Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Hypervelocity cannons: Why not centrifugal?

  1. Mar 10, 2016 #1
    After reading a lot about railguns, light gas guns and other approaches to build an hypervelocity/hypersonic cannon, I wonder why the "centrifugal cannon" option hasn't been more thoroughly explored by any military power in the world. Yes, I know the old versions had a precision problem and were unreliable, but I can't see how this couldn't be corrected using modern technology. Sure there must be an obvious reason, but my ignorance doesn't let me grasp it. Please keep with me for a minute:

    Let's build a cylindrical vacuum chamber with an inner diameter slightly over 2 m, on a gimbal to provide for azimuth and elevation. Inside, we install a well-balanced disk 2 m in diameter and an engine (or transmission, or magnetic engine...) able to make it spin at 250,000 rpm (like not a few turbochargers or even heavier industrial machinery.) Up to this point I can't see anything intrinsically impossible or even too expensive or complicated. Any ship and even probably some large airplanes could handle this.

    OK, now let's load a projectile on the border of the disk and spin it up. At just 20,000 rpm, the linear/tangential velocity of this projectile will already be fully hypersonic at sea level: over 2 km/s. At 75,000 rpm, it's close to 7.9 km/s ---Mach 23, almost ground-level orbital speed at Earth's equator! At 125,000 rpm, the rotational speed of a lame turbocharger, it's 13.1 km/s ---way over Earth's escape velocity. If we go up all the way to 250,000 rpm, we get over 26 km/s, over Uranus' escape velocity. Since it's spinning in a vacuum, aerodynamic forces, drag and heating are nil. With proper refrigeration at any other source of heat, you could possibly keep it running for awhile until the appropriate moment to attack.

    So just choose your rotational speed, azimuth and elevation and make a precise computer-controlled system release the projectile at the appropriate instant. Maybe we'd need a system of locks to transfer it from vacuum to atmosphere without damage, and/or a final but simple magnetic "drive" for extra precision. And we have a projectile streaking towards its target at highly hypersonic speeds. Sure, the accelerations are going to be hellish, but that is expected in most if not all hypervelocity designs. I feel this could be way "simpler", maybe more robust and ultimately more inexpensive than railguns, etc. Additional acceleration-resistant on-board guidance maybe could be developed for a second stage of such project, just like they're trying with "conventional" approaches.

    Sure this is going to be impossible, unfeasible, far-fetched and/or plainly ridiculous by some reason(s), since nobody seems to be doing it and certainly I'm not smarter than the top military engineers around the world. But... would you help me to understand these reasons, please? :smile:

    (If you think the higher rotational velocities are too far-fetched, please keep it under 50,000 rpm; Zippe-type uranium centrifuges need at least 90,000 rpm and they've been manufactured for ages, that's '50s tech. Or even lower rotational velocities with a larger disk diameter ---in both cases, we'd still have vastly hypersonic speeds above 5 km/s, around Mach 15.)
    Last edited: Mar 10, 2016
  2. jcsd
  3. Mar 10, 2016 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    Directional control (aiming) requires two axes.
  4. Mar 10, 2016 #3
    Thank you but... would you be so kind to expand, please? As I stated we have the "thing" mounted on a gimbal to provide for azimuth and elevation, and once the projectile is released, it will naturally tend to fly tangential to the release point at tangential initial velocity ---basically following the trajectory you aimed for using the gimbal and the precision release. I'm sorry, but I didn't understand you... :frown:
  5. Mar 10, 2016 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    It takes time to spin it up, and targets do not hold still while you spin it up.
  6. Mar 10, 2016 #5
    It sure does (actually I was calculating how much it would reasonably take to spin the thing up... turbochargers do it real fast), and they sure doesn't (except static targets.)

    But as I said, since it's spinning in a (near-)vacuum, there're close-to-zero aerodynamic effects inside the chamber (turbulence, drag, heating) and you could possibly spin it up upon first alert before entering combat and keep it spinning until you need it or disengage (or overheats, but without aerodynamic effects that would take a while if it's properly refrigerated and/or most components are magnetic and such.) You could even have several "stacked" disks (very much like a multi-disk hard disk drive) spinning at "stand-by" velocities, or deccelerating to reload, or accelerating to shoot again. Add a moving (or multiple) "expel system" if needed and actually that would make kind of an "hypersonic autocannon." All of this could obviously be thoroughly calculated and tested prior to actual deployment during the R&D programme. Don't you think so...?

    Heck, as I see it, it's not so different from building a large, fast precision gyroscope with a couple extras. :smile: The flywheel/rotor of not a few aviation and spacecraft gyroscopes perfectly work for hours nonstop (or days, or more...) at 15,000 - 20,000 rpm with negligible degradation, and even the most basic air-driven attitude ones are up, running and stabilized in about 1 minute or so. Granted my "gyrocannon" is larger and rotates faster, but only by maybe 1 order of magnitude or so. In 1937, the Nazis were already developing 20,000-rpm gyroscopes for their V-2's. And I'd say modern battles involving this kind of weapons would not take hours to develop.
    Last edited: Mar 10, 2016
  7. Mar 10, 2016 #6


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    "Aimed," note the past tense.
  8. Mar 10, 2016 #7
    I still don't get you on these ones, sorry. :frown: We have computers today. The thing is fully aimed in azimuth and elevation (and rotational, thus tangential velocity) when the projectile is released. Before release, your sensors and computers keep it constantly following your target in azimuth and elevation with a pre-computed release speed. As soon as you hit the "arm" switch, the disk accelerates up (or deccelerates down) to the computed release speed (if needed) while the whole system is keeping the aim nonstop. And when you press the "fire" button, it just releases the projectile in the correct moment of the next rotation, then slows down to reload and spin up again. Very much like the launch procedure of any other modern weapon.

    If it's the "multi-disk" version, you can pre-select a "burst" mode and it will launch all the projectiles sequentially, then they'll slow down to reload and repeat. Or, if in "single shot" mode, immediately the next disk will spin up (if needed) to get ready to fire again, while the "shot" one slows down to reload and repeat. In both cases, while constantly keeping the aim computationally.
    Last edited: Mar 10, 2016
  9. Mar 10, 2016 #8


    User Avatar

    Staff: Mentor

    What's balancing the disk after it fires and is still moving at 20,000+ rpm?
  10. Mar 10, 2016 #9
    Yes, I thought about this. Maybe a counterweight at the opposite side of the disk which is also released in the opposite direction? You could think of it as an "empty shell" ---just a worthless, cheap piece of tungsten or DU or plain old lead or the like with the same mass being "counter-released" simultaneously (not very difficult to implement if the release system for both the projectile and the "counter-mass" is electromagnetic ---just cut the power, or apply power from the same capacitor to both releases if it's a fail-safe system, and both objects will be released in the very exact instant. The detonators of a simple two-point levitated pit implosion atomic bomb are WAY more difficult to synchronize, and it's been done since way before I was born.)

    It could introduce a small limitation in the angle of fire to prevent a self-hit with this "balancing shell" (not very relevant if the system is located on top of the tallest mast for instance), but that's all. If I'm not wrong, the disk will stay perfectly balanced with zero eccentricity in mass.
    Last edited: Mar 10, 2016
  11. Mar 11, 2016 #10
    I look at it a different way, instead of a long range hypervelocity cannon, here is my 2 cents worth.

    The inherent lack of precision and accuracy IMO would be more useful for a short range rapid fire, where one would want a spread of projectiles, something like a shotgun blast of pellets. Use 100g ball bearing size projectiles filled into tubes of the centrifuge and launch the whole batch at incoming missles or planes.
    Size up the centrifuge to launch a 1000 or so pellets is the general direction of the incoming theat. The spread would be difficult for the incoming to maneuver against. With the centrifuge mass made to be much larger than the mass of pellets launched, the loss of mass would not be as great a design problem for balancing.
  12. Mar 11, 2016 #11


    User Avatar
    Science Advisor

    High speed turbos and gyros are certainly not 2 m in diameter.

    Peripheral velocity is proportional to the product of diameter and rate of rotation. That is limited by the strength of the material necessary to hold such a disc together.

    For cast iron flywheels you are limited to about 5000 ft/min = 83.33 feet/sec = mach 0.075
    You can do better with carbon fibre but not that much better.
    Last edited: Mar 11, 2016
  13. Mar 11, 2016 #12
    Sorry but there must be something wrong here. The fan of a CFM56 jet engine is 1.735 meters in diameter and rotates at 5,200 rpm, meaning its tips have a linear/tangential velocity of 472.4 m/s, or Mach 1.3761 in standard conditions: http://theflyingengineer.com/flightdeck/pw1100g-gtf/ Same with the RR Trent, causing its typical "buzzsaw noise": http://www.pprune.org/tech-log/92703-large-turbofan-noise.html In normal operation, they're usually "capped" to prevent the tips from going supersonic, but only because it uses to be aerodynamically inefficient.

    Many years ago some Russians decided that supersonic was good for their propellers too and they developed the Kuznetsov NK-12 in the '50s, the most powerful turboprop ever built. It's deployed in the world-famous Tu-95 Bear and variants. The tips of its up-to-6.5 meters in diameter double counter-rotating propellers go supersonic as standard procedure. That's why it's so noisy.

    And we're talking vintage aerodynamics here. Now go imagine in a vacuum...
  14. Mar 11, 2016 #13
    Yes, I've thought in this too, but this is a parallel project... :-p
  15. Mar 11, 2016 #14


    User Avatar
    Science Advisor

    So calculate the tensile strength of a carbon fibre disc with detachable loads along the circumference, then from that compute the maximum circumferential velocity. That will be the maximum speed it can launch a projectile.
  16. Mar 11, 2016 #15
    Certainly! And I'd certainly love to know how to do that... but unfortunately I don't. :frown: I'm just your average person with some scientific background, some imagination and lots of curiosity. But I'm good at Math, and if someone pointed me in the right direction, I'd sure love to give it a try. :wink:
  17. Mar 11, 2016 #16
    If you designed it like the Hot Wheels car launchers


    You could probably just have the motors running continuously and feed the projectiles in as needed. Drag wont be that big of a deal, but bearing friction will be a concern. It also seems like a device like this would lack durability. The mass of the projectile would have to be much less than the disks to ensure they don't slow down significantly.
  18. Mar 11, 2016 #17
    Heck, I would have never thought in that one. :-p But I don't think that would be better than a "conventional" railgun or light gas gun or the like. :frown: As you say, too much friction and I'd add too harsh accelerations involved to achieve hypersonic speeds in a reasonable length. The "trick" in "my gyrogun" is that I don't see any "nasty forces" in action other than the most basic expected in all hypervelocity designs: kind-of-smooth acceleration (programmable as needed), no friction if using levitated disks or the like, no rails to degrade (this seems to be a major problem in railguns), no "firing" of anything but just an electromagnetic release... actually the thing just gets aimed in azimuth and elevation, spins up, releases as needed, then spins down to reload and repeat, all of it in a vacuum with minimum wear under precise computer control. If needed, add an "expel system" for smooth transition from vacuum to atmosphere and maybe a final "magnetic correction drive" for extra precision (you already have the speed and the general trajectory, you'd just maybe need to improve precision a bit.) I hardly can think of a "simpler" and more robust (and maybe inexpensive) machine to do this.

    There must be something that I'm not seeing if people way smarter and better educated than me like the top military engineers around the world are not exploring it. But I still don't know what is it. :confused:
    Last edited: Mar 11, 2016
  19. Mar 11, 2016 #18
    I don't think it is anything as complicated as you think. If you have something like a weapon it needs to meet criteria beyond killing power/coolness-factor. If you have a machine gun that can shoot really accurate and really far with a high rate of fire, but it weighs 150lbs, it wont be useful for troops to carry onto the battlefield. If you have a gun like the one you are describing, maybe the shells fire at Mach 10, but if the equipment needs to be carefully calibrated it wont be able to traverse rough terrain. Do you understand my point?
  20. Mar 11, 2016 #19
    Sure and agreed. :smile: It's only that in the current state of the art, all approaches to hypervelocity guns are kind of "naval", if only (and not only) because of the required power source. Actually they are thought more for extended indirect-fire (ballistic) range than for direct fire power (if sure an hypersonic projectile hitting a nearby target with direct fire would cause really nasty effects!) Using current or proximal technology, I can't envision a "battlefield infantry-like hypersonic gun." Even a heavy tank or the like would have problems handling that, and it would possibly need an external (and quite massive) power source. As I see it, it would be more like a mobile-ICBM launch complex (Topol-style) with its auxiliary vehicles (maybe not so much, but close.) On the opposite, a warship (including some large surfaced submarines) could easily handle all of that, and maybe large airplanes with massive payload capacity could do it too (or static facilities on the ground of course, but those are easily detected, tracked and counter-fired.)

    Your idea of an "infantry-like" hypervelocity weapon is certainly interesting, especially as a particularly powerful anti-materiel / anti-armor gun. But while "heavy" hypervelocity guns are being activelly pursued by several countries right now, I'm not aware of anyone pursuing "light" hypervelocity guns with the current or nearby state of the art.
  21. Mar 11, 2016 #20


    User Avatar
    Science Advisor

    The limitation on flywheels is disintegration due to hoop stress from centrifugal force.
    See; https://en.wikipedia.org/wiki/Cylinder_stress#Hoop_stress

    Flywheels for energy storage are limited by material tensile strength.
    See; https://en.wikipedia.org/wiki/Flywheel
    Which gives the following optimised design parameters for carbon fibre reinforced epoxy.
    Material; Composite: CFRP (40% epoxy). Radius = 1.964 metre. RPM = 3382.

    Compute; Circumference = 2 * Pi * 1.964 = 12.34 metre
    Compute; Angular velocity = 56.367 rev/sec
    Compute; Circumferential velocity = 12.34 * 56.367 = 695.5 m/sec for a CFRP flywheel.

    The speed of sound in air is about 343.2 m/sec
    That gives a mach number of 695.5 / 343.5 = 2.025 mach
    That velocity is a significant over-estimate since projectiles on the periphery will have mass but no tensile strength.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Discussions: Hypervelocity cannons: Why not centrifugal?