Team MeltDown Build Diary

[Giles]

Pretty much moving my old thread here, with a better title that isn't limited to just one robot.

Having completed our first feather and competing at GSL/RoboChallenge champs, me and Dave feel like we've learnt so much that a new build is required!
We are hoping to make it to Germany in September, which gives us much longer than 3 weeks to build the robot unlike last time...

We were thinking something long the lines of biteforce/lockjaw grab/lifter. Here is a first pass design throwing some ideas around:







Will no doubt change tons before I start actually doing proper CAD files for it.

[Andy]

Looks really cool, you could do something similar to biteforce and make it so you can add a wedge and take the forks off the front for fighting spinners

[cnc_machinist_andy]

Interesting looking bot Giles. Only drawback i can forsee is that the barbs may cause u to become locked under a bot or barrier?

[robotic_ants]

You've spent too much time watching the Overhaul 2 build for sure.

[Giles]

Mayyyybeeeee It looked a right mess in the first season but I must say, it looks awesome now :P

@Andy, The plan with this is that both sets of forks are able to rotate in both directions (like Biteforce), so if they got stuck you could retract them in the opposite direction. So in theory, no they wouldn't get stuck, but in actuality, yes, they will probably get stuck at some point :P

[Flag Captured]

The main bother you'll have for a grab/lift is the weight - you've essentially got two lifters in there as well as 4wd. Perfectly doable but you might have to get a bit clever with it! brushless motors would make it a lot easier on weight consumption too!

Also stop making cool machines, you're not being a newbie correctly lol

[Giles]

Yea, we came in heavy with our first bot, weight is definitely an issue. We'll be trying to fit all of the internals into an extremely tightly packed format to keep external surface area to a minimum. Having seen how well laid out bots like Tormenta 3 are, we know how well organised things have to be! And this time, considering the 6 month gap, we should have plenty of time to get titanium parts made to save weight.

At the moment we're focusing on our speed controller dev, so CAD might be on hold for a while until we get some awesome speed controllers going, will probably do a big post about those when they are done. Custom ESC's are totally not needed, but very fun :P

[Rapidrory]

I'm glad you're enjoying ESC development; it's very satisfying when they finally stop producing smoke and start behaving properly :P
With regards to building a lifter, you want all the weight as far back as you can so it won't fall forwards instead of lifting the other robot. The scoops will help keep the pivot point forwards, but worth baring in mind. If you want to go brushless, I have some blank VESC PCBs if you want to try putting some intelligent controllers together (google VESCs for more info). Would mean you could use some really small brushless and still get plenty of power out of them.

[Giles]

I think we're going to try and stick to brushed for now, just because the ESC's are much simpler to make and we want to do some logic/current limt clutches and stuff for the weapon controls built into the ESCs. Maybe we'll change to brushless later on. We're really trying to make a robot that we understand down to the smallest detail, with no elements that we can't repair ourselves (like debugging and replacing components in cooked speed controllers), I feel like jumping straight into brushless will be a step in the wrong direction for now, but maybe in the future!

[Rapidrory]

Fair enough. The VESCs do have current limits and a load of other settings that you can change via a GUI over USB, so if you do decide you need the weight in the future then they'd do the job nicely. Here's what I mean about it being able to use smaller motors; at the end of the vid you can see how small the brushless is compared to the normal drill motor!

[Giles]

That is indeed much smaller!

3D printing is great for making two part moulds btw! Tonight I cast this 60A Polyurethane onto the hub of an old castor wheel using a PLA printed mould. Feels very much like a banebot with treads now! To de-mould it easily I coated the mould in vaseline before casting, and then it popped apart very easily after the polyurethane had cured (about 2 hours). I like the idea of being able to design my own tread patterns and casting them onto whatever hubs I want.

20160408_232913.jpg20160408_232400.jpg20160408_232349.jpg

[Giles]

Looking into grabber mechanism designs now. I'm thinking that the top jaw should be a linear actuator and the bottom powered by a sprocket. A 50mm stroke actuator at 5:1 lever ratio would give me 250mm range of motion on the tip of the arm along with plenty of speed. Looks like you can get a 50mm Gimson GLA-750-S with around 200Kgf max, so that would result in 40KG of holding force at the tip of the arm, which seems plenty to grapple a 13.6kg robot.

[Giles]

Long time no post! In the middle of moving house so not much progress... I've 3D printed a small scale test for the grabber arm though to test the mechanism. It seems to work well and should be able to open/close rather quickly with reasonable torque.

13288219_10154054929885867_601339940_o.jpg

13282814_10154054929535867_1584818236_o.jpg

I'm using a beetle size gear motor in this test, but its a 100RPM one, so its extremely slow! That paired with the small pitch of the threaded rod makes for a very boring video, so I'll spare you all from that. Will post a video/pics of a full scale test once I get my hands on appropriate gear motors (Looking at you Ellis ).

[Coolspeedbot]

Looking into grabber mechanism designs now. I'm thinking that the top jaw should be a linear actuator and the bottom powered by a sprocket. A 50mm stroke actuator at 5:1 lever ratio would give me 250mm range of motion on the tip of the arm along with plenty of speed. Looks like you can get a 50mm Gimson GLA-750-S with around 200Kgf max, so that would result in 40KG of holding force at the tip of the arm, which seems plenty to grapple a 13.6kg robot.

Did you make that animation in Maya?

[Giles]

Yep. I use Maya for all my thinking/testing ideas before I make CAD files with proper dimensions in Inventor. Its just so much faster to make stuff and try out things.