Is an electromagnetic power hammer a feasible idea?
May 16, 2021 5:39 AM Subscribe
I'm curious about the possibility of using an electromagnet to make a power hammer, similar to one employed in smithing work but instead of a motor to drive the hammer up and down you used an electromagnet.
Sounds like a fun project to try and build. StuffmadeHere on youtube does projects like this. I think to truly know the answer you'd have to try, arduously, to learn to code, smith, and build one yourself. Being a maker is hard work!
You could reach out to a local makers club for advice and to share some routers/3d printers for a prototype though!
posted by bbqturtle at 6:58 AM on May 16, 2021 [1 favorite]
You could reach out to a local makers club for advice and to share some routers/3d printers for a prototype though!
posted by bbqturtle at 6:58 AM on May 16, 2021 [1 favorite]
to learn to code
This is hardware. There's absolutely no need to involve software, because the core of this is storing the energy that's going to be dumped into that electromagnet, then doing so instantaneously.
Which involves power electronics, not arduinos or stuff like that.
posted by Stoneshop at 7:40 AM on May 16, 2021 [7 favorites]
This is hardware. There's absolutely no need to involve software, because the core of this is storing the energy that's going to be dumped into that electromagnet, then doing so instantaneously.
Which involves power electronics, not arduinos or stuff like that.
posted by Stoneshop at 7:40 AM on May 16, 2021 [7 favorites]
Most reciprocating power hammers utilize the built up rotational inertia of the motor and flywheel to carry the impact and then quickly reverse out of the way, letting the smithy reposition the piece they work on.
An electromagnet would have to use a very strong solenoid to impart the same energy. Perhaps in conjunction with a spring to store the force. Then the hard part is timing the energization of the windings if the hammer should bounce off the material.
On top of that, you'd have to deal with the magnetic field around your shop and materials.
posted by nickggully at 7:57 AM on May 16, 2021 [1 favorite]
An electromagnet would have to use a very strong solenoid to impart the same energy. Perhaps in conjunction with a spring to store the force. Then the hard part is timing the energization of the windings if the hammer should bounce off the material.
On top of that, you'd have to deal with the magnetic field around your shop and materials.
posted by nickggully at 7:57 AM on May 16, 2021 [1 favorite]
The problem with electromagnets is you need something that's extremely strong or something with a long distance to impart a large force, then you have to go into damper design to deal with rebound and such, which limits its "speed". Then the head becomes magnetized and you need to degauss it periodically lest it imparts a magnetic field on the finished product.
Basically, you're looking at a repeated use "gauss captive bolt gun".
And what advantage does it have over the more... "mechanical" cousins? I can't think of anything.
posted by kschang at 8:26 AM on May 16, 2021 [2 favorites]
Basically, you're looking at a repeated use "gauss captive bolt gun".
And what advantage does it have over the more... "mechanical" cousins? I can't think of anything.
posted by kschang at 8:26 AM on May 16, 2021 [2 favorites]
I mean electric motors are electromagnets, just arranged in a circle to rotate things rather than pull them forward and back.
And there's no reason you can't make an electric motor that goes in a line rather than in a circle. They are called, unsurprisingly, linear motors, most commonly linear induction motors.
Honestly, I've mostly hear about them in terms of maglev trains, and wouldn't be able to give a whole lot of details on how they're actually used. But I suppse I can wildly speculate on why power hammers use rotational motors instead: I imagine it's because you want to keep up a fairly steady rate of hammering, and that's easier when you are using up only a portion of the inertia of a big flywheel's inertia, than when you are using up all the inertia of a linear hunk of metal.
I imagine, though, if you wanted to hit something really hard with a very specific amount of momentum, and didn't care about keeping up a terribly regular/quick pace, you could certainly use a linear motor to do that.
posted by Zalzidrax at 8:38 AM on May 16, 2021 [3 favorites]
And there's no reason you can't make an electric motor that goes in a line rather than in a circle. They are called, unsurprisingly, linear motors, most commonly linear induction motors.
Honestly, I've mostly hear about them in terms of maglev trains, and wouldn't be able to give a whole lot of details on how they're actually used. But I suppse I can wildly speculate on why power hammers use rotational motors instead: I imagine it's because you want to keep up a fairly steady rate of hammering, and that's easier when you are using up only a portion of the inertia of a big flywheel's inertia, than when you are using up all the inertia of a linear hunk of metal.
I imagine, though, if you wanted to hit something really hard with a very specific amount of momentum, and didn't care about keeping up a terribly regular/quick pace, you could certainly use a linear motor to do that.
posted by Zalzidrax at 8:38 AM on May 16, 2021 [3 favorites]
The reason power hammers use flywheels and rotary-transformed-to reciprocating motion is because that design conserves almost all of the energy used to make such a large mass move at such high speed, and can also recover some of energy used in the rebound. Also, it can have energy imparted to it gradually over time, e.g. by increasing the speed of the flywheel.
Any solely-linear solution is going to have a much higher impulse power requirement, which would add greatly to the cost of the power supply and storage system.
You could do it, but would it be cost effective or energy efficient? No.
Probably the smartest linear design would be to use an electric motor to raise the weight, and let gravity impart the drop -- like a pile driving machine. These are pretty clever in using a rope wourd multiple times around a smooth pulley to facilitate easy controlled raising of the weight as the rope is cinched, and then the weight drops as the rope is loosened.
posted by seanmpuckett at 9:09 AM on May 16, 2021 [3 favorites]
Any solely-linear solution is going to have a much higher impulse power requirement, which would add greatly to the cost of the power supply and storage system.
You could do it, but would it be cost effective or energy efficient? No.
Probably the smartest linear design would be to use an electric motor to raise the weight, and let gravity impart the drop -- like a pile driving machine. These are pretty clever in using a rope wourd multiple times around a smooth pulley to facilitate easy controlled raising of the weight as the rope is cinched, and then the weight drops as the rope is loosened.
posted by seanmpuckett at 9:09 AM on May 16, 2021 [3 favorites]
There's a Chinese patent from 1990 you may find interesting.
posted by aramaic at 9:48 AM on May 16, 2021
posted by aramaic at 9:48 AM on May 16, 2021
power electronics, not arduinos or stuff like that
My electronics designer friend calls anything that needs more than ten amps "plumbing".
This would pretty definitely be a plumbing project on the electrical side, but the main design challenges would be mechanical.
My first instinct would be to design it around a suitably heavy vertical steel rod in a couple of sliding bearings, with a couple of opposed pairs of tension springs pulling on it sideways. This structure would naturally want to bounce, and you could set the natural frequency at which it wants to do that as well as the resonant stroke length by manipulating the spring tensions. The bottom of the rod would have a suitable hammering surface attached and there's be an anvil of some sort at some adjustable height below it.
I'd have the top of the rod plunging in and out of a hollow solenoid coil connected to a foot pedal and lever arrangement that would let me raise and lower it. I'd power the coil from the mains, via a suitably rated zero-crossover solid state relay controlled by a circuit that senses the rod position with a couple of opto-interruptors. That circuit would make the relay switch the solenoid current on when the top of the rod just enters the bottom of the coil and off when it exits from the top or the bottom, whichever happens first. This would make the solenoid add a pulse of kinetic energy to the mass-and-springs assembly on every upstroke, synchronized to the actual bounce rate.
With the solenoid positioned fairly high, the solenoid pulses would occur near the ends of the upstrokes and the springs would pull the rod down and out while the solenoid was still pulling it upward; in effect the solenoid would simply be acting more like an additional spring than an actual shover, and its main effect would be to bias the centre of the bounce cycle upwards. But the further down the solenoid gets pushed by the foot pedal, the closer to the centre of the upstroke the upward pull pulses would occur, the better timed their upward pull would be with respect to the existing upward motion of the rod, and the more energy they would therefore add to the oscillation.
I would expect to be able to tune this contrivance for quite good controllability. Might want some kind of spring and damper arrangement on the foot pedal to limit the extent to which the solenoid tries to push itself down further than I'm asking for.
posted by flabdablet at 11:25 AM on May 16, 2021
My electronics designer friend calls anything that needs more than ten amps "plumbing".
This would pretty definitely be a plumbing project on the electrical side, but the main design challenges would be mechanical.
My first instinct would be to design it around a suitably heavy vertical steel rod in a couple of sliding bearings, with a couple of opposed pairs of tension springs pulling on it sideways. This structure would naturally want to bounce, and you could set the natural frequency at which it wants to do that as well as the resonant stroke length by manipulating the spring tensions. The bottom of the rod would have a suitable hammering surface attached and there's be an anvil of some sort at some adjustable height below it.
I'd have the top of the rod plunging in and out of a hollow solenoid coil connected to a foot pedal and lever arrangement that would let me raise and lower it. I'd power the coil from the mains, via a suitably rated zero-crossover solid state relay controlled by a circuit that senses the rod position with a couple of opto-interruptors. That circuit would make the relay switch the solenoid current on when the top of the rod just enters the bottom of the coil and off when it exits from the top or the bottom, whichever happens first. This would make the solenoid add a pulse of kinetic energy to the mass-and-springs assembly on every upstroke, synchronized to the actual bounce rate.
With the solenoid positioned fairly high, the solenoid pulses would occur near the ends of the upstrokes and the springs would pull the rod down and out while the solenoid was still pulling it upward; in effect the solenoid would simply be acting more like an additional spring than an actual shover, and its main effect would be to bias the centre of the bounce cycle upwards. But the further down the solenoid gets pushed by the foot pedal, the closer to the centre of the upstroke the upward pull pulses would occur, the better timed their upward pull would be with respect to the existing upward motion of the rod, and the more energy they would therefore add to the oscillation.
I would expect to be able to tune this contrivance for quite good controllability. Might want some kind of spring and damper arrangement on the foot pedal to limit the extent to which the solenoid tries to push itself down further than I'm asking for.
posted by flabdablet at 11:25 AM on May 16, 2021
Possible, but the heat and energy management alone would be exciting. High-impulse linear drives over distance are difficult. :/
posted by introp at 1:58 PM on May 16, 2021
posted by introp at 1:58 PM on May 16, 2021
This thread is closed to new comments.
posted by Stoneshop at 5:48 AM on May 16, 2021