www.robowars.org

RoboWars Australia Forum Index -> Technical Chat

Yuppie seeks wisdom
Goto page 1, 2, 3  Next

Post new topic   Reply to topic
  Author    Thread
Cognisant



Joined: 20 May 2012
Posts: 24
Location: Brisbane


 Reply with quote  
Yuppie seeks wisdom

Hello, noob here, a little introduction first, I've been seriously interested in AI theory for years now and that's what brought me to robotics, I'm currently building a quadcopter for fun (hopefully profit) and I'm thrilled to meet all you wonderfully knowledgeable people, please share with me your wisdom as I endeavour to find a place for myself in this little community.

So here's m'plan, since I have no formal engineering background of any sort I'm just going to tell you guys what I'm trying to do, how I'm trying to do it, and you can tell me how I'm doing it all wrong, sound good?

Ideally I would like to enter my first tournament this year, the bot I have in mind is a roughly "T" shaped thing with two wheels taller than the bot itself so it can be flipped over and still drive, the weapon will be a internal flywheel powered oscillating blade, angled downward but not scraping the floor, two height adjustable ball-bearing-like wheels (I forget what they're called, I suppose a couple of screws would do the same job with a little more friction) will let me adjust the blade's height.

Sorry if that's confusing, the blade isn't mounted downward it's just the wedge like shape of the bot itself with the tall wheels at the back that puts the whole frame on a forward tilt which means in operation the forward pointing blade points downward at an adjustable angle.

As for the blade itself I'm thinking titanium with TeflonĀ® PTFE coating, serrated, and before each match I'll spray on some WD-40, the idea being that there's no restrictions (as far as I know) on internal flywheel speed and the blade itself is only going back and forth about a millimetre so I can get a metric shitload of power behind the tips of those titanium teeth at a reasonable speed of oscillation. This will hopefully go through anything softer than titanium like a hot knife through butter, perhaps even punch through titanium plate if I can get a head on strike, although I have no idea how much getting the blade made for me is going to cost, a lot I guess.

Concerning parts I've got a couple of Black & Decker drills, I'll happily take apart the bigger one if it comes to that (I've found it's not very practical for my purposes) and enough of the AXI2814/22 Droidworx motors to open my own store, and LiPo's, speed controllers, balsa wood, foam board, tin plate, paper thin aluminium plate, y'know flying stuff.

Ah balsa wood, that would be good for a laugh.

I guess I should start by downloading Rhino3D, although I've got TurboCAD it just pisses me off but I guess that's just because I don't know how to use it properly (can't even remember the one used in high school), anyway so I'll model the design, get the a quote on the blade, probably have the load bearing flywheel components done in steel or aluminium (advice? sales pitch?) then go physically shopping for scrap I can make the rest of the chassis from.

After watching a number of videos I'm thinking rubber and wood for the armour, a wood CNC machine seems like a good investment and I've got a wooden baton my father made in his youth, it's old varnished hardwood, actual hardwood, very strong and very light, if I can make 3-4cm armour of wood like that and put a sheet of rubber over the top it should stop/slow just about anything. Of course it's still going to get chewed up, I have no illusions about that, it's just that I've noticed how much raw kinetic energy is involved in combat and even if a weapon doesn't puncture the armour the energy is still being transferred through the frame, even throwing the bots around, so designing armour to absorb that energy seems fairly sensible.

I figure if I'm up against a spinner or something like that I just turn my bot around, lower the transmission ratio and use Cartman's sumo ass technique to shunt the opponent into a corner whereby I can slash'n'stab them at leisure, also by happy coincidence ablated wood chips and rubber bits may present an unintentional mobility hazard to low clearance designs, which appear to be the norm.

For the transmission I've been looking at the double cone CVT, I figure it would be the easiest to make, maybe three in total, with one I can run both wheels off the same motor, saving weight and increasing torque in a turn, another to control the overall speed/power ratio (for both the tactical advantage and making overall control easier) and the last is for the flywheel, either to let the motor push the flywheel beyond it's own maximum RPM or for more power to quickly bring the flywheel back up to speed, both if possible; the disadvantage of this is clearly the three servos I'll need but if they're not load bearing I suppose they can be quite small.

Now my latest idea might be going too far (just had it while writing this) y'see the flywheel is essentially dead weight, unless, I were to somehow make the flywheel itself powered, mounting motors inside to be powered by a brushed setup.... Or turn the entire flywheel into one giant brushed DC motor, indeed maybe I can already buy, oh right duh, and I should just put the third CVT between the "flywheel" and the oscillation mechanism, but could it take that much torque...

Post Mon May 21, 2012 6:24 am 
 View user's profile Send private message
Cognisant



Joined: 20 May 2012
Posts: 24
Location: Brisbane


 Reply with quote  

I could reduce the torque on the CVT with gear ratios, although then the problem becomes it going so fast that it might catch fire.

Oh yesss Twisted Evil

Post Mon May 21, 2012 6:33 am 
 View user's profile Send private message
Nick
Experienced Roboteer


Joined: 16 Jun 2004
Posts: 11802
Location: Sydney, NSW


 Reply with quote  

First off, what weight class will this bot be - it sounds like a feather weight from the description? I wouldn't bother with a titanium weapon; it makes excellent armour but does not hold an edge at all - sawing wood or soft plastic would be it's limit.

The weapon sounds like you are reinventing the oscillating sabre saw, perhaps you should drop by a Bunnings to see how they work and get some ideas - you could even use a battery powered one as the weapon, or adapt the blades to your own mechanism. Having said that, saw weapons are never very effective unless you can immobilise the opponent for some time.

Post Mon May 21, 2012 7:35 am 
 View user's profile Send private message
Knightrous
Site Admin


Joined: 15 Jun 2004
Posts: 8511
Location: NSW


 Reply with quote  

Some simple advice.

Start off with a simple, two wheel drive pushy bot, build it and take it to a local competition.

Once you've completed your first event and you haven't had any wheels fall off or battery leads come off during your fights, you are now ready to start building those big dreams.

CVTs, plasma nitride coated beryllium, cold fusion batteries and self aware AI sounds awesome on paper, but 99.9999999999999% of the time, the big complex ideas end up failing before they even get off the ground.

In summary, start simple, work your way up. You'll thank yourself in a few years time Smile
_________________
https://www.halfdonethings.com/

Post Mon May 21, 2012 8:44 am 
 View user's profile Send private message
Cognisant



Joined: 20 May 2012
Posts: 24
Location: Brisbane


 Reply with quote  

Yeah feather weight.

I'm surprised titanium doesn't hold an edge, actually what do you mean, it doesn't really need a sharp edge, I just need something tougher than whatever it's being rammed into. I've got a tempered steel awl...

Anyway I understand then problem with saws, you're essentially dealing with sloped armour, y'know the saw tooth doesn't go straight in, ideally it gouges into the material at an angle and each tooth shaves off a piece at a time until eventually the saw goes in, but that only works if the enemy stays still, or is clamped to a table. With an oscillating blade however the tip ideally goes straight in, so it's like a saw but instead of scratching along the surface of the armour the tip puts all it's energy (plus the collision inertia) into a single point and when head on the serration just keeps those contact points as small as possible.

Of course on an angle a serrated oscillating blade is essentially a saw, but with the advantage over other saws of being effective unlimited (there's no RPM limit because it's not spinning externally) and when being used in a slash the force of the bot's rotation is being added to the impact, also if the serration is all forward pointing and the encounters anything but a flat surface to it's slope then if that first tooth penetrates it stops being a saw and goes back to being an armour piercing penetrator (on sloped armour) I think that would be most likely if the blade was shaped like a shark's tooth, triangular with an outward bulge along it's edge, and again the serration being comprised of forward pointing spikes.

Finally whereas a saw does mainly cosmetic damage if an oscillating blade does manage to penetrate it may not do much cosmetic damage to the armour but it could seriously damage the internal components, just cutting a wire or two would be devastating, hitting a fully charged LiPo battery would be catastrophic.

@Knightrous
Thank you for your advice, but I'm not afraid of failure or making mistakes, if both wheels fall of and the whole thing catches fire I'll be the first to start laughing, before I put it out, take it apart, figure out what went wrong and how to make sure it doesn't happen again.

Learning is a process of humiliation and failure, best to get it over with.

Post Mon May 21, 2012 9:25 am 
 View user's profile Send private message
Nick
Experienced Roboteer


Joined: 16 Jun 2004
Posts: 11802
Location: Sydney, NSW


 Reply with quote  

Titanium is a strange metal; its not particularly hard and its main use as armour is due to its shock absorbing springyness. It you are thinking of a pointed weapon like a sabre saw blade that can penetrate and then cut, a Ti point will quickly round over and stop working on even mild steel. For the money, get something made from tool steel or perhaps bisalloy.

The stroke length should be much more than 1mm unless the saw teeth are microscopic. Reciprocating saws typically have a 5 to 10mm stroke.

Post Mon May 21, 2012 10:16 am 
 View user's profile Send private message
dyrodium
Experienced Roboteer


Joined: 24 Aug 2004
Posts: 6476
Location: Sydney


 Reply with quote  

Most featherweights have armour that could stop bullets cold, eg from titanium up to 6mm aermet steel, so you're right with the fact that most kinetic impacts cause them to be thrown around rather than majorly damaged externally.

However, where saw type weapons can do well is targeted attack to wheels, or take a gamble and aim for the top plate... some robots have little more than duct tape and others would be near impregnable.

Definitely chuck an MDF baseplate together with some drives, battery and esc just to get a feel of things first before anything further, it'd take one afternoon and you can reuse all the parts. Smile Welcome!

Post Mon May 21, 2012 10:34 am 
 View user's profile Send private message Visit poster's website MSN Messenger
Valen
Experienced Roboteer


Joined: 07 Jul 2004
Posts: 4436
Location: Sydney


 Reply with quote  

Anything that relies on hitting a bot more than once in the same place to work generally wont do anything. Imagine your saw thing hitting at a piece of 6mm bizalloy (very hard steel) that's moving past it at say 15km/h.
The big spinners have about the same energy as a .50cal armour piercing bullet.
Their goal is to transfer that energy in one hit so the other bot can't get away or absorb it.

We aren't hating on your ideas at all here, but we want you to be a successful builder

Build a decent box first, its harder than you think and they have a good chance of winning, well in NSW they do anyway ;-P
_________________
Mechanical engineers build weapons, civil engineers build targets

Post Mon May 21, 2012 10:36 am 
 View user's profile Send private message Visit poster's website AIM Address Yahoo Messenger MSN Messenger ICQ Number
Nick
Experienced Roboteer


Joined: 16 Jun 2004
Posts: 11802
Location: Sydney, NSW


 Reply with quote  

The CVT sounds interesting! If you use one motor for drive, how will the bot steer? If you go for a reciprocating weapon, don't bother with teflon coating, just do what the commercial saws do and use a round shaft in a bronze bearing - it will cost just a few dollars.

Post Mon May 21, 2012 6:54 pm 
 View user's profile Send private message
Cognisant



Joined: 20 May 2012
Posts: 24
Location: Brisbane


 Reply with quote  

@Nick
The dual cone CVT is two cylindrical cones with a little wheel in between them that goes back and forth so the distance it's traveling over each cone's surface changes depending upon where it's located. At one end the being the thickest part of the top cone and thinest part of the bottom cone the wheel makes the top cone spin slower but with more torque on the driveshaft while the bottom cone spins faster but with less power. In this way the power from the motor can be split between the two wheels, of course both wheels are receiving the same amount of power from the motor it's just their speed/torque ratio that's being altered.

Most of the time when turning you want more torque on the outside wheel, y'know so it actually turns, and in this setup I guess the hope would be that using one large motor instead of two smaller one's turns out to be that much more efficient to make the mechanical disadvantage worth it. But if you're pushing your opponent then the steering is effectively reversed, the greater torque on one wheel means it will have more of an effect while the other, just gets in the way, unless the cone only tapers down so far so that when in "full turn" the fast outer wheel stops receiving power altogether.

Hmm, a bit messed up.

With three servos the two wheels could have their torque ratio set separately and be disengaged at any time (I'm not exactly sure how but I'm sure there's a way) and if both were disengaged at the same time the power could be redirected to the weapon system (again, in theory) though that would leave the whole bot dangerously immobile.

Huge thanks for the bronze bearing advice, the Teflon was to make sure the blade didn't stick so there wouldn't be energy wasted on pulling it out before it goes back in again, and I've heard that over in Somalia they were spraying AK47 rounds with Teflon to give them armour piercing qualities.

@Valen
Hmm, okay, screw the serration, I'll just go with a spike (perhaps with a sharpened philips head shape to keep the contact point low) and make the oscillation distance smaller, which will increase the oscillation rate to some truly ridiculous speed and gearing that down a bit will bring it to some ridiculous force. The duration of contact wouldn't have to be long, indeed I'm hoping that the initial contact (y'know, from ramming) would be enough to penetrate, even if only a little bit and from there it only has to make the hole big enough to let the diameter of the spike slip through.

Surely it would be just as good if not better than a weighted spinner, it's more or less the same thing just on a very small contact point, ideally head on instead of on an angle, and unlike an external spinner the internal flywheel has no mandated RPM limit, if this can't penetrate then I can't conceive of anything that can.

@Dyrodium
I don't have a well ventilated workshop and the possibility of getting reportedly carcinogenic MDF dust around the place is somewhat off-putting, for much the same reason I've been avoiding carbon fibre for my quadcopter.

Can I use HDPE or would that be too soft?
Actually for structural non-armour purposes do you think I could get away with using thick HDPE instead of aluminium for most things, I've never handled the stuff so I don't know how much it weighs but I get the impression that it's quite light.

Post Tue May 22, 2012 1:35 am 
 View user's profile Send private message
Glen
Experienced Roboteer


Joined: 16 Jun 2004
Posts: 9481
Location: Where you least expect


 Reply with quote  

Hey welcome! ^^

If you have the equipment to make it then why the hell not! Making something work that generally wouldn't is half the fun of it. Having the tools to make it all is probably going to limit what you make in the end though Smile

CVTs are especially awesome. Would love to implement one in one of my spinners with centrifugal adjusters build into each pulley to work them. Just don't have to equipment to make such a beast yet. Need a real mill for that Sad But that would be the way to go for powering a flywheel.

With the hdpe, yeah for internal bulkheads it would probably be okay in thicknesses 20mm and over, plus be easier to work with than aluminium. Definitely wouldn't use it any place its likely to be hit though. Engineering plastic is pretty light. Most of its quite a bit under half the weight of Aluminium for an equivalent thickness.


quote:
use Cartman's sumo ass technique to shunt the opponent into a corner


Hahaaaaa, You must overcome the ass with your mind! >__>
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz

Post Tue May 22, 2012 2:04 am 
 View user's profile Send private message Send e-mail Visit poster's website MSN Messenger
Jaemus
Experienced Roboteer


Joined: 01 Apr 2009
Posts: 2674
Location: NSW


 Reply with quote  

Welcome dude, loving the ideas there, i think everyone else has pretty much summed it up, but looking forward to seeing whatever you do produce Smile
_________________
<Patrician|Away> what does your robot do, sam
<bovril> it collects data about the surrounding environment, then discards it and drives into walls

Post Tue May 22, 2012 4:42 pm 
 View user's profile Send private message Send e-mail MSN Messenger
Cognisant



Joined: 20 May 2012
Posts: 24
Location: Brisbane


 Reply with quote  

So Glen by a "real mill" do you mean something like this?
http://www.actmachines.com/milling-machine-p-1.html

And "centrifugal adjusters" ...what's that?

The cone CVT would only need milled HDPE cones with hobby shop metal rods threaded through them (and secured somehow) then you would weld gears to the end of the rods, put the rods in brass castors (or buy bearings), just not in that order, and have the whole thing held together in a couple of milled HDPE (or maybe polycarbonate) blocks, and it will be screwed in such a way as to have the cones and wheels being pressed together, maybe with nuts & bolts and rubber washers.

Anyways I've until this weekend to draw up some designs before I go shopping for parts, now I don't have a mill (as much as I would love to) but with the drills and coping saw I do have (I could buy a tougher saw) I should be able to cute crude shapes and make holes in HDPE and light metal, I think a HDPE frame with bolted on (again with the rubber washers, maybe stubby cylinders in this case) aluminium plate armour (steel if I can manage it) shouldn't suck too much.

I wonder how much having parts milled for me would cost, and do I have to send them a CAD file or can I just send them something like a paint drawing with the measurements marked, I guess the extra time required to convert that into something the machine would recognise would cost me extra.

@Jaemus
Your quote is hilarious, can you tell me more about it?

Post Wed May 23, 2012 3:04 am 
 View user's profile Send private message
Knightrous
Site Admin


Joined: 15 Jun 2004
Posts: 8511
Location: NSW


 Reply with quote  

Variators are CVT's, but use centrifugal counter weights in to change the ratios of the pullies. However they are generally limited in the available ratios and cannot do over drive ratios.


_________________
https://www.halfdonethings.com/

Post Wed May 23, 2012 8:10 am 
 View user's profile Send private message
Jaemus
Experienced Roboteer


Joined: 01 Apr 2009
Posts: 2674
Location: NSW


 Reply with quote  

quote:
Originally posted by Cognisant:


@Jaemus
Your quote is hilarious, can you tell me more about it?


It's something I just stumbled across one day when bored reading quotes from IRC on http://bash.org/ Smile
_________________
<Patrician|Away> what does your robot do, sam
<bovril> it collects data about the surrounding environment, then discards it and drives into walls

Post Wed May 23, 2012 5:23 pm 
 View user's profile Send private message Send e-mail MSN Messenger
  Display posts from previous:      

Forum Jump:
Jump to:  

Post new topic   Reply to topic
Page 1 of 3

Goto page 1, 2, 3  Next

Forum Rules:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Last Thread | Next Thread  >
Powered by phpBB: © 2001 phpBB Group
millenniumFalcon Template By Vereor.