Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!lll-winken!sun-barr!newstop!sun!amdcad!amdcad!military From: abennett@athena.mit.edu (Andrew Bennett) Newsgroups: sci.military Subject: Re: Power Armor Message-ID: <1991May29.011311.7350@amd.com> Date: 24 May 91 20:14:41 GMT References: <1991May24.030024.8386@amd.com> Sender: military@amd.com Organization: Massachusetts Institute of Technology Lines: 77 Approved: military@amd.com From: abennett@athena.mit.edu (Andrew Bennett) > If we really wanted to, could we actually produce effective > PA? If not, what advances still have to be made and what is > the estimated time frame for developement of the respecive > technologies to arrive? > > What real use would PA have in combat? Would they be a cost > effective alternative to convential weapons? A quickie estimate (backgrtound - I've had about a year of robotice/man-amplifier systems experience, as well as about 3 years of weapons design). PA requires three main features: a dense, long-lasting power supply, a robust control/feedback system to manuver it, and a reason to build it. 1) Power supply - you need enough power to keep the systems running for a few hours if you want to make it worthwhile (a suit that only runs for an hour or two between refuelings isn't going to be able to range very far afield). Figuring that it has to propel the user quickly (to avoid being a target), carry a considerable weight in weapons, air conditioning/purifying systems, radiation hardening (optional), shock-proofing (otherwise a near miss is as good as a hit), and supplies (water is probably the most important), it is going to need a very dense, relatively light power source. That means some form of combustion like an internal combustion engine or turbine system (possibly, some new form of fuel cell might work, too). All of these tend to get large quickly. 2) Control system - It has to be able to balance as well as a human, or at least offer feedback good enough for the human's sense of balance to work. Either way, that's a lot of real-time processing and sensing going on. So, figure each suit requires a processor in the 10-100's of mflops, maybe more. Add to that the threat sensing, target acquisition systems and whatever other targeting/weapons goodies you want, and the total processor is definitely not a trivial device. 3) Reason to build it - What does PA offer to make it worthwhile? It will definitely be an expensive proposition to build it. You'll need the latest in strong, light technologies for it, and those aren't cheap (see: B-2). You'll also need state of the art computer and sensing systems, their accompanying software, and all of it has to be robust and *very* reliable. In return, the PA offers a smaller, more nimble target than a tank. That, unfortunately, is about it. The economics tend to point away from such a device. Think of it this way: you generally don't use a weapons system against a target unless the return on your efforts is greater than the loss if you lose your weapon. i.e. you don't attack individual tanks with cruise missiles (unless it's one **** of a dangerous tank), since the effort to produce the missile is much greater to you than the loss of the tank would be to the other side. So, you'd only employ PA against very valuable targets. BUT, if they're valuable, then they're probably well defended. In that case, why risk trained soldiers when an automaton (smart bomb, cruise missile, etc.) may do the job just as well (and cheaper)? 3a) Tactical consideration - PA has one operator, so he/she has to do *everything*: maneuver, targeting, threat monitoring, etc. This can quickly become overwhelming. So, you'll have to develop even *better* sensing/warning systems to help. More cost, sigh. Finally, for the cost of the PA, the other side may deploy a swarm of infantry with small missiles. They can afford more losses then the PA side can (much like your PA vs. tank scenario in the original posting). Summary: I don't think it's worth the $$. Sorry. -- Andrew Bennett MIT Network Services abennett@mit.edu MIT Room 11-124H abennett%athena@mitvma.bitnet 77 Massachusetts Ave. Phone: (617) 253-7174 Cambridge, MA 02139