### Author Topic: Project a object at variable distances/trajectory  (Read 1642 times)

0 Members and 1 Guest are viewing this topic.

#### entity333

• Beginner
• Posts: 4
##### Project a object at variable distances/trajectory
« on: March 26, 2013, 05:13:29 PM »
Hi There,

I am wondering what kind of mechanical device would be necessary to project say......a small rock a distance between 5-20 feet? Would it be something that uses air pressure or something else? I understand that the weight of the object would have to be factored in. As far as firing at a random trajectory i suspect some kind of electronic swivel device would be also neccessary.

My thought is that the working components of the device would fit in the palm of your hand.

Any ideas would be greatly appreciated.

Thanks
Kevin

#### waltr

• Supreme Robot
• Posts: 1,944
##### Re: Project a object at variable distances/trajectory
« Reply #1 on: March 26, 2013, 07:37:38 PM »
Sling shot
catapult
rubber band across two fingers
coil spring in a tube

Doesn't seem that any electronics is needed.

#### entity333

• Beginner
• Posts: 4
##### Re: Project a object at variable distances/trajectory
« Reply #2 on: March 27, 2013, 09:11:52 AM »
Please forgive me, i don't have any experience on electronics or mechanical engineering.

I want it to be controlled electronically. As i mentioned in my initial post.

I'm just wondering what kind of device would be served best for the purpose of projecting an object between 5-20 feet, the size of a small rock. Would this be some kind of actuator? If so what kind?

thanks

#### newInRobotics

• Supreme Robot
• Posts: 1,015
• N.I.R.
##### Re: Project a object at variable distances/trajectory
« Reply #3 on: March 27, 2013, 09:55:52 AM »
You gonna have to give us more information. Do You want that specific device to account for weight of a projectile and output proportional push? Or are all projectiles going to be same weight and output force can be pre-defined and hand tuned? Does device have to calculate output force based on desired distance and launch angle? Is device going to be hand-held, or will it be stationary? Does it have to reload automatically?

To make a small rock (what do You define as small rock: diameter, weight?) fly 20ft You can use any of the methods mentioned by waltr, plus a motor, or two to make it electronic (or semi electronic at least  ). You have to be more specific on what You want to achieve.
"Computer games don't affect kids; I mean if Pac-Man affected us as kids, we'd all be running around in darkened rooms, munching magic pills and listening to repetitive electronic music." - Kristian W

#### entity333

• Beginner
• Posts: 4
##### Re: Project a object at variable distances/trajectory
« Reply #4 on: March 27, 2013, 10:44:39 AM »

The device needs to account for the weight and output proportional force that would randomly project the object a distance between 5-20 feet.
The device has to calculate output force as to not exceed predefined distance of 20 feet.
Launch angle is not relevant but needs to be randomized.
The device will be stationary but can be picked up and moved.
The device does not need to reload itself. However when reloaded manually it would have to sense that it is loaded and proceed projecting the object at random distance/angle within the predefined limits.
object weight would be around 45 grams.

#### jwatte

• Supreme Robot
• Posts: 1,345
##### Re: Project a object at variable distances/trajectory
« Reply #5 on: March 27, 2013, 11:55:13 AM »
If you can fix the object weight, this will be simpler. If you can say "object will weigh 45 grams plus minus 3 grams" for example, then you don't need to include scales in the launcher.

If the object is ferromagnetic, you can build a railgun-like launcher. If not, you will need to either build a slingshot-like launcher with a spring of some sort, an actuator to pull it, and a release to release it, or you will need to do something pneumatic. Pneumatic might actually be simpler. You'll need to measure the kinds of force that can be produced, and calibrate your control appropriately.

Another kind of actuator that might work is one or more rubber wheels that give something an initial velocity; this is used by some auto-firing Nerf guns for example. It depends on the shape of the object. Or you might be able to use some kind of sled that mates properly with the actuator.

The pivoting is simple with two servos (heading and pitch.) Make sure to get reliable, strong servos with good control, such as Dynamixel AX-12 or MX-28. The math for how to change the firing force based on angle and distance is very simple trigonometry and calculus, which is easy to put into the coding microcontroller.

To build a system like this and make it robust, expect to spend a lot of time doing trial-and-error and go back to re-work things based on learnings. Also, if the cost of the final components is \$X, expect to spend at least 3 * \$X in various research/testing.

#### Azraels

• Full Member
• Posts: 57
##### Re: Project a object at variable distances/trajectory
« Reply #6 on: March 27, 2013, 11:59:32 AM »
The easiest way to do is would be to mount a compressed air cannon on a turret.
Have an MCU controlling everything.
A light sensor would tell it when the object is loaded. You place a infrared or laser emitter to one side of tube then the sensor is on other side. When object is placed in tube it breaks the signal letting the MCU know object has been loaded.
You will need an air compressor actuated by relay, pressure sensor, and a solenoid valve. The relay and solenoid valve are controlled by the MCU and the pressure sensor tells the MCU when the right pressure is achieved.
Turret position could be achieved by servo or stepper depending on how large your cannon will be.
The air cannon assembly can be made of PVC tubing or metal pipe. KNOW your PSI safety limits. DO NOT EXCEED!
Google DIY air cannon or just go to instructables.

Getting the object to go the distance you want, you will either have to do the math or just keep firing your object and then changing your programming to get the desired PSI and inclination of the barrel.

Your programming would look something like this:
2. True. Turn on air pump. Pump. Receive PSI from sensor. Pressure reached? True- Turn off Pump
3. Move turret to random position within specified scope.
4. Fire. Activate solenoid valve.
5. Repeat.

This sounds fun, might actually build one myself.
« Last Edit: March 27, 2013, 12:02:29 PM by Azraels »
If your first post is, "I want to build a super complex robot with object recognition, etc..but I have never done programming or electronics...etc."
Your doing it wrong. Start Simple and Work Up.

#### entity333

• Beginner
• Posts: 4
##### Re: Project a object at variable distances/trajectory
« Reply #7 on: March 27, 2013, 12:22:18 PM »
This is awesome guys! Thanks for the advice.

@jwatte - Yes the weight can be 45 grams plus minus 3 grams.

@Azraels - The whole "Unit" I expect to be no more than 8" x 8" and height not an issue....would this be possible with the components you describe?

#### Azraels

• Full Member
• Posts: 57
##### Re: Project a object at variable distances/trajectory
« Reply #8 on: March 27, 2013, 12:30:44 PM »
I dont see why not. Just take what designs you find on Google and scale down.  But I would go with the metal piping instead of PVC for something that small. Make sure you check the rated PSI. Also do not allow your programming to turn the turret to face you. Allow for a safe zone where it cannot point.
If your first post is, "I want to build a super complex robot with object recognition, etc..but I have never done programming or electronics...etc."
Your doing it wrong. Start Simple and Work Up.

#### jwatte

• Supreme Robot
• Posts: 1,345
##### Re: Project a object at variable distances/trajectory
« Reply #9 on: March 28, 2013, 12:05:12 PM »
An air cannon works great if you have reasonable uniformity in the object size/shape. If not, then you can use an air cannon driving a sled that you load the object in -- more like a piston at this point.

#### Duane Degn

• Jr. Member
• Posts: 40
##### Re: Project a object at variable distances/trajectory
« Reply #10 on: April 21, 2013, 09:08:57 AM »
As jwatte suggests, the irregular shape of the rocks will likely cause trouble in an air cannon unless you use some sort of sled.

An alternative which may be easier to implement would be to use a servo catapult.

Here's one made by a friend of mine.

Servo Catapult