Looking for some insight: Mechanical hammer linkage

Technical Technical Discussion
1

Hey all-
I’ve been looking over a design for an upcoming project, and I decided I’d come here for some inspiration on the best method of doing something.

Here’s my issue: I have three pounds total, for the entire project. Exotic materials (like Titanium and Carbon Fiber) aren’t really available to me, unless there’s no alternative. I need to power a hammer coming down, with as much force as possible. This hammer will most likely be a 1/8th inch thick plate of some sort, with a spike/point at the end. The hammer can be as long as necessary, but I only have three pounds for the hammer assembly, and two drive motors, and some rudimentary armor.

Now the question I have for you all: How would you power it? I don’t have weight for pneumatics, or for huge motors that need big speed controllers. The lighter the mechanism, the better. So what would you use?

I’ve been thinking of directly driving it to an old RC car motor I own, but that option has very little control for the hammer… And I’d prefer to be able to have a button that I’d hit which would send the hammer down, HARD.

2

Do you need to be able to power the hammer back up? If not, the simplest solution might be some sort of trigger-released spring.

3

The hammer will need to come back up, and down again, many times over.

For those that are wondering, I’m looking for some ideas for an upcoming combat robot, and figured I’d ask here.

4

springs to drive it down, motorized winch and cable to pull it back up. Fast down, slower up.

5

I’ve never seen an effective beetleweight with a hammer type weapon…

6

I hope to change that.

I’ve got a few ideas for this, mostly along the lines of direct driving the hammer to one of a few motors I’ve got. It’s either that or build some weird heavy linkage, unless someone has a better idea.

7

I would check our MARS’s 2006 robot. I’m certain you could adapt its catapult design to a battle bot arm.

8

I’m picturing…a small motor (small but fast…remember, F=ma, and if you can’t have mass, go with speed) with a pair of wheels driven directly from the motor. The assembly would run along the inside of a pair of rails with thin plates acting as motor mount, wheel mounts, and guides along the rails to keep the hammer aligned for repeatable contact, with stops at both ends for containment. The rails could be integrated into the frame, if you have one, for improved rigidity. You may also be able to wire or cable in place of rails.

I’m not sure which direction the hammer has to go (laterally, vertically, etc.), but another thing you might think about is the direction the hammer travels in relation to your drive train if it has to go laterally (which is what I’m assuming if you’re attacking an opponent). When the hammer strikes, there will be an equal and opposite force on your robot. Not knowing how strong your drive train is and using the above hammer concept, if it travels parallel to the drive train, you may want a simple brake on your wheels (or a drive motor that can’t back-travel) to resist the opposite force, putting more impact on your opponent. Another option is running it perpendicular to the drive train so the opposite force is parallel to the axes of the wheels, which will offer more resistance without improving the drivetrain (unless you’re using something like omni-wheels).

This may all be too elaborate or too crazy, but it’s a brainstorming session, right?

9

Just what I was thinking.

I also like sanddrag’s idea, springs to drive it (high ‘a’), motor to pull it back (can be slower).

Maybe electromagnetics? (Think solenoid, but longer). A relatively light ‘hammer’ of magnetized steel, passing through two coils axial with each other -= first one accelerates it a bit, second one the rest of the way, all in a short time. Use capacitors to store the impulse energy. The first coil can also be sued to bring the plunger back.

The idea is that you switch off the coil just as the driving force starts to fall off (center of plunger hits center of coil), then switch on the second one (and off again at the right moment). Problems will be switching a very inductive load on and off fast, getitng the timing right, and heat management. (Force is proportional to the current in amperes times the turns of wire. High current implies thick wire, but thick wire implies fewer turns).

Hope this whets the imagination.

Don

10

I’m liking all of these ideas, but the electromagnets and coils are a bit out of my range, as I have mechanical knowledge, and very little electrical.

However, I’ve got a CAD design of a direct linkage to a high speed motor I own, and will be prototyping. I figure that if I can’t impart huge force with a tiny, light mechanism, maybe I can smash rapidly with little force, and in the end do more damage?

I’ll post some pictures once I prototype this system.

Thanks for the help!

11

Just curious, what event is this for?

I skipped robogames due to school, but I have an antweight (finished), a beetleweight spinner (Son of Whyachi style, WIP) and hobbyweight (finished) for some norcal tournaments. Hopefully it’s the same one(s) you’re attending.

12

This will probably be for the Halloween bot terror in Gilroy.

I’m not even sure if I’ll go for a hammer, but I think it would be cool to be the only one out there with one…

13

Nice, hopefully I’ll see you there.

Will you only be competing in beetleweight? or are you bringing any other bots?

Also, I would recommend fronting some cash to get atleast a small amount of Ti for armor. May cost $20-30 or so for enough to make better armor. With the amount of spinners now, you really need good armor.

14

I’m probably going to compete in just Beetleweight, as I don’t have much experience with anything but FIRST, and so I figure I’ll start here and work ym way up or down, depending.

15

Maybe you could make the hammer spring-loaded then use a small motor with a cam mecanism to release and then raise the hammer back up? The problem with this would be control, but if you want to hit rapidly it could work.

16

DISCLAIMER: I am not a Mech E (IANAM?)

It might be fun to see if you can rig your drive motors to power the hammer. Perhaps with a funky transmission you can save the weight of additional powerful motors.

Start with a stored energy approach - the springs sound good, but I am sure there are many more options.

Then add some sort of winch/ratchet. The point is to store the energy and then be able to disconnect the motor.

At this point we have a system that feels a lot like many of the other posts. To make it fun, lets add a cool transmission. The goal is to power the winch sometimes and the wheels other times. You could simply go with a tranny that selects one of two output shafts.

I believe a straight up differential would do the trick, with one output going to the motors and one going to the winch. When the hammer is unsprung, the differential will split the motors torque between the two. As the spring stores energy, the back torque will increase. Eventually it will be “infinite” load such that all the power is transmitted to the wheels. The ratchet will then hold the energy until you release it.

The obvious down side to this system is that you are immobile immediately after firing.

Good luck!

17

The easiest way to do this would be to use a “quick return crank” mechanism with the fast stroke set down with the hammer and assisted with springs or surgical tubing to help pull it down. The up stroke could be set so that it has enough power on the up stroke to re-stretch the springs. there would only be one extra part in this mechanism over a direct drive and could get very fast down strokes.


that website explains the mechanism. For your hammer application just remove the slider part and use the beam with the slot as your hammer shaft.

18

You might want to try and search for time-tested mechanisms like a fly press, or a blacksmith’s power hammer.

19

So, amazingly, 57 had something like this on the robot this year. For punching pesky tubes out of the way. Or out of reach of other bots. Admittedly, it only worked for 2-3 rounds at the end of LSR and then the plate was broken off from a ram delivered by another robot, but the idea proved sound eventually.

It was a stored energy system much like people are considering here, high stiffness springs attached to a PVC shaft that was pulled back. The trick we discovered is that pulling back something under high tension is (relatively) easy. But the more stress a release mechanism is under, the more difficult it is to release. Usually.

Our final solution was to retract the shaft via spectra cord wound onto a pulley. The pulley was driven by a worm gear with the worm on a BB tranny/motor combo. Now here’s the trick. The BB tranny + worm assembly was pivoted and sprung so it would naturally pull away from the worm gear. Unless it was pushed into the worm gear by a cam on a servo.

So, starting with the shaft extended it went: turn servo cam to engage worm, start motor to retract punch, stop motor via micro switch when retracted, wait for the right moment… turn servo cam to release worm gear and pulley, AND reverse motor briefly to break static friction.

The two important details we discovered were that last bit there and that turning the worm in one direction made it tend to pull into the worm gear, while turning the other direction pushed away. It’s pretty important to use the former for retraction and the latter for release, else you might put too much pressure on your servo cam. The other trick is simply deciding on how to balance retraction speed and rate of fire with the speed and power of the punch/hammer.

I can take some pics of the silly thing if all of that was about as clear as mud. But I think you could make a pretty hard hitting hammer with this even at beetleweight. Making a swinging, Deadblow style hammer would just substitute torsion springs and direct driving of the hammer shaft.

Now you do have to consider good-ole Mr. Newton here. Hitting something very hard is also going to transfer a good amount of force to your own robot. For a swinging hammer, this is going to translate into your robot jumping everytime you fire. For the punch, hitting another robot will push you backwards, though the force there in on the springs and spring supports which should already be able to handle it. If you miss, though, something on your robot is going to need to arrest that punch pretty rapidly, and that could be as rough on your robot as your punch is on another. So I wouldn’t recommend missing.

20

I was considering building a hammer bot myself when I was watching the carnage at Robogames the other day.

My basic idea is to store a bunch of energy in a flywheel, and use a small servo shifter to transfer the power into the hammer when I want to strike. The hammer could then be returned to the stowed position with some kind of spring.