2
$\begingroup$

I am posting this here because of the multidisciplinary knowledge needed to tackle this issue.

Problem: I have to make an electromechanical device that is required to have controllable (downward) piston strokes, that can up to deliver 500N of force upon impact with a body. The piston's stroke length can be flexible, but it needs to deliver at least 500N when it hits something and go back up to its starting position.

The real difficulty: The first problem is meeting the required frequency of this operation. Each full cycle (starting position > travel down > strike > come back up) needs to happen at least twice a second (100 - 120 cycles per minute).

Work done so far: Basic math tells me that this should be doable with 200-300W of power accounting for losses. However, I have an electrical engineering background with very little mechanical/kinematics/dynamic loads knowledge.

I have a 1hp DC brushed motor, that's driven through PWM and I have control over the motor's operation. However, I would really appreciate some input on what kind of set up you would use to transfer the rotary motion to linear motion. The project has a $1000 budget and currently there are the following options: 1. Crank piston style (Rotor turns the crank connected to a linear shaft 2. Carry ball and screw system

Motor Specs: Operated at 24V, 5600RPM no load, 0.6Nm Torque.

I would gladly take any mathematical suggestion too! Theoretical analysis tends to get quite complex as the piston does not have to constantly move with 600N (No load to pull back) but only when it's on the downward stroke.

| improve this question | |
$\endgroup$
  • $\begingroup$ 500-600 N isnt terribly useful if we dont know the impact time it would be much better to know the impulse needed. Also understand that all of this is a nasty problem for your bearings. $\endgroup$ – joojaa Mar 1 at 22:12
  • $\begingroup$ Do you mean that it needs to travel freely down, touch the "body", keep going down until it gets to 500N, and then retract? How exact does the 500N need to be? How big does the face of the actuator need to be? What is the distance over which you expect this 500N to be exerted? Can you include a sketch of the desired operation? $\endgroup$ – TimWescott Mar 2 at 1:30
2
$\begingroup$

Obviously cranks and pistons have been used for hundreds of years for this. Your other option would be fast-acting motors or hydraulic rams.

But a 500N (or is it 600?) force isn't what happens when something hits an object. Hitting an object imparts kinetic energy, and a 1 HP motor just doesn't have a lot to give because you have to speed something up so it can be slowed down when it hits. Newtons would be a force, like compressing a spring, and even then you need to say how much power you are going to put into whatever you're hitting. So 5/600 N over what distance? Also, if you are really planning on slinging something around that hits like 5 sledgehammers even a little bit a tiny 1 HP motor isn't going to cut it. Whatever you're banging around is going to beat the mechanism to hell.

This doesn't seem like a good application for an electric motor, because all the power is in the downstroke, which will slow the motor down and drive up the amps. Maybe a large flywheel would help.

| improve this answer | |
$\endgroup$
1
$\begingroup$

If budget was not an issue you could design a crank / piston assembly from scratch...

As it is an issue then consider re-purposing a small single cylinder engine - lawnmower or chainsaw perhaps. Consider lubrication as well...

| improve this answer | |
$\endgroup$
1
$\begingroup$

Use a screw press mechanism and reducing gearbox. Then it becomes an issue of calculating the force multiplication of the screw and gearbox mechanism to get 500 newtons.

It won't be fast, but it will provide the force you require.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.