How to Make Mecha Work in a realish setting

Discussion regarding the Outsider webcomic, science, technology and science fiction.

Moderator: Outsider Moderators

raistlin34
Posts: 270
Joined: Sun Nov 08, 2015 3:46 pm

Re: How to Make Mecha Work in a realish setting

Post by raistlin34 »

I once asked an engineer if It would be possible to build something like AMEE from Red Planet.



He answered the main problem would be the battery. I'm confident however drone technology will eventually find a way around.

discord
Posts: 629
Joined: Thu Mar 24, 2011 7:44 am
Location: Umeå, Sweden

Re: How to Make Mecha Work in a realish setting

Post by discord »

The only 'handwaved' plausible reason I have found for larger mecha is actually neural interface with most humans incapable of 'piloting' any vehicle through it, and almost all of the remaining are incapable of piloting anything except humanoid shapes.

Now a neural interface could actually give a lot of those advantages needed to 'replace' tanks, increased situational awareness as you are tied into the sensor systems, response times also improved, with a good HUD and such intuitive control it is actually possible to run a 'one man tank', to explain why you need a onboard pilot? I'd say any kind of lag or packet loss induces severe nausea in pilot, bonus though, since the pilot is 'jacked in' there is very little need for control surfaces or other movement leading to very small crew compartment.

now what about those exceptions to the rule? these few would not be wasted on a plebian thing such as a tank, those would be ship 'pilots', as in pocket battleships(or perhaps smaller but more dedicated 'fast ship killers) running automated systems and 'bridge crew' only, probably space based, or perhaps aircraft pilots, very specialized aircraft pilots.

raistlin34
Posts: 270
Joined: Sun Nov 08, 2015 3:46 pm

Re: How to Make Mecha Work in a realish setting

Post by raistlin34 »

discord wrote:The only 'handwaved' plausible reason I have found for larger mecha is actually neural interface with most humans incapable of 'piloting' any vehicle through it, and almost all of the remaining are incapable of piloting anything except humanoid shapes.

Now a neural interface could actually give a lot of those advantages needed to 'replace' tanks, increased situational awareness as you are tied into the sensor systems, response times also improved, with a good HUD and such intuitive control it is actually possible to run a 'one man tank', to explain why you need a onboard pilot? I'd say any kind of lag or packet loss induces severe nausea in pilot, bonus though, since the pilot is 'jacked in' there is very little need for control surfaces or other movement leading to very small crew compartment.

now what about those exceptions to the rule? these few would not be wasted on a plebian thing such as a tank, those would be ship 'pilots', as in pocket battleships(or perhaps smaller but more dedicated 'fast ship killers) running automated systems and 'bridge crew' only, probably space based, or perhaps aircraft pilots, very specialized aircraft pilots.
Like Shadowrun'¥s riggers?

kiwi
Posts: 129
Joined: Mon Aug 12, 2019 7:35 am
Location: New Zealand

Re: How to Make Mecha Work in a realish setting

Post by kiwi »

Incinerator wrote: I'm curious to know what would happen if a defense contractor decided to throw minor scruples like 'operator safety' and 'nuclear proliferation' out the window and miniaturized a fast fission reactor as much as they possibly could.
I’m no nuclear engineer, but I understand that there would be issues with neutrons - which are produced in large quantities by the fission reactions - and gamma rays - which are emitted by some fission products. Both of these radiation species don’t have electric charge, which means they are hard to stop. In a traditional reactor design, large quantities of water and concrete are used for this (like meters thick).

I think the physics involved boils down to “they actually have to run smack into an atomic nucleus (or maybe electron, in the case of a gamma ray) before they can exchange energy and slow down.” And since atoms are mostly empty space, you need a whole lot of them to give adequate probability of absorbing a given neutron / gamma ray. (Someone please let me know if I’m wrong!)

So we could maybe have a very thin and very special handwavium (possibly quantum handwavium) material which packs atomic nuclei in like brickwork, and stops all the radiation. It would also get hot.

Voitan
Posts: 214
Joined: Wed Apr 06, 2011 3:04 am

Re: How to Make Mecha Work in a realish setting

Post by Voitan »

discord wrote:The only 'handwaved' plausible reason I have found for larger mecha is actually neural interface with most humans incapable of 'piloting' any vehicle through it, and almost all of the remaining are incapable of piloting anything except humanoid shapes.

Now a neural interface could actually give a lot of those advantages needed to 'replace' tanks, increased situational awareness as you are tied into the sensor systems, response times also improved, with a good HUD and such intuitive control it is actually possible to run a 'one man tank', to explain why you need a onboard pilot? I'd say any kind of lag or packet loss induces severe nausea in pilot, bonus though, since the pilot is 'jacked in' there is very little need for control surfaces or other movement leading to very small crew compartment.

now what about those exceptions to the rule? these few would not be wasted on a plebian thing such as a tank, those would be ship 'pilots', as in pocket battleships(or perhaps smaller but more dedicated 'fast ship killers) running automated systems and 'bridge crew' only, probably space based, or perhaps aircraft pilots, very specialized aircraft pilots.
Check out a new series called "Obsolete". Basically compact Votom's are being offered to any human for some cheap resources for reasons unknown and the interface is much like you described.

User avatar
icekatze
Posts: 1399
Joined: Sat Mar 05, 2011 8:35 pm
Location: Middle of Nowhere
Contact:

Re: How to Make Mecha Work in a realish setting

Post by icekatze »

hi hi

By the time you've got neural interfaces of that level, I think you'd probably be better off leaving the meat out entirely.

Overkill Engine
Posts: 135
Joined: Tue May 17, 2011 9:51 pm

Re: How to Make Mecha Work in a realish setting

Post by Overkill Engine »

aleks wrote:
I'm curious to know what would happen if a defense contractor decided to throw minor scruples like 'operator safety' and 'nuclear proliferation' out the window and miniaturized a fast fission reactor as much as they possibly could. Could they get small enough to fit on a set of powered armor?
I doubt they can get more power even if they neglect all safety procedures. The amount of equipment needed to make it work for your suggested reactor type is still considerable and can't be miniaturized noticeably with currently available materials.
You may shrink reactor part but how small can you get heat exchanger, steam generator, coolant and power systems to make it man portable (30-50kg)?
Using it to power powered armor will add extra kg which will significantly add on needed W count.
If even you manage to make it smaller power output will suffer considerably. Currently we may build something that can fit on truck trailer but that doesn't fit your question.

The simplest design follows power inefficiency as seen in RTGs https://en.wikipedia.org/wiki/Multi-mis ... _generator giving measly 125W for 45kg resulting with 2.8W/kg. I know this is completely different type of reactor from one you proposed but this is example of smallest and simplest working reactor.
Even if we neglect all safeties for something that small with current materials we won't get better power output. We simply lack more efficient ways to convert that heat into electricity.
Having hot tube with fissionable fuel inside surrounded with thermocouples cooled from outside to make temperature difference as big as possible won't get many Watts.

The only way I could see current RTG's being workable is with some handwavium tier battery/capacitor tech to allow them to slowly store up the sheer wattage output needed for even basic powered armor.

Which still leaves an unacceptable level of limitations for something that needs to be operationally flexible in the field for military use. You'd have to have the battery packs be hot swappable while the suit wearer is deployed without a full engineering crew and bay. At which point having a trickle charging generator is pointless weight to have integrated in the suit, it would be better to just have a fast charging central station with some sort of mobile distribution method.

Which of course becomes a single point of failure/logistics for the enemy to attempt to destroy or cripple.

User avatar
Ithekro
Posts: 261
Joined: Tue Sep 03, 2019 3:55 am

Re: How to Make Mecha Work in a realish setting

Post by Ithekro »

My preference are the regular type "Arm Slaves" from the anime Full Metal Panic. Mainly because they start out as almost a plausible idea using existing 1980/90 era technology before getting into the more sci-fi related techs. That and their weaponry are mostly redesigned real world weapons repurposed to into rifle type mounts for the mecha to use. (40mm rifle. 57mm shotcannon or a 57mm smoothbore gun)...though the anti-tank dagger is different.

User avatar
Werra
Posts: 840
Joined: Wed Jun 06, 2018 8:27 pm

Re: How to Make Mecha Work in a realish setting

Post by Werra »

icekatze wrote:hi hi

By the time you've got neural interfaces of that level, I think you'd probably be better off leaving the meat out entirely.
Unless meat is cheaper than hardware.

User avatar
Arioch
Site Admin
Posts: 4486
Joined: Sat Mar 05, 2011 4:19 am
Location: San Jose, CA
Contact:

Re: How to Make Mecha Work in a realish setting

Post by Arioch »

Werra wrote:
icekatze wrote:By the time you've got neural interfaces of that level, I think you'd probably be better off leaving the meat out entirely.
Unless meat is cheaper than hardware.
The hardware needed to keep a human occupant alive is almost certainly going to be more expensive (and much heavier) than pretty much any remote control or autonomous control mechanism.

Sarlith
Posts: 6
Joined: Thu Feb 08, 2018 4:50 pm

Re: How to Make Mecha Work in a realish setting

Post by Sarlith »


User avatar
Arioch
Site Admin
Posts: 4486
Joined: Sat Mar 05, 2011 4:19 am
Location: San Jose, CA
Contact:

Re: How to Make Mecha Work in a realish setting

Post by Arioch »

It's cool to see that the commercial version of the Guardian looks very sleek and futuristic, a bit similar to Spot. Unfortunately the product page doesn't provide any specs. In particular, I'm curious what the power source is and what kind of endurance they can get out of it.

I notice that the Guardian XO is positioned "for the industrial manufacturing, oil & gas, maritime, construction, automotive, military & defense, distribution & warehousing, aviation, and aerospace sectors" for lifting heavy objects. I presume any military version would be primarily useful for the same thing... I don't see any combat role for it. Contrary to what Hollywood believes, super-strength is really not a huge asset in modern infantry combat, except perhaps to aid in the carrying of heavy weapons or extra supplies... but you don't need a complicated exoskeleton for either of those roles. And with the limited dexterity of an exoskeleton like the Guardian, it's hard to imagine that any useful task that it could do couldn't eventually be done by an unpiloted robot like Atlas. A Guardian that fell over might have a difficult time even getting up (never mind going prone or doing other things that you would need to do in combat), whereas the Atlas can do bloody backflips.

In the short term, I think the biggest problem is the power source. Spot is a technological marvel, but with a battery endurance of only 90 minutes, I don't see what useful work it can do outside of university experimental programs or perhaps some limited hazardous environment tasks.

The other exoskeleton mentioned in the article, the Japanese "Skeletonics Arrive", is a toy that doesn't look like it can do anything useful. The "arm slave" model only works if the exoskeleton is an open armature; the moment you try to enclose it in armor, you severely limit the range of motion of the arms. (Unless the mech is large enough to completely enclose the operator with space to move his arms, but you need a Gundam-sized mech for that.)

User avatar
anticarrot
Posts: 43
Joined: Sat Apr 02, 2011 8:45 pm

Re: How to Make Mecha Work in a realish setting

Post by anticarrot »

I think the best you could hope for is not a walking tank but a walking Apache helicopter, but with inferior weapons, durability, and protection for the pilots. 90% of the armour woudl be rated against rough-handling and blow-through damage only, with only about 10% being rated to stop light cannons at the major joints and cockpit. (Assuming there is a cockpit.) In the open, such a machine would need to be able to use its main weapon in CIWS mode to force enemy aircraft to keep away, and (try to?) shoot down incoming smart munitions.

Speaking of which, we often see mechs in urban settings, where they have convenient chest high walls buildings to hind behind. But smart artillery shells have been able to go round corners for a while now. Mounting warheads sideways and then firing them at the appropriate time is also old tech. Hiding behind things isn't actually protection any more.

Even 'realistic' mech like this have to compete against the same weapons mounted on small 'jeeps' or light tanks like the CVR. Problem there is power and endurance. Because any kind of 'magic' power plant that gives a mech endurance of more than a few hours will also be energy and power dense enough to give a light ground vehicle the power of limited flight, which is arguably more useful than a pair of legs.


The other realistic (but less popular) is the more human sized uploaded mind / AI 'combat android'. You can easily structure such a thing like a mech from a storytelling PoV (it's piloted by a team of super-sentai uploadeds from a 'bridge' in the 'head') but you're not going to get the same towering engine of destruction that most people want from a mech setting. Unless all the regular characters are the size of mice or something...

User avatar
Arioch
Site Admin
Posts: 4486
Joined: Sat Mar 05, 2011 4:19 am
Location: San Jose, CA
Contact:

Re: How to Make Mecha Work in a realish setting

Post by Arioch »

I guess one use case for a tall, walking mech could be one in which the combat scenario is so asymmetrical that being able to see the enemy is much more important than preventing the enemy from seeing you (such as in a case like the Martian war machines in War of the Worlds). Though in this kind of scenario you would think that air power would dominate, and you'd only need ground forces to mop up. And the hovering versions in the 1953 movie kind of make more sense and feel higher-tech.

Incinerator
Posts: 26
Joined: Thu Jun 20, 2019 12:59 am

Re: How to Make Mecha Work in a realish setting

Post by Incinerator »

I did think of one scenario where a really tall walker might be useful, albeit this is really stretching things. Think of a medieval siege tower with legs. This could be a good option for breaching the upper floors of a skyscraper in an urban environment, where landing troops on the roof isn't an option. If you tried to put wheels or tracks on such a thing, it would either be too unstable to stay standing or too big to fit through the streets. A helicopter's rotors would really blow around anyone trying to rope across the gap between the cabin and the building, plus it would be a sitting duck for any fire from the building's defenders. The latter would also apply to any other sort of hovering aircraft.

Of course, this assumes the attacker doesn't simply decide to shell the building until it collapses.

User avatar
orion1836
Posts: 399
Joined: Mon Feb 06, 2017 11:38 pm

Re: How to Make Mecha Work in a realish setting

Post by orion1836 »

Relevant video.

The necessity to right angle your mech out of the path of incoming 20mm fire and pull out non-conventional weapons to be able to beat an airborne fighter highlights the weaknesses of a mech in today's battlefield. Gasaraki was pretty realistic in how they approached near-future technology.

User avatar
SaintofM
Posts: 163
Joined: Sat Sep 28, 2019 8:01 pm
Location: In a Galaxy Far Far away

Re: How to Make Mecha Work in a realish setting

Post by SaintofM »

orion1836 wrote:Relevant video.

The necessity to right angle your mech out of the path of incoming 20mm fire and pull out non-conventional weapons to be able to beat an airborne fighter highlights the weaknesses of a mech in today's battlefield. Gasaraki was pretty realistic in how they approached near-future technology.
Tanks are pretty weak to to helicopter and airplanes.

Would a ground to air missile system help?

Voitan
Posts: 214
Joined: Wed Apr 06, 2011 3:04 am

Re: How to Make Mecha Work in a realish setting

Post by Voitan »

anticarrot wrote:I think the best you could hope for is not a walking tank but a walking Apache helicopter, but with inferior weapons, durability, and protection for the pilots.
Something like the helo troops in Bubblegum Crisis?
https://www.youtube.com/watch?v=v6DNqjCyA20

User avatar
anticarrot
Posts: 43
Joined: Sat Apr 02, 2011 8:45 pm

Re: How to Make Mecha Work in a realish setting

Post by anticarrot »

orion1836 wrote:The necessity to right angle your mech out of the path of incoming 20mm fire and pull out non-conventional weapons to be able to beat an airborne fighter highlights the weaknesses of a mech in today's battlefield. Gasaraki was pretty realistic in how they approached near-future technology.
I have to disagree. If you're in a fighter, there is a technical term for a ground target that can point a big gun back at you: It's called an anti aircraft platform. Strafing an anti aircraft platform from the front, as happened repeatedly here, will go against all training and common sense. If you must use your gun, attack from the side or rear. Also not familiar with the plot or exact scenario, but that mech seemed to be missing a lot of avionics you'd expect to be there. Or it's there but failed or turned off, so the pilot can open the hatch and 'use the force' instead.

A while back America did a series of combat tests of helicopters VS fighters.
During the two-week exercise, the helicopters proved devastating to the fixed-wing aircraft. In most cases the fighter pilots had no idea they were being "attacked" until they returned to base for debriefing. This led to a series of claims and counter-claims, so for the second week the helicopter pilots were instructed to follow Air Force procedure and call out "guns-guns-guns" when "firing". The kill ratio in favour of the helicopters climbed even higher during this period. Over the entire two-week period, the outcome was a 5-to-1 ratio in favour of the helicopters.[6]

[Aircraft] with 20 mm cannons were less fortunate, stacking up a 0.7-1 kill ratio. < snip> Due to their maneuverability, helicopters are very dangerous opponents when matched against fixed winged aircraft. To this day, the basic lesson is that fighters should stay away from helicopters, and only attack from high altitudes or at long ranges (beyond visual range with missiles) and only if the situation presents itself.
Frankly I'd expect a future fighter's manual will contain an entry...
Mech.
1) Do not engage with guns
2) Especially not from the front
3) No, not even if you 'just see one of them', because there is never just the one mech
4) No, not from the back either, because your last great revaluation will be that a mech's waist can turn quicker than your fighter can.
5) If you absolutely have to kill a mech, use a missile at range if we're not in a coalition, of strike from high altitude with a smart bomb if we are.
6) Just be quick about it. Because if a Mech can see you, then so can the enemy air forces.

Incinerator
Posts: 26
Joined: Thu Jun 20, 2019 12:59 am

Re: How to Make Mecha Work in a realish setting

Post by Incinerator »

At the risk of being pedantic...
anticarrot wrote:
< snip> Due to their maneuverability, helicopters are very dangerous opponents when matched against fixed winged aircraft. To this day, the basic lesson is that fighters should stay away from helicopters, and only attack from high altitudes or at long ranges (beyond visual range with missiles) and only if the situation presents itself.
In the Wikipedia article you are quoting, this particular portion you quote has no citations. In fact, nothing in that section "Phase IV and on" has any citations. It is, at best, a conclusion drawn by whomever wrote it. This doesn't make it wrong, but it does make it an unreliable source.

It should be noted that these tests took place in the late 1970s. Relative to modern times, these were (somewhat) early days for aircraft sensors and missile guidance systems. Further, quoting the same unreliable section, it states "Beyond-visual range combat wasn't practiced at this typical exercise." This implies that the fixed-wing aircraft were required to approach to visually identify their targets before firing, which would be suicide for a modern fighter.

That all said, if a fighter is flying low in a CAS role or for some other reason, a helicopter that is hiding behind a hill or some other kind of obstacle may find itself in a good position to ambush that fighter, due to being in a radar shadow. Against a fighter that knows the helicopter is there, though, I think that helicopter is toast.

Locked