If all you want is something that can in effect oscillate (waving a flag), then all you need is a core rod, a spring on either end of the rod, and a coil inbetween the springs. Charge it one way, charge it another. Let the springs dampen the effect so it mechanically makes the snappy solenoid action more sinusoidal, and then it's just a matter of swapping polarity on the coil. You also need a snubber diode on the coil to eliminate field discharge when swapping polarity during a pause between polarity swaps, so it will be a bit snappier and not heat so much.

 

If all you want is something that can in effect oscillate (waving a flag), then all you need is a core rod, a spring on either end of the rod, and a coil inbetween the springs. Charge it one way, charge it another. Let the springs dampen the effect so it mechanically makes the snappy solenoid action more sinusoidal, and then it's just a matter of swapping polarity on the coil. You also need a snubber diode on the coil to eliminate field discharge when swapping polarity during a pause between polarity swaps, so it will be a bit snappier and not heat so much.

Thanks for the suggestion, Boba. But my application is for a battery-powered mechanism, and it's critical that I find the most efficient way of activating the solenoid. That means that LOTS AND LOTS of experimentation will have to be involved. And yes, a snubber has always been on the table ... I don't like the smell of refried nFets much...

 

Thanks for the suggestion, Boba. But my application is for a battery-powered mechanism, and it's critical that I find the most efficient way of activating the solenoid. That means that LOTS AND LOTS of experimentation will have to be involved. And yes, a snubber has always been on the table ... I don't like the smell of refried nFets much...

What I suggested could be battery powered, not sure why you are thinking it can't be. If anything I've simplified your problem significantly. All you need is a tank circuit, a couple of OpAmps, a couple of transistors and you have a self-sustaining cycler for your inductor.

You need a short, wide coil with many turns:

https://www.allaboutcircuits.com/textbook/direct-current/chpt-15/factors-affecting-inductance/

 

A battery, a 555 based pulse generator and a piece of Nitinol. Pretty simple.

I’m committed to bring this up again and again. It’s a commonly used technique for animation. You apply a voltage to a piece of nitinol wire, the flag moves smoothly one way. You remove the voltage, the flag smoothly moves back in the other way.

Repeat as desired.

No worries about driving a solenoid. No high current transistors. Timing is but a simple calculation for s 555 pulse generator. Current requirement calculation is but one equation. Typically, a D cell with run for hours.

Search for nitinol. Everything you’ll need to know will appear in the first few results.

 

LOL! ... thanks for the suggestions, guys. But I don't know where you got the idea that I needed a flag-waving cycler thingy... my device is one-shot only, but it's power source is so limited that I need to make sure its configuration is optimal. It's essentially an ultra low power mcu waiting for an event to happen before it triggers the solenoid, and it's powered by three AA batteries. It's important that those batteries last for as long as possible, in the order of several years. Triggering events should not be more that three or four per year.

 

LOL! ... thanks for the suggestions, guys. But I don't know where you got the idea that I needed a flag-waving cycler thingy... my device is one-shot only, but it's power source is so limited that I need to make sure its configuration is optimal. It's essentially an ultra low power mcu waiting for an event to happen before it triggers the solenoid, and it's powered by three AA batteries. It's important that those batteries last for as long as possible, in the order of several years. Triggering events should not be more that three or four per year.

I thought the TP was 'Hamlet', and he said he wanted a flag-waver in post #11....?

As for your needs, making a solenoid isn't all that difficult. Not really. Straight math, materials, and a little physics to help it, thrown in. A solenoid is a current device, not a a voltage device. Your limiting factor is your power-source, so choose that wisely (check the datasheets for your battery source). If you want to power a 1-shot solenoid a few times a year, using batteries, one way to do it is to charge a capacitor, and then dump it into the solenoid when you need to fire. If you can use a small solar cell to assist in reducing drain on the battery so much the better, and the cap can be charged over time.

Another way, slightly more complicated is to use a spring as your rod-driving force, a motor to turn a screw to re-charge the spring, and a solenoid to release a catch that lets the spring discharge. That way the only time the battery is used is a to fire the solenoid and b to turn the motor. Using a threaded shaft makes the motor work the least to restretch the spring against a catch, so it could be tiny, and recharging the battery, or reducing drain on it, could be managed by a small attached solar cell.

 

Three or four events per year is good information. Three AA batteries is more data. How long does each event last? Seconds, minutes, hours??

At 650 mAh, that’s conservatively 160mA/event/year. For “several” years of operation, we need to know two things. Current draw of the microprocessor and what “several” is?

 

Three or four events per year is good information. Three AA batteries is more data. How long does each event last? Seconds, minutes, hours??

At 650 mAh, that’s conservatively 160mA/event/year. For “several” years of operation, we need to know two things. Current draw of the microprocessor and what “several” is?

Enough to just trigger the solenoid and release a spring loaded mechanism. So that would be just a couple of seconds.

 

Here's a document I found on the subject with very useful information.

 

I thought the TP was 'Hamlet', and he said he wanted a flag-waver in post #11....?

You're absolutely right, Boba. I didn't mean to hijack the thread, but it seems that I've unwittingly done so. @Hamlet , if this bothers you, I'd be more than happy to open my own thread. I've been posting my questions here because I thought my interests were the same as yours and that I could add important contributions to this discussion.

 

Here's an interesting bit I just noticed. When using a magnetic plunger of the same length as the coil, the force exerted by the solenoid is largest when the plunger is inserted halfway through it.

I'm pretty sure that is true with just a iron plunger too. Been many years and a couple of strokes since working with this stuff. But just seems to me that is how a solenoid acts.

 

Here's an interesting bit I just noticed. When using a magnetic plunger of the same length as the coil, the force exerted by the solenoid is largest when the plunger is inserted halfway through it.

RIGHT! the math explains it better than I can. But it is not super intuitive unless you visualize the lines of magnetic flux.And normally magnetic flux is not visible.

 

RIGHT! the math explains it better than I can. But it is not super intuitive unless you visualize the lines of magnetic flux.And normally magnetic flux is not visible.

Yeap ... in fact, magnets can do pretty weird stuff sometimes. There's a very interesting document a the KJmagnetics website explaining some strange behavior.

 

Here's an interesting bit I just noticed. When using a magnetic plunger of the same length as the coil, the force exerted by the solenoid is largest when the plunger is inserted halfway through it.

I don't have any material to cite, but I've noticed the same with iron on open-ended solenoids. If the plunger is made longer, say twice
as long as the coil, then the plunger wants to center itself, but 1/2 half of travel is the strongest force. If your coil is much longer than the magnet, like with a coin sized neo mounted on a wooden plunger, inside of a two inch coil, then it will also center itself, and go no further.
Adding a neodymium magnet to either end of a iron plunger will offset the centering somewhat. All depends on the shape of the magnetic
field for coil and the shape of the induced magnetic field inside the plunger (you're creating an electromagnet with a floating core after-all.)
Both will act to reach an equilibrium.

 

But using the PM for the plunger also make the coil pole sensitive, where the iron one isn't. I have also seen a true "push-pull" solenoid done with a magnetic plunger. Two PMs with opposite poles and none magnetic center section, energize the coil one direction and it pulls, energize the other way and it pushes.

I've used a cylinder magnet and H-bridge to create a powerful solenoid that can both pull, and push. No return spring required.

(Edit: no hijack offense taken)

 

@strantor , I think you will find this post of your interest. I remember you mentioning something related to this subject in the past, if I'm not mistaken.

​

Yes, thank you for thinking of me. What you're remembering is my attempts to build linear "torque motor"-type voice coil/solenoid devices for my prosthetic hand project. This is closely related. That project got squashed before I figured out how to do what I wanted, but I still believe in it and I may revisit the concept even without the larger project that originally bore its necessity. I believe it would be of use to the world in many applications.​

Here's an interesting bit I just noticed. When using a magnetic plunger of the same length as the coil, the force exerted by the solenoid is largest when the plunger is inserted halfway through it.

I found this true as well. It was an undesirable attribute in my case. Getting a linear force:current relationship regardless of stroke position was one of my main goals, and the potential fix that I was testing when I abandoned it, was a metal outer armature to guide the flux past the ends of the rod at max stroke in both directions.