectileās trajectory in 3d. The same apparatus, wired for realtime video feed, then can be used for tracking of both the target and the projectile for a real tactical deployment.
Raspi would likely not be enough to handle the two video streams fast enough, but a lower-end gaming rig should be able to cope. Or we could outsource the image processing to a FPGA thatād isolate only moving objects from the image and report directly their coordinates.
1 Like
The drone army wonāt stand a chance! 
1 Like
What if the Pi sampled images from more synchronized cameras on a timer over a longer period of time? Still inaccurate but lots of fun close calls ā¦ esp. if the height doesnāt vary as much. Hours of entertainment.
On the other hand, converting the gaming machine would be extra fun for coolness.
I like that you already know exactly which tools to use for the job.
1 Like
For finding the 3d position we need at least two cameras taking the image at the same moment. (We could get away with different moments but that adds lots of computing complexity.)
The projectile is fairly fast. This gives us rather harsh limitations in terms of bandwidth and processing requirements.
Random thought. We could possibly handle it by dedicated hardware. Four sensors, sensing the projectile, mechanically tracked to keep it in the center (all four signals equal). Or we could use conically scanned tracking radar, also mechanically aimed. This device then would provide cooked trajectory data to the raspi, realtime enough and without the analysis overhead; a standard serial line should do the job.
Perhaps even a single camera or tracker/radar could do the job, assuming that the projectile stays in the plane of the catapult arm swing and doesnāt go significantly to a side.
1 Like
Itās fast, but the operators tend to linger and repeat their movements a lot. If there was a longer sampling period before aiming ā¦ I dunno.
What about the dedicated hardware? That sounds interesting. Does that mean aiming multiple signals at the target and then aiming based on how the targetās behavior alters the signals? And does ācookedā data mean that the raw input has been processed with our code already or somehow prepped to fit it into the serial line?
Hmm ā¦ I will need to recruit some interns to meet my deadlines on this project.
Also ā¦ couldnāt the FCC send more and different drones after us if we used the best sorts of ā presumingly regulated ā signals? Then we might need a bigger boat.
Hey, Iām ready ā¦ Iām just sayinā ā¦
Good lord do you have any idea how expensive, wasteful and energy intensive preparative mass spec is? It is only good for ultra-valuable isotopes. It isnāt even practical for nuclear weapons anymore. Why do you need to print sensor arrays in your garage anyway? I mean it sounds like fun, but ā¦
Drone catapult!
The drone is there as a moving or quasi-stationary object in free space. It is also usually making significant noise, which can be also used for target discrimination, at least for the multicopters.
The radar one, yes. The conical scanned one is sending a cone of signal via a parabolic antenna, and sweeps it around in a way that there is an overlapping area in the center. If you get stable reflection from the target, it is in the middle of the scanning pattern. If not, the reflection is amplitude modulated, with the amplitude dependent on how far off center the target is, and phase corresponding to the phase of the spinning part of the antenna that tells you in which direction you should move the tracking to get the target back to the center. A nice simple control loop.
If we use a pulse radar for this, we can also get the target speed (via Doppler) and distance (via delay). Then we wouldnāt even need two stations for reconstructing the 3d trajectory.
Cooked often means processed. Raw data is e.g. a pair of photographs of the target space, with the target. Cooked data is the coordinates of the target.
Interns good! Share! 
We can go for the ISM band. Or we can use exotic low-probability-of-detection spread-spectrum pulses and never (in a reasonable timeframe) send one at the same frequency. Should be doable, at least at the relatively low frequencies up to a gigahertz or three, with few-100s-$ software-defined radio transceivers.
But weād definitely need interns for that, who can do the math.
Or we can go for the optical processing, outside of the frequency range of the FCC authority. The same conical scanning approach could work for a laser beam. Modulate the beam so the electronics can look for just one specific frequency that the sky and other background is not providing, and use a notch filter so the other wavelengths from the background wonāt swamp the sensor.
Yes. But for growing a couple nanometers thick layer you need only a minuscule amount of the material, which compensates for this factor.
It was never practical for the nukes to start with. It was a quick-and-dirty we-know-it-works-so-letās-do-it step to cover the bases before the K-25 plant got up to speed. Even the Russians wondered, if I remember correctly, why Americans did not just use centrifuges, and considered it to be an intentionally supplied bad intel.
Fun, and properly decentralizes the technological power currently held by nation-states and corporations. All the goodies on eBay are merely secondhand crumbles from their feasts. We the People deserve more.
I want the world where I can order a $200 scanning electron microscope from China, and attach it to a $100 turbomolecular pump with rotors made to order from titanium on a friendās 3d-printed closed-loop optically guided EDM machining rig with submicrometer accuracy.
2 Likes
Something that fires artificial hailstones. I actually think that might be practical.
On the principle that your freedom to hit should stop a short distance from my nose, I detest drones as much as I detest Facebook (which tries to find stuff about me by inspecting the accounts of people who know me.) Personally I feel if they are close enough to shoot out of the sky easily from my house, I should have the right to deal with them. But they might fall on someone else.
If only we could make contact with the local buzzards and teach them that thereās a nice juicy pigeon waiting in exchange for each one of these things that you bring in. And equip them with nets.
My brain is hanging on this one ā¦ is there more than one cone signal that overlaps? Or does the cone overlap with itself and change in a predictable way after bouncing off the object?
And Iām sold on the low-tech, spread-spectrum pulses. With linux machines and, yes, in the garage.
One of those interns is going to end up making all my money on this, er, Freedom Catapult. 
So ā¦ if Freedom Catapult R&D incorporates as an educational 501c3 doing a pilot DIY after-school program and leases some space near the right campus, I can have the MOU signed for those greedy, IP-thieving interns to start next fall. We can do MOUs with the school district, local Continuum of Care, Head Start, CPS, courts and law enforcement too.
We need an academic partner for the program evaluation and to gather data for funding applications too. And a fundraiser.
The drones are causing all the problems in northern Idaho? Hmm ā¦ I suppose Moscow, Idaho could work. Thereās a school there. Or, better! Missoula, Montana ā between May and October. Missoula! Thatās the ticket.
And ā¦ if we presell ads for the Freedom Catapult payloads at a nominal cost, I think the 501c3 can take the income without risking itās exempt status.
Can we also teach the kids to make drone kits as targets for the Freedom Catapult? Is that too busy?
Think about having a flashlight, and spinning its light cone a bit. And sensing how much light reflects from the white ball you are tracking.
If the ball is in the center, it is illuminated equally. If it is moved off-center, the flashlight illuminates it more or less, depending on its rotation angle. The flashing intensity and phase tells you how much and in what direction is the ball off the light.
This is the principle of a conical scan radar. Described here at about minute 24. They are describing more methods of radar tracking, too.
Another nice video here, but it doesnāt deal with tracking directly, just about radar maintenance. Good post-WW2 vintage piece of art: https://www.youtube.com/watch?v=64LUeQ4DAqg
Another good one here, 1962, āDefensive electronic countermeasuresā. https://www.youtube.com/watch?v=d5T1vPmA-l4
With more software-defined radio boxes, we could even rig up a phased array radar with electronic scanning. Look ma, no moving parts!
But the fast ADCs are expensive. Chinaman, where are you when youāre needed?
MOU? 501c3? This sounds like Paperwork, the art so dark that vintage Caribbean voodoo looks like a blank paper sheet in comparison, so difficult to understand that quantum physics pales with envy!
shudder!
Other side of the same coin, Iād say. Close enough in the problematics. Artillery and targets live in a symbiosis; the former cannot exist without the latter.
1 Like
I promise not to explain any of it.
1 Like
Clearly the ammunition shortage is over if the market is supplying novelty items.
2 Likes
What if we tabled Freedom Catapult and did a steampunk rpg about drones invading medieval eastern Europe from the future.
Playerās job? Go back to eastern Europe from the future and defend the house from the new drones that threaten to rewrite history ā and not in a good way. How? Thatās right. Catapults!!
No ready industrial capacity available except cranks and levers and whatever you can remember from your life in the future. Which is now the present. And also the past ā¦ unless the drones change the pastā¦
Never mind. Catapults!
1 Like
