That's a good way to avoid machining costs! Part of the attraction of using a thick rim is to have a very wide tooth - two layers of 12mm with a tooth on both parts might be even better. Wouldn't there be distortion problems from welding?

I am just starting a quick CAD of a bar weapon the same diameter (320mm) as the disk. One advantage that I still have enough 10mm Ti sheet to make two bars, reducing that cost to just the waterjetting. To cut costs further, it could be a 1 tooth design with a regular steel counterweight; that cuts the material, CNC and hardening costs by about half.

You shouldn't get much distortion. The only issue would be heat effected zones, but that is easily fixed but heating the whole disc up to 200 degrees or more, welding the metal while its hot and then let it air cool. That stops the cold steel sucking the heat out of the weld. That's what I attemped to do on Huntsman except I only own a small tourch and had a lot of steel to heat. I think I almost got it Last Rites resistant.

I have a huge propane torch but I don't have a large welder so the job would probably have to be sent out to a pro. Would it be better to weld the inside edges together to avoid having the HAZ on the outside where is can be hit?

ummmm, just keep the weld away from the tooth. Outside or inside is fine, but not on the tooth, and as long as there is good pre and post heating and slow cooling. I'm pretty sure Glen could do it.

Just preheat the whole thing in a oven to 200C. It's easier to do properly compared to using a propane torch and running around it forever trying to keep the heat consistant.
My dad use to weld up diffs for our speedway car in the oven so we could weld them with low hydrogen welding rods. Without preheating the diff, we ended up breaking a couple of diffs due to the welds letting go. When done, you just turn the over down to 50C and then turn it off after a couple of hours, this lets it cool down slowly. Never broke a diff again after we started doing that
_________________
https://www.halfdonethings.com/

whats the radius of the teeth? It'd require a fair bit of re-jigging etc but I think our CnC could taper down the spokes in a few hours lol. just need to balance the disk again afterwards. Alternately Russell is a magician with an angle grinder he could do it with "old school" methods and be within a mm or so.

note many grinder disks may be required lol.

Alternately I'd be interested to see how a large number of fine spokes would compare. (like 20x 5mm thick spokes). The "scary" load we need to watch for is the horizontal vs vertical spinner battle. Hitting stuff in the plane of the disk with pretty much any spoke design is a non issue, but its the off axis hits that cause the thing to bend. As such keeping the spokes as thick as possible would be a good idea.

In my copious spare time I'll look at knocking up a fancy pants design.

Also things like adding radius's between the tooth and disk and such like will make things considerably stronger.
_________________
Mechanical engineers build weapons, civil engineers build targets

The main problem with many small spokes is the cutting cost, its surprisingly expensive in thick steel. I couldn't see a way to calculate the edge length of a complex curve in Rhino, but doubling the number of spokes almost doubles the edge length and therefore the cost.

Rather than spend time & money designing a fancy disk to withstand horizontal spinners, it is probably more effective to have interchangeable weapons and swap to a bar for those fights.

Before we get too carried away on the design, we should think about how much KE is enough to make a winning lightweight bot without going overboard. With the Scorpion motor I have on hand, the likely disk speed will be 5,000 RPM. Doing a quick simulation with a 25mm thick disk weighing around 6Kg gave a KE of 13,000 joules. I have no idea what KE values typical lightweight spinners produce, howerver its over 4 times what Scissorhands produces with only double the weight.

Supply voltage

Picking the supply voltage(s) for Orge is harder than it looks. The drive motors could need between 4 to 6S packs, while large brushless motors like 8 to 12S voltages. That would mean either separate packs or a centre-tapped system and either way it means more complexity and ways to fail.

I just read today that the Robot Power Vypre ESC is finally getting near production; This ESC works up to 45V and opens up a new possibility. Using a Dewalt 36V motor with a 2 stage, 16:1 P80 gearbox and a 10S lipo pack gives these figures:



Pretty sweet numbers IMHO. The 36V motor is a little larger & heavier than the 18V version, with a much larger commutator - it should take more abuse and run cooler. While the motor is heavier, the 2 stage gearbox is smaller & lighter so it all balances out.

At the same time, the Scorpoin S5535 weapon motor does just over 7K rpm at 37V, so something around 5K is likely under load. That makes a 10S battery pack (or more likely two 5S packs) a very easy choice and keeps the design as simple as possible:

Fewer chargers to ship and operate.
One power switch for everything (less weight and wiring)
Fewer packs to buy, ship & maintain.
The overall pack weight should be a little less.

Did I miss anything?

Agreed. Run the same voltage through everything and have one set of batteries. Huntsman had 6s on the drive and 10s on the weapon and it just ruined the robot in my opinion. Also consider how you will transport the batteries. Or chose some that can be brought in america and then re-sold/left behind, ie. cheap.

Yeah, that amount of Lipo batteries would be WAY over the airline's official limits! Perhaps the best thing is to order cheap nanotech packs for US delivery and sell them off after the event - seems a bit of a waste though.

There is no limit if the batteries are under 100Wh. 5s 5000mAh batteries are 92.5Wh so you can take as many as you want in carry on only. The fun part is trying to convince airport security that carrying 12 batteries in your backpack is legal.

Here is the latest disk design, the size may change but I think the general shape is about right:



This disk has a 320mm diameter and only weights 5.7 Kg. The two bearings are wider than the disk, so the idea is to weld in a steel tube to hold the bearings and to keep them in place with the aluminium pulley. I noticed that some Bisalloy grades are available in 32mm, so if they are obtainable the disk could be made thicker rather than wider to increase the weight. Its not too obvious from the render, the single tooth is counterweighted by making the rim thicker on the opposite side.

Looks like a good enough start Got any renders of the rest of robot in plan?
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz

sure does look sweet
_________________
<Patrician|Away> what does your robot do, sam
<bovril> it collects data about the surrounding environment, then discards it and drives into walls

@ Glen: I need to work out the sizes of major components before I can really start on the frame design - will post a couple of ideas though.

I was looking at the video of Catastrophy vs Cobra - any idea why it can self-right with the disk while Demon needs a srimech? I think its a torquey weapon motor and a short (front to back) frame.