[FW] Tracked 4WD Lifter: Aegis

[R9000]

Ok, carrying over from this thread, this is the build diary for my first combat robot. For the past couple of weeks I've been researching and collecting parts, and just now I've got something like a preliminary design down. I'll be building it over Easter, and it'll be mostly HDPE for the chassis. I will however use some 3mm steel (or titanium, if I can get hold of it) for some parts, including the shovel. The robot will be fully invertible, and the weapon will work either way up. I was originally going to use Heng Long Snow Leopard tracks, but the sprockets have a diameter that is smaller than the motors, so I'm probably going to end up re-selling them, and instead I'm using 100mm castors, with custom-made tracks made from two connected bicycle chains. However, at the current design, the robot will run on the castors should the tracks come off or become too unreliable.

I have made some rough CAD diagrams on SketchUp, and the current ones at least give an idea of the layout, but are missing several parts including axles, the tracks themselves, and the current lifter motors are not the kind they should be. This is because I haven't finalized them yet. The chassis also may become shorter, because I have a lot of extra space, and I'll add extra braces between the side plates, rather than just the wheel and lifter axles. I can't finalize the sizes of these until I can measure the tracks, and I need to know how they'll fit around the castors for that.

R6a0auW.jpgt8tDTUu.jpg
eK0zi8O.jpgDodezrB.jpg

Does anyone know any reasonably-priced gearmotors I could use for the lifter? It doesn't have to be very fast (though I don't want it to be too sluggish), just strong enough. I'm using 18V drill motors for the wheels, but obviously they're way too fast.

[PJ-27]

Nice CAD.

For reference my lifter has a max RPM to the arm of 70 (I think...) and that feels fast. I use a scrounged motor/gearbox, that is essentially a wiper motor (from a car) with a 24v motor on it.

[R9000]

Thanks. Wow, yeah 70 RPM is fast. More than 1 rev. per second. I was gonna go slower than that, since the shovel will basically orbit the whole robot. In the case of using a 24V motor however, what voltage LiPo battery should I go for? 5S at 18.5V for the wheels, or 6-7S for the motor to have its full power? I know you can over-volt motors to an extent, but I obviously want to be careful here. One more thing: the shovel arm needs to be powered on both sides of the robot, so I either need two motors to share the load, or a gearbox with two outputs (if they exist). I was thinking of two of these (ebay link) at 12V and 40 RPM. Any thoughts?

I also calculated I'll need about 40Nm of torque overall in the motors (~0.3m (arm length to tip of shovel) * 13.6Kg * 9.81) but two of these motors should only give around 0.43 Nm under load, so I'm guessing I need something bigger. Either that or I've made some error in calculation.

[Phill]

Your arm length seems huge. If it is 30cm axle to tip your robot will be longer than some heavyweight bots. Due to leverage I'd recommend going wider rather than longer. This might screw with your tracks idea though.

Yeah it is 40Nm at the axle as is. But if you get that arm down to 20cm it is 26Nm for eg.

I got there using this calculator
http://www.sensorsone.com/force-and-...ue-calculator/

You also won't need to hold all the weight at the exact length you seem to be designing to scoop and lift so you need a lot less power. Iirc the ballpark figure is half the weight to just topple over a robot.

You will need a torquier motor(s) to achieve this, yes. But the difference seems to be less than you thought, especially with a design tweak.

[R9000]

Yeah, I know some component of the opponent's weight will always be supported by whatever reaction forces it has, I suppose my calculations were more of an overestimate to ensure enough power. I'm going to make the chassis and arms shorter and recalculate. However I really need to ascertain how much room I'll need in the middle for the motors first. Also, remeasuring, strictly the arm length from axle to shovel bracket is around 0.23m, it's the lip of the shovel that adds the extra length, and I rounded up even then.

PJ, I know you said you used a windscreen-wiper motor, but what's the approximate torque and size of those, and is there some variant I could use where I could have one on each side? That's my main problem here, getting motors that fit properly so that the lifter axle is powered on both sides.

[R9000]

Ok, shortened the chassis to get the arm length down to roughly 0.18m, 0.23m from tip of shovel to axle. Not sure if this'll allow me to fit my motors in though. If I can get one with a gearbox in-line with the motor (but slower and more powerful than the drills) I should be good. Only about a 16mm gap between the shovel and tracks though, so I may have to make the arm slightly longer to accommodate.

Here's the new CAD diagram:
hvpW7uE.jpg

I also threw on some measurements for you to get an idea of the dimensions.

[daveimi]

Hi, this design will be fine for whiteboards but if you plan to enter spinner events (Robochallenge) your arms will get chewed off. There are other lifter pusher designs such as Minimoth that achieve something similar whilst being spinner proof.

[R9000]

How does Minimoth manage it? Does it just have thicker lifter arms?

[daveimi]

It's more of a bulldozer and the arms are up and out the way. It's not invertable but can roll itself over, it's the feather version of behemoth. Vid below of Minimoth surviving Decimator. You'll recognise the shape when you see it. It's not the same as you're trying to achieve, but the raw energy in the feather spinners will just pull pieces off anything that's not thick deflective armour. The trick to Minimoth is that he has a "skirt" round the bottom for spinners to chew on, and if all else fails can lift the blade up out of the way. As I say, it's a different design to you. Also look at Sewer Snake for inspiration.

My point being, imagine Decimator chewing on your arm axles where they come out the body...

https://www.youtube.com/watch?v=UGghO2ymWDU

[R9000]

Yeah I know what Behemoth and Minimoth look like, but they both have pneumatic powered flippers. Sewer Snake's design is clever, but yeah, not quite what I'm going for. Is there not any way I could protect or strengthen the arms sufficiently to resist spinners? If I'm gonna lose this design, I might as well go for a generic flipper.

In other news, I'm still really struggling to find gear motors with high enough torque. The highest (reasonably sized motor) torque I could find was 50Kgcm (~5Nm). And I still need about 26Nm. There must be lower-geared, more powerful motors somewhere, at least that aren't too big. Anyone know of any?

-Edit- Looks like someone has trodden practically the same path as me in 2010. I might end up redesigning, considering this one seemed to just disappear. :/ Then again, his beetleweight seemed to turn out fine.

[daveimi]

You've stumbled upon the reason no one uses gearmotors for a lifter lol. You could build a bulldozer with an electric actuator instead of gas. Actuators are a bit slow so you'd need summat meaty near the fulcrum so youd have speed as well as enough push to lift.

[overkill]

Ah, Those crazy Aussies and their spinners! You are safe from Decimator this year but it will be back with a much larger weapon motor next year.

Have you looked at the Banebots P80 or BB220 gearboxes? I did a CAD design with an NTM 5060-380 brushless motor, a P80 64:1 gearbox and a 2:1 final chain drive, which takes much of the strain off the GB. It had a no-load speed of 65 RPM but I don't know how much torque it would generate.

[R9000]

Those gearboxes look fantastic, although expensive, and I'd rather not import if I can avoid it. Excellent suggestion though, and one I might look into in future. However at the moment I think I'm looking at just redesigning the weapon completely - I just bought a pneumatic cylinder from Danny B with the idea of using it later - maybe I'll use it here? Idk. It's probably gonna be a chassis redesign whatever happens, unless I just make it a pusher. Ive got some thinking to do, I guess. Problem is, the majority of top-mounted weapons defeat the purpose of it being invertible.

[harry hills]

Dewalt or similar beefy drill in low gear add another stage on. Its what most other people with gearmotor powered lifters do. Hardox and careful design and execution will mean its survivable in full combat.

[plargen]

You do see the odd banebot gearbox on ebay