Asking about the viability of a design

[Shooty]

Hi there. So I've wanted to build a combat robot for years now, and having been inspired by the revival of Robot Wars (and attending the Series 10 filming) and months listening to Inside The Bot, I've decided I'd like to try to design a featherweight. I've already designed some antweights, and have spent the last few weeks frantically searching the internet with tips on building featherweights. I'm very wary about running before I can crawl, and so I don't want to start a design I won't be able to finish. The main reason I've made this post is that I have a concept for a machine, and I wanted to ask folks who have experience in the field if A: the design is actually practical, or at least functional, and B: If it would be possible for someone like me to build it.

Screenshot (4).pngScreenshot (5).pngScreenshot (6).png

My main concept for the machine is that one of the largest problems FW's face is that spinners have gotten so powerful that many machines simply cannot withstand continued blows to their bodies. An interesting solution I saw with the heavies was Gabriel, a machine with plastic wheels designed to absorb hits, while all of the important electronics were kept up high in (relative) safety. And it seems that this idea can work in feathers too: a machine called Huge from the USA placed 2nd in a tournament, as seen here: .

In fact, the messages I sent and received from Huge's builder inspired me to take this route.

Screenshot (7).png

Apologies for the s**tty solidworks aside, the draft pictured here features simply massive 50cm diameter wheels, and has been designed to keep many of the electronics (batteries, ESC's, motors), above 20cm high. All of the body in the draft currently uses 10mm HDPE, although this can be changed. Vertical spinning discs still pose a major threat, but the theory is that flippers won't get as much purchase on such an awkward body, and some weapons such as drums may not even be able to reach the vitals at all!

I know from the designer of HUGE that with some cuts, the wheels can be reduced to weighing 3 pounds each. I've tried to model a hypothetical pneumatic system off of Chris Bonnici's (Liftoff and Blastoff) breakdown here: http://www.fightingrobots.co.uk/thre...weight+flipper, though I will freely admit that this is the part of the machine I know the least about.

Screenshot (8).png

Here are some of the measurements in detail. Now obviously, the thing's ginormous, and I'm hoping that by going an all plastic route, I'll be able to cut some weight via that approach. I'll probably need some sort of steel for the bottom of the machine, and a chunk to act as a medium between the ram and the flipping arm. Another idea would be to tighten the flipper down, taking up less space. The only thing I actually have to build (discounting tools) are the Argos drill motors. I'll probably use rory's Feather ESC's (Between the ESC's and the Nanotwo antweight Kits, it often feels like Rory's built half of my bots already!), but the rest has yet to be decided.

Now obviously, there are many challenges to the design at the moment:

• The drill motors might not have enough torque to propel the machine with those great big wheels.
• The machine is very wide, to the point that it runs a serious risk of running overweight.
• the jagged edges on the sides of the flipper are a liability if facing horizontal spinners (I already have a few ideas to combat that.
• Without a vertically mounted ram, the machine will lose power out of it's flip.
• I live in Ireland, so most parts will be harder to find and more expensive to purchase.

Any feedback would be appreciated. Thanks for reading!

[Theo]

I think the large wheel design is really good. It can withstand spinners better, keeps everything safe and great against flippers. However, their drawback is stability; if you intend to use it upright (flippers aren't really good both ways up), then it's going to roll itself over quite easily, not to mention control issues, they aren't exactly the most manoeuvrable and good positioning is vital or a pneumatic setup or else you're going to waste a lot of CO2. The design itself is good, but for a torque reaction axe mostly, which is what I'd recommend. Pneumatic systems are also quite difficult, especially for a first time builder (on this scale). Regardless, I wish you good luck and welcome!

[Ocracoke]

You could potentially use a linear actuator rather than a pneumatic ram for the sake of simplicity though wherever they be fast and light enough is another matter. Just a thought

[Runsler]

Well...
i had about the same ideas about half a year ago, just wanted (for simplicity) to make a thwackbot, not a flipper.
And it was only for the raptor class, so just 6kg and limited to 12V. Still wheels with 30cm diameter.

Encountered a few problems there (actually a few more, but will try to mention those that might matter most to you.

1. controlling.
It is really, really hard to maneuver such a thing. I had a quiet heavy "tail" to drag around, that also came in, but still... any wedge (or non-wedge, that is) bumping into you from any direction make you drive over it. Good to not actually harm you, but not good for controlled, precise driving. Look the fights of Gabriel and HUGE, they are all over the arena just smashing around anything they could reach. Not because they want, but because they are hard to control once a wheel is lifted just a bit.
Another thing is, you'll need a kinda wide stance. I had 30cm wheel diameter on a bot 28cm-ish wide. That stance was not wide enough, might have to do with the tail, too, but whenever i tried to do fast maneuvers, i flipped myself because the center of gravity was too high.

2. Landing on your sides.
You WILL need some spikes of some sort on your wheels to make you selfright when you land on your sides. Which will happen quiet often. I hoped to just spin on a wheel then, using how unbalanced the bot would be as spinner to flip myself over again. Doesn't work (or not good...). Also, While on your sides, your precious electronics would be in reach of any spinner. Might be the flipper could help on self-righting, but i don't think so, with it working sideways then...

3. rotating yourself.
As i see your design, you'll constantly rotate yourself around your axis when you try to drive forward.
That already is a problem with many flippers: front really low to the ground, but if they accelerate to charge, the front lifts up a bit. in your case... that problem will be multiplied if you want some good acceleration. You'd need something like those stabilizers on childs bikes, just to your back. Leaving you to have stabilizers in the back on the ground, flipper on the front, and the wheels in between easily loose contact to the ground. or make the stabilizers more flexible, still leaving the front to rise a bit.

4. Torque.
You mentioned it yourself...
in my setup, i used the Ranglebox Neptunes wit 18V Motors on 12V (just out of time problems and rules...), so there is more potential with a better setup. Still... even undervolting the motors that much gave me a smoking robot at the end of the last fight, burning one motor. Not because of enemy hits, but because i was on one wheel and tried to rotate the hole bot to selfright. And they got pretty hot after each fight. The wheels were slipping like hell, but there i mostly think it was the center of gravity being quiet a bit away from them, should be better on your bot. So while you possibly will have enough torque to drive yourself, you might get problems with overheating the motors, since it will be like (almost) continuous (almost) stall current through them. Don't know the gearing of the argos drills without searching, but that is something to keep in mind. Some cooling could possibly help here, too...
Anyway, that is a big possible problem, and since i did an awkward setup, you might have less problems with that. Or more, since your bot will be twice the weight, and since i used solid wheels, the percentual mass you want to move is a bit bigger, too

5. Fixing the wheels.
The usual easy method of "just hammer/melt a nut into HDPE-Wheels to make them fit on Drill shafts etc." does NOT work with such big wheels.
Had 5 fights with my bot using that method. in 3 of them the wheels came off, the other 2 were lost because i landed on my sides. Was my first bot, and i did some design errors i've seen right in the first fight but couldn't fix all of them while at the event... But this was the biggest flaw.
You will need a hub of some sort, and a good one, too.
And don't underestimate the forces on the gearbox shaft. since you can't stabilize it from the other side, but possibly jump a few times with that flipper, there will be a lot of force and leverage on the shafts. I used some 40mm square tube of Ali to push the gearboxes in (after some filing), and also push in some ali squares with fitted in ball bearings to further support the shafts. This way i had a rather sturdy construction with the shaft not only fixed in the gearbox, and it was one of the few parts actually working really good.


Besides that, it is really fun to drive such a thing, and it is something not so common as the usual box/wedge-shape.
Don't know about the flipper part, but i did my simple thwackbot including design and stuff within one week on evenings/night with nothing more than a jigsaw, a cordless drill and a file (and ordering a few things^^).

[maddox10]

Living in ireland won't make the parts more difficult to source. Most will be found on this site. Or you just ask individual roboteers by PM.
P&P. Can't be more expensive than in Belgium. And I survive that for about 17 years now, robot wars related stuff that is.

As Runsler mentions, the torque on the axles with wheels that big is huge.
Gabriel and compatriots solve that by mounting the wheels on very very strong shafts, and use drive setups that aim for a reasonably low top speed. Also a huge improvement on controlability .


But, looking at your design, you're building Wheely Big Cheese with wheels on steroids.

[Shooty]

Hi there, quick update. So I've been reading all of the responses posted so far, and they've given me a good perspective of the challenges this bot would face, as well as the difficulties of designing and building it as well. I still think the machine has potential, and I've already made a few adjustments to it (spikes on the sides (don't know why I forgot that in the original CAD), re-designing the wheels). I realize though that there are some things that I simply don't know enough of, such as pneumatics and gearing, that would make this type of build impossible for a first build. So I've decided that that I'll probably follow a different, far simpler path for my first featherweight, and put this one on the backburner. Linear actuators are something I didn't really consider, and I'd love to make a good control bot using them (I particularly like the design of the FW Aegis, building something like that would be nice). Hopefully I'll be able to learn a lot more at my first event, which I'm hoping will be before the end of the year. Thanks again everyone for the posts, they were of great help!