Comment by jgrahamc | Hacker News Reader

Comment by jgrahamc | Hacker News Reader

My stab at how the bullet measures its own rotation:

Suppose the bullet leaves the muzzle rotating at 300,000 RPM (quite reasonable for a rifle). That is 5kHz. It's easily within the capabilities of a Hall Effect sensor to measure a 5kHz rotation and so I'd make the very centre of the bullet contain a free rotating magnet. The rifle itself sets the magnet spinning at some known rate opposite to the rifling. When the bullet leaves the rifle it is able to measure the time between detections of the magnet and given the known rifling, the known magnet rotation speed it would be able to calculate rotations.

Alternatively, the inner magnet could be spinning in the same direction as the rifling but faster and detect complete rotations of the bullet. Either way I think Hall Effect + spinning magnet at bullet core would work.

Also, 5kHz is well within the operating range of a micro controller so this wouldn't need much computer power. Lastly the article says that the weapon is accurate at 500m and that the shrapnel kills everyone within a few meters. Also says that the soldier estimates the target distance.

That sounds like there's plenty of room for the measurement to be not too accurate. Suppose 'several meters' is 5 then you can get away with an error of 1% or so.

I suspect it would be easier to make the front spin and fin-stabilise the rear than try to make a rotating core. That being said, given a known muzzle velocity I don't understand why it's not a simple timing problem.

Muzzle velocity varies quite a lot based on ambient temperature, air pressure, altitude, and propellant grain size.

Ah, so knowing the muzzle velocity might actually be a red herring - the turns per metre travelled is going to be a more stable metric.

Still easier to have temperature and pressure sensors at the weapon( they fit in a wrist watch so there should be no problem to do the same at a high tech gun) and send a rectified time input to the bullet IMHO.

Ten years ago, I had a $2 top with LED lights which cleverly displayed words in a fixed position by using a sensor to use the earth's magnetic field as a reference.

If the top cost $2, it probably was not a magnetometer clocking the LEDs in that toy.

Any take on why a simple timer isn't sufficient?

For 500 meters against the wind it would not guarantee a stable velocity, thus not a stable path length over certain time (it would probably be non-linear to calculate).

So, it seems to exist a correlation between the RPM of the projectile and its velocity over wind's influence that guarantee a linear calculation over the RPM of the projectile.

>Also says that the soldier estimates the target distance.

If the target is behind a rock, the soldier uses a laser rangefinder built into the gun to get the distance to the rock, and then inputs into the gun "the target is 2m behind the front of the rock". So, unless the explosion radius is small, the soldier has to make sure the bullet explodes behind whatever is between the rifle and the target.

It appears to use a Ballastic computer (http://en.wikipedia.org/wiki/XM25_CDTE), like this one for iPhone: http://ballistic.zdziarski.com/

Given that this is basically an ultra-fast artillery device, the tech in this gun more likely represents a natural evolution of applied external ballistics (http://en.wikipedia.org/wiki/External_ballistics) - incidentally this is a purpose for which the ENIAC was designed. I bet the magic has more to do with being able to collect and characterize environmental data through sensors and plug these meaningfully into Ballistic models.

They could just put a coil in the bullet and a permanent magnet along the barrel, then measure frequency produced by the temporary dynamo to calculate velocity (based on known rifling). Then use a timer to measure distance.

You don't need a separate magnet, the earth's field would be adequate.

That's a plausible explanation! It may also be a small accelerometer able to calculate the rotation. The free rotating theory seems very correct though. I'd go with magnet but, I think it ma all be solid state so, accelerometer came to mind.

Reply on news.ycombinator.com