Ion Systems - Build progress log

This is the build log for the Ion Systems team. We are a group of students studying MEng Computing and Electronics at Heriot-Watt University.

Our first attempt at building a fighting robot is a heavyweight to enter robot wars, name to be decided.

Currently we have a few things we want to achieve:

• Fastest spinup of any spinner in the competition (Hoping to get it less than 2s to full speed) using 3 150cc motors. (30kW)
• Most manoeuvrable/fastest spinner (using 3 of the Turnigy 50cc motors for drive, 15kW drive)

So for our initial design we had a full body ring spinner designed as a giant bearing, but due to advice from people here it was decided to go with an external bearing design instead.

This was the initial concept:
Attachment 6126

And after the redesign it looks like this:
Attachment 6127Attachment 6128

Its going to be a squeeze to get everything to fit, but I think we will manage. I will update this thread regularly with our progress in the CAD design.

Why no meltybrain?

I mean that's a cool idea but its kinda boaring to watch from what i have seen and a proper spinner can do alot more damage and spinup faster. Maybe at somepoint in the future though!

In theory a meltybrain is the better option, as the whole machine is spinning mass.

And you take out a few motors and bearings out of the equation.

But its way more fun when the thing can utterly destroy itself :lol:

Latest progress, finished designing the ring and roughly worked out where everything is going.
Attachment 6183Attachment 6182Attachment 6181Attachment 6184

Looks nice.
Are your omniwheels large enough to run inverted, or are you adding a srimech?

How is the power transferred from the three Rotomax motors to the gears driving the ring - belts? The motors don't like side loads on the top of the motor; I had to modify mine to make it strong enough. Also, why have three gears? You can save plenty of weight & space by having all the motors belt drive one gear.

Quote:

---

Originally Posted by proffles

Looks nice.
Are your omniwheels large enough to run inverted, or are you adding a srimech?

---

We have a set of wheels for each side to keep the space usage down.

Quote:

---

Originally Posted by overkill

How is the power transferred from the three Rotomax motors to the gears driving the ring - belts? The motors don't like side loads on the top of the motor; I had to modify mine to make it strong enough. Also, why have three gears? You can save plenty of weight & space by having all the motors belt drive one gear.

---

We are making some pulleys that will be attached to the outside of the main can, then we will take off the bit that extends out the top. Either using V-belts or timing belts with centrifugal clutch.

We have 3 gears.. because we have 3 gears. I will rearrange it to have one.

Having pulleys at different heights around the can will make belt driving the one gear much easier. Have a look at micro-V belts; they have excellent grip without needing much tension and the pulleys are no harder to make than a regular V pulley, particularly if your lathe has a DRO.

This is going to be great, you've been looking at 'The Ringmaster' bear in mind the ring will be super costly to make. I'm looking forward to seeing the finished beast.

Quote:

---

Originally Posted by Gabe01

This is going to be great, you've been looking at 'The Ringmaster' bear in mind the ring will be super costly to make. I'm looking forward to seeing the finished beast.

---

Yeah last quote we had was ~£1300 for just the hardox part

Lots of progress since the last update (mostly prompted by the ridiculous timescale).

We have now finished the armor design and how to do the teeth on the ring.

Attachment 6304



Now just to work out how all the internals will be mounted!

Our application should be in on Tuesday.

It looks like a demonic Oreo :lol:

You could make the teeth stronger and easier to mount with a small alteration to the ring:

http://www.nswrfc.org/Nick/Triforce-1.jpg

If you add some ramps (light grey) around the teeth to form a flat bottomed pocket, the teeth will be easier to fit and weld to the ring and should be a little more resistant to tearing out.

you should try sponsorship from Oreo's

Attachment 6305

+1 for demented swirling murderbiscuit. =]

Quote:

---

Originally Posted by overkill

It looks like a demonic Oreo :lol:

You could make the teeth stronger and easier to mount with a small alteration to the ring:

http://www.nswrfc.org/Nick/Triforce-1.jpg

If you add some ramps (light grey) around the teeth to form a flat bottomed pocket, the teeth will be easier to fit and weld to the ring and should be a little more resistant to tearing out.

---

For some reason i never thought to cut the ring with flats on it lol.

Are those ramps part of the cut ring i take it?

Yes, I just made them a different colour to make it more obvious the ring was thicker.

I'd really just dump the omni drive system and go conventional with 2wd. You've got enough of a pain in the arse building a ring spinner powered by brushless motors.

Quote:

---

Originally Posted by typhoon_driver

I'd really just dump the omni drive system and go conventional with 2wd. You've got enough of a pain in the arse building a ring spinner powered by brushless motors.

---

I don't think it's going to be that bad. I would rather try something completely off the wall now and if it doesn't work great then go to 2wd for the next series. (Also we need something academically interesting for the university to help fund it.)

Up to you guys of course. I'd say that going 2wd now with the proviso of an omni drive system down the line gives scope for the project to grow (something academic types all love). Trying to design a machine to survive 50kJ plus impacts and continue to dish them out is no trivial matter in itself.

Omni mixer, job done - http://www.robotlogic.com/product_omx3.html

I use their standard mixer and its a doddle, plus invert at the flip of a switch

Adam, I'm trying to set up a IMX-1 and having a bit of trouble, did you get yours set up easily? I can get it to work roughly, but its not centred, so when the stick is centre, the wheels still spin.

Yes, it worked right away, perhaps some trimming on your stick position to cure that dead band area ?

Found out the robot was almost 20kg overweight so had to ditch the Hardox top and bottom armor, reduce tooth thickness to 25mm and take out a lot of material from the skateboard wheel supports. We also are going to 3d print pretty much everything internal that isn't a motor or axle mount.

Spent about 15 hours on the CAD over the weekend, and ordered the VESCs, drive motors and OMX-3 mixer.

Attachment 6307

I like the profile on the teeth, its a shame you had to thin them down. So the top & bottom armour rings are just screwed into the 8mm aluminium plates now? Given how easily I had 10mm ali bent in featherweights, its not going to last long. 3D printed internal parts can work well if they are thick and printed with 100% in-fill, but its risky to use them for anything structural. If possible, prototype parts with 3D printing and then mill them from HDPE, nylon or acetal for extra strength.

Its 15mm aluminium now, 25.6Kg for them both. I am quite worried about bending, so there will be lots of support pillars going between them, mostly in areas where things are attached to them.

We have some Nylon Alloy910 which is pretty ridiculous how strong it is. Should be fine at 100% as long as it doesn't suffer any major impacts.

We also have a major price problem, the current estimate is like £6500 for the whole thing. Which is just a tad more than we want to spend...

I wonder if adding a thin layer of HDPE would help at all, to act as a spoil plate for the aluminium underneath. Might help to absorb some of the impacts from axes.

I agree with Gary, keep it super simple. By all means use 3D printed parts, just remember that they break easily.

The taulman 910 needs to be kept incredibly dry both during storage and during printing. I'm not just talking about throwing a few silica packets in a bag and sealing it for a while. I use a heated vacuum chamber which causes all the water to evaporate at around 60 deg C. Then during printing i keep it in a dry atmosphere until it hits the hot end. If you don't then it will cause weakness and won't print well. Even then there are a few other tricks with it but i can't be giving away all my secrets ;)

Your humidity must be really high then. I just keep it in a sealed bag with 2kg of silica. Then the spool sits in the same bag while printing with the filament coming out of a little gap. Works great. Last time I had to dry out some nylon 6, I made the mistake of putting the whole spool in the oven... and it melted lol.

Our application is in, fingers crossed that we get picked!

We have now started ordering all the parts and putting them together! This is our latest set of renderings for those who like those things. The robot has been on a diet to allow us to get another set of batteries in.

Attachment 6316Attachment 6317

Certainly is a beautiful rendering - you really seem to know what you're doing with that. Can't wait to see this thing for real.

Here is our latest progress on Triforce.

On the CAD side:

• The pillars which support the skateboard wheels are now a single piece to make them stronger.
• The top and bottom plates have a nice weight saving pattern applied to them.
• 4 support pillars have been added to strengthen the center of the robot.
• The attachment of the armor rings to the aluminium top and bottom plates has been redesigned to make them easier to attach. They are now welded to steel tabs that bolt on either side of the aluminium.
• Added hardox covers to protect the sprockets on the wheels.
• Changed batteries from Turnigy Nanotech to Optipower 30C. Now with slightly lower capacity but less chance of fire, lower weight and smaller.

Build progress:

• Drive motors now running with VESCs. Thanks to Rory for his help.
• Started work on software to provide controls conversion for inverted driving and automatic orientation to make driving easier (Turns to face the same way all the time).

Some nice pictures:

Initial layout for weight saving structure
Attachment 6340

Final weight saving design, saves about 6.6Kg in total:
Attachment 6341

Attachment 6342

New skateboard wheel support pillars. They also provide most of the structure of the robot.
Attachment 6343Attachment 6344Attachment 6345

1000Kg Horizontal force.
Attachment 6346

1000Kg vertical force.
Attachment 6347

Thas vewy nice, I'd personally be a bit worries about those FOS for the vertical load, are you expecting the equiv of 1000kg force? Also, out of interest, are you doing this as a uni project with the uni's resources or just on the side?

Quote:

---

Originally Posted by bitternboy

Thas vewy nice, I'd personally be a bit worries about those FOS for the vertical load, are you expecting the equiv of 1000kg force? Also, out of interest, are you doing this as a uni project with the uni's resources or just on the side?

---

Yeah, the idea with those pillars is that the bolts or skateboard wheels will fail before the pillar. Which means we might lose one or two support wheels but the chassis will remain intact and we just replace the broken parts.

Its on the side, but being sponsored by the university. So we get access to their resources. We actually have everything we need to do it ourselves anyway but we cannot afford the ludicrous price tag >£7K.

Quote:

---

Yeah, the idea with those pillars is that the bolts or skateboard wheels will fail before the pillar.

---

Ah, very sensible.

Are they giving you that money based on the Uni promotion opportunities alone? Seems generous.

Quote:

---

Originally Posted by bitternboy

Are they giving you that money based on the Uni promotion opportunities alone? Seems generous.

---

Yeah. I ain't complaining lol.